8000 SPU2: step stopped voices without the output math by inada-s-ai[bot] · Pull Request #610 · inada-s/pcsx2 · GitHub
[go: up one dir, main page]

Skip to content

SPU2: step stopped voices without the output math - #610

Merged
inada-s-ai[bot] merged 1 commit into
zdxsv-masterfrom
ai/spu2-stopped-voice
Nov 16, 2026
Merged

inada-s-ai[bot] merged 1 commit into
zdxsv-masterfrom
ai/spu2-stopped-voice

Conversation

@inada-s-ai
@inada-s-ai inada-s-ai Bot commented Nov 16, 2026
Copy link
Copy Markdown

Description of Changes

SPU2 MixCoreVoices: a stopped voice runs a dedicated step (volume slide, decode position/NAX, block header, pitch, consume, voice 1/3 write-back) and skips the output math and the dry/wet accumulation of a silent sample.

Rationale behind Changes

In GGPO battles most of the 48 voices are stopped but still stepped every sample. Same state changes as before (same calls in the same order); only computations whose result is zero are skipped. Part of speeding up rollback frames (#606).

Suggested Testing Steps

  • Verified (private rig, windows-mini, headless replay bench, 2 runs per side against a clean build of the base): rerun frames about 2% faster; save/load unchanged; bench completed all rollbacks.
  • Not run: windowed play with audio, macOS/Linux builds (CI).

Did you use AI to help find, test, or implement this issue or feature?

Yes: Claude Code profiled the bench, wrote the change and ran the A/B.

🤖 Generated with Claude Code

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@inada-s-ai
inada-s-ai Bot enabled auto-merge November 16, 2026 12:57
@inada-s-ai
inada-s-ai Bot merged commit 7a93768 into zdxsv-master Nov 16, 2026
1 check passed
@inada-s-ai
inada-s-ai Bot deleted the ai/spu2-stopped-voice branch November 16, 2026 13:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants

0