You're quite welcome Jeffrey. I hope it helps you out. That whole drifting thing can really be annoying and when you've been a drummer first as your weapon of choice and have paid close attention to your meter, this can really drive you crazy. I've honestly never considered myself anything special as a musician and I mean that...but man, one thing I got is timing. Many times I found myself beating myself senseless with Sonar while talking to myself (that's just only child syndrome lol goes with the territory) really wondering if it was me.
I'd sit there with my guitar just strumming in time to my click to see if the strums were landing right on the beat...and I was positive they were, yet the grid showed me different. That's when I knew something wasn't quite right. When doing this with other DAWs, it showed me that I was actually as close to spot on as a human could be with the majority of my strums being dead on while the ones that weren't were so close to on, they weren't worth fixing as no one can really be a metronome. But the results I was seeing in Sonar were scaring me.
I bought a new soundcard one day and it helped leaps and bounds of course, but there were still a few things that just never made sense to me. I either left them alone or tried to compensate if something bothered me by purposely playing something a little early. Or sometimes, and I know some of us have HAD to experience this....but I'd record something in Sonar and that particular take would be WAY off. I'm talking like so off you had no choice but to redo the whole track. This anomaly would happen to me about every 100 times I'd press record and was in every version of Sonar I have ever used. This would happen if you pressed play and then pressed record while playback was already rolling. Certain times when doing this, it would be so out of time it was like I was recording from another country through a 56k modem. LOL!
But upon learning of this clock sync thing from Eric, none of that happens to me anymore and all my tracks are now spot on when I record them. The only thing that sucks about this method is I must use a few different clock sync numbers per situation. I'll explain as it may be helpful/useful t others that may be in my situation.
I use consoles here for all my stuff. So, this means that input monitoring is not needed on my behalf since everything I do plays in real time into my mixer and my soundcard input monitoring is disabled as it's just not needed. This in turn allows me to not have to touch my ASIO buffers and get down super low like everyone else has to when they record. I can leave my buffers set to 2048 or 4096 while recording. The problem here is, if you do the sync test at THOSE buffer settings, you get a different "nudge number".
So I have sync numbers for 32, 64, 128, 2048 and on one of my older cards, 4096. Sometimes due to the project we may be working on, 32 or 64 buffers can't be used on my end if I've already loaded up a bunch of things that are cpu intensive and I find out I need to re-record something that forces me to lower my ASIO buffers. I normally don't have to do this for audio, but if I'm recording using a guitar sim, I can't leave my buffers at 2048 or the latency is too insane as you know. So for that, I do a buffer change and may end up at 128 for a loaded up project. If the project isn't loaded up with stuff, I'll use 32 or 64. Each one of those buffer settings will give you new nudge numbers, so I have them all written down and just change when I need to. It's rare for me to use anything other than 64 or 2048 though, but I have the others covered should I need them.
So if you (or anyone else) finds themselves in this situation, it's not a bad idea to do a few sync procedures using different buffer sizes if you ever need them. I hope this stuff helps you out...make sure you post a message back and let us know if this helped you or not. Good luck with the Sonar re-install too! :)
-Danny