No Freerun on LFO - Surge XT

[hellerdejanuar]

Bug Description:
Voice LFO Freerun doesn't behave as (I) expected. It seems to work almost like retrig, making it impossible to keep it phase synched with tempo.

Surge Version
Surge XT 0.9.nightly.7323973

• Version:
• Plugin Type: VST3
• Bitness: 64-bit

Reproduction Steps:
Steps to reproduce the behavior:

1. Simple patch with LFO (not SLFO) to cutoff. Optionally set rate to Sync.
2. Select Freerun
3. Play two notes spaced half a LFO cycle
4. The two notes have now different phase. And none is synched to session BPM

Expected Behavior:
In Surge (no XT) it does work as intended.

Computer Information (please complete the following information):

• OS: Macos Mojave
• Host: Ableton
• Version: 11.0.2

[baconpaul]

Ha so, the freerun works as expected if you run transport, but doesn't if transport is paused. (at least in LPX)

Believe it or not, that's a return of #146

Just in JUCE. Note that that bug was one I fixed almost 3 years ago. Funny!

If there's no transport we don't advance ppqpos artificially, and freerun VLFOs use song position.

I'll tag this into the milestone and apply the fix. Thank you for the excellent bug report.

[hellerdejanuar]

forgot to mention it also doesn't work with the transport running. El mar., 30 de noviembre de 2021 13:04, Paul ***@***.***> escribió:

…

Ha so, the freerun works as expected if you run transport, but doesn't if transport is paused. (at least in LPX) Believe it or not, that's a return of #146 <#146> Just in JUCE. Note that that bug was one I fixed almost 3 years ago. Funny! If there's no transport we don't advance ppqpos artificially, and freerun VLFOs use song position. I'll tag this into the milestone and apply the fix. Thank you for the excellent bug report. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#5534 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ARRTCWACDAD6LWU3ENR4OYDUOTY2LANCNFSM5JCFJY5Q> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.

[baconpaul]

huh so then live is different than logic, or my test was bad. (Your nightly is also quite a bit older than mine but that's probably OK - I don't think this code has changed much)

I'll dig. Something odd is definitely afoot. Very much appreciate the bug report.

[baconpaul]

Anyway I found it! Turns out we had an internal counter update which was placed in a block from when we looped by block, not sample, in juce; so it's just moving one line. The reason transport made it 'better' for me in logic was because my buffer size (64) resets the ppqpos really frequently so the drift was not that bad although now i retest it, not perfect; but with long buffers the result would be worse

PR in and this issue will auto close when we merge it, and an hour or so after that there will e a new nightly with the fix.

Thanks for the bug report. Excellent and long standing bug in XT we hadn't identified. Really appreciated.