Aquilon Layer Calculator

Analog Way LivePremier™ resource & layer planning — unofficial tool
In- and outputs follow the system rate (“Sys”) and default blanking (“Def”); each connector can override both.
Inputs
Give every connector a name and a resolution (e.g. 3840x1080) — capacity 1/2/4 is calculated automatically from the pixel clock at the connector's rate. A capacity-4 input is only allowed on connector 1 or 3 of a card and disables the next connector. Tick 🔗 to group a connector with the one above it (input group: consecutive connectors on the same card, same format, max 4) and pick the canvas layout. Change the card type per slot via the dropdown.
Outputs
Max capacity 2 per output connector — grouping does not change the capacity of the individual outputs. Tick 🔗 to build an output group (consecutive outputs on the same card, max 4): the group forms one canvas (e.g. 4× 3840x2160 → 15360x2160) and attaches to a screen as a whole.
Screens
Create screens and assign outputs via the dropdown (click an assigned chip to remove it; each output belongs to one screen). Assign outputs to regions where useful, then build layers: Mixer or Split, capacity 1/2/4, optionally Cut&Fill, and the scope. The bar below the builder shows how many layers still fit. Screens scroll horizontally.
Calculation
Assumptions & rules used
  • Capacity is derived from the pixel clock including CVT blanking — FULL (standard CVT, ≈35% horizontal overhead, 550 µs vertical), CVT-RB v1 (+160 px, 460 µs) or CVT-RB v2 (+80 px, 460 µs) — against dual-link steps of ≈330 MHz: cap 1 ≤ 330 MHz · cap 2 ≤ 660 MHz · cap 4 ≤ 1320 MHz. Examples (RB2): 3840×1536p50 (308 MHz) = cap 1 but p60 (372 MHz) = cap 2; 4K@30 = cap 1; 4K@120 = cap 4. With FULL blanking 4K60 (≈695 MHz) becomes cap 4.
  • Connectors follow the system rate ("Sys") and the default blanking ("Def") unless overridden per connector.
  • Capacity-4 inputs: only on connector 1 or 3 of a 4-connector card; the following connector is disabled.
  • Groups (input or output): consecutive connectors on the same card, max 4, all the same format. Grouping does not change the capacity of the individual connectors; the group bandwidth is simply the sum. Capacity-4 signals cannot be grouped.
  • Each VPU (video processing card) has 8 source links (mixing-layer capacity) and 8 output links (program-output capacity). Per VPU: 8×cap1 / 4×cap2 / 2×cap4 mixing layers, and the same for PGM outputs.
  • Cut & Fill doubles the layer resource; C&F is not available on capacity-4 layers. Split layers come in pairs: one pair costs the same as a mixing layer of equal capacity and gives 2 non-mixing (cut-only) layers.
  • Screen regions (manual §6.3.7): a region is a set of outputs of the screen; one output belongs to exactly one region (default: 1 region with all outputs). Layers can be full-screen or restricted to a region.
  • Layer cost: a full-screen layer is instantiated on every VPU the screen spans (capacity × number of VPUs); a region-restricted layer only consumes source links on the VPU(s) rendering that region. That is why regions free up layer capacity on large screens.
  • Scaling-engine boundary (§5.5.4): a layer spread over more than 4 output links costs ceil(links/4) × its capacity in layer links on that VPU. Optimized mode (§5.5.6) removes this boundary for the whole VPU when a screen uses ≥5 output links and has a capacity-2 layer.
  • Regions: consecutive outputs on the same card, max capacity 4 per region — a larger group/screen is split into multiple regions over multiple VPUs.
  • Screens are spread over VPUs to maximise the layer pool per screen ("ideal spread"); if that doesn't fit, the calculator falls back to densest packing. VPUs join automatically for screens larger than one VPU.
  • Dedicated multiviewer outputs are separate and cost no VPU resources; they are not part of this calculation. Values assume 4:4:4 10-bit.