docs: tested configurations: RX 5600 XT 6 GB (RDNA1, no mesh shaders), Windows 11, main `de866017` #2715

Open

github-sync wants to merge 1 commit from github/pr/2715 into main

github-sync commented

2026-11-12 09:31:53 +00:00

Owner

@Waryxxful opened this pull request on GitHub · 2026-11-12 08:58 UTC · original

Why. The page has no RX 5000 row yet, and RDNA1 is one of the cards without mesh shaders (#1001) that runs the compute fallback. These are the first measured runs on one, made from the same save each time so they can be compared later.

Two rows:

Main de866017: Kokiri Forest and Hyrule Field over four runs, the first start's preparation times on this card (shader programs 120 s, predicted pipelines 3,603 s, noted pipelines 2,355 s; later starts 11 s), and VRAM at the driver's allowance.
One switch a run: RX_VK_MIRROR=0, RX_VK_PARALLEL=3, RX_VK_PRECOMPILE=0.

All runs were scripted through the control bridge: teleport, 15 s to settle, 60 s standing, from a copied save restored before each run. The fps are the medians of the log's N presents lines.

Docs only; no checks read this page.

🤖 Generated with Claude Code

> [@Waryxxful](https://github.com/Waryxxful) opened this pull request on GitHub · 2026-11-12 08:58 UTC · [original](https://github.com/NotRexed/Hookshot/pull/2715) **Why.** The page has no RX 5000 row yet, and RDNA1 is one of the cards without mesh shaders (#1001) that runs the compute fallback. These are the first measured runs on one, made from the same save each time so they can be compared later. Two rows: - **Main `de866017`:** Kokiri Forest and Hyrule Field over four runs, the first start's preparation times on this card (shader programs 120 s, predicted pipelines 3,603 s, noted pipelines 2,355 s; later starts 11 s), and VRAM at the driver's allowance. - **One switch a run:** `RX_VK_MIRROR=0`, `RX_VK_PARALLEL=3`, `RX_VK_PRECOMPILE=0`. All runs were scripted through the control bridge: teleport, 15 s to settle, 60 s standing, from a copied save restored before each run. The fps are the medians of the log's `N presents` lines. Docs only; no checks read this page. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

github-sync added 1 commit

2026-11-12 09:31:53 +00:00

docs: tested configurations: RX 5600 XT 6 GB (RDNA1, no mesh shaders), Windows 11, main de866017 c7f199410b

The first RX 5000 rows: Kokiri Forest and Hyrule Field over four runs from the same save, the
first start's preparation times on this card, and three switches one at a time.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

github-sync referenced this pull request

2026-11-12 14:41:09 +00:00

fix(vk): stop requiring vertex strides for mesh fallback draws #2796

github-sync commented

2026-11-12 14:44:35 +00:00

Author

Owner

@Joey-Magnus approved on GitHub · 2026-11-12 14:34 UTC · original

I merged this onto current main (51da0464) and it applies cleanly at the top of the Runs table. The rows follow the page's format: build string from the log, driver, settings, place, and fps from the "N presents" lines. They fill a real gap, the first RX 5000 card on the compute mesh fallback. No tests read this page.

Two findings in the rows deserve their own issues rather than staying only in a table. First, a first start of about 1.6 hours of pipeline preparation on this card (3,603 s predicted plus 2,355 s noted). Second, RX_VK_MIRROR=0 being 3-5 fps faster than the default here, and RX_VK_PRECOMPILE=0 cutting private memory from 27 GB to 20 GB. The 27 GB alone is worth a look on a 24 GB machine.

#2757, #2771 and #2778 also add rows at the top of the same table, so whichever lands second needs a trivial rebase.

> [@Joey-Magnus](https://github.com/Joey-Magnus) approved on GitHub · 2026-11-12 14:34 UTC · [original](https://github.com/NotRexed/Hookshot/pull/2715#pullrequestreview-5876238355) I merged this onto current main (51da0464) and it applies cleanly at the top of the Runs table. The rows follow the page's format: build string from the log, driver, settings, place, and fps from the "N presents" lines. They fill a real gap, the first RX 5000 card on the compute mesh fallback. No tests read this page. Two findings in the rows deserve their own issues rather than staying only in a table. First, a first start of about 1.6 hours of pipeline preparation on this card (3,603 s predicted plus 2,355 s noted). Second, RX_VK_MIRROR=0 being 3-5 fps faster than the default here, and RX_VK_PRECOMPILE=0 cutting private memory from 27 GB to 20 GB. The 27 GB alone is worth a look on a 24 GB machine. #2757, #2771 and #2778 also add rows at the top of the same table, so whichever lands second needs a trivial rebase.

github-sync referenced this pull request from a commit

2026-11-12 19:00:55 +00:00

fix(vk): stop requiring vertex strides for mesh fallback draws (#2796)

Commenting is not possible because the repository is archived.

No reviewers