andyjohnmm Posted April 18, 2021 Share Posted April 18, 2021 Frustrating stuff. Noticed my recordings were being displaced late weeks back. Did the "ping" test based on suggestions here and got a result of +58 samples. Set delay slider to -58. Now things seemed too early. Did an actual cable out of 1 to input 1 with a transient and got a similar result so I've left the slider as is. Seems my percussion and claps are all just slightly behind the beat though (Monitoring the input from SSL2+ interface and click up nice and loud - it's not my playing that is the issue). Can't seem to find a once and for all answer to getting the audio track laid down to be the actual performance. Could it have something to do with my buffer size always set to 1024 because I'm monitoring from the direct input and don't use the monitoring from the program? Thanks very much. Quote Link to comment Share on other sites More sharing options...
JakobP Posted April 18, 2021 Share Posted April 18, 2021 Buffer size probably is the cause, but in case you haven't, you should also set software monitoring to off. Quote Link to comment Share on other sites More sharing options...
triplets Posted April 18, 2021 Share Posted April 18, 2021 Could it have something to do with my buffer size always set to 1024 because I'm monitoring from the direct input and don't use the monitoring from the program? With such a high buffer, the recordings will always be late. Put the buffer at 32 and monitor from Logic and see how that works. And put the recording delay slider back to 0. Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 19, 2021 Author Share Posted April 19, 2021 Thanks to all replies. Being deeper into the session, with multiple software synths and many tracks, running the session with the buffer that low causes overload. Is there a way to get a definitive delay adjustment using my direct monitoring from the interface (which I'm most accustomed to) but use say, 512 as my buffer selection? Seemed the audio was actually ending up ahead of the beat, not late, as previously stated (even with the delay slider set to zero!). The Large session can still run at that 512 selection, though I'm unsure if my recordings are actually what I'm recording, if you will. They seem close, but how can I be sure they are dead on, using the direct monitoring from the SSL+? Very much appreciate your knowledge on this. I have been reading other posts etc., Though I'm still unclear on the matter, admittedly. All Quote Link to comment Share on other sites More sharing options...
triplets Posted April 19, 2021 Share Posted April 19, 2021 Have you tried Freezing tracks or bounce in place to reduce CPU strain? Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 19, 2021 Author Share Posted April 19, 2021 Yes, went through the "heavier" tracks and froze them. So that is in place, though, I'm still wondering if I say, set my buffer to 512, and do a test with running a signal out and back in - and delaying the recording with the slider one way or another to accommodate the discrepancy - will this be a final say in how I should have that slider set - and then using that buffer size from now on I'll know to leave it there and be comforted knowing my recordings are "true" and displayed/playing correctly down to the sample? All of that using the direct input monitoring on my interface? I'd like to be able to continue using it. Many thanks again. Quote Link to comment Share on other sites More sharing options...
triplets Posted April 19, 2021 Share Posted April 19, 2021 Did you turn off Software Monitoring in Logic? Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 19, 2021 Author Share Posted April 19, 2021 Will turn this off and attempt a test with 512 buffer size and using direct input monitoring. Thanks. Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 19, 2021 Author Share Posted April 19, 2021 With the 512 buffer size selected, I did a "ping" test with the i/o plugin and adjusted the delay slider accordingly (to -50 samples). Using the direct input monitoring from the interface, with Software Monitoring off, It seems the recording is still being placed early (clapping to the click). Same thing happens when I use Software monitoring and no direct input sound at all. Is it possible the recorded audio is still somehow being placed early? Quote Link to comment Share on other sites More sharing options...
triplets Posted April 19, 2021 Share Posted April 19, 2021 Did you try with 0 samples to see if it's off? Because Logic calculates the latency based on the interface driver and using Logic as monitor, so with direct monitoring the ping value might be irrelevant? Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 19, 2021 Author Share Posted April 19, 2021 Tried the slider at 0 samples. Click is blaring in the phones and my clap timing is fine - still seems all hits are consistently being placed ahead of the beat, early, all of em. Quote Link to comment Share on other sites More sharing options...
triplets Posted April 19, 2021 Share Posted April 19, 2021 Then it has to do with direct monitoring, but that's not Logic's fault I think. Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 19, 2021 Author Share Posted April 19, 2021 Okay - unsure how to proceed. Are there other tests or methods by which I can solve this issue? I still don't know if Logic is placing my audio correctly. Thank you. Quote Link to comment Share on other sites More sharing options...
triplets Posted April 19, 2021 Share Posted April 19, 2021 One thing you can try is to turn software monitoring back on in Logic. Use a lighter project, engage all your synths with buffer at 64 and see where Logic puts the recording. Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 21, 2021 Author Share Posted April 21, 2021 Okay. Will attempt that. Wondering, is there any instance where Logic will put a recording early - "ahead" of the grid - even when the delay slider is set to zero? Really focusing in on getting the claps on beat, with slider set to zero, though they still seem consistently placed early, just slightly. This is at 512 buffer setting. Quote Link to comment Share on other sites More sharing options...
triplets Posted April 21, 2021 Share Posted April 21, 2021 If you are monitoring thru Logic and not your interface, I seriously doubt that Logic will place a recording early with a buffer at 512. The physics don't add up for that. Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 23, 2021 Author Share Posted April 23, 2021 I see. Okay. Using direct input monitoring from the interface - is there a chance the audio recording would somehow end up being placed early that way more so than monitoring from logic? Also, do samples I created and am using within drum machine designer remain unaffected if changing the tempo the session? Thanks. Quote Link to comment Share on other sites More sharing options...
triplets Posted April 23, 2021 Share Posted April 23, 2021 It can happen. When you monitor thru Logic, the signal has to go thru Logic and back out. So there's always latency. Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 23, 2021 Author Share Posted April 23, 2021 I see, yes. Recent test, using 512 buffer size, with direct input monitoring from interface, the audio seems to be placed correctly. Thing is, if I do the ping test or a test using a recorded transient - it seems to suggest that I should be setting the delay slider to place the audio early. Are you saying this test is not applicable if I'm using direct monitoring? And the audio clip samples in drum machine designer - I can find information on regular track clips being affected when changing tempo of session, but what about the samples I've assigned to cells - do they stay at the original quality and pitch I originally recorded them at if moving the tempo around? Thanks again. Quote Link to comment Share on other sites More sharing options...
triplets Posted April 24, 2021 Share Posted April 24, 2021 The ping is based on buffer size and monitoring thru Logic. So if you can set Logic to place the file correctly with the slider and not change the buffer size, it should work. Quote Link to comment Share on other sites More sharing options...
andyjohnmm Posted April 29, 2021 Author Share Posted April 29, 2021 I see, yes, okay. Using the direct monitoring from the interface, or in effect "no" monitoring by just hearing what's happening in the room outside of the headphones, is it safe to say that file will be placed sample-correct using - in this example - a 512 buffer size? Quote Link to comment Share on other sites More sharing options...
triplets Posted April 29, 2021 Share Posted April 29, 2021 I see, yes, okay. Using the direct monitoring from the interface, or in effect "no" monitoring by just hearing what's happening in the room outside of the headphones, is it safe to say that file will be placed sample-correct using - in this example - a 512 buffer size? The only way to tell is by testing it and record with those settings. You can zoom in to the sample to see where the transient starts. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
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.