mirror of
https://github.com/allaunthefox/Research-Stack.git
synced 2026-08-20 14:07:29 +00:00
5ad2e7f8bb
Cross-referenced against our prover orchestration layers: - Plan-Execute-Verify-Replan ↔ L0-L3 pipeline - Agents as specialists ↔ 11-agent swarm - Guardrails ↔ ProverWatchdog - Sandbox testing ↔ Virtual FPGA tests - Trajectory-aware eval ↔ BFS audit trail 5 gaps identified, 4 strengths confirmed
14 lines
1.1 KiB
Text
14 lines
1.1 KiB
Text
created: 20260507000000000
|
|
modified: 20260507000000000
|
|
tags: ResearchStack Hardware HDMI GPU Compute
|
|
title: HDMI Compute Fabric
|
|
type: text/vnd.tiddlywiki
|
|
|
|
! HDMI Compute Fabric
|
|
|
|
The HDMI compute fabric (`5-Applications/scripts/hdmi_computational_shell.py`) turns display interfaces into general-purpose computation channels. Includes `score_route_damage` for damage scoring along routes, DSB cluster detection for broken invariant grouping, and the computational shell architecture that treats HDMI as a data bus. Integrates with [[FPGA Warden]] for hardware-level compute offload and the [[Virtual Display Blitter Compute Surface]] for blitter-based rendering computation. Part of the broader hardware-as-compute strategy alongside DisplayPort computational controller (`5-Applications/scripts/displayport_computational.py`), voltage computational substrate, and PCIe computational controller. This strategy treats every physical interface as a computational resource.
|
|
|
|
* [[FPGA Warden]]
|
|
* [[Virtual Display Blitter Compute Surface]]
|
|
* [[ASIC Topology]]
|
|
* [[Motherboard Computational Substrate]]
|