jasonwagner Posted October 1, 2015 Share Posted October 1, 2015 Apogee said Logic changes it's latency based on CPU usage. I called them to ask about this and they said all I can do is keep repining for all my I/O. Does anyone else use the I/O plugin and find that the ping value can be different at times? Mine will go between -7 and +49-52. If one re-pings at a different value, for example -7 to +49, all of them will change to +49-52 when I reping, so the latency seems to be session wide. I say 49-52 because it varies from track to track. Can anyone confirm what Apogee said is true and is there a way to resolve this besides just re-pinging? Thank you! Quote Link to comment Share on other sites More sharing options...
Eric Cardenas Posted October 1, 2015 Share Posted October 1, 2015 Apogee said Logic changes it's latency based on CPU usage. This is not true. They probably meant that latency change depending on the amount of latency inducing plug-ins. Quote Link to comment Share on other sites More sharing options...
jasonwagner Posted October 1, 2015 Author Share Posted October 1, 2015 Thank you, was just reading this thread. Going to try and unplug everything. To be fair to apogee it was a chat support so I might have misstated the issue. viewtopic.php?f=1&t=109603 Quote Link to comment Share on other sites More sharing options...
jasonwagner Posted October 1, 2015 Author Share Posted October 1, 2015 Restarted it and unplugged replugged all the digital cables, it didn't change this time. Hopefully it stays put. If it happens again, I'll try what you said in that other thread. "What you can test next time is to go into Preferences > Audio > Devices. Uncheck the Audio Engine box and apply. Re-check it and apply and that should take care of it." Would you say that this issue is definitely not supposed to be happening? Quote Link to comment Share on other sites More sharing options...
Eric Cardenas Posted October 1, 2015 Share Posted October 1, 2015 It should definitely not happen just because of a change in CPU load. Changing sample rate or I/O buffer size will of course require re-pinging. Quote Link to comment Share on other sites More sharing options...
The Beatsmith Posted October 1, 2015 Share Posted October 1, 2015 I believe it can happen for sure with firewire, and possibly USB. it doesn't happen with my PCIe SSL card, it's always 80 samples, although thinking about it, a couple of times i've seen it at 81 samples, but it's very rare Quote Link to comment Share on other sites More sharing options...
stardustmedia Posted October 2, 2015 Share Posted October 2, 2015 That does happen with some of my hardware. But only with my effects that do some kind of time modulation and only when I forget to set the effect to 100% dry (no effect) for pinging Otherwise, the ping is always stable, as long as I don't change/add plugins to the session, change buffer, etc. I use Apogee converters (Symphony, DA16X) thru Thunderbolt Bridge. So pretty similar to your system. What hardware do you connect with tie I/O-plugin? Quote Link to comment Share on other sites More sharing options...
Eric Cardenas Posted October 2, 2015 Share Posted October 2, 2015 That does happen with some of my hardware. So you mean that you are getting different ping values on different tracks due to different CPU loads? Quote Link to comment Share on other sites More sharing options...
stardustmedia Posted October 2, 2015 Share Posted October 2, 2015 That does happen with some of my hardware. So you mean that you are getting different ping values on different tracks due to different CPU loads? Nope, sorry for the confusion. It was because the used effect hardware has some time modulation going on. I get with every ping different values. Was pretty confused the first time I tried the IO, until I realized (duh!) I should measure the ping "dry" Quote Link to comment Share on other sites More sharing options...
Eric Cardenas Posted October 2, 2015 Share Posted October 2, 2015 Nope. Because that effect hardware has some time modulation going on. I get with every ping different values. Ah, that explains it. The issue I'm having is with statements like this: Apogee said Logic changes it's latency based on CPU usage. This is just nonsense. Quote Link to comment Share on other sites More sharing options...
stardustmedia Posted October 2, 2015 Share Posted October 2, 2015 Definitely nonsense. But the more plugins you load the more/different latency you get and at the same time more CPU usage. There is some kind of indirect truth in it. Nevertheless the statement is crap Quote Link to comment Share on other sites More sharing options...
Eric Cardenas Posted October 2, 2015 Share Posted October 2, 2015 Correlation vs causality. The two things are correlated but high CPU is not the cause. Quote Link to comment Share on other sites More sharing options...
stardustmedia Posted October 2, 2015 Share Posted October 2, 2015 Correlation vs causality. The two things are correlated but high CPU is not the cause. Voilà You nailed it Quote Link to comment Share on other sites More sharing options...
Eric Cardenas Posted October 2, 2015 Share Posted October 2, 2015 Quote Link to comment Share on other sites More sharing options...
jasonwagner Posted October 2, 2015 Author Share Posted October 2, 2015 That does happen with some of my hardware. But only with my effects that do some kind of time modulation and only when I forget to set the effect to 100% dry (no effect) for pinging Otherwise, the ping is always stable, as long as I don't change/add plugins to the session, change buffer, etc. I use Apogee converters (Symphony, DA16X) thru Thunderbolt Bridge. So pretty similar to your system. What hardware do you connect with tie I/O-plugin? Thanks for the replies everyone. I'm using just EQ and Compressors for the I/O, no reverbs or effects. Using a PCI-e card with the symphony. It was like this even before I added the DA/AD16xs, so just the symphony and the modules in the chassis going right to the PCI-e card. That should eliminate the DA/AD16xs as a culprit. Is a negative number even possible, like the -7 I get? Maybe I should leave them at the +49-52 range once I get it to ping there. I'm 90% sure I don't notice a change in ping value from the 49-52 range even if I change the i/o buffer, does that seem weird? Last night I didn't see it change but it was a consistent session until we stopped, so not the normal situation where I see the change. I'll check again this evening and report back, It's not a huge deal to re-ping but I'm just concerned I forget and if this is not supposed to be happening it would be nice to resolve it. Would the PCI slot in the computer I use make a difference? Can hardware cause any latency, even a sample or two that might explain the 49-52 range? The ping function seems really straight ahead, as in it's not requesting the apogee stuff to report anything, it's just calculating the time it takes to ping back right? So if that is right, it means the apogee is actually changing to be more or less quick the in the I/O, as opposed to reporting bad information. Last week I loaded a SSD for my system drive, started from scratch so the system drive is really fresh and bare bones as far as software goes at the moment. It was doing this before the new hard drive. Quote Link to comment Share on other sites More sharing options...
stardustmedia Posted October 2, 2015 Share Posted October 2, 2015 Hard disc doesn't influence the ping. What happens when you connect the DA-out directly into the AD-out and ping again? Quote Link to comment Share on other sites More sharing options...
jasonwagner Posted October 2, 2015 Author Share Posted October 2, 2015 Hard disc doesn't influence the ping. What happens when you connect the DA-out directly into the AD-out and ping again? I'll check that this evening. I mentioned the hard disk as it seems to eliminate the possibility of some software issue since it's a clean, stripped down install. Quote Link to comment Share on other sites More sharing options...
jasonwagner Posted October 3, 2015 Author Share Posted October 3, 2015 It was the 49-52 range when I booted up. I'll update if it occurs again and I find a consistent resolution. Quote Link to comment Share on other sites More sharing options...
RedBaron Posted October 3, 2015 Share Posted October 3, 2015 I'm using just EQ and Compressors for the I/O ... Are all of those compressors/EQs fully analogue? Any outboard that's digital will have AD+DA conversion latency at least, and will probably have some DSP processing latency too. Also, it's possible to have negative latency if you're connected to a digital processor using SPDIF because there isn't any conversion latency - if you do an SPDIF or ADAT loopback you'll almost certainly get a negative Ping. Core Audio drivers even have the ability to publish different latency figures for the SPDIF and other streams, but I don't think any of the audio drivers bother to implement the feature. It wouldn't make any difference anyway because Logic only has one individual Recording Delay preference. Quote Link to comment Share on other sites More sharing options...
jasonwagner Posted October 13, 2015 Author Share Posted October 13, 2015 Sorry I missed your reply Redbaron. All are analog, my I/O in mixing all runs through AD16x and DA16x, clocked to the symphony. Thank you for the run down on that! It seems I have found the issue, it's when the computer goes to sleep, which I have set at three hours but I forgot to shut off last light and just loaded up a session and pinged, it was off again. I tried unchecking and applying core audio in preferences but that didn't do the trick. It took a shut down of the computer and symphony to get back to normal. Now I know. Thanks for the help folks. Quote Link to comment Share on other sites More sharing options...
Eric Cardenas Posted October 14, 2015 Share Posted October 14, 2015 Thanks for getting back to us with your new findings. Setting the computer to sleep can be problematic when using PCI audio interfaces. Set the screens to sleep instead. 1 Quote Link to comment Share on other sites More sharing options...
Zachchapel Posted March 4 Share Posted March 4 On 10/14/2015 at 1:24 AM, Eric Cardenas said: Thanks for getting back to us with your new findings. Setting the computer to sleep can be problematic when using PCI audio interfaces. Set the screens to sleep instead. I know this is a super old post...but I'm running into issues as well. I used to just run a Focusrite Red8 with my setup but then I expanded with Native HD Thunderbolt (which I assume is just the PCI in a box) with 2 additional HD/IO's and the Red8 since it is protools compatible. However, my ping for hardware inserts in-and-out (with nothing in the chain, just from one out of a converter into one of it's inputs) always changes as well. So if I'm understanding what you're saying, putting the computer to sleep can change this due to the fact that I'd a PCI-based system? It would make sense because I never had to worry about it when I was just using the Red 8. It's pretty big differences, too...like right now it's +304 but before that it was -102...it's never consistent. Anyways, I appreciate any and all help. 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.