AI Fixture CREATOR for GrandMA2: Advanced AI Parsing & 3D Integration

Unlike simple text parsers, our neural network analyzes a manual like a living lighting engineer. When generating a fixture profile for MA2, it doesn't just extract channel numbers; it forms a complex physical and optical model of the device so that it works perfectly on your console.

Pan / Tilt: Extracts the head's rotation and tilt angles, as well as physical limit values (e.g., 540° and 270°).

Zoom: Determines the minimum and maximum Beam Angle. This directly affects the light spot width in the 3D visualizer.

Focus: Locates the channel responsible for the sharpness of the beam edges or the projected gobo pattern.

Prism: The AI identifies the presence of a prism and its number of facets (e.g., 3-facet, 8-facet, linear).

Prism Rotation: Captures the prism's rotation speed and direction, as well as the ability for precise indexing (stopping at a specific angle).

Gobo: Accurately recognizes both static and rotating gobo wheels.

Color Rendering: Collects precise data for the Color Wheel or CMY/RGBW mixing systems.

Control: Finds service DMX values for remote Lamp On/Off and motor Resets.

Physics: Determines weight, RDM support, connectors (PowerCON, DMX), and Daisy Chain limits.

Our neural network automatically assigns the correct behavior type to each parameter. This is critical for the flawless operation of console effects (Phasers / FX Engine):

Snap (Instant): Value transitions happen instantly, ignoring Fade Time. This is automatically applied to color wheels and gobos so that the audience doesn't see «scrolling» intermediate colors during scene changes.

Fade (Smooth): A smooth transition from one value to another. This is applied to the Dimmer (intensity), Pan/Tilt movements, and CMY color mixing.

The fixtures we generate already include built-in baseline behavior algorithms for fast and comfortable operator workflow:

Default: The baseline state of the fixture (what happens when you press Clear). The AI assigns logical default values: beam centered (Pan/Tilt 50%), dimmer closed (0%), shutter open, and color/gobo in the open/white position.

Highlight: When the Highlight button is pressed on the MA2, the fixture temporarily ignores the current scene, resets the color to white, removes the gobo, and opens the dimmer to 100%. This allows you to instantly locate the fixture on stage.

In addition to standard channels, we automatically generate and add ready-made Presets directly to the fixture library—even if they weren't explicitly stated in the manual. This greatly accelerates show programming:

Pan / Tilt: The AI calculates the center position and creates Center, Max, and Min presets.

Zoom: Automatically adds Narrow, Normal, and Wide presets.

Focus: Generates Far, Normal, and Near focus values.

Manufacturers often describe channels (specific colors, gobos, etc.) in detail only in the largest DMX mode (e.g., Mode 1), leaving other modes as empty tables.

The neural network can fill these gaps.

Your fixture will work fully with all effects, regardless of the DMX mode you patch it in!

Modern lighting fixtures are more than just a single beam. LED Bars, Pixel Washes, and hybrid strobes often consist of dozens of individually controlled light sources. If you've ever tried to manually build a multi-instance profile for a 40-pixel LED bar in the MA2's built-in editor, you know the pain and the risk of addressing errors.

Our AI GrandMA2 profile generator takes care of this routine for you. The neural network correctly recognizes the fixture structure in the manual and automatically creates full Multi-Instance profiles.

The algorithm deeply analyzes the DMX map and understands the device's segmentation logic:

Pixel-by-Pixel Control: The AI generates the exact number of Instances based on the pixel count (e.g., a center LED and surrounding rings, or an LED matrix).

Segments: If the device is controlled by macro-zones (e.g., 4 sections of 10 LEDs on an LED bar), the AI recognizes this and creates instances for these specific logical groups.

Main Module: The AI automatically isolates the Main Module, which contains global control channels (Pan/Tilt, Master Dimmer, macros, Lamp/RDM control), separating them from the pixel channels.

Crucially, we do not generate «dirty» code. Simple parsers just copy pixel channels one after another (Pixel 1 Red, Pixel 1 Green… Pixel 40 Red…), bloating the fixture XML file to a massive size. This leads to console lag and crashes.

Our AI uses the professional Smart Reference method built into the GrandMA2 architecture:

First, the neural network creates one master Base Module describing the structure of a single pixel (e.g., just R, G, B, W, and Dimmer channels).

Then, the generator forms all other pixel instances by simply referencing this base module, using DMX Offsets for the mathematical address shifts.

Thanks to this approach:

The fixture file remains incredibly light and clean.

The profile fits perfectly in the Layout View.

The fixture is handled correctly by the console's matrix effects (MA Tricks / Selection Grid).

The console does not lag during complex Pixel Mapping, as the processor is not overloaded with redundant lines of code.

If the fixture is operating in a basic mode (e.g., Basic or Standard) where there is no individual LED control, generating a complex structure is unnecessary—for these, a standard single file is downloaded.

The program creates a complex multi-part profile only for modes that contain pixel segments or for extended modes (usually labeled in manuals as Extended, Pixel Mode, Advanced, Pattern, etc.).

How to get the correct file:

Select the desired extended mode from the DMX Mode dropdown menu at the top.

If the selected mode supports multi-segment or pixel control, a bright yellow «DOWNLOAD MULTI .XML» button will appear at the bottom of the screen next to the standard green button.

Click this yellow button to download the ready-to-use reference-based (using Offsets) multi-instance profile for GrandMA2.

Our profiles are generated with full consideration of the strict requirements of virtual show programming. All physical parameters—chassis dimensions, rotation limits, and optical characteristics (min/max beam zoom, realistic color rendering)—are automatically translated into the fixture's 3D geometry.

This ensures that the created fixtures will work and look correct in MA3D (the visualizer for GrandMA2) and the built-in GrandMA3 visualizer. You get a physically accurate and beautiful image, allowing you to program your light «blind» long before arriving at the venue.