Jump to content

Percolating synth timing differs depending on how I start playback


exmachina

Recommended Posts

I cannot figure this out -- I'm producing a friend's tune that's built around a percolating synth track where the starting point in the percolation appears to shift depending on how Logic playback is initiated:

 

1. It starts way off sync when the song loops back to the beginning in cycle mode

2. If I start playback using the space bar the percolation/sequence starts a little ahead of the beat

3. BUT it plays as intended if I first hit the the space bar and then the Return key to re-initiate playback

 

This sync problem does not appear in my friend's source Logic project. Well, actually it does appear, but only when cycle mode is on and the song loops back to begin again.

 

Exporting from my project file results in an off sync performance akin to the way it sounds when I start playback using the spacebar. Exports fine in the original project file. Similarly, freezing the percolating track doesn't help because it appears to follow the performance I get when starting playback by spacebar.

 

Percolating synth is an altered Alchemy patch called Bright Majestic Waves, if that's a helpful clue.

 

I cannot upgrade to 10.7 from 10.6.3 because I'm stuck on Catalina until the M1 iMac refresh happens and all my plugins are supported (I need my 27" form factor). I've added automation, processing, a few additional synth tracks, but my processor doesn't appear to be overloaded. It doesn't fall out of sync -- it either starts in sync or it doesn't.

 

Thanks in advance for any tips or suggestions. Had I any hair left I would be pulling it out.

Edited by exmachina
Link to comment
Share on other sites

So it's the arpeggiator. While its speed is synced to the song tempo it is otherwise free running.

Affected are both Source B and D.

72990555_Bildschirmfoto2021-10-24um11_00_06.jpg.8a2baeed0e5bca28888fa782d836f114.jpg

 

Set those for both Sources to "On" and see if the results are getting more consistent. There might be additional LFO modulations that also affect the sound (Source B at least has also filter modulation).

Link to comment
Share on other sites

Thanks, gacki, I really appreciate your looking into this.

 

Setting both B and D sources to Sync On change the sound but not the start timing issue, which still behaves in the manner described in my original post.

 

It's very perplexing -- why would a sequence be in the pocket when I hit return on a playing project, but not when I initiate playback by hitting the space bar or play btn?

Link to comment
Share on other sites

Can't say what it is, but as I said the sequence sounds different even when played manually. So I would expect to sound different during playback as well.

Because the arpeggiator pattern is very simple there are very few possible ways it will come out, even when it is misplaced. So the behavior might be coincidental.

Perhaps you could attach the original Alchemy setting used in the song?

Link to comment
Share on other sites

Can't say what it is, but as I said the sequence sounds different even when played manually. So I would expect to sound different during playback as well.

Because the arpeggiator pattern is very simple there are very few possible ways it will come out, even when it is misplaced. So the behavior might be coincidental.

Perhaps you could attach the original Alchemy setting used in the song?

 

Further, why would it play consistently in my friend's original project, but not my version? I'm using the same patch, which I'll try to attach to this post.

Bright Majestic Waves - Roll.acp.zip

Link to comment
Share on other sites

OK, so this is completely different from the original "Bright Majestic Waves" which was deriving its "sequency" sound mainly from a downward arpeggio in Source B.

Here it seems to come mainly from the "sequencer" modulator in Source D instead which modulates coarse tune. (This element is also present in the original preset but is mostly overshadowed by Source B there). The problem is similar though: the sequencer is free running so playing a note will start at different points within the sequence.

I'll have to look into this again in more detail; this might take some time.

Link to comment
Share on other sites

So here are a few observations. Whether this helps or not I don't know, but it's kinda interesting for myself. I'm not presenting solutions (yet). Maybe I'll look into the manual later on but I always find it interesting to analyze the status quo first and draw my conclusions from there (which are usually false, but hey...).

 

So the "bubbling" comes mainly from the Sequencer modulator for Source D which modulates Coarse Tune.

There's also the arpeggiator involved but only to re-trigger the AHDSR1 envelope. I think we can discount this for now. There's also a whole bunch of effects involved; we can turn those off for our test purpose.

 

Here's the "naked" Source D only preset:

BMW - Roll Source D naked.acp.zip

Now let's observe the behavior of the sequencer (and the arpeggiator) in general. This is somewhat weird IMO.

We load Logic (this is important: Logic must not run already!), create a new project with an Alchemy track, load that preset, go to Advance mode and observe the Sequencer and Arpeggiator. Both are NOT running.

Once we hit Play or play a note the Sequencer and Arpeggiator will start. The odd thing for me is: They never stop. Never. I can close that particular song, create a new one from scratch again with the Alchemy preset - and the Sequencer and Arpeggiator will already run. The only thing you can do to stop them is to quit Logic.

Kinda strange, I think.

 

Now, how does the sequencer sync itself to Logic? This is getting a bit longer.

When Logic is stopped it is running free in the song tempo. Any notes you play will end up "somewhere" within that sequence; it will not reset with a newly played note.

The polyphony for Source D is set to "1", so it's actually monophonic. If we raise that value we can play several notes at once - but those notes will "play" the sequence always in sync even if those notes are played slightly apart timing-wise; the sequence does not affect single notes but Source D as a whole. This actually makes sense since the Sequencer modulates Coarse Tuning; and if you turn Coarse tuning manually it will also affect all sounding notes the same.

 

Here's a 4 voice polyphonic Source D preset:

BMW - Roll Source D poly.acp.zip

When Logic is running things get a bit more interesting.

The Sequencer is reset every time "play" is executed - either by hitting the button or using one of the key commands. The Sequencer does NOT care at all whether the play position is on a downbeat or not!

So yes: If we start Logic from anything but a downbeat the Sequence and hence the "bubbling" will be out of sync. This will immediately become a problem once we rapidly jump around during editing and the Song Position Pointer may have landed anywhere timing-wise.

What happens in Cycle mode?

The Sequencer is NOT reset during repeats in Cycle mode. It is reset when "play" is executed (as described above). So if the Cycle starts on a downbeat and its length is a multiple of the Sequencer length things should be ok. In all other scenarios: not ok.

 

So in theory one solution would be to always start either from the beginning of the song or from a multiple of the Sequencer length (which in 4/4 would be the downbeat). When using Cycle the start and length must be observed.

 

I think I'll stop here for the moment. Whether the Arpeggiator also plays a role (and which one...) remains to be seen.

Link to comment
Share on other sites

Thanks again, gacki, this is really great analysis.

 

I've confirmed that the cycle length requirement is indeed the reason why the sequence goes way out of sync after looping back to the beginning. Good work -- one mystery down.

 

We can make the sequencer and arpeggiator restart with every keyDown by enabling Trigger mode on the Sequencer and turning on sync for the Arpegiator. That doesn't help me because in my buddy's patch there's something else going on that causes the sequencer to loop over the first few notes until a keyUp event is seen.

 

HOWEVER, in testing your patch along with a copy of the original percolating region in a new project I could no longer see the problem -- now the correct rhythm is maintained when I use the space bar/play btn and no longer changes when I hit the Return key to restart the song.

 

Remembering that the original project didn't exhibit the problem, I decided to copy all my tracks into a new project and everything works -- there is something corrupt in my previous work project file that was causing the timing variance!

 

I'll need to set up some bus configurations to recreate my work mix, but unless adding these missing elements retriggers the issue I think we can consider the mystery solved.

 

Thanks again for all your help on this, gacki, I really appreciate the effort you put into it and I learned quite a bit from your analysis. Alchemy is a deep, dark beast and the spotlight you shone on it illuminated some of that mystery. Cheers!

Link to comment
Share on other sites

Meh, I may have spoken too soon -- while things seemed good with my copied tracks, after recreating my subs and copying over channel settings things went south again, which seems to suggest a resource starvation problem though the Performance Meter seems within bounds. ¯\_(ツ)_/¯

 

I reset the channels and recreated the settings one at a time and I think there was an improvement, but honestly I've been at this so long today I'm at the point where I'm second guessing myself and need to rest and reassess.

 

Maybe what I need to do is recreate that percolating synth part on a physical synth so I can rule out performance issues? Or at least redo the modulation approach BMW uses in a more traditional manner? Should probably first figure out how to make the sequencer sync work for me. Ugh. Well, I do need to gain a better grip on Alchemy.

 

In the meantime I've been able to capture the mix using Soundflower, so I'm not dead in the water, but it sure would be nice to figure this out.

Link to comment
Share on other sites

Sadly, freezing, bouncing or exporting as audio and reimporting the problem track doesn't solve the issue.

 

Freezing and bouncing both have the same offset timing problem while exporting/reimporting almost gets me there, timing-wise, but introduces a weird note flam that's not in the Alchemy track. ¯\_(ツ)_/¯

 

Maybe I could use Soundflower to capture the track while soloed and then bring it in as an audio track.

Link to comment
Share on other sites

  • 3 months later...
Just to close the loop on this, I did go ahead and use Soundflower to rewrite Logic's audio out into QuickTime Player where I saved the resultant mix. I did have to comb through the QT-saved file to remove several very short audio gaps that presented as clicks, not sure if those were Soundflower artifacts or some other ghost in the machine thing, but I finally got a mix I could use. Thanks again to gacki for the help.
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...