Well, I have two fireface 800’s. There is no buffer size in the control panel. Where does this param live?
Not trying to be snarky, but it gets tiring receiving peoples two cents. This issue needs repro cases, not conjecture. If you think there is a better way to repro this please give it a try.
Feel free to try my repro steps for yourself and post your results.
And again, there is no need to go outside the DAW to change buffer sizes. Nuendo should never loose sample accuracy no matter what i do.
If you click the little hammer thing on the Taskbar then when the control panel pops up it has all the different possible sample buffer sizes. It is the same as the one that pops up when Nuendo is open. When I’m in the studio tomorrow I’ll take a screenshot and post it here.
I think that you have already proved that there is a bug. However, it is not yet clear where the fault lies and if there is a suitable workaround in order to avoid it.
I’ll have a go tomorrow.
Says you…! I still maintain that the best way to change buffer size is to close the project, change the buffer size and then open the project. Some other applications force you to re-start the whole application every time. Whilst I don’t disagree that it would be much better if you could just change the buffer size on the fly, I have always run into problems with big mixes by doing this, so my fix is just not to do it. I never close Nuendo though, just the project.
So then how do I do it outside Nuendo? RME setup does not have that feature!?
Besides your comments are still only conjecture, a guess. You have no proof thy this will solve the issue yet you recommend this. I’ve even stated this issue survives DAW restart and system reboot. So it is very likely, even if I were able to change the buffer size outside Nuendo that that would fix the issue.
Where in the Nuendo manual does it say I would need to do what you are suggesting anyway?
In the fact that you are not using the driver of your RME’s. You are using the build-in audio card of your computer.
Secondly, I see that you are using stereo audio files on a mono channel.
Not trying to prove that your repro is flawed, but it sure adds a ton of variables, and it might explain why no-one else can reproduce your problem.
Changing buffers outside Nuendo:
I’m on Mac, I see no way to change RME buffer size outside Nuendo. If it’s possible let me know.
Fredo, I know I’m using internal sound card for this test. I even stated that in the post! Wanted to eliminate rme variable. Either way it happens with both Again, if there is a way to change the internal sound hardware buffer size outside nuendo let me know.
And if you have not tested every combination of sample rate, and buffer size you have not performed my full repro test. Perhaps it only happens when changing from a specific rate or buffersize? You would not find that out by only testing one change. You may need to run the gamet.
If you can’t reproduce this with Steinberg plug-ins then it looks like more of a UAD (and other plug-in manufacturer) issue. As I have mentioned before, plug-ins can have bugs where they behave differently at different sample rates/buffer sizes.
Id just like to say, as a long time user of Nuendo, I’d love to find root cause of this issue. And great if it’s user error, or not a Nuendo bug or a plugin bug. I’m fine with that outcome. I’m not trying to paint Nuendo in a bad light or anything.
Ive provided repro steps that demonstrate the issue. I’d like to know why these repro steps cause this problem. I know there are probably 17 other ways I could test this, but let’s stick to THIS repro that works. Why does it occur here under these fairly vanilla circumstances?
I will try to find additional plugins that cause this same issue.
For now it would be great to know if anyone out there that has Cambridge can reproduce exactly the way I demonstrated.
Secondly I can’t reproduce your problem at any buffer/sample rate that my sound card will do. I tried with Altiverb, Sonnox plugs and quite a few other plugs that I have. In all cases there was no problem.
-I create two identical tracks
-put one out of phase so they cancel out completely
-add altiverb to the first track
-bypassed altiverb so 100% dry signal is output
-playback shows both tracks cancel out (plugin delay is accurately compensated for)
-change hardware buffer size to 768
-BUG: playback shows both tracks NO LONGER cancel out (plugin delay is incorrect)
-change buffer back to 1024 both tracks cancel out 100%
changing the buffer to 1024/512/256/128/64/32 it works great (100% cancellation). only 768 and 384 don’t work properly (here with rme).
Max-- just curious-- have you checked the problematic buffers under different sample rates? I’m guessing you did your test at 48k. Does it do the same thing at 44.1 (or vice versa if I guessed wrong)…
You and I may be smelling the same overheated wire here! I’ve had questions about those things for years, and never actually sat down and did the kind of testing that would really nail it one way or the other… I’m tempted to run the same test on N4 just to see if we all missed this in the previous cycle… but will have to wait until the current project is closer to done! In the meantime, am very interested in what you’re talking about…
Conjecture : It is a mac specific issue. (I’ve read the thread, and can’t see a pc repro)
I can test on various metric halo interfaces, and internal audio. The old fireface is no longer in service.
I have 3 different macs here to try - I will attempt to get at least one of them done today.
