Get started

ELM is all about displaying your video content on LED fixtures of any shape, in the most convenient way. It also helps you set the rig up and keep it running: Device Manager finds the ENTTEC Art-Net nodes on your network and configures them — addresses, names and per-port universes — and Sentry watches your RDM fixtures and your gateways and tells you when part of the rig stops answering.

Key features

• 2D vector-based stage editor
• 3D (voxel) stages — build a solid cube, a hollow cube, a sphere or a cylinder and drive it as a live volume that 3D effects sweep through by depth
• Up to 32,768 DMX universes per protocol (Art-Net, sACN and KiNet each have their own range). That is the engine's ceiling; how many actually output is set by your licence — 512 on a stock Professional or Architectural licence, raised by universe add-on packs. See Licenses. Getting near the top of the range also means spreading output across several network cards — see Computer requirements.
• Art-Net, sACN, KiNet output protocols
• Scheduled sequences
• Control via DMX, OSC, OS2L, MIDI and HTTP
• CITP/MSEX for consoles and visualizers
• HD video + audio playback
• Spout + NDI integrations for live video inputs
• Stream LED previews to visualizers via NDI
• DVI outputs for projectors, TVs and Martin P3 system
• Over 100 high quality animations included
• Audio-reactive effects
• Automatic BPM and beat detection from any audio input — a microphone, a line-in or your computer's own sound — driving one master tempo for effects, shaders and sequences. Set it up on Settings → Audio.
• RGB, RGBW, RGBA, RGBAW colour support
• Tunable white: single, dual or triple white emitters, each with its own colour temperature in Kelvin, plus a CCT fader per stage and an optional dim-to-warm behaviour
• 16-bit colour support
• Import DXF, CSV and xLights xModels into a 2D stage (Settings → Project → Import). Point-cloud import for 3D stages is not in this release.
• Sentry: RDM monitoring of your fixtures and your ENTTEC gateways, with alert rules and notifications
• Device Manager: discover and configure ENTTEC Art-Net nodes, one at a time or a whole install at once
• Interface in English, French, German, Italian, Korean, Portuguese, Russian, Spanish and Chinese (Simplified and Traditional)
• Extreme robustness for permanent installations

Sentry, Device Manager and Remote control are licensed features. Sentry and Device Manager come with an Architectural licence, Remote control with Professional and above; on a plan that does not include them the buttons and pages stay visible but open an upgrade message instead of the feature. An unlicensed copy is the exception — demo grants all of them, along with unlimited universes, in exchange for a periodic output blackout. See Licenses.

The older fixed white LED types (White, WhiteWarm, WhiteNeutral, WhiteCool, and the WWA / WAW / AWW permutations) are retired. They are no longer offered when you pick an LED type, but projects that use them still open and output exactly as before — when you load such a project ELM tells you how many fixtures use a retired type and suggests you re-select your real LED type — White (single), White (dual) or White (triple) — to opt into the tunable-white features.

Computer requirements

ELM runs on 64-bit Windows 10 (version 1607 or later) and Windows 11. Windows 7 and Windows 8 are no longer supported, and there is no 32-bit build. ELM installs everything it needs to run — you do not have to install the .NET framework, and ELM will never prompt you to.

Recommended

• Windows 10 or 11, quad core 3.5 GHz CPU, 8 GB RAM
• NVIDIA GeForce 900 series or NVIDIA 10 series or better
• SSD hard drive

The computer requirements depend on your media type. Playing multiple HD videos requires a good computer. A rule of thumb is you can play one full HD video at 30 fps for each 2 GHz CPU core. So if you have an Intel i7 4 GHz (4 cores), you should be able to play 8 full HD videos. Increasing the playback speed to 2x means the fps also increases two times and the number of videos you can play simultaneously is roughly divided by two.

The table below shows the approximate number of videos you can play simultaneously at 30 fps depending on your CPU.

For live video performances where you want to change playback speed up to 10x, the recommended video resolution is 640×360.

To use the built-in effects, particularly the eye candy effects, the video card will be doing the hard work. If you're using a laptop, make sure it can sustain continuous use of the effects without overheating. Usually, an integrated video card can render two eye candy effects. For more effects, you need a discrete video card.

Measure your own computer

The figures above are rules of thumb. To measure the machine you will actually use, open Settings → Benchmark.

The results are a guide. Nothing in your project or settings is changed by a benchmark run. One warning: your live lighting output is paused while the network test runs, so do not run it during a show.

Network

About DMX output: a 1 Gigabit network should be able to handle 2048 universes at 44 fps. For more universes, add a supplementary network card and a second Ethernet network.

ELM has controls for spreading output across cards. On each output protocol page (Settings → Art-Net, Settings → sACN, Settings → KiNet):

• Set the Default NIC — the adapter used for any universe that is not assigned to a specific card.
• Open the advanced output-assignment section to assign individual universes to a specific card.
• Click Auto-balance to spread all universes evenly across the available cards.

Most setups send everything out the Default NIC and never touch the rest. Use the per-universe assignment only when one card cannot keep up. Whichever card a universe is assigned to, every fixture on that universe must be reachable from that card.

Licenses

Everything to do with licensing lives on one page: Settings → License. It shows what you have, what it covers, and how to activate this computer.

Without a licence: demo mode

A computer with no licence on it runs ELM in demo. Demo grants every licensed feature — Art-Net output, sACN output, KiNet output, Live mode, Remote control, Sentry and Device manager — with no universe cap, no Sentry device cap and no Device Manager node cap. The Plan card on the License page reads Unlimited for DMX universes, and every feature is marked Included. Nothing in the app is hidden, greyed out or read-only because you have not bought yet.

The restriction is on the output, not on the features: the rig goes dark for about ten seconds every five minutes. ELM sends real zero frames during that window, so the blackout is genuine, and the moment it happens shifts by up to 30 seconds each cycle so it is not perfectly predictable. The splash screen carries a DEMO watermark, the status bar reads Trial mode, and the trial window says Full-featured trial and In trial mode, the app takes a short break every 5 minutes. That makes demo fine for building and proving a show, and unusable for running one.

Demo grants more than the Free and Professional plans do. This surprises people, so it is worth stating plainly: the moment you activate a licence, the demo grant is replaced by that licence's entitlements. Activate a Free licence and Remote control, Sentry and Device manager switch off and output is capped at 8 universes — all three of which worked on the same machine five minutes earlier, unlicensed. Nothing is broken; demo is an evaluation state, not a plan, and every statement elsewhere in this manual about a feature being "not included in your licence" is about a plan that lacks it, never about demo.

A lapsed subscription lands in the same place. Once the grace period ends the status card reads Subscription lapsed and ELM runs in demo — every feature back on, with the blackout — until you renew.

On the License page an unlicensed machine reads Demo mode. Activating a licence removes the blackout and the watermark.

Plans

ELM ships in three plans. The universe count and the feature list on your License page are the authority for what you have bought — the table below is what each plan includes by default. Demo is listed first for comparison; it is not something you buy.

The Plan card lists every licensed feature with Included or Upgrade beside it:

The Sentry and Device Manager buttons stay on the top menu bar when your plan does not include them, drawn in gold. Their tooltips read Sentry is not included in your licence. and Device Manager is not included in your licence., and clicking one shows Feature not included in your licence — Your licence does not cover the Sentry feature. Please upgrade your licence to use it. instead of opening the workspace. In demo this does not happen: both workspaces open normally (see Without a licence: demo mode above).

If you discover more devices than your allowance covers, a banner tells you exactly where you stand: Sentry is monitoring 1,200 of 1,450 discovered devices — your licence cap. Upgrade to monitor them all. and the equivalent for Device Manager nodes.

Going over your universe count does not stop ELM. The universes within the cap keep outputting, the ones beyond it go dark, and ELM warns you: Your license is limited to 512 DMX universes — any universes beyond that will not output until you upgrade.

One thing to know about the Free plan: Share anonymous usage data on Settings → About is locked on. The note beside it reads Included with the free licence. Upgrade to Professional to turn this off.

Activating

There are two routes, both on Settings → License.

If the computer is online: under Activate with a voucher, type your voucher code and the email address the licence was bought with, then click Activate.

If the computer has no internet connection: under This computer, click Copy to copy your machine ID, send it to ENTTEC support, and they send back a licence file. Click Import license file… and select it. Some builds have online activation switched off altogether and say so: Online activation isn't available in this build — use "Import license file" below.

There are no hardware dongles to lose, break, or fail in the middle of a show.

Staying activated

What happens after activation depends on which kind of licence you hold. The status card names the state:

ELM syncs the licence quietly at launch. The Last synced row tells you when the licence server last issued your licence; if it looks old you have simply been offline, which is fine. If a licence is deactivated or revoked on the ENTTEC server, ELM returns to demo mode at the next launch — a network fault or an offline venue never does that.

Each license includes 2 activations, so you can set up ELM on two computers, for example a main and a backup, as long as only one runs at a time. The licence is locked to this computer's processor and system drive. You can change any other component freely, but replacing or re-imaging the boot drive changes the machine ID, so deactivate the licence first if you plan to. Volatile things like MAC addresses and drive letters are deliberately not part of the fingerprint.

To move ELM to another computer, deactivate it in the app on your current machine to free up an activation, then activate it on the new one. You can view your license and its activations any time on your ENTTEC account page.

Add-on packs

To add more universes later — or more Sentry devices or Device Manager nodes — buy an add-on pack and apply it to your existing licence. There is no need to buy a second licence or run several side by side.

How many universes you can be licensed for

Two different numbers get quoted, and they are not in conflict once you know which is which:

• 32,768 universes per protocol is what the ELM engine can address. Art-Net, sACN and KiNet each have their own range of that size. It is a capability of the software, not an entitlement.
• 512 universes is what a stock Professional or Architectural licence grants (8 on Free). That is the number ELM actually enforces, and universes beyond it go dark.

Universe add-on packs raise the licensed number; the engine's 32,768-per-protocol ceiling is the most any licence can usefully grant. ELM cannot tell you which pack sizes are for sale — packs are bought from ENTTEC and applied in your ENTTEC account, and the app never sees the catalogue. If you are sizing an install past 512 universes, ask ENTTEC sales what packs apply to the count you need, and treat Settings → License as the authority on what you actually hold: the DMX universes figure on the Plan card is the number ELM enforces, add-ons included.

A large licence is not on its own enough to drive a large rig. Near the top of the range the output has to be spread across several network cards — see Computer requirements.

Do not type an add-on code into ELM's activation box. Add-on packs are applied to your licence in your ENTTEC account, not in the app. Sign in at enttec.com/my-account and apply the code under My Licences. ELM picks up the extra capacity the next time you start it. If you enter one in the voucher field, ELM tells you the same thing rather than failing silently.

Software updates

Updates are included with your licence: ELM offers you every version your licence entitles you to, at no extra cost. You can find ELM and everything about it on its product page at enttec.com.

How ELM tells you

ELM checks for a new release when it starts, and again every four hours while it is running. A mid-session check never opens a dialog — interrupting a live show is not acceptable — so all it does is light a small coloured dot beside Settings in the title bar. Hover it and it names the version. The dot clears once you open Settings → About; a newer version found by a later check lights it again.

To check yourself, go to Settings → About and click Check for updates.

The update window

Installing

What Install now does depends on how ELM was installed.

• The auto-updating desktop installer — the update downloads inside ELM. ELM asks you to save your project, then restarts into the new version.
• An MSI install, or ELM Server — there is no in-place update. ELM tells you a new version exists and opens the download page in your browser so you can install it yourself.

What changed

After ELM updates itself it shows a What's new window listing the changes. You can reopen it at any time with the What's new button on Settings → About, which also gives you the release-notes history for earlier versions.

Definitions

Setup

How ELM connects to your LED fixtures, and a high-level overview of the stages concept.

Getting started with LED strips

ELM can control any DMX-enabled lighting fixtures, not only LED strips. This is made possible because ELM outputs industry standard protocols like Art-Net, sACN and KiNet. ELM can therefore control traditional DMX dimmers and any RGB, RGBW, RGBA and RGBAW lighting fixtures, as well as complex fixtures like moving heads.

Controlling LED strips is a little bit different from traditional DMX fixtures. The main difference is that you need to use LED controllers to drive your strips. They replace the Art-Net nodes you use for traditional DMX fixtures. The LED controller receives DMX via Art-Net/sACN and converts it to the special protocol the LEDs understand.

The signal path is: ELM sends Art-Net/sACN over a network switch to your LED controllers and Art-Net nodes, which drive the strips and DMX fixtures. Video output to projectors or TVs goes out of the computer's own DVI/HDMI ports.

The next step is to get LED strips and DMX fixtures. Since there are many LED protocols out there, you need to make sure your LED controller fits with your LEDs. Common protocols for RGB LEDs are WS2811 and WS2812/B. For more information about getting compatible LEDs, contact your ENTTEC representative.

ENTTEC makes both the pixel controllers and the pixels themselves. The Pixelator Mini receives Art-Net/sACN and drives your addressable strips, and the ENTTEC Smart Pixel range provides the LEDs, with pixel dots and pixel tape in a range of resolutions, colours and IP ratings.

From install to first light

This is the whole path from a new computer to LEDs that light up, in order. Each step names the section that covers it in depth; nothing here replaces those sections, it just tells you what to do next. Steps 1 to 3 happen outside ELM, and skipping them is the most common reason a correctly patched project stays dark.

1. Install ELM and start it. ELM is a 64-bit Windows application and installs everything it needs — see Computer requirements. Until you activate a licence it runs in demo mode, which blanks the output for a few seconds every few minutes; see Licenses.
2. Put the computer and the gateway on the same network. Give the computer a fixed IP in the same subnet as your Art-Net node or pixel controller — in practice the first three numbers match and both use the mask 255.255.255.0. If the Windows firewall is on, open Settings → Project and click Firewall (ELM must be running as administrator) so Windows lets ELM send and receive. KiNet is the exception: Philips controllers expect the computer on 10.x.x.x with mask 255.0.0.0 — see KiNet.
3. Configure the gateway or pixel controller itself. ELM does not do this for you except on the models Device Manager supports. The node needs its own IP, the universe each of its ports listens to, and the protocol it listens for; a pixel controller also needs the LED chip type (WS2811, WS2812B and so on) and the number of pixels per port. For a supported ENTTEC node, do it in Device manager; otherwise use the node's own web page or the manufacturer's utility. Write the port universe numbers down — you type the same ones into ELM in step 6. If you will use Sentry later, set the ports the fixtures are on to Output+RDM now.
4. Choose the output network adapter in ELM. Open Settings and go to the page for the protocol you are using: Art-Net (the selector is Default NIC), sACN or KiNet (Adapter). Pick the card your lighting network is on. Each protocol has its own selector, so set it on every page you intend to output on.
5. Create a stage. Click New stage on the home screen, name it and give it a size with the same aspect ratio as your media — see New stage dialog. The size is in pixels and can be changed later without breaking the mapping. A rig built as a volume rather than a flat surface is created differently: see 3D stages.
6. Patch your strips. In the 2D stages workspace, on the Strips tab, click Add. In the Add strips dialog set the number of strips and LEDs per strip, pick the Type that matches your fixture's colour order and capability (Pixel types), pick one of the Shapes and, for lines and arches, the Patching direction that matches how the strips are actually wired, then set DMX protocol, Start universe and Start address to match the port you noted in step 3. Art-Net universes are numbered from 000 upwards, so a gateway port set to universe 0 is Start universe 0 in ELM.
7. Send the universes to the node. With Art-Net and KiNet, every universe is broadcast out of the box, so a single node on its own lighting network needs nothing typed here. Enter node IP addresses in the universe grid only when you want to unicast instead — see Art-Net. Check the coloured block behind each universe number on that page: green means the universe is patched and being sent.
8. Load a media. Go to Media (Alt+M), click the round + on an empty slot and pick a video, image or effect — see Media library.
9. Play it on the stage. Go to Live (Alt+L) and select the stage's tab. Double-click the media in the A bank — that plays it and runs the A-B fader all the way to the A side — then raise the stage Intensity. See Live mode. This is the step that puts light on the rig.
10. Prove the patch. Back in the 2D stage workspace, open the Testing tab and switch Test patterns on. The stage stops showing your media and sends a rectangle you can move across the strips, which is how you check that the LEDs come on in the order you patched them and that red really is red. Switch it off when you are done — while it is on, this stage outputs nothing else.

If nothing lights up, work through Troubleshooting. The usual causes are in steps 2 to 4: the wrong network adapter, a firewall rule that was never added, or a computer and a node on different subnets.

Quick overview

ELM allows dividing your installation into multiple zones and layers called stages. You control the media content independently for each stage. Stages can overlap and they are merged (blended) to generate the result.

A common scenario is to create a stage spanning the whole installation to map media across the board. To enable precise control over specific zones, you create smaller stages inside the big stage, then activate these zones whenever you want. For example, you can play a video on the overall stage (which acts as the background layer), and when there's a special event, activate the logo zone and make it flash.

Stages come in two kinds, and they have separate workspaces reached from the 2D stages and 3D stages entries at the top of the window. They share the same media library, the same intensity, colour and speed controls, and the same merging behaviour.

• A 2D stage is a flat mapping surface you draw your strips onto. This is what the rest of this chapter describes.
• A 3D stage is a volume. You generate it from a shape — a solid cube, a hollow cube, a sphere or a cylinder — and ELM projects your 2D media onto it. You choose which face of the volume the media lands on (Front, Top or Side) and whether it fills the volume's depth (Extrude) or lights only the front plane (Surface).

To represent your LED strips and DMX fixtures inside a 2D stage, ELM has a vector-based editor. You draw many types of shapes, including matrices, loops and curves, and ELM computes the position of each LED based on your drawing. A 3D stage is not drawn: its LED positions are generated from the shape parameters you enter when you create it.

Application tour

A guided tour of every screen and control in ELM.

Home screen

ELM opens on the home screen: Welcome! Start by creating a stage, which is the mapping surface where your media meets your LEDs. Duplicate your stage to get multiple media layers. Start a new project or load an existing one. Hold Alt to underline the shortcut key of each entry in the top menu bar.

If you have not activated a licence yet, ELM runs in demo mode and a banner appears across the top of the window: You're running ELM in demo — the output blanks briefly every few minutes. Activate to remove it. Click Activate in the banner to enter your licence, or the ✕ button to dismiss it for now. It comes back once, with wording that explains the blackout, the first time the output actually blanks — so you know the dark moment is the demo nag and not a fault. See Licenses.

The top menu bar

The menu bar in the window's title bar has eight entries. They are always visible, whatever screen you are on.

These are access keys, not global shortcuts, so they can collide with a button on the screen you are looking at. One collision ships in ELM 2026: the Duplicate button on the Strips tab of a 2D stage also answers to Alt+D. While that tab is showing, the two share the key, so use the menu button itself if you specifically want the 3D workspace. Everywhere else Alt+D reaches 3D stages. See Duplicate strips dialog.

Sentry and Device manager are never hidden, but they are drawn in gold if your licence does not include them — the tooltip then reads Sentry is not included in your licence. or Device Manager is not included in your licence. Clicking one explains how to upgrade.

To the left of these entries the bar carries the Master int. and Master speed faders, which scale every stage at once. The master intensity does not affect the monitor — it only modifies the output data.

New stage dialog

This is the dialog for a 2D stage, opened by New stage in the 2D stages workspace. A 3D stage is created from a different dialog, documented under 3D stages.

A rule of thumb is to create a stage with dimensions respecting the aspect ratio of your media. Then you'll position your LED strips on it to create the mapping. For example, you can create a full HD stage (which has a rectangular aspect ratio) to fit your media sources, and then map a 20×20 square LED array on it.

You can easily change the dimensions later, so don't worry. The minimum recommended size is 320×180. Smaller than this, you'll need to constantly work at a very high zoom level, which isn't very convenient. So it's better to use a higher size and let ELM scale your media.

The same window opens when you duplicate a stage from the Stages manager. Its instruction line then reads Duplicate your stage to get new layers. and four extra options appear, marked below.

Pro tip: To help you position strips like they are in the real world, use a picture of your installation as the stage's background (see Stage Tab). No pictures handy? Use your phone's camera to get one.

New 3D stage dialog

A 3D (voxel) stage is created from its own dialog, not from this one. Go to 3D stages in the top menu bar and click New 3D stage, either in the toolbar at the top right or on the welcome panel when you have no 3D stages yet. There is no drawing step: you describe the rig with numbers, a live preview redraws as you type, and ELM generates the LED positions and patches them in one step.

Every control of that dialog — shape, LED counts and sizes, the hollow cube's Faces option, run axis, wiring, DMX addressing and the Advanced — depth slicing section — is documented once, under 3D stages.

Stages

Shortcut key: Alt+S

This is the 2D stages workspace. 3D stages have a workspace of their own, reached from the 3D stages entry in the top menu bar, with its own tabs and its own Manage window.

You can create as many stages as you want. Generally, you'll want an overall stage corresponding to the whole installation and multiple individual stages to target specific zones. The overall stage allows you to map content across the board. Then the other stages allow you to override specific parts of your installation whenever you want. Each stage is a tab: select the current stage by clicking its name at the top.

The controls at the top right of the tab bar are where every stage operation starts.

Before you have created any stage, the workspace shows the welcome panel instead, with New stage and Load project buttons.

Pro tip: Different effects may demand different mapping layouts. One goal of the mapping is to ease content creation. Don't hesitate to create multiple stages with the same LED strips but positioned in different ways.

Status indicator

At the left of the stage's name, a small mark tells you the current control mode and whether the stage is activated. Most of the time there is no mark: a stage under manual control that is activated carries nothing, which is the normal state. A mark appears only when something is worth telling you about.

Read it in two parts. The shape is the control mode — a letter for remote (R), schedule (S), the audio reactive mix (A) or testing (T), and a plain dot when the stage is under manual control. The colour is the state: blue for activated, red for deactivated, and yellow for testing, which is always yellow and always wins over the other marks.

A stage is activated when a media is selected and its intensity is above 0, when it carries at least one effect layer that is switched on and above 0 intensity, or while its test patterns are running.

The same marks are used on the 3D stage tabs.

Stage syncing

Everything using the same media slot is synced — stages, and any effect layer pointing at that slot. You can still set each stage's own output parameters (intensity, colour filter, speed and CCT), but the media content is the same everywhere the slot is used. When applicable to the media type, the playback speed is the highest speed requested by anything using that slot, so a fast stage speeds up the slower ones sharing the same media.

Pro tip: To play the same video file on multiple stages but at different speeds, load the video file in multiple media slots. Then use a different media slot for each stage, or for each layer.

Stage merging and layering

There are two ways to layer content. This section is about how whole stages combine when they share LED strips. To layer several media on a single stage, use effect layers instead — see the end of this section and Effect layers.

Stage merging

When multiple stages contain the same LED strips, a merge occurs following the order as shown in the Stages Manager. This allows creating complex visuals by layering multiple stages on top of each other. You can target specific zones by creating stages with only a subset of all strips or with parts of the strips outside the stage.

A stage takes part in the merge only while it is activated. A stage is activated when a media is selected and its intensity is above 0, when it carries at least one effect layer that is switched on and above 0 intensity — even with no base media at all — or while its test patterns are running. A stage that is not activated is treated as transparent and is left out of the merge entirely.

Which stage is on top

The Stages manager list is ordered bottom layer first. The stage in the first row is the bottom of the stack; each stage further down the list merges on top of the result built so far. Drag and drop rows to change that order.

Every mode below therefore combines two things: this stage (the one whose merge mode you are setting) and what is below it — the result of merging every activated stage that comes before it in the list. There is no left and right, only up and down.

You set a stage's merge mode in the Stages manager, and in the New stage dialog when you duplicate a stage. The default for a new stage is overwrite. The names below are exactly as they appear in the drop-down.

multiply, subtract and overlay need something underneath. These three only ever modify pixels that another stage has already written. If a stage using one of them is the lowest activated stage on a strip, it contributes nothing at all and those LEDs stay dark — the stage looks broken. Put it above the stage it is meant to act on.

Pro tip: To easily target specific zones of an installation, first create the overall mapping then select the strips corresponding to the individual zones and right-click in the strip list to access the "new stages from strips" menu.

Effect layers

You no longer need extra stages to stack content. Each stage carries its own stack of effect layers, edited in the Layers panel of the Live view — the per-stage screen described later in this chapter, under Live mode, and reached with Alt+L. In that panel: click + Add layer, then right-click the layer to give it an effect or edit an existing effect's parameters. Every layer has its own blend mode, opacity fader and Area, and can be hidden (H) or soloed (S). Clear removes every effect layer at once and keeps the base media, and Save as new media bakes the whole stack into a single media you can re-select later to get the layers back. This is per-stage, and none of it changes the merge described above. See Effect layers for the panel in full.

The two mechanisms use different blend-mode lists, but the same underlying modes. Stage merging offers all twelve modes in the table above; an effect layer's drop-down offers a curated five, under plain-language labels:

The five behave on a layer exactly as the table above describes them on a stage, reading "what is below it" as the base media plus every layer under this one. A layer saved in an older project with a mode that is no longer offered still loads and renders unchanged, and that mode is appended to that layer's drop-down so you can see it and keep it. The remote-control API can also set a layer to any of the twelve — see the remote-control HTTP section for the exact strings it accepts and for the modes a layer renders differently from a stage.

Edit zone

Position your strips as they are in the real world, or based on the effect you want to create with your media. A strip has a start (green) and an end (red) control point showing the direction of the patch. Add intermediate control points to build other shapes. Bundle several strips together to handle a complex shape as one item.

You can position parts of your strips outside the stage. The LEDs outside the stage are not mapped. This is a common scenario when you use the same strips in several stages and want to target specific parts in each one.

When any LED sits outside the stage, a warning appears under the edit zone: Some LEDs are outside the stage, followed by the number of LEDs in brackets. Click the warning to select the strips those LEDs belong to. If it is deliberate, ignore it.

Strips tab

Shortcut key: Alt+P

Strip list

The strips for the current stage are shown in this list. Select one or several strips in the list to also select them in the edit zone. Click the column headers — Group, LEDs, DMX start, End — to sort. Type text in the filter box to filter on the group names. The buttons under the list are Add, Duplicate and Edit.

Right-click in the list for more options. The menu holds three groups of commands:

• Selection — Select all, Deselect, Select even, Invert selection and Go to selection (which scrolls the edit zone to what is selected). There is no "select odd" item: use Select even followed by Invert selection.
• Editing — Copy, Paste, Delete, and Bundle / Unbundle.
• New stages from strips — a submenu that builds new stages out of the selected strips. It does not move or delete the strips you started from.

Pro tip: Use the group and sub-group fields wisely in order to use the filter box and quickly select the strips you want.

Add strips dialog (patching)

Open it with Add at the bottom of the Strips tab. Patch your strips and lighting fixtures here to create arrays and any shape you need.

Multiple LED types (or colour types) are supported, including RGB, RGBW, RGBWmax, RGBA, RGBAW/RGBWA, and white fixtures with one, two or three white emitters, each with its own colour temperature. Most colour component orders (GRB, BGR, WRGB, and so on) are supported. Bit depth is a separate setting: any type can be driven at 8- or 16-bit, chosen in the Bit depth box. 16-bit uses two DMX channels per colour component for smoother colour transitions, and doubles the number of channels each LED takes. RGBWmax activates all channels (RGB + white) when white is needed, for maximum brightness. See Pixel types for what each type does.

Pro tip: Got the patching direction wrong? No problem. You can easily flip your strips horizontally and vertically later. For snake mode, right-click in the strip list and use Select even; there is no "select odd" item, so follow it with Invert selection to get the other half.

Pro tip: Want to control more complex DMX fixtures like moving heads? Let's say your moving head has 20 channels and the RGB channels start at 3. In ELM, patch it with a start address of 3 and a LED address offset of 20. ELM will send 0 for the unused channels, which allows merging the DMX data from ELM with the DMX data from a console. The console provides the values for pan, tilt and all other channels except the RGB.

Pro tip: Whites! For an accurate preview, pick the white type that matches your fixture — White (single), White (dual) or White (triple) — and enter each emitter's real colour temperature in Kelvin. The stage monitor then tints the strip with the mix ELM is actually sending, and it follows the stage's CCT fader in the Live view.

Pixel types

The pixel type tells ELM what emitters your fixture has and in what channel order they are wired. Set it in the Type box of the Add strips and Edit strips dialogs. The letters give the channel order, so a strip wired blue-green-red is BGR, not RGB.

Bit depth

Bit depth is set separately from the type, in the Bit depth box of the Add strips and Edit strips dialogs. Choose 8 or 16. Any type can be 16-bit. 16-bit uses two DMX channels per colour component (most significant byte first), so it doubles the channel count in the table above — an RGB strip at 16-bit uses 6 channels per LED instead of 3, and a White (triple) strip uses 6 instead of 3. It gives smoother fades and more headroom for colour corrections, at the cost of twice the universes. Projects made with the old RGB16 / GRB16 / White16 style type names open as the plain type set to 16-bit — there is no longer a separate "…16" entry in the Type list.

Tunable white

A white fixture is described by how many white emitters it has and the colour temperature of each one, in Kelvin, in the order the channels are wired. Enter those values when you add or edit the strip (or, for a 3D stage, in Edit LED arrangement). They are saved with the project.

However many emitters the fixture has, its overall level comes from the perceived brightness (luma) of the media colour mapped onto it — roughly 30% of the red, 59% of the green and 11% of the blue. The emitters then share that level between them to reach the target colour temperature.

The Live view then has a per-stage CCT fader, to the right of the Speed fader, which mixes the emitters toward a target temperature between 1800 K and 10000 K. It works on 2D and 3D (voxel) stages, and drives the wire output and the 2D/3D previews together. It can also be set from a remote over OSC or the HTTP API. Pull the fader to the bottom (Off) and every emitter runs at the same level — the fixture's native white, which is what ELM did before this fader existed.

For a dual or triple white fixture, When dimming chooses what happens as the fixture fades: Constant colour holds the mixed temperature where you set it, and Dim to warm drifts it toward your warmest emitter as it fades. The fixture can only reach temperatures between its warmest and coolest emitter — a target outside that range is clamped to the nearest one.

Older projects can still contain the retired white types (White, WhiteCool, WhiteNeutral, WhiteWarm, WWA, WAW, AWW). They keep working exactly as before — nothing is re-mapped — but they are no longer in the Type list. When you open such a project ELM tells you how many fixtures use one. Edit those strips and select White (single), White (dual) or White (triple), then set each emitter's colour temperature, to get the tunable-white behaviour.

Pro tip: For the colour temperature settings to get a more accurate stage monitor preview, see Settings → White balance.

Duplicate strips dialog

Copy the selected strips and create new strips by applying certain transformations. Select the strips, then click Duplicate on the Strips tab. Click Duplicate in the dialog to create the copies.

Alt+D is the button's access key, but the 3D stages entry in the top menu bar uses the same letter and that bar is always on screen, so while the Strips tab is showing the two share the key. Click the button when you want to be sure which one you get. See Home screen.

Each copy is made from the previous one, so every offset below stacks: with a count of 4 and an X offset of 100, the copies land 100, 200, 300 and 400 pixels to the right of the original.

ELM pre-fills the two patch offsets from what you selected, so the copies are patched directly after the originals: if the selection sits inside one universe you get an address offset the size of the selection, and if it spans several universes you get a universe offset instead and an address offset of 0. Change them if your gateways expect something else.

Pro tip: When in the design phase of the lighting installation, keep the duplicate options in mind. They allow creating complex shapes by copying simple elements, which is a real time saver.

Pro tip: To make sure certain strips always stay together, use the bundle feature. You can create complex shapes by combining multiple simple elements. Select the strips, right-click in the strip list and choose Bundle or Unbundle — Ctrl+B and Ctrl+Shift+B do the same.

Pro tip: To copy strips to another stage, select your strips and use Ctrl+C and Ctrl+V. You can also right-click in the strip list to see all options.

Edit strips dialog

Edit the selected strips and change their patch. Select the strips on the stage or in the strip list of the Strips tab, then click Edit. With that tab showing, Alt+E does the same.

Everything you change applies to all the selected strips at once. Some fields disappear depending on what you selected: when the selection contains a bundle, the dialog shows There are bundles in the selection making some options unavailable.; when it contains strips whose LEDs you have positioned individually, it shows There are strips with individually editable LEDs in the selection making some options unavailable.

Buttons

The buttons along the bottom left apply the rest of your changes at the same time as they do their own job. Reverse LEDs and Edit LEDs pos act at once and close the dialog. Re-patch, Split and Join ask you something in a second window first and close the dialog only once you confirm it — cancel that window and you are back in the dialog with nothing changed. Locate leaves the dialog open when you close it, unless you use Set last LED, which applies the new LED count and closes.

If ELM told you when the project loaded that some fixtures use a retired white LED type, this is where you fix it: select those strips, open this dialog, and set Type to the White (single), White (dual) or White (triple) entry that matches the fixture, then set each emitter's colour temperature. Nothing is re-mapped for you — the old white types keep working exactly as before until you change them.

Offset / resize strips dialog

Precisely change the position and size of the selected strips. Select the strips, then click the resize button in the middle of the Offset arrow pad on the Strips tab, or press Alt+O.

The dialog changes the bounding rectangle of the selection. What that rectangle covers depends on the As a group checkbox next to the Rotate box: when it is ticked, the rectangle is the one around all the selected strips and they move and scale together; when it is cleared, every selected strip is moved and scaled on its own to the values you type.

Nothing moves while you type. The stage draws the rectangle you are describing so you can see where it lands; the strips are only repositioned when you click OK. The two spacing checkboxes and the Linked state are remembered for the next time you open the dialog.

Pro tip: Right-click in a numeric box and move the mouse left/right to quickly change the value. Each box also has a calculator button if you would rather type a sum.

Stage tab

The Stage tab holds the settings that apply to the whole stage: its size, the media preview, the background image you trace your strips over, the snapping grid, and sticky notes.

Pro tip: When you zoom in very closely, ELM switches the media preview off by itself and shows High zoom: media preview disabled. on the stage, so you no longer have to do anything about it while editing. If the preview is costing you CPU at normal zoom levels, set Media opacity to 0.

Monitor window

The monitor window lets you see the result of the stage only or after all stages have been merged. When you don't have access to the real fixtures, this is an indispensable tool while creating effects. Alternatively, you can use any 3rd party visualizer able to receive the LED rendering via NDI (see NDI output in the Stages manager) or supporting Art-Net or sACN.

Open it with the Monitor button on the stage's Stage tab, or with the small screen icon on the stage's row in the Stages manager.

Merged, in the window's title bar, switches between the two views. It is ticked when the window opens, so what you see by default is the merged result of every stage. Untick it to see this stage on its own — that is the one to use when you are chasing down which stage is contributing what.

Monitor window for a 3D stage

A 3D (voxel) stage has its own monitor window, titled 3D voxel stage and showing the live volume. Open it with the Monitor button at the bottom of the 3D stage page's Stage tab, or with the Monitor button on the stage's row in the Manage 3D stages dialog.

It shows the output and nothing else — there is no Merged checkbox and no options. The projection, wiring and testing controls all live on the 3D stage page itself, so keep that page open beside the monitor while you commission a rig. See 3D stages.

Pro tip: Try reducing the stage's size and see if the output is still good. Also reduce your videos and other media size accordingly to potentially save a lot of CPU. High resolution media is not always a good thing.

Testing tab

Quickly test the mapping by generating a rectangle that you can move over the strips. Turn Test patterns on and the stage stops showing your media: the rectangle is what goes out on the wire, so what you see on the rig is exactly what the test is sending.

Pro tip: Reduce the width and height of the testing rectangle and drag the rectangle around the stage. This way, you can see if the patch order is right. For example, if dragging the rectangle from left to right makes the LEDs go on from right to left, then you know you need to flip them horizontally. To do so, go to the Strips tab and click the Flip X button. If you'd rather watch the rig than the screen, click the moving-bar button beside Width and let ELM sweep the bar for you.

Pro tip: Quickly test the red, green and blue component order of your LEDs by generating a pure red, pure green and pure blue test. For each test, note the colour of your LEDs. If you're not getting a RGB order, edit your LED strips and select the LED type corresponding to the order you've observed (BGR, GRB, GBR…).

Pro tip: While the testing mode is active, the DMX is constantly being output at the full output rate, even when there's no change. This is handy while you're configuring your LED controllers and want to see if it's working. This is also a good way to test whether your network supports the load.

Pro tip: Use the "Test selected strips only" option, go in the Strips tab and change the selection in the strip list. This is similar to traditional lighting consoles' locate function.

Testing a 3D stage

A 3D (voxel) stage has its own Testing tab, on the right of the 3D stage page. It does not use a rectangle — it runs one of eight patterns through the volume, sent as real DMX, so you can check that the rig in front of you is wired, addressed and oriented the way you think it is. The patterns and their controls are documented under 3D stages.

Stages manager

Select multiple stages in the list to modify them all at once. The first stage of the list is the bottom media layer and the following stages can override any preceding stages. Drag and drop stages to reorder.

Type in the filter box at the top right to narrow the list. Right-click the list for Select all (Ctrl+A), Deselect (Ctrl+D) and Invert selection (Ctrl+I); the shortcuts also work anywhere in the window. On a show with many stages, that is how you build the multi-selection the buttons at the bottom act on.

Each row carries its own controls. Three of them sit at the end of the row and are easy to walk past: a small screen icon with no label at all, a DVI button and a ... button.

Pro tip: You can use ELM to output video to a projector or TV without controlling LEDs. Simply create a stage and don't put any LEDs on it. Then in the stages manager, click on the DVI button.

3D (voxel) stages are not listed here. They have their own Manage 3D stages dialog, where each row has Monitor, Edit LEDs and Delete buttons and a New button at the bottom. Merge order, test patterns and NDI don't apply to them, so those columns are absent.

3D stages

A 3D stage is a volume of LEDs — a cube, a hollow cube, a sphere or a cylinder — that ELM drives as one rig. ELM slices the volume into parallel depth planes, evaluates your media once per plane, and sends the result to the LEDs. Everything you already know about stages still applies: media, effect layers, intensity, colour, speed, sequences and remote control all work on a 3D stage.

Open the workspace with 3D stages in the top menu, next to 2D stages. Each 3D stage is a tab. With no stages yet the screen reads No 3D stages yet. Build a cube, sphere, cylinder, or imported point cloud, then drive it as a live volume that 3D effects sweep through by depth. with a New 3D stage button under it. Ignore the words "imported point cloud" there — point-cloud import is not in this release and there is no Import button. Build your rig from one of the four shapes.

The old 3D tools screen is gone. In ELM 2025 a 3D structure was built by stacking 2D stages. That workspace has been retired from the menu in ELM 2026 and is replaced by 3D stages. If you open a project that still contains legacy 3D content, ELM shows a Legacy 3D content warning: This project contains legacy 3D (3D tools) content, which is no longer supported in ELM 2026. To view, edit or convert that content, please keep this project in ELM 2025.

The 2D stages the structure was stacked from load normally and keep outputting on their existing DMX addresses — it is the 3D structure on top of them that is gone. Nothing converts it for you. There is no conversion command in ELM 2026: if you need that content as a 3D stage, keep the project in ELM 2025, or rebuild the rig here as a 3D stage and re-patch it.

How the volume is oriented

Read this before you create a rig. One set of world axes governs the shapes, the run axis, the depth axis, the projection faces, the test sweeps and the viewport:

So the Front projection face is the X–Y plane with depth receding along Z, which is why picking Front and then clicking the Front camera button shows your media the same way round as on a 2D stage. The viewport draws the same axes, Y up.

Two tooltips in the New 3D stage dialog still describe an older, Z-up convention: Run axis says "Z = vertical columns", and Faces says the two open ends are the top and bottom under Z. Both predate the axis correction and are wrong. The vertical axis is Y. Trust the table above and the live preview, which are correct.

Create a 3D stage

Click New 3D stage (top right, or on the empty screen). The dialog builds the rig from a parametric shape and patches it to DMX in one step. The preview on the right redraws as you type, so check it before you click Create.

Give the stage a Name (up to 30 characters; it must be unique), then pick a Shape:

The size fields carry no unit. They set the rig's proportions and how big it is in the 3D view; the LED counts, not the sizes, decide how many fixtures you get.

Faces appears for the hollow cube only. Ticking Close the two ends gives a closed box — all six sides, for a cube people walk around. Clearing it leaves the four walls with both ends open, which is what a room, a booth surround or a floor-standing cube needs.

Which two ends are open is decided by the Run axis: they are the two faces that axis points at. With the default Z the open ends are the front and back. For the usual booth or room — walls all round, open at the top and bottom — set the run axis to Y, which also makes the wall strips run vertically. Set it before you click Create: the run axis cannot be changed afterwards. The preview redraws as you change it, with Y up, so you can see which two faces are open before you commit.

A closed hollow cube looks identical to a solid one from outside, because the LEDs it drops are the ones nobody could see. The preview therefore states what you saved — for example 216 fewer than solid (512) — under the LED count.

Addressing

If the rig does not fit in the universes you have, ELM says so and creates nothing: The 3D rig does not fit in the available DMX universes. 16-bit uses twice as many channels per LED as 8-bit, so try 8-bit, fewer LEDs, or a lower start universe.

Advanced — depth slicing

Expand Advanced — depth slicing only if you want to control how the volume is cut up. The rig is decomposed into parallel planes along the Depth axis, and your media is evaluated once per plane. Depth slices (1–256) is how many planes there are.

Depth axis defaults to Z for every shape, whatever the run axis is. It is the same setting as Projection face on the Stage tab, named for the axis instead of for the face: Z is Front, Y is Top, X is Side. The two cannot disagree, and unlike the run axis this is not a permanent decision — change it from the Stage tab whenever you like and the volume re-slices and the media re-projects, without ever re-addressing the rig.

ELM derives the slice count from the shape you are building and from the depth axis — a 20 × 8 × 4 slab is 4 deep seen front-on but 20 deep from the side. The preview states the result next to the LED count: 16 depth slices — one per layer of the rig, or 8 depth slices for 16 layers — layers will share content when the two do not match. Every LED snaps to its nearest plane, so a mismatch means pairs of physical layers show identical content. If you type a value yourself it is kept and no longer follows the shape; a coarser count is a legitimate way to buy back rendering cost on a deep rig. You can change the count again later in Edit LED arrangement.

The preview

The right half of the dialog draws the rig as points, with the LED count, the saving line for a hollow cube, and the slice line. Show LED wiring (on by default here) draws the chain order over the rig, and ◀ / Play / ▶ steps or animates a highlight along it so you can see where each run starts and which way it goes. Very dense rigs are thinned on screen only; the stage you create uses every point.

The dialog's introduction ends "You can re-patch or import a point cloud later." Only the first half is true in this release: point-cloud import (CSV / PLY / OBJ) is not available, there is no Import button anywhere in the 3D workspace, and a 3D stage is always built from one of the four shapes above. The DXF, CSV and xLights imports on Settings → Project → Import load into a 2D stage, not into a volume.

The 3D stage page

Each tab shows the volume on the left and a Stage / Testing panel on the right, laid out like a 2D stage page. Drag the divider to widen the panel.

Moving the camera

Drag to rotate, scroll to zoom, Shift+drag to pan. The same hint is printed along the bottom of the viewport. Four buttons sit above the volume:

The viewport uses the axes above — Y up, X right, Z front-to-back — matching the rig model, the projection faces and the New 3D stage dialog's preview. A 3D stage built in an earlier ELM 2026 build appears rotated compared with before: that is the correction — the rig, its addressing and its output are unchanged, only which way the preview points the camera.

Stage tab

Projection face and Depth apply as soon as you pick them — there is no Apply step — and neither one re-addresses the rig. Flip between faces freely on a patched installation to find which way round your content wants to be.

Scan has been removed. Earlier builds offered a third Depth mode called Scan, described as a moving cross-section. It replaced the media's vertical axis with the depth position, so the picture did not vary vertically inside a plane and its own top-to-bottom detail was spent on depth; it also only ever did anything for shader effects. Projects saved with Scan open on Extrude.

Testing tab

Turn Testing mode on to send a known pattern to the rig. This is real DMX and it overrides any effect on the stage, so use it to check a physical install, not during a show. Turning it on also switches Live colours on so you can see the override in the preview. The rest of the fields appear once the switch is on.

Edit LED arrangement

Open it with Edit LEDs on the Stage tab, or from Manage. This window is about addressing only — LED positions do not change here, and the projection face and depth mode live on the Stage tab.

The header states what the rig actually is, for example 512 LEDs in 64 strips of 8, running along Z. and, under it, the DMX footprint of your current choices: 3 channels per LED — 1,536 channels, about 4 universes. That second line updates live as you change LED type, bit depth or universe packing, so you can see whether 16-bit still fits before you commit.

The run axis is shown in the header but cannot be changed. Strip membership is fixed when the rig is created, so to run the strips along a different axis you have to build the stage again — see How the volume is oriented for what each axis means. Nor is there a per-strip re-address here: the controls above apply to the whole rig. If the Address chase test shows the generated order does not follow your cable, Snake (serpentine) and New universe per strip are the only two things you can change; anything else means re-wiring the rig to match, or rebuilding the stage.

Manage 3D stages

Click Manage (top right) to open Manage 3D stages — the list of your 3D stages, with the same actions per row.

Driving a 3D stage

3D stages take their media from the normal media and Live workflow — there is no effect control on the 3D stage page. Go to Live, select the 3D stage, and click a media in the bank exactly as you would for a 2D stage. Intensity, speed, the colour picker, effect layers, layer solo and the CCT fader for tunable-white fixtures all apply to a 3D stage and update the volume immediately.

In the media bank an effect marked with a small 3D badge varies with depth: it is evaluated once per depth slice and fills the volume in true 3D. Anything else is flat, and the Depth setting on the Stage tab decides whether it fills the volume (Extrude) or lights only the front plane (Surface). A layer stack you saved from a 3D stage carries the badge too, but there it only records where the stack came from: a saved stack is flattened to one image, so it extrudes like any other flat media.

3D stages also appear in the stage picker when you create or edit a sequence group, alongside your 2D stages, and a group can mix both. Recording a step captures the 3D stage's look — effect, intensity, colour, speed and transition — and replaying it brings the look back.

Media library

Shortcut key: Alt+M.

You can manage your media and see what's active. Up to 255 media items can be loaded.

To add a media, click the round + button on an empty slot and pick a type from the menu. To load several video or image files in one go, use Batch add.

Video files (with or without audio)

ELM decodes video with its own built-in video engine, so you do not need to install codecs. Installing a DirectShow codec pack changes nothing for video files. ELM can play these files: .mov, .avi, .wmv, .mpeg, .mp4, .mp2, .mkv, .webm, .ogv and .ogg. There's no video size and resolution limits. The audio track is played back on the default sound device. If not needed, you can remove the audio by using the transcode button.

The recommended video format is MPEG4 in an avi file. For smaller files at the expense of a higher CPU usage, you can use H264 in an avi file.

Picture files

Use images like bmp, jpeg, gif (animated or not) and png.

Effects

The built-in effects give you pixel-perfect and configurable visuals. ELM ships 21 basic animations and 151 shader effects, each one configurable in multiple ways. Many effect parameters are controllable via the DMX remote.

Choosing Effects opens the Effect library window. It has four tabs:

Type in the search box at the top right to filter the list by name; a tab with no match shows No effects match the filter.. Hold Ctrl to pick several effects at once, or double-click one to select it and close the window. Tick Grayscale at the bottom left to preview the shaders without colour — the box appears on the three shader tabs, not on Basic. An effect carrying a small 3D badge varies with depth and fills a 3D stage in true three dimensions instead of being painted on it as a flat picture.

You can add your own shaders. Put a GLSL or ISF fragment shader in %ProgramData%\ELM\shaders with a .frag extension and it appears in the library next to the built-in ones. If the shader starts with an ISF header comment, its inputs become the effect's parameters; a shader without one simply has no parameters. To give it a thumbnail in the picker, drop a JPEG beside it named after the shader, for example myshader.frag.thumb.jpg.

3D effects

Effects that fill a 3D (voxel) stage in true three dimensions rather than being projected onto it as a flat picture. Picking 3D effects from the + menu opens the same Effect library filtered to those effects only, so the Basic tab is hidden and the window opens on Eye candy. The same effects also appear in the normal Effects list, marked with the 3D badge.

Capture devices

Use live videos captured with a video card input or a connected camera like a webcam. All DirectShow compatible video cards, cameras and devices are listed.

Adding one opens the Select a video capture device window, where you choose three things: Device, Stream and Physical input. Stream lists every format the device offers, for example 1280x720 @ 30 fps MJPG 24-bit. Choose it deliberately: most cameras offer the same resolution in several formats at very different frame rates — a typical webcam lists 1280x720 as both MJPG at 30 fps and uncompressed at 10 fps. The format you pick is applied when the device opens. If you don't pick one, ELM uses the fastest format available at the chosen resolution. If a format saved with your project is no longer offered — after a driver update, or with a different camera on the same port — ELM falls back to the best available and notes the substitution in the log.

A device that is unplugged and plugged back in recovers on its own; ELM keeps retrying with a growing delay. A device that genuinely cannot be opened, because it is gone or held by another application, is reported once rather than every few seconds. On a computer with no camera attached the window shows No devices detected. with the controls greyed out.

Screen capture

Capture any part of your screen and use it as media in ELM. For example, play a video on YouTube and map it to your LEDs. Another utility is to capture your video editor (like Adobe After Effects) preview zone and let ELM generate a LED preview in real-time to assist your content creation. This avoids having to render your video before trying it in ELM.

Screen capture dominant colours

Capture any part of your screen and extract the dominant colours. This is useful to do colour matching between any kind of content and your LEDs.

Texts

Generate scrolling text with various fonts.

Spout (v2.007)

Use live videos generated by the most common video frameworks and VJ software. This includes Resolume, Ableton Live, Processing, Max/MSP, TouchDesigner, Cinder, OpenFramework, VVVV, Isadora, After Effects, Mapio, Unity3D and more. The Spout integration gives you total control over the content and is the way to go when you want interactive visuals.

Network Device Interface (NDI v6.0)

Receive live video streams via the network, as commonly used in professional live show productions. Easily connect ELM to a multitude of NDI compatible hardware and software, including capture cards, IP cameras and video over the cloud. NDI has been developed by NewTek.

MJPEG streams

The most basic way to receive video streams via the network. If your IP camera doesn't support the more complete NDI protocol, it should at least support MJPEG streaming.

DMX bitmap

This special media allows receiving DMX from a lighting console and converting it to a video. You can then use this video like any other media and map it on your LEDs. A common usage scenario is to control groups of LEDs with one RGB or intensity value.

Pro tip: To quickly test your installation without having the final videos, just use the built-in effects. You can create your schedules and test everything right away. Then when you get your videos, replace the built-in effects.

Pro tip: Media file paths are saved relatively to your project file. Use the consolidate feature to copy all files in a directory next to your project. Then copy everything to your other computer.

Sequences

Shortcut key: Alt+U.

The sequences section allows you to create sequences and steps (cues) for your stages and launch the perform panel for live performances. First you create groups for your stages either based on the different zones of your installation and/or the different layers. A stage can only be in one group. You can duplicate a stage (see Stages Manager) if you want to put it in multiple groups or if you'd like multiple layers like a background and a foreground. This way, you'll be able to create sequences for the background group and other punchier sequences for the foreground group.

3D (voxel) stages can be used in sequences too. A 3D stage appears in the stage picker next to your 2D stages, and one group can mix both. Recording a step captures a 3D stage's look — effect, intensity, colour, speed and transition — exactly as it does for a 2D stage. A stage that is hidden or locked is not offered; if the picker is empty it tells you so with No stages available. A stage can only be in one group and cannot be hidden or locked.

Select several groups with Ctrl or Shift, or right-click the list for Select all, Deselect and Invert selection.

Edit sequences

Go in the live panel and configure the stages of the group as you want and then store the result as a step of a sequence. The live panel is where you pick the media and set the intensity, speed, colour and layers for each stage; it is described later in this chapter, under Live mode, and reached with Alt+L. The settings stored in a step, for each stage, are: the media, the stage intensity (a stage stored at zero intensity comes back dark), the stage speed, the stage colour filter, the transition duration, the transition effect, the transition curve, and whether the audio mix is on together with its whole configuration. Media parameters are stored as well unless you untick Media params in the stage filters. An interpolation is made for transitions between steps and sequences for all numeric parameters, including the media parameters. For example, you can change the zoom level of an eye candy effect to animate it between steps.

A step stores one media per stage. If a stage is showing a stack of effect layers, ELM saves the whole stack — each layer's media, blend mode, opacity and region — as a new media called Layer stack in the media library and stores that instead, so playing the step back reproduces the full look rather than just the base media. Only the layers you can actually see are baked in: a layer that is switched off or at zero opacity is left out, and if any layer is soloed only the soloed ones are kept. ELM tells you when this happens. The look stays editable: select that Layer stack media on the stage again to get the individual layers back. Re-recording an unchanged stack reuses the same media instead of piling up copies. Each stack does use one of your 255 media slots.

Each group can have up to 255 sequences and each sequence can have as many steps (cues) as you want. You'll then be able to recall these sequences during a live show or via any remote mode.

Overview section (upper part)

Pro tip: The keyboard shortcuts for the add, update, insert and delete steps work even when the focus is outside the edit sequence window. So you can press Ctrl-space-a to add a step while working in the live panel — and Ctrl-space-u, Ctrl-space-i and Ctrl-space-d to update, insert and delete.

Details section (bottom part)

The Steps columns

The first four columns describe the step. The last six override what the step recorded: tick the box in the cell to reveal the value and apply it, untick it to fall back to what was stored.

Schedule

You can schedule your sequence so that it starts playing automatically. Each sequence carries its own schedule: open Edit sequences, select the sequence, then open the Schedule tab at the bottom right, next to Steps. A group has its own scheduler, allowing each group to play one sequence at a time.

Turn the switch at the top of the panel on first. Nothing below it does anything while it is off.

Perform panel

The perform panel is a touchscreen, keyboard and MIDI controller friendly interface to control your show. You can trigger your sequences and play live with the intensity, speed, strobe, tap tempo and more. Most of the controls can also be controlled via HTTP and OSC.

The panel is dead until you switch Active on — it reads The performer is inactive. Activate it in order to enable the controls. Once active, the performer overrides the stage parameters in the Live view — the per-stage screen described later in this chapter, under Live mode.

When another controller is driving the group, Remotely controlled appears beside the group name in the title bar.

Live mode

The main goal of the live mode is to see what's playing on each stage and configure the effects. For example, if a playlist is running, you'll see what's happening in real-time, including the transitions. If nothing is controlling the stage, you'll be able to manually select what's playing.

Each stage is a tab across the top. 3D stages appear here too, beside your 2D stages, and use exactly the same controls — media banks, faders, Layers panel and A-B fader. A 3D stage's tab tile draws a small structural 3D view of the rig instead of a flat output image, and the large preview shows the live 3D volume when the preview source is set to Output. See the 3D stages chapter for building and mapping the rig itself.

Above the A-B fader the panel tells you when something else is driving the stage: Under remote control, Under scheduler control or The testing mode is active. A stage set to Read-only in the Stage tab shows The stage is read-only and its live controls are disabled — change that on the stage itself, not here.

Audio mix

Audio mix drives one stage from the music. It takes its beat from ELM's master tempo — whichever source you picked on the Beat detection page (Settings → Audio): tap, MIDI clock, OS2L or audio beat detection. That keeps it in step with your sequences and tempo-synced effects, and it means Audio mix works on a MIDI clock or a tapped tempo with no audio input at all. It still listens to the music for the feel: level and frequency bands drive intensity, colour and the automatic strobe.

Choose your audio input once on the Beat detection page; every audio-reactive feature in ELM uses it. There is no per-stage audio input picker. Projects made before this open set to Auto, and because Auto now follows the master tempo, a show that never set a tempo source runs at the default 120 BPM instead of chasing the kick drum.

Turning Audio mix on replaces the transition effect drop-down and the Execute button for that stage with the controls below. Turning it off puts them back.

Pro tip: Using black and white media allows you to use the colour filter and get exactly the RGB colour you want as output.

Pro tip: Use the live mode as a monitor to help you see what's going on while remotely controlling ELM with another lighting console. You can also open multiple DVI outputs (one per stage) and use them as monitors.

Effect layers

An effect layer is an extra media blended over the stage's base media. Build a look by stacking layers: a base video, a strobe over it, a colour wash on the left half. The stack belongs to the stage, plays live, and is saved with your project.

The Layers panel sits in the centre column of the Live view, directly under the Intensity, Speed and CCT faders. The bottom row is always the base — the media you clicked in the A bank — badged Base and marked (Base — no blend). Effect layers stack above it, and the badge beside the Layers title counts every row including the base.

Build a stack

Click a media in the A bank to set the base. Then click + Add layer: ELM adds a layer on top and immediately opens the media picker for it. Pick the effect you want. If you cancel, the layer stays but is empty — right-click it and choose Choose effect… when you know what you want in it.

Media tiles only ever set the base. A layer's effect is chosen from that layer's own right-click menu, never by clicking the bank. The panel says so on screen: Media tiles set the base. Right-click a layer to give it an effect or edit an existing effect's parameters.

Drag a layer row up or down to reorder the stack. The base row stays at the bottom and is never a drop position.

Right-click a layer

Right-clicking an effect layer gives you:

• Choose effect… — pick the media this layer plays.
• Edit parameters… — open the layer media's parameters in a popup, the same editor the media banks use. Nothing else is needed to make a layer effect look right.
• Move up / Move down — reorder without dragging.
• Delete — the same as ✕, with the same confirmation.

The base row has a shorter menu. Its first item is Choose media…, and that is the one way to swap the base media while keeping your layers — a click in the media bank replaces the whole look and clears the stack. Both actions are undoable with Ctrl+Z.

Save as new media

Save as new media bakes the base plus every visible layer — with its blend mode, opacity and area — into the next empty media slot. You need a base media and at least one layer, or ELM tells you there is nothing to save. Hidden layers and layers at zero opacity are not baked, so what you save is what you see.

The saved stack behaves like any other media: click its tile to play it, put it in a sequence step, or drive it remotely. Clicking it in the bank also un-bakes it back into editable layers, so you can pick a saved look up months later and change one layer. Its tile is the highlighted one while it plays, and a stack saved from a 3D stage carries the same small 3D badge as a 3D effect.

While a sequence is driving the stage and the step is playing a saved stack, the panel expands the stack so you can see what is in it, but the rows are read-only and have no right-click menu. The hint reads Baked layer stack (under sequence control). Re-select it on the base media to edit the layers. Editing mid-playback would change the live output, so ELM does not allow it.

Sentry

Sentry is ELM's fixture monitoring for permanent and architectural installations. It watches the RDM fixtures on your DMX lines — their sensors, and whether each one still replies — and the ENTTEC gateways they plug into. When a reading crosses a threshold you set, or something stops answering, Sentry raises an alert in the app and, if you configure it, by email or SMS.

The two halves run on different clocks, and this decides whether Sentry suits your site. Watching the gateways is continuous. Polling the fixtures on a schedule only happens while the rig is dark, because RDM shares the wire with your DMX output — so a rig that plays content around the clock is never polled on a timer. Before you plan a permanent install around Sentry, read RDM shares the wire with your DMX output and If your rig never goes dark below.

Open Sentry from the top menu bar. Sentry is part of the ELM Architectural plan. On a copy that is not licensed for it the Sentry button is drawn in gold and clicking it offers the upgrade instead of opening the workspace.

Sentry or Device Manager?

The two screens look alike — both list things called "devices", both have a Queue tab, both can Identify — so it is worth stating the split plainly:

• Device Manager sets the ENTTEC boxes up. Network settings, per-port protocol, universe and mode.
• Sentry monitors the fixtures plugged into them, and now the boxes themselves.

A gateway port has to be set to Output with RDM enabled in Device Manager before Sentry can find anything behind it. A gateway that cannot do RDM, or has it switched off, says so on its row in the Devices list.

RDM shares the wire with your DMX output

Everything else in this section follows from that one fact. Discovery briefly holds Art-Net output on its last frame, so it is best done between cues. Scheduled monitoring waits until the rig is dark and stands down the moment content plays. Watching a gateway is the exception: that is a small network heartbeat rather than an RDM scan, so it runs continuously, including during a show.

"Dark" means no light on the wire, not ELM shut down. ELM watches its own Art-Net output: the rig counts as dark while every channel it sends is zero, and counts as lit the moment any channel is above zero. A blackout, a stage with nothing playing, master intensity at zero, a schedule that ends with Black at end, or DMX output switched off on Settings → Project all read as dark within about a second. Content starting up reads as lit just as quickly, and a pass already running is abandoned.

Two consequences worth knowing before you rely on this. ELM judges it from Art-Net output only, so a rig driven over sACN or KiNet always looks dark to the schedule and passes will run during the show, sharing the DMX line with your content. And the daily hours window narrows this rule, never overrides it: monitoring runs when the rig is dark and the clock is inside the window.

Get monitoring running

Nothing is monitored until you do these three things. Sentry ships with scheduled monitoring off and an empty device list.

1. Discover your fixtures. On the Devices tab, click Discover devices. ELM asks you to confirm first, because discovery clears the current device list, re-scans every Art-Net node and holds your Art-Net output on its last frame for a few seconds while it runs. Other protocols are not interrupted.
2. Turn scheduled monitoring on. Discovery does not start a timer. The Sentry monitor: chip at the top of the workspace reads Scheduled monitoring off until you set one up. Click Set up monitoring… beside the chip (or Monitoring… on the Devices toolbar) to open the Monitoring settings window, tick Monitor discovered devices automatically (while the rig is dark), set the Monitoring cycle (seconds) and click Save.
3. Add rules. Discovery and monitoring collect readings; they never alert on their own. Go to Rules & alerts and create at least one rule, then tick Enabled on it.

Discover devices and Refresh work whether or not scheduled monitoring is on. Once it is on, Set up monitoring… is replaced by a Hold button that pauses the schedule and a Resume button that starts it again; discovery and refresh still work while it is held.

If your rig never goes dark

A site that plays content 24 hours a day never satisfies the gate, so scheduled monitoring never runs. You can complete all three steps above, tick the master switch, build your rules, and still have nothing polled: the chip sits on Content active — monitoring on hold indefinitely, no sensor readings are collected, and no sensor-threshold rule can ever fire. Decide how you are going to handle this before you rely on Sentry for a permanent install.

Three things still work on a rig that is always lit:

• Gateway rules. The gateway heartbeat is not RDM and is not gated, so a Watch: Gateway rule reports a controller that dies mid-show, whatever the content is doing. On a 24/7 site this is your only continuous cover.
• Manual passes. Discover devices and Refresh ignore the gate entirely — a maintenance visit can inspect the rig with the show running. Discovery still holds Art-Net output on its last frame for a few seconds while it runs.
• The Live tick on a sensor row, which polls that one sensor about once a second while you have the fixture open.

To get scheduled monitoring back, give the rig a genuinely dark window and point the schedule at it: leave a gap between scheduled sequences, tick Black at end on the last schedule of the night, or drop master intensity to zero for a maintenance period. Even a few minutes of black a night gives the monitor several passes at the default 30-second cycle. You can then tick Only monitor during set hours and set the hours around that window, so an unplanned dark moment during a show — a blackout cue, a gap between cues — does not start a pass at a time you would rather keep the wire clear.

Watch the offline rules on a rig like this. Offline detect defaults to MissingOrStale, and "stale" means nothing has been heard from the fixture within the staleness timeout — 30 seconds out of the box. With nothing polling, every fixture goes stale shortly after your last manual discovery and the rule reports the whole rig offline when nothing is wrong. On a site that is never dark, set fixture offline rules to Missing, which only changes when a discovery actually runs, and put the continuous alerting on gateway rules instead.

What the monitor chip is telling you

Devices

The Devices tab lists every fixture Sentry has found, with a detail panel underneath for the one you select.

While a sweep is running a Communication underway strip appears above the list with counts of Background requests (the discovery or monitoring sweep still to run) and Priority requests (your own clicks, which jump the queue). A half-filled list during a large discovery is still working, not missing data.

The list columns are UID, Manufacturer, Model, DMX address, Footprint, Port and Last seen, plus Presence with a coloured dot:

Right-click a device for Create notification rule…, Alert when this device goes offline, Create group from selection… (which uses the whole multi-selection), Identify device and Stop all identify. The rule shortcuts create the rule already scoped to that device and switch you to Rules & alerts to finish it.

Grouping by gateway

Set Group by to Gateway and the list becomes a collapsible tree, one header row per discovered gateway, including gateways with no fixtures behind them. Gateways start collapsed so a large rig opens instantly; right-click anywhere in the tree for Collapse all gateways and Expand all gateways.

Each gateway header carries its name, IP, port summary, device count, its average RDM response time and a health dot — green when it answers reliably and quickly, amber with slow to respond — check the gateway, red with not responding to RDM — check the gateway, not ELM. That points you at a misbehaving gateway instead of leaving you wondering whether ELM is at fault. A gateway with RDM switched off is badged RDM disabled and one that cannot do RDM at all is badged RDM not supported; neither can report fixtures, and the row explains that rather than claiming nothing was found.

When the gateway itself stops answering Art-Net, the whole header row turns red. Hover it for the reason, and right-click it for Alert when this gateway goes offline.

The device detail panel

Select a fixture and the panel below the list fills in. Until you do, it says Select a fixture above to see its sensors, identity and DMX address.

The Identity / Gateway / DMX section is collapsed by default — it is reference data you read once, and the chart below needs the height. Expand it for manufacturer, model and model ID, device firmware, last seen, presence, the gateway's name, IP, MAC and firmware, and the fixture's DMX address, footprint, personality and universe. A model name or firmware string shown dimmed and italic is a numeric fallback: that fixture did not report the descriptive value over RDM. A warning triangle beside the fixture name means it refused a required RDM parameter; hover it to see which.

Addresses above 512 are accepted with an amber warning, not rejected: an ENTTEC line running Overdrive DMX carries up to 8 universes (4096 channels). Make sure Overdrive is enabled on the gateway or the fixture will not receive data — it is a per-port setting on the gateway itself, set in Device manager's settings form (DMX mode → Overdrive) or on the node's own web page, never in Sentry. An address below 1, above 4096, or one whose channel footprint would run past 4096, is a hard error — the note turns red and Apply is disabled, so nothing goes on the wire.

Sensors and the chart

The Sensors grid lists each sensor the fixture declares, with columns Live, Chart, Name, Value, Unit, Range, Normal and Updated. Units are shown as symbols (°C, V, A, Hz, Ω) rather than raw RDM codes.

Charted sensors are stacked into one band per quantity — temperature, voltage, current, power — each with its own labelled value scale up the side, sharing a single time axis. Each band auto-scales to the readings it holds and will not magnify a reading past the resolution the sensor can actually report, so a rock-solid 230 V rail looks flat instead of appearing to swing wildly. Sensors measuring the same quantity share a band so they are comparable; a declared normal band is shaded behind the trace only when that band holds a single sensor, because a shaded region cannot honestly stand for two.

The gestures are printed under the chart:

While you are zoomed or panned the chart shows a PAUSED badge and keeps recording in the background.

Rules & alerts

This tab has three sub-tabs: Rules, Groups and Alerts & notifications.

Click Add rule and fill in the editor on the right. A new rule starts disabled — it does nothing until you tick Enabled, so you can finish configuring it without arming a pager halfway through.

The fastest way to build a rule is not this editor. Right-click a device in the inventory for Create notification rule… or Alert when this device goes offline, right-click a gateway header for Alert when this gateway goes offline, or right-click a sensor row in the detail panel for Create threshold rule for this sensor — which seeds the threshold from the current reading. Each opens the new rule on the Rules tab, scoped correctly and disabled, ready for you to adjust and enable.

Watching the gateways themselves

Set a rule's Watch to Gateway and it reports on the ENTTEC controllers rather than the fixtures behind them. ELM sends a quiet Art-Net heartbeat to every discovered gateway every 10 seconds and treats one as unreachable after 30 seconds of silence — three missed polls, so a single dropped broadcast on a busy network cannot raise a false alarm.

Because that heartbeat is one small broadcast rather than an RDM scan, it keeps running during your show, unlike scheduled RDM monitoring which waits for the rig to go dark. A controller that dies mid-show is reported while it matters.

When a gateway goes quiet ELM also checks whether the box still answers a network ping and says so in the alert and in the header-row tooltip:

• Still answers ping — it is powered and on the network, but its Art-Net has stopped. Usually a controller that needs restarting.
• Does not answer ping either — check its power and cabling.

A gateway restored from a saved project starts as unknown rather than assumed online, and is given time to answer before anything is reported. Debounce, cooldown, severity, email and SMS routing and recovery alerts all work exactly as they do for fixtures.

Groups

On the Groups sub-tab, Add group creates a named set of devices — "North Facade", "Bridge Span" — and you point a rule's Scope at the whole group. Membership is a list of device UIDs; the quick way to fill it is to multi-select devices in the inventory and right-click Create group from selection…, then rename the group here. Groups scope rules; they do not regroup the Devices list, which uses Group by.

Alerts & notifications

Alert history on the left lists every alert Sentry has raised — Time (UTC), Severity, the Rule that fired and the Message — and Export CSV saves it.

Email and SMS use your own accounts; ELM sends nothing on your behalf until you supply credentials.

Enabling a channel here does not route anything on its own. Each rule chooses its own channels with its Email and SMS tick boxes, so you decide which alerts go out and which stay in the app.

Queue

The Queue tab is the RDM traffic log — every request Sentry has put on the wire, including your Identify and set-address commands. It is not a list of alerts; those live in Alert history.

Counts along the top show how many requests are Pending, Running, Succeeded and Failed. Narrow the list with the Status filter (All, Pending, Running, Succeeded, Failed, Cancelled) and the free-text box, which filters by name or gateway. Select several rows — filter to Failed, then Shift or Ctrl-click — and right-click Retry or Cancel to act on all of them at once. Export CSV saves exactly the rows currently shown, after the filter. Clear completed tidies finished rows away and Cancel all empties the queue.

Monitoring settings

Everything on this window applies to scheduled RDM monitoring only. Gateway watching is a separate mechanism and is not affected by any of it.

Monitoring settings are saved with the project, so a show file carries its own monitoring cadence.

How many devices your licence covers

Sentry monitors up to the number of devices your licence covers. An Architectural licence issued without an explicit figure carries a baseline of 1,000 devices; Settings → License shows the number your licence actually grants beside the Sentry entry.

Discovery still finds and lists everything on the wire. If you discover more devices than your licence covers, an amber banner appears above the device list — for example Sentry is monitoring 1,000 of 1,100 discovered devices — your licence cap. Upgrade to monitor them all. The devices beyond the cap stay visible in the inventory but are not evaluated and cannot raise an alert. The monitored set is chosen by lowest UID so it stays the same between passes rather than shifting as discovery reorders the list. Contact ENTTEC to raise the limit; the banner and the monitored set follow the new entitlement as soon as the licence updates, with no restart.

Device manager

Device manager is where you discover and set up the ENTTEC gateways and nodes on your network from one place. It finds the ENTTEC Art-Net nodes on your network and sets them up — addresses, names and per-port universes — one at a time or a whole install at once. Instead of configuring each unit by hand through its web page, you can configure a batch of nodes in one pass, which makes it the fastest way to bring a large or permanent rig online and to keep its gateways consistent.

Open it from the top menu bar (Alt+V). Entering the workspace starts a network scan on its own, so the inventory usually fills in before you touch anything.

Device manager is part of the ELM Architectural plan. On the Free and Professional plans the Device manager entry stays in the top menu but is locked — shown in gold — and clicking it gives you an upgrade message instead of opening the workspace. The refusal is also enforced below the interface, so nothing that reaches the workspace by another route can write to hardware: an unlicensed copy reports Device Manager isn't included in your licence. instead of sending the change.

Device manager and Sentry are two different jobs on the same hardware. Device manager sets up the ENTTEC boxes — addresses, names and per-port universes. Sentry monitors the fixtures plugged into their DMX ports. A gateway port has to be set to Output+RDM here before Sentry can find anything behind it.

The workspace has four tabs:

Finding nodes

Click Refresh in the header to scan. It broadcasts an Art-Net poll and lists every node that answers. Nodes also arrive on their own from ELM's normal background polling, so the list can grow while you are looking at it.

Before the first scan, the inventory shows a short paragraph saying what the screen is for, plus a Scan for nodes button that runs the same scan as Refresh. The paragraph stays until you scan; after that it is still available from the ⓘ next to the header.

If nothing is found, the network adapter is almost always the reason. ELM scans on the adapter chosen for Art-Net, not on every adapter the machine has. Go to Settings → Art-Net, pick the adapter your lighting network is on, come back and press Refresh. Also check that the nodes are powered and on the same subnet — a node on a different subnet will not answer the poll.

Inventory tab

The Supported column decides how much you can do with a node:

A node running firmware older than its profile requires reads as ENTTEC — not yet, not as Configurable. Update it through its own web page, then press Refresh. A node running firmware newer than the profile is treated the same way, because a newer firmware can use a different configuration interface.

Working on one node

Right-click a node in the inventory:

The Name field below the grid is the install name. It works for any node, supported or not, and is stored against the node's MAC address, so it survives an IP change and a re-scan. It is saved with the project. This is also the name written to the device as its own name when you press Configure or apply an auto-configure run.

The settings form

Select a Configurable node and ELM reads its current configuration and builds a settings form in the pane below. The form is built from the model's own profile, so a Storm 10 shows ten port rows and a DIN Ethergate MK2 shows two, each with the fields that model actually has.

DMX mode is what the port does. On a Storm 10 the choices are Off, Input, Output, Output+RDM and Overdrive; a DIN Ethergate MK2 adds Address. Set it to Output+RDM if you want Sentry to discover the fixtures on that port, or to Overdrive if that one line has to carry more than one universe (see Overdrive DMX), and set Count to the number of universes the port carries. The auto-configure recipe does not touch DMX mode — a bulk run leaves every port in whatever mode it was already in — so both RDM and Overdrive are switched on here, per port, or on the node's own web page.

Auto-configure tab

Bulk configuration is not "select some rows and press go". You build an install list — the order the boxes are installed in — and every auto-increment walks that order. Position comes from the list, not from the order nodes happened to answer the scan.

1. Build the install list

One device per line, MAC, then an optional label and zone:

00:11:22:33:44:55, Stage-L, Zone A

Blank lines and lines starting with # are ignored. Position is assigned by line order, 1-based, across the valid lines. There are three ways in:

The list appears below with #, MAC, Label and Zone columns. Importing again re-numbers positions by the new line order; a MAC that is already listed is updated in place rather than duplicated. Lines with a malformed MAC are skipped and counted as such.

Beside the buttons, a summary reconciles the list against what is on the network — for example 12 matched · 1 missing · 3 unlisted. Missing means the MAC is on your list but nothing on the network answered to it: a dead box, an unplugged cable, or a typo on the paper. Unlisted means a node is out there that your list does not mention. Fix missing nodes before applying; a listed node that is offline is simply left out of the run.

The install list is saved with the project, so the paperwork survives a save and reopen, and the next visit to site starts where you left off.

2. Write the recipe

The recipe ships filled in with static addressing from 192.168.0.10 step 1, mask 255.255.255.0, router 192.168.0.1, protocol sACN, merge HTP, universe base 1, one universe per port, continuous. Change what you need and leave the rest.

3. Regenerate the preview and read it

Press Regenerate preview. The table does not update on its own — press it again after every change to the recipe or the list, or you will be reading a stale plan.

The preview is a dry run. Nothing has been sent at this point.

Only nodes that are both on the install list and resolve to a configurable model appear in the preview. A listed node that is offline, or one ELM cannot configure, is not in the batch.

Beside the buttons is a summary reading, for example, 18 changes · 2 warnings · 0 blocked:

One blocked row disables Apply selected → Queue for the whole run. Unticking the blocked row does not re-enable it — fix the cause in the recipe or the install list and regenerate.

4. Apply

Apply selected → Queue writes the settings to the ticked devices and moves you to the Queue tab.

Applying cannot be undone, and it restarts hardware. Each node restarts as it is configured and drops off the network for a few seconds, so anything running through it stops. Changing addressing between DHCP and static restarts every node in the run. Plan a bulk run for a time when the rig can go dark, and check the preview table before you press it.

Applying the same run twice is safe: a device that already has the settings shows no changes and is skipped.

The licence node cap

How many nodes one run can configure is set by your licence. Every node still shows up in the inventory and in the preview, but the rows beyond your limit are marked over capacity, coloured amber and skipped, and a banner above the preview tells you how many are covered — Managing 20 of 25 configurable nodes — your licence cap. Upgrade to manage them all. The rows that fit are still applied; the cap does not block the run.

Rows are kept in install order, so the first N on your list are the ones that get configured. Architectural licences start at 20 nodes; Settings → License shows the number yours includes, as up to 20 nodes beside the Device manager row. Capacity add-on packs are applied to your licence in your ENTTEC account, not in ELM, and the new number arrives the next time ELM starts.

Queue tab

Everything Device manager sends to hardware goes through this queue, whether it came from Configure on one node or from a whole auto-configure run. Counts across the top show Pending, Running, Succeeded and Failed.

Filter by status or by text to find one node in a long run. Right-click rows to Cancel or Retry them — the action applies to the whole selection. Cancel all stops the pending work, Clear completed tidies the list, and Export CSV writes the rows currently shown to a file you can open in Excel, which is the commissioning record for the run.

If a row fails, check whether the node is back on the network first — a node that is still restarting cannot answer. Press Refresh on the Inventory tab, then retry the row.

Supported devices tab

Not every node can be configured from ELM. This tab is the list of models Device manager can configure and the firmware each one needs. As of ELM 2026 it holds two models:

Firmware matching is exact, because the configuration interface is firmware-specific. A Storm 10 on firmware 1.4 is listed in the inventory but is not configurable; update it through the device's own web page, then press Refresh.

Every other node on your network is name-only: you can give it an install name, open its web page and see it in the inventory, but its settings have to be changed on the device itself. If you need a model added, use Request support — contact ENTTEC in the pane below the inventory.

Reference

All settings panels, remote control options (DMX, OSC, OS2L, HTTP), troubleshooting, and the end-user licence agreement.

Settings: project

This is the overall panel, displaying the most important info about your project status. The top of the page holds the project buttons and the project counters; below them, a grid of status tiles shows what every subsystem is doing right now.

Project buttons

Counters and output settings

Status tiles

The lower part of the page shows one tile per subsystem: CPU, Media, Art-Net, sACN, KiNet, NDI out, Audio, MIDI, Remote and CITP. Each tile has a coloured square and a line of text beside it. Read the square first:

This is the first place to look when something is wrong on site: a red Art-Net tile or a grey MIDI tile tells you which page to open next.

Backups

A project backup file is created every 5 minutes. To avoid slowdown during a live show, you need to manually save the project to trigger the backup creation. To open a backup, select the Backup file type in the file browser when opening a project. To set the backup to be the current project, simply save the project.

Import xLights / xModels

Select one or multiple xModel files to import. Strips will be automatically created to best fit the models.

Import CSV LEDs

Select one or multiple CSV files to import. Strips will be automatically created to best fit the LEDs. Each row of the CSV file can contain the following columns:

A stage will be created to contain all LEDs based on their X and Y positions. To export the LEDs of a stage as a CSV, see Stages Manager.

Import strips from AutoCAD DXF file

Select a DXF file to import using the AutoCAD 2000 and up format, either in binary or text. Strips will be created based on the entities. Notes are also created based on the Text and MText entities.

Art-Net

Settings for the Art-Net universes. Configure them by entering the IP addresses of your nodes in the grid at the bottom of the page. ELM provides 32768 Art-Net universes; how many you may actually use depends on your license.

Page controls

The universe grid

The coloured block behind each universe number tells you its state: green means the universe is patched and being sent, dark yellow is a warning, red is an error, and grey means the universe is not in use. Red also appears when a universe is in use but falls beyond your license limit — ELM keeps the show running and blacks out that universe rather than stopping output.

Advanced: several network cards, and auto-numbered node IPs

Above the universe grid, on the left, is an expander headed Advanced — send output across several network cards, or auto-number node IPs. It only appears when the computer has more than one network card (loopback doesn't count), and it starts collapsed. Most setups send everything out the Default NIC and never need it. Use it if one card can't carry all your universes, or to fill in a block of sequential node IPs without typing them.

Each fixture must be reachable from the card its universe is assigned to. A universe sent out a card that can't reach the fixture fails silently — nothing arrives and no error is raised. ELM asks you to confirm before it re-routes anything.

Once more than one card is available, each universe box grows a small coloured chip on its right edge showing which card it goes out of (no chip means the Default NIC), and hovering the chip names the card. A universe whose IP address falls outside its assigned card's subnet gets a red outline — ELM warns but still sends. You can also right-click a page tab and pick a card to send that whole page out of it.

sACN

Settings for the sACN (E1.31) universes. Configure them by entering the IP addresses of your nodes in the grid at the bottom of the page. ELM provides 32768 sACN universes; how many you may actually use depends on your license.

Page controls

The universe grid

The coloured block behind each universe number shows its state: green when the universe is patched and being sent, dark yellow for a warning, red for an error or for a universe in use beyond your license limit, grey when it is not in use.

Advanced: several network cards, and auto-numbered node IPs

Above the universe grid is the same expander as on the Art-Net page, headed Advanced — send output across several network cards, or auto-number node IPs. It only appears when the computer has more than one network card, and it starts collapsed.

Each universe box then shows a coloured chip for the card it goes out of (no chip means the default adapter), and gets a red outline when its IP falls outside that card's subnet. You can also right-click a page tab to send that whole page out one card. Remember that each fixture must be reachable from the card its universe is assigned to — universes sent out the wrong card fail silently.

KiNet

Settings for the Philips Color Kinetics devices (power supplies). Both protocol versions are supported, and the KiNet page has a sub-tab for each: V1(DMXOUT) and V2(PORTOUT). The two sub-tabs are configured completely independently — each has its own adapter, its own Locate uni. and Monitor buttons, its own quick mapping panel and its own universe grid. Use Philips QuickPlay Pro to configure and get the IP addresses of your KiNet devices, and close it once the configuration is done to avoid flickering.

Philips controllers generally expect the computer to be on the 10.x.x.x network with a 255.0.0.0 subnet mask. If the adapter you select is not on that network, ELM shows a warning on the page — change the computer's IP address before troubleshooting anything else.

Controls on both sub-tabs

KiNet V1 (DMXOUT)

On this sub-tab each universe carries only a power supply IP address; there is no port. One power supply drives one universe.

Quick power supply mapping is the practical way to fill the grid — without it you are typing IP addresses into boxes one at a time.

KiNet V2 (PORTOUT)

On this sub-tab you map each KiNet device's port to a universe in ELM. ELM provides 32768 KiNet universes; how many you may actually use depends on your license.

Quick power supply mapping maps one supply per press, port by port:

Quick power supply mapping reaches the first 2048 universes only. Beyond that, set the IP addresses and ports in the grid directly, or use the IP — universes range fill under Advanced.

Scheduler

The scheduler is responsible for activating your sequences for those with scheduling options. The sunrise and sunset times are automatically updated throughout the year based on your location.

The top of the page shows the date and time ELM is currently using and today's Sunrise and Sunset for the position set below. The clock ticks once a second, and the sunrise and sunset are recomputed whenever you change the latitude, longitude, a preset or a test helper — so you can confirm the times before relying on them in a schedule. The times shown already include the sunrise and sunset offsets set below.

Audio

ELM listens to one audio input for the whole app, and this is where you choose it. Everything that reacts to audio follows that one input: the audio reactive effects, the audio reactive mix and beat detection. Pick System (loopback) to listen to what is playing on this computer, or pick a microphone, a line-in or an ASIO driver.

Settings > Audio is a single page, headed Beat detection. There is no sub-tab and no list of audio input slots — the eight-input panel from earlier versions has been removed, together with its activation switches, falloff speed, frequency-range boxes, auto-gain delay, slot arrows and the WaveIn / ASIO mode switch. The controls at the top of the page, described below, are the whole audio input chain. The rest of the page — spectrum, beat readout and master tempo — is covered in Beat detection.

If you open a project made with an older version of ELM, any per-effect audio input choice it carries is ignored — the effects listen to the input you select here. The audio reactive effects look exactly as they did.

Beat detection

Beat detection listens to live audio, works out its tempo (BPM) and beat in real time, and uses that beat to drive ELM's master tempo clock. Every tempo-synced feature (shader and effect time, sequence playback and live faders) can follow the detected beat, with no MIDI clock, DJ software or extra hardware needed. It works with any audio input: a microphone, a line-in, or the computer's own output (loopback).

Open it from Settings > Audio. That page is the beat detection page — the heading Beat detection is printed at the top of it, it is not a sub-tab you click. The page shows a live spectrum analyser, a pulsing beat indicator with the detected BPM, and the master tempo panel where you choose the tempo source.

The Input source, Input level, Auto-gain and Gain controls across the top of the page are the app-wide audio input; they are described under Audio. The controls below are the rest of the page.

ELM detects onsets across six frequency bands (roughly 20 Hz to 12 kHz) instead of reacting to raw volume, so it locks onto real musical material, not only four-on-the-floor. It confirms a new tempo before switching, so a passing drum fill will not pull the BPM around, holds time through breakdowns and quiet passages, and keeps the last good reading until its confidence is high, so the on-screen BPM stays steady rather than flickering. It searches roughly 60 to 200 BPM and reports in a stable octave; use the ÷2 · 1× · ×2 switch when you want the other one. The whole engine runs on your machine at low latency, with nothing to install and no cloud.

MIDI

The detected MIDI controllers are listed here. Activate the ones you want to use for your performance. The configuration for which notes and CC messages to use is done in the Perform panel. Multiple controllers can be used at the same time as long as they are transmitting on different channels or at least sending different note or CC messages.

White balance

Adjust the colour temperature and intensity of the various LED types in order to get a more accurate preview in the stage monitor. The page lists one card per entry — White, White cool, White neutral, White warm/amber and Default — each with a short line saying which part of the mapped media colour drives it. Default covers every other LED type: RGB, RGBW, RGBAW and so on.

Check which white type your strips use first. These white entries drive the preview only for the retired fixed white types — White, White warm, White neutral, White cool, WWA, WAW and AWW. Those types still load and render so old shows keep working, but they are no longer offered when you add or edit a strip. The picker now offers White single, White dual and White triple instead, and for those types the stage monitor colour comes from the emitter temperatures you set on the strip and from the stage's CCT fader on the Live view, not from this page. See Edit strips.

These settings are saved with your project. They change only how ELM draws your rig — the stage monitor, and the previews ELM streams to a console over CITP or to the remote control view. They never change what is sent to your fixtures.

License

Manage your license. ELM holds one license at a time: importing another replaces the one already stored, it does not add to it. To get more DMX universes, buy an add-on and apply it to your existing license in your ENTTEC account — see below.

Status

The card at the top shows a coloured dot with the current state and a plain sentence explaining it. It also names your plan, but only when your license carries a plan name — older licenses and demo mode do not, and then the line is simply absent. If no plan name is shown, read your plan off the Plan card below instead: Remote control marked Upgrade means Free; Remote control Included with Sentry and Device manager Upgrade means Professional; all three Included means Architectural. A demo machine also shows all three as Included, but its headline reads Demo mode. Underneath the status, when a license is present:

Plan

The Plan card shows the DMX universes your license grants — including any add-on you have applied — then every feature ELM gates, each marked Included or Upgrade: Art-Net output, sACN output, KiNet output, Live mode, Remote control, Sentry and Device manager. The two features with a capacity also show the included quantity beside the name, for example Sentry up to 1,000 RDM devices and Device manager up to 20 nodes.

This card is the authority on what you hold. The universe number here is what ELM enforces — universes beyond it go dark — and it is not the same as the 32,768 universes per protocol the engine can address; that is a capability, this is your entitlement. If you need more, universe add-on packs raise this number (see below and Licenses).

On a computer with no license the Plan card reads Unlimited for DMX universes and marks every feature Included, because demo grants everything. That is correct, not a display fault: the restriction in demo is the periodic output blackout, not the feature list. Activating a license replaces that grant with the license's own, so the numbers on this card can go down the first time you activate — a Free license reads 8 universes with Remote control, Sentry and Device manager marked Upgrade.

Activate with a voucher

Enter your voucher code and the email you bought with, then click Activate to activate this computer. If the build you are running cannot activate online it says so and points you at Import license file… instead.

Add-on pack codes are not entered here. Extra universes, extra Sentry devices and extra Device Manager nodes are applied to your existing license in your ENTTEC account under My Licences. The extra capacity appears the next time you start ELM. If you enter an add-on code on this screen ELM tells you what it is and where to use it, and leaves the code in the box so you can copy it across.

This computer

Buttons at the bottom

Benchmark

Benchmark sizes your rig against reality. It tests your actual network, CPU and graphics card and tells you how many universes this computer can truly sustain at your chosen frame rate, so you can plan a show against what the hardware can do rather than a guess. The tests are a guide only and do not change any of your settings.

The network test pauses your live lighting output. It sends real data on the wire, so ELM pauses output for the few seconds it runs and asks you to confirm first: "This sends real test data over your selected network adapters for a few seconds. Your live lighting output will be paused while it runs, so don't run this during a show. Continue?" A banner shows while it runs, and output is restored when it finishes, is cancelled or fails. The CPU and graphics card tests do not touch the network and do not pause anything. Test everything starts with the network test, so it asks the same question and pauses output too. Do not run either during a show.

Set a Target frame rate (FPS) first, since it affects the calculated capacity. Choose 25, 30, 40, 44 or 60. Use Test everything to run all three tests in turn, or run each one on its own.

Once a test has run, a banner appears at the top of the page. It tells you roughly how many DMX universes your hardware can handle (a 20% safety margin is already included), what today's build is limited to and how many you are using, and lists the figure for every standard frame rate — higher frame rates carry fewer universes. The banner is outlined green when your hardware covers all the universes you have configured, amber when it covers at least 80% of them, and red below that.

About

About is the last tab in Settings. It tells you which build of ELM you are running, gives you the two things support will ask for — your system info and your log file — and holds the privacy choices for what ELM sends back to ENTTEC.

Language

ELM ships in ten languages. The language selector is not on this tab — it is the button in the bottom-right corner of Settings > Project, labelled Language with the current language named under it. Click it and pick a language from the menu; the interface changes straight away, with no restart.

Remote control: DMX

ELM can be remotely controlled in many ways. Multiple inputs can be active at the same time to give you even more flexibility. The DMX input always takes priority.

Remote control is a licensed feature. If your licence does not include it, the Remote settings page stays visible but is locked — its tab turns yellow and the page is covered by an upgrade message — and none of the remote inputs run. DMX input, OSC, OS2L and HTTP all stay off, including when you open a project that already has them switched on, and remote control stops being applied if the entitlement lapses while ELM is running. If a working integration suddenly does nothing at all, check Settings > License first.

Settings for remotely controlling ELM with a lighting console (Art-Net, sACN or an ENTTEC USB Pro interface). Open them from Settings > Remote > DMX.

The DMX sheet showing what can be controlled and by which channel is integrated in ELM. It is dynamically generated based on the current remote settings and your stages, so you can't lose it.

The groups created in the sequences panel are automatically added to the remote control.

When no DMX data is received for more than 5 seconds, remote control of the stages is automatically deactivated. This is a failsafe feature in case the remote console has a problem. The stages go back to the schedule right away; if there are no schedules, the current media continues to play.

Groups behave differently, on purpose. Losing DMX releases a group's remote control but does not stop the group: it keeps playing the sequence it was last told to play. A short dropout from a console that only sends on change, or a few seconds of network trouble, therefore holds the triggered look instead of dropping your show to the schedule. To hand a group back to the scheduler, send an explicit value on the group's channel 1 — 0-245 (remote deactivated) or 246-250 (group inactive).

Basic fixture mode (8 channels per stage)

Extended fixture mode (15 channels per stage)

Group mode (12 channels per group)

CITP/MSEX

Activate this option to enable ELM to communicate with other CITP/MSEX compatible consoles and software. CITP/MSEX versions 1.0, 1.1 and 1.2 are available. ELM can:

• Send the media thumbnails to your console.
• Stream the LED preview of every stage to your console or visualizer.
• Send the stage statuses, including the current playing media.

CITP has its own tab and its own network adapter selector on the Remote settings page. Set the frame rate in your console or visualizer, up to 20 FPS. Make sure firewalls are off. The CITP tab is on the same settings page as the other remote inputs, so it is locked too when your licence does not include remote control.

OSC

Settings for remotely controlling ELM via OSC (Open Sound Control) over the network. ELM listens for incoming OSC on all network adapters. Open the settings from Settings > Remote > OSC and turn the switch on.

Remote control is a licensed feature. If your licence does not include it, the Remote settings page stays visible but is locked — its tab turns yellow and the page is covered by an upgrade message — and the OSC input will not run, including when you open a project that already has it switched on. If ELM ignores every message you send, check Settings > License before you debug the network.

The full address list is also built into ELM, below the settings, along with the numbered list of transition effects.

Stages

/elm/stages/{stage name}/
  live/
    intensity           0..1
    rgb (or r, g, b)    0..1, 0..1, 0..1
    media               0..255
    speed               0..10
    targetcct           1800..10000 (Kelvin), or 0 to turn it off
    transitionFx        0..43
    transitionDuration  0..9999 seconds
    audioMixControlled  {0: no, 1: yes}
    remotelyControlled  {0: no, 1: yes}
    mix/
      position          0..1
      A/
        media           0..255
        speed           0..10
      B/
        media           0..255
        speed           0..10

  merge                 1..12, or a merge mode name

targetcct is the target colour temperature for tunable-white fixtures on the stage — the same control as the CCT fader in the Live view. The value is in Kelvin and is clamped to 1800-10000. Send 0 (or any negative value) to turn the mapping off and drive the fixtures at their native white.

merge sets the stage's merge mode. Send a number from 1 to 12, in this order: overwrite, multiply, screen, overlay, darken, lighten, difference, add, subtract, blackkey, intensityCrossFade, intensityWhiteFade. You can send the mode name as a string instead.

Groups

/elm/groups/{group name}/
  performer/
    active           {0: no, 1: yes}
    paramTransitionDuration
                     0..9999 seconds
    intensity        0..1
    blackout         {0: no, 1: yes}
    bumpingIntensity {0: no, 1: yes}
    strobing         {0: no, 1: yes}
    strobeOnDuration 0..10 seconds
    rgb (or r, g, b) 0..1, 0..1, 0..1
    speed            0..10
    freeze           {0: no, 1: yes}
    bumpingSpeed     {0: no, 1: yes}
    tempoBPM         0..300
    tap              press the tap tempo button
    cleartempo       deactivate the tap tempo
    tempoDivider     1..128
    autoGo           {0: no, 1: yes}
    playMode         {0: loop, 1: all, 2: random, 3: row,
                      4: row random, 5: column, 6: column random}
    play             sequenceId, [stepId]
    go               play the next step
    back             play the previous step

/elm/groups/{group name}/
  sequences/{sequenceId}/schedule/
    enabled          {0: no, 1: yes}

tap and tempoBPM only work when the master tempo source is Manual. ELM now runs one master tempo for the whole show, and you pick its source on the Beat detection page (Settings > Audio) or from the Tempo source selector in the performer view. When an external source is driving — MIDI clock, OS2L or audio beat detection — these two messages are accepted and then do nothing. The same applies to the HTTP performer tap and tempoBPM.

Media parameters

/elm/media/
  slots/{slotId}
    parameters/{name}             value in the parameter's range (as in the live panel)
    parameters/{name}/normalized  0..1

{name} is the parameter label shown in the live panel, and every parameter type can be driven, not just sliders: a number takes its value (or 0..1 with /normalized), an on/off toggle turns on at 0.5 or more, a colour is addressed one channel at a time as {label} R, {label} G and {label} B, a position as {label} X and {label} Y, and a trigger or bang fires once on any rising value. {slotId} is the media slot number as shown in the media library, 1 to 255.

Settings

/elm/settings/
  masterIntensity    0..1
  masterSpeed        0..2
  dmxOutputActive    {0: no, 1: yes}
  schedulerActive    {0: no, 1: yes}

Bindings

/elm/binding/{id}             value in the target's own range
/elm/binding/{id}/normalized  0..1

This one pair of addresses reaches anything in ELM that can be mapped, by its global id, so a single OSC surface can drive both worlds without knowing which is which. {id} is either a performer or stage shortcut id — for example master.intensity, master.speed, tempo.bpm, tempo.divider, step.go, step.back, sequence.play.index, performer.active, or a per-group id such as group0.intensity — or a media parameter id in the form media.slot.{slotId}.param.{label}. The slot id in a binding address counts from 0, so the first media slot is media.slot.0, unlike the /elm/media/slots/ addresses above which count from 1. There is one id per performer action, and the actions are the ones listed in the shortcuts editor under the sequences menu. An id ELM doesn't know is reported in the Monitor.

Notes

Addresses and names are case-insensitive. Media index 0 is an empty slot; the slots are numbered 1 to 255. RGB can be sent in 3 parts at the /r, /g and /b addresses. audioMixControlled controls the audio mix activation. remotelyControlled determines whether OSC overrides the scheduler. DMX takes priority over OSC: a stage or group already held by the DMX input ignores OSC until DMX releases it. Use * (wildcard character) in the stage name to target multiple stages at once. For example, /elm/stages/*/XYZ targets all stages. /elm/stages/background*/XYZ targets all stages with a name starting with background.

OS2L

Use the Open Sound to Light (OS2L) protocol to receive commands from VirtualDJ. This allows using the timeline editor in VirtualDJ to trigger your lighting effects in sync with the song that is playing. Open the settings from Settings > Remote > OS2L and turn the switch on.

Remote control is a licensed feature. If your licence does not include it, the Remote settings page stays visible but is locked — its tab turns yellow and the page is covered by an upgrade message — and the OS2L listener will not run, including when you open a project that already has it switched on. VirtualDJ will simply fail to connect. Check Settings > License first.

To activate OS2L in VirtualDJ, go in the options and check yes for os2l and set os2lDirectIp to the ELM IP address (or 127.0.0.1 if the same computer) and port. For example: 127.0.0.1:9666.

Mapping a message to an action

OS2L messages drive the perform panel through its shortcuts. Open the perform panel from the sequences menu, click Shortcuts, and select an action in the list. Each action has a keyboard, a MIDI and an OS2L shortcut row; the OS2L column in the list shows what is mapped today.

In the OS2L shortcut row, choose the message type:

• btn — a button in VirtualDJ. Type the same Page and Name VirtualDJ sends; both must match exactly. The action follows the button: on when the button goes on, off when it goes off. Leave Page empty to match messages that carry no page.
• cmd — a command in VirtualDJ, matched on its id alone. Enter the id in the numeric box. VirtualDJ's param (0-100) becomes the value of the action, so a fader in VirtualDJ can ride an intensity or a speed in ELM.
• none — no OS2L mapping for this action.

ELM fills the list in with a default mapping the first time: every on/off and preset action gets a btn named after the action, and every continuous action gets a numbered cmd id starting at 1. Read the names and ids off the shortcuts list, or use the Monitor to see what VirtualDJ actually sends, and change whichever you prefer.

A btn message that matches no shortcut can still play a sequence directly, as long as the group is active: ELM matches the message's page to a group name — or uses the current group when the page is empty — and the message's name to a sequence name in that group, or to a sequence number. So a VirtualDJ button named 3 on page background plays sequence 3 of the group "background" with nothing mapped at all.

Beat and tempo

OS2L also carries the beat and BPM of the playing track, and it is one of the sources for ELM's single master tempo. Select OS2L as the tempo source on the Beat detection page (Settings > Audio) or from the Tempo source selector in the performer view, and VirtualDJ's BPM drives the master clock — every tempo-synced effect and every playing sequence follows it. While OS2L is driving the tempo, ELM's own tap tempo and Type BPM controls are greyed out, and the OSC and HTTP tap and tempoBPM messages are ignored.

There is no fallback. If VirtualDJ disconnects, or nothing is sending OS2L, the master clock holds the last tempo it was given and keeps running at it — nothing switches to audio beat detection for you, whatever the falls back to audio text printed beside the OS2L status dot on the Beat detection page suggests. Switch the tempo source to Audio or Manual yourself. See Beat detection.

HTTP

Settings for remotely controlling ELM via HTTP, which is a protocol based on TCP/IP. Use it to build a custom web interface, or to drive ELM from other software. Open the settings from Settings > Remote > HTTP and turn the switch on.

Remote control is a licensed feature. If your licence does not include it, the Remote settings page stays visible but is locked — its tab turns yellow and the page is covered by an upgrade message — and the HTTP server will not run, including when you open a project that already has it switched on. Nothing listens on the port, so your client's connection is refused rather than answered with an error. If your integration does nothing at all, check Settings > License before you debug your client.

Where the API is served

The HTTP API is served on this computer only. Use http://localhost:port/elm (or http://127.0.0.1:port/elm) from software running on the same machine as ELM. The port can be omitted if it's 80, which is the default for HTTP. Requests coming from another computer do not reach the server. The URL shown on the settings page is built from this computer's name, but it only answers on this computer.

You can restrict access to the API by activating the Basic auth feature. For more details see Basic access authentication.

Requests

Effect layers

A stage's effect layers are part of the API. Every stage state — from GET /elm/stages, GET /elm/stages/{name} and GET /elm/stages/{name}/live — carries a layers array, one entry per layer, with its index, name, mediaId, blend, opacity, isEnabled, isSolo and regionX / regionY / regionW / regionH. A stage with no layers returns an empty array.

POST /elm/stages/{name}/layers/{index} sets one layer and returns it. Layers are numbered from 0, unlike media slot ids which are numbered from 1. Send only the parameters you want to change.

regionX / regionY / regionW / regionH are a fraction of the stage, so the API can place a layer on any rectangle. The Layers panel's Area drop-down only offers five of them — Full is 0,0,1,1, Left is 0,0,0.5,1, Right is 0.5,0,0.5,1, Top is 0,0,1,0.5 and Bottom is 0,0.5,1,0.5. Post a rectangle that is none of those and the layer is masked to it correctly, but the drop-down reads Full, and changing it in the UI throws your rectangle away.

The blend value

blend takes one of these thirteen tokens, matched without regard to case. They are the internal merge-mode names, not the plain-language labels the Layers panel shows:

overwrite  add  subtract  multiply  screen  overlay
darken  lighten  difference  blackkey
intensity  intensityCrossFade  intensityWhiteFade

The Layers panel's drop-down offers five of them under friendlier labels. Post the token, not the label — Normal and Multiply (mask / tint) are not accepted:

The other eight tokens are accepted and stored, and the layer's drop-down grows an extra entry so the mode is visible in the UI and is not silently reset. Two things to know before you use them:

• A layer renders add, subtract, multiply, screen, overlay, darken, lighten and difference the way Stage merging and layering describes them. blackkey, intensity, intensityCrossFade and intensityWhiteFade are stored but render as overwrite — a layer has no intensity-driven merge, because its opacity fader already does that job. intensity is a retired alias of intensityCrossFade; use intensityCrossFade on a stage.
• An unrecognised value is ignored, not rejected. A misspelt token still returns 200, and the layer keeps the blend it already had. Compare the blend field of the response with what you sent to confirm the change landed.

The blend field in every layers array comes back as the same token, so a layer showing Normal in the UI reads "blend": "overwrite" over the API. The merge parameter of POST /elm/stages/{name}, which sets a whole stage's merge mode, takes the same thirteen tokens — but that one rejects an unknown name with 400 instead of ignoring it.

Media parameters

POST /elm/media/slots/{id}/parameters/{name} works for every parameter type, not just sliders. Address the parameter by the label shown in the live panel, and send either value in the parameter's own range or normalized between 0 and 1. A number takes its value; an on/off toggle turns on at 0.5 or more; a colour is set one channel at a time as {label} R, {label} G and {label} B; a position as {label} X and {label} Y; a trigger or bang fires once on any rising value. A name the media doesn't have returns 404.

Worked examples

To start sequence #1 of the group "background", the request is:

POST /elm/groups/background/performer?active=1&sequenceId=1 HTTP/1.0\x0D\x0A
Content-Type: application/x-www-form-urlencoded\x0D\x0A
Content-Length: 0\x0D\x0A
\x0D\x0A
\x0D\x0A

To get info about the "background" group, the request is:

GET /elm/groups/background/performer HTTP/1.0\x0D\x0A
\x0D\x0A
\x0D\x0A

Notes

• Each line ends with \r\n characters which are \x0D\x0A in hexadecimal.
• There are two empty lines at the end of a request.
• Use * (wildcard character) in the stage name to target multiple stages at once for POST.
• All data is returned in JSON.
• Use the returned HTTP status code to do proper error handling. When a request fails, the reason comes back both in the body and in the X-Api-Error response header.
• DMX has a higher priority than HTTP. If the DMX input already controls a stage or group, the live, layer and performer POSTs are refused with 423 Locked and the message "Another remote method with a higher priority is controlling the stage/group of stages <name>."
• tap and the tempoBPM parameter of the performer POST only take effect when the master tempo source is Manual (Settings > Audio, Beat detection). While MIDI clock, OS2L or audio beat detection is driving the tempo, they are accepted and ignored.
• Over HTTP the crossfade is capped at 9.999 seconds, and every POST to /live re-applies that cap, even when you don't send transitionDuration — so a longer crossfade set in the Live view or over OSC is shortened the next time you post to that endpoint. Set longer crossfades over OSC.
• Addresses and names are case-insensitive.
• Parameters can be provided in the query string or form data.
• Media id 0 is an empty slot. The slots are numbered 1 to 255.

Troubleshooting

Before you contact support

Errors and warnings appear in the notification panel at the bottom-left of the ELM window. Some operations — opening a project, importing a CSV or DXF, exporting a patch, loading, consolidating or transcoding media — collect their problems into a Warnings window instead; it has a Copy to clipboard button, so copy that list before you close it.

ELM writes a log of every session to %ProgramData%\ELM\logs. A new file is started each day, named elm-<date>.log (for example elm-20260729.log). Go to Settings → About and click Open log folder to go straight there, or click the log path printed underneath the button to open today's file.

When you write to support, click Copy system info to clipboard on the same screen and paste the result into your message. It carries your ELM version and system details, which is the first thing support will ask for.

If support asks for RDM detail, start ELM with the --verbose command-line option. This adds one decoded line to the log for every RDM sensor reply — sensor number, type, unit, range and value. It is scoped to RDM only: it does not turn on general debug logging.

Output: Art-Net / sACN / KiNet

General

Sentry and Device Manager

EULA

This End-User License Agreement (EULA) is a legal agreement between you (either an individual or a single entity) and ENTTEC, the author, copyright holder and distributor of this Software, for the software product identified above, which includes computer software and may include associated media, printed materials, and "online" or electronic documentation ("SOFTWARE PRODUCT").

By installing, copying, or otherwise using the SOFTWARE PRODUCT, you agree to be bound by the terms of this EULA. If you do not agree to the terms of this EULA, do not install or use the SOFTWARE PRODUCT.

Software product licence

1. Grant of licence. This EULA grants you the following rights: Installation and Use. Each license is registered to one person and covers the use of the SOFTWARE PRODUCT on a fixed number of activated computers, determined by the license type. Activating the SOFTWARE PRODUCT on a computer uses one activation; you release it again with Deactivate this computer in Settings → License, or from your ENTTEC account. Each activation is tied to an identifier derived from that computer's processor and system drive; replacing or re-imaging the system drive changes the identifier and requires a new activation. ENTTEC's licensing server is contacted to activate a license and to refresh it from time to time; a license that is deactivated or withdrawn on that server causes the SOFTWARE PRODUCT to return to demo mode the next time it starts.

2. Description of other rights and limitations. Limitations exist on Reverse Engineering, Decompilation, Disassembly and changing (adding, deleting or modifying) the resources in the compiled assembly of the SOFTWARE PRODUCT. You may not reverse engineer, decompile, or disassemble the SOFTWARE PRODUCT, except and only to the extent that such activity is expressly permitted by applicable law notwithstanding this limitation.

Updates and maintenance. Whether you receive software updates depends on your license. A license with an active subscription receives updates for as long as the subscription is current; when it lapses, the software you have already installed keeps working but no further updates are offered. Updates are not charged for separately. Your expiry date, any grace period, and when your license was last synced are shown in Settings → License.

Separation of components. The SOFTWARE PRODUCT is licensed as a single product. Its component parts may not be separated for use on more than one computer.

Software transfer. You may not rent, lease or sublicense the SOFTWARE PRODUCT on a temporary or permanent basis.

Termination. Without prejudice to any other rights, ENTTEC may terminate this EULA if you fail to comply with the terms and conditions of this EULA. In such event, you must destroy all copies of the SOFTWARE PRODUCT and all of its component parts.

Copyright

Except for the third-party components identified below, all title and copyrights in and to the SOFTWARE PRODUCT (including but not limited to any images, photographs, clipart, libraries, and examples incorporated into the SOFTWARE PRODUCT), the accompanying printed materials, and any copies of the SOFTWARE PRODUCT are owned by ENTTEC. The SOFTWARE PRODUCT is protected by copyright laws and international treaty provisions. Therefore, you must treat the SOFTWARE PRODUCT like any other copyrighted material.

Limited warranty

No warranties. ENTTEC expressly disclaims any warranty for the SOFTWARE PRODUCT. The SOFTWARE PRODUCT and any related documentation is provided "as is" without warranty of any kind, either express or implied, including, without limitation, the implied warranties of merchantability, fitness for a particular purpose, or noninfringement. The entire risk arising out of use or performance of the SOFTWARE PRODUCT remains with you.

No liability for damages. In no event shall ENTTEC be liable for any special, consequential, incidental or indirect damages whatsoever (including, without limitation, damages for loss of business profits, business interruption, loss of business information, or any other pecuniary loss) arising out of the use of or inability to use this product, even if ENTTEC is aware of the possibility of such damages and known defects.

Third-party components

The SOFTWARE PRODUCT includes third-party and open-source components that are licensed separately. Notwithstanding the Copyright section above, those components remain the property of their respective owners and your use of them is governed by their own licences, not by this EULA.

They include, among others:

Copies of the licence texts, and — for the LGPL components — the corresponding source code or the written offer required by those licences, can be obtained from ENTTEC on request. NDI is a registered trademark of Vizrt NDI AB.

9. Data collection

9a. This section describes what the installed software sends to ENTTEC. You control both switches on Settings → About, under Statistics & privacy.

9b. Anonymous usage data — on by default. Share anonymous usage data sends anonymous usage statistics and crash reports so ENTTEC can find and fix problems. No personal information, no project names and no media names are included. You can turn it off at any time. On a free licence the tick box is locked on — sharing anonymous usage data is a condition of that licence, and the screen says so underneath: "Included with the free licence. Upgrade to Professional to turn this off."

9c. Identified diagnostics — off by default. Share identified diagnostics is an explicit opt-in. Turn it on only when you are working with support on a problem you have reported. While it is on, crash reports also carry your licence email address, your Windows user name, your computer name and your full, unfiltered diagnostic logs, so support can tie a crash to you. Turn it off again at any time; from that moment crash reports are anonymous again.

9d. Your choices are stored on this computer under %ProgramData%\ELM\telemetry and take effect as soon as you change them. Withdrawing consent stops further collection; to ask about data already sent, contact ENTTEC support.

How licensing works in the software

The notes below describe what ELM does on your computer. They are a description of the product, not additional licence terms.

Activations. Your licence covers a fixed number of activated computers. Activating ELM on a computer uses one activation. To move a licence, release the computer first: Settings → License → Deactivate this computer, or from your ENTTEC account. Trying to activate beyond your allowance is refused with a message saying the licence is already activated on the maximum number of computers.

Machine identifier. Each activation is tied to an identifier derived from the computer's processor and its system drive, hashed so that no raw serial numbers leave the machine. Replacing or re-imaging the system drive changes the identifier and you will need to activate again, so deactivate the computer first. Network adapters and MAC addresses are deliberately not part of it. Your machine identifier is shown on Settings → License under This computer, with a Copy button.

Contact with the licensing server. ELM contacts ENTTEC's licensing server to activate, and refreshes your licence in the background from time to time. Being offline does not affect a licence already on the computer — Last synced on Settings → License simply goes stale, which is expected in a venue with no internet. A licence that is deactivated or withdrawn on the ENTTEC server is a different matter: ELM returns to demo mode the next time it starts and tells you why. Only a confirmed server deactivation does this; a network problem or a server outage never does.

Subscriptions and updates. A subscription licence reads Licensed, with a line saying it is active until a date, and an Expiry row underneath. When it expires you enter a grace period — full features continue, the status becomes Reconnect to renew and a Grace until row appears with the date the grace ends. Past the grace period the status becomes Subscription lapsed, a Renew subscription button appears, and ELM runs in demo (the output blanks briefly every few minutes) until you renew. Your installed copy is never removed.

Running without a licence. ELM without a licence runs in demo mode. Demo grants every licensed feature — Art-Net, sACN and KiNet output, Live mode, Remote control, Sentry and Device manager — with no universe cap and no device caps, but the output blanks for about ten seconds roughly every five minutes and the splash screen carries a DEMO watermark. This is intended for evaluation, not for a show. Demo is not a plan: once you activate a licence, that licence's entitlements replace the demo grant, so a Free or Professional licence covers fewer features than the unlicensed copy did. See Settings → License for what yours grants.

Note: Due to constant innovation, information within this document is subject to change.