SynthstromAudible / DelugeFirmware

https://synthstromaudible.github.io/DelugeFirmware/
GNU General Public License v3.0
552 stars 87 forks source link

Performance/improve render caching, fix unitialized variable use #2200

Closed m-m-adams closed 19 hours ago

m-m-adams commented 6 days ago

Significant improvement in voices by tagging likely/unlikely branches. Can likely improve a lot more by caching some repeated checks.

Also removes rendering directly to the output buffer - the extra time to check and set up for that is worse than an extra memcpy

Fixes a bug where stereo wavefolding used the non-anti aliased fold function

same before as here - #2199

Synth after:

image

Multisample after:

image
github-actions[bot] commented 6 days ago

Test Results

56 tests  +8   56 :white_check_mark: +8   0s :stopwatch: ±0s  9 suites ±0    0 :zzz: ±0   9 files   ±0    0 :x: ±0 

Results for commit c183d8bd. ± Comparison against base commit 99281139.

:recycle: This comment has been updated with latest results.