board
OCuLink Riser for the Supermicro AIOM Slot
Public Unreviewedby John Lauer
Passive OCP NIC 3.0 SFF card that breaks a Supermicro AIOM slot's PCIe x16 into four OCuLink 4i ports (4 x4, Gen4): 6-layer, 85 ohm, length-matched, DRC-clean KiCad project with Gerbers, BOM and CPL.
← Commit history
docs/clips.md+14−1@@ -3,7 +3,7 @@ Every step of the adom-aiflow run records its own clip. This page is rebuilt and pushed as each clip stops, so it can stay open in a browser tab while the build runs. -Last updated 2026-10-13 02:28 UTC. 7 clip(s).+Last updated 2026-10-13 02:32 UTC. 8 clip(s). | # | step | visit | recorded | links | state | |---|---|---|---|---|---|@@ -14,6 +14,7 @@ Last updated 2026-10-13 02:28 UTC. 7 clip(s). | 5 | routing | 1 | 2026-10-13 02:15 UTC | [clip](#clip-5) | suspect | | 6 | placement | 2 (back) | 2026-10-13 02:21 UTC | [clip](#clip-6) | suspect | | 7 | routing | 2 | 2026-10-13 02:25 UTC | [clip](#clip-7) | suspect |+| 8 | placement | 3 (back) | 2026-10-13 02:28 UTC | [clip](#clip-8) | suspect | ### Clip 1: libraries @@ -99,3 +100,15 @@ Step `routing`, visit 2. Recorded 2026-10-13 02:25 UTC, 305 min after the prompt Raw recording: `window-854470-20261013-022518.mp4`, 224 MB, kept with the run (not uploaded). +### Clip 8: placement++<a id="clip-8"></a>++Step `placement`, visit 3. Recorded 2026-10-13 02:28 UTC, 308 min after the prompt, 3.9 min long. Back because: the start board carried pre-trim copper of the PERST# buses; rebuilt from the final routes so the live plan and the start board split cleanly.++> suspect: no frames++> no 10x cut++Raw recording: `window-854470-20261013-022812.mp4`, 308 MB, kept with the run (not uploaded).+