Latency reporting to DAW

Yeah, there's always USB latency no matter what you do. Some amount is unavoidable. The issue is that the latency is not accurately reported to the DAW for recording and playback compensation and that latency is not consistent so manually measured and set up offsets end up not always being correct.

For example, if you reamp a DI track the new recorded track will often be slightly misaligned relative to the DI track and project timeline. Because the USB latency can vary and the driver's reported latency is not always correct, the DAW's attempt to compensate for the expected latency can place the new track either early or late relative to the other tracks depending on how off the reported latency figure actually is.
 
Even if you are going through a mixer or other gear wouldn’t you still have that latency as it needs to be hooked up to your machine? Or does it come down to USB vs Thunderbolt vs other connectivity standards? I’m really showing my ignorance here, and I’m a computer guy 🙄 just not necessarily a recording gear guy. And I will buy additional gear to eliminate this problem too. We all know how frustrating it is
 
Forever Again said:
Even if you are going through a mixer or other gear wouldn’t you still have that latency as it needs to be hooked up to your machine? Or does it come down to USB vs Thunderbolt vs other connectivity standards? I’m really showing my ignorance here, and I’m a computer guy 🙄 just not necessarily a recording gear guy. And I will buy additional gear to eliminate this problem too. We all know how frustrating it is
Click to expand...
For an explanation of the issue, see "Latency Compensation" here:
https://forum.fractalaudio.com/threads/axe-fx-for-the-recording-musician.177592/post-2162323
 
Forever Again said:
Even if you are going through a mixer or other gear wouldn’t you still have that latency as it needs to be hooked up to your machine? Or does it come down to USB vs Thunderbolt vs other connectivity standards? I’m really showing my ignorance here, and I’m a computer guy 🙄 just not necessarily a recording gear guy. And I will buy additional gear to eliminate this problem too. We all know how frustrating it is
Click to expand...
I would assume it all comes down to the Asio- or USB-driver written for the Axe. You might have noticed that the PC-editer also have quite a lag, but since it's purely function as a control interface, it's not really an issue. All processing is obviously happening inside the AxeFX.
 
Me too, but if I used a Mac (so I could have multiple sound interfaces running concurrently) I would want this feature!
Thanks
Pauly
Fashion Foul said:
I send the AF3 through my mixing board and have never had latency issues, at least it runs fast enough that I don't notice it.
Click to expand...
 
Rasmus Toftlund said:
I would assume it all comes down to the Asio- or USB-driver written for the Axe. You might have noticed that the PC-editer also have quite a lag, but since it's purely function as a control interface, it's not really an issue. All processing is obviously happening inside the AxeFX.
Click to expand...
Yes this is my understanding as well that it is down to the Asio-USB drivers - it is the drivers that tells windows and subsequent programs what to do. It's quite common for hardware devices to have upgraded drivers but I don't think I've seen that as of yet for the FM3
 
Rasmus Toftlund said:
+1 for this. But I do suspect it will be somewhat difficult to solve, at least when it comes to doing recording thru a processed signal chain. The latency shifts around depending on what kind of preset you have constructed, and the combinations are almost endless. On the other hand, it is very disappointing that when using the Axe (or FM# for that matter) for direct unprocessed recording via USB (say for re-amp purposes), it is still not able to report correct latency back to the DAW. THAT should be quite simple and a slam dunk. Btw, we are talking about more than 10 ms offset (492 samples in my current preset. In Cubase on a FM3), from the already reported latency compensation - that is quite a bit!
Click to expand...

Fwiw with my Axe Fx II and XL+s I NEVER had any issues, so those devices had no problems in this regard so I'm hoping they can fix it with a fw

Rasmus Toftlund said:
Did you change preset? Because my testing showed it will change based on the actual preset. Though 1160 to 750 is quite a heavy change. I also use Cubase btw.
Click to expand...

Hmm I don't think that matters as I've reamped and recorded with the Axe Fx sooo many times. It now seems to be stable at 750...
 
Mines usually pretty stable as well. I have a manual offset in Reaper and it's typically close enough for rock n roll. It has changed on me in the past though.
 
mr_fender said:
Mines usually pretty stable as well. I have a manual offset in Reaper and it's typically close enough for rock n roll. It has changed on me in the past though.
Click to expand...
That is what is most strange to me...a lot of people report that it works fine, while others cannot maintain any sort of stable value.

I'd be curious...do you reamp prior to the mixing stage? In other words, not yet having plugins on your tracks? One series of tests I did seemed to indicate that plugins with delay compensation (PDC) caused the instability...which seems odd, but I'll admit my knowledge of how all this works is limited.

I'm wondering if anyone else would be willing to perform similar tests to see if that is the "outlier". To explain the methodology I used for the PDC test:
  • blank project, no plugins.
  • test reamp
  • adjust your latency offset value so the reamp is on time
  • test reamp
    • reamp should be on time
  • add 100 instances of a "non PDC plugin" (i used Waves CLA-76)
    • I was just trying to go overkill with the # of plugins
  • test reamp
    • reamp was on time
  • start adding plugins with PDC (I think I used Izotope Ozone 9)
  • test reamp
    • reamp was no longer on time and getting it back to "on time" wasn't exactly working
I don't recall exactly how many instances of PDC plugins it took, but the reamp started to get off time...and finding a new accurate latency offset value didn't seem to work. (One reamp would be off a certain number of samples, setting the offset to correct for that amount would still result in an inaccurate reamp). What's worse is that removing the PDC plugins did not return the offset to its original state (my original setting no longer worked and I could no longer get the timing offset to "hit").
 
sixtystring said:
That is what is most strange to me...a lot of people report that it works fine, while others cannot maintain any sort of stable value.

I'd be curious...do you reamp prior to the mixing stage? In other words, not yet having plugins on your tracks? One series of tests I did seemed to indicate that plugins with delay compensation (PDC) caused the instability...which seems odd, but I'll admit my knowledge of how all this works is limited.

I'm wondering if anyone else would be willing to perform similar tests to see if that is the "outlier". To explain the methodology I used for the PDC test:
  • blank project, no plugins.
  • test reamp
  • adjust your latency offset value so the reamp is on time
  • test reamp
    • reamp should be on time
  • add 100 instances of a "non PDC plugin" (i used Waves CLA-76)
    • I was just trying to go overkill with the # of plugins
  • test reamp
    • reamp was on time
  • start adding plugins with PDC (I think I used Izotope Ozone 9)
  • test reamp
    • reamp was no longer on time and getting it back to "on time" wasn't exactly working
I don't recall exactly how many instances of PDC plugins it took, but the reamp started to get off time...and finding a new accurate latency offset value didn't seem to work. (One reamp would be off a certain number of samples, setting the offset to correct for that amount would still result in an inaccurate reamp). What's worse is that removing the PDC plugins did not return the offset to its original state (my original setting no longer worked and I could no longer get the timing offset to "hit").
Click to expand...
Damn, that's more than a little crazy! 😳
 
sixtystring said:
That is what is most strange to me...a lot of people report that it works fine, while others cannot maintain any sort of stable value.

I'd be curious...do you reamp prior to the mixing stage? In other words, not yet having plugins on your tracks? One series of tests I did seemed to indicate that plugins with delay compensation (PDC) caused the instability...which seems odd, but I'll admit my knowledge of how all this works is limited.

I'm wondering if anyone else would be willing to perform similar tests to see if that is the "outlier". To explain the methodology I used for the PDC test:
  • blank project, no plugins.
  • test reamp
  • adjust your latency offset value so the reamp is on time
  • test reamp
    • reamp should be on time
  • add 100 instances of a "non PDC plugin" (i used Waves CLA-76)
    • I was just trying to go overkill with the # of plugins
  • test reamp
    • reamp was on time
  • start adding plugins with PDC (I think I used Izotope Ozone 9)
  • test reamp
    • reamp was no longer on time and getting it back to "on time" wasn't exactly working
I don't recall exactly how many instances of PDC plugins it took, but the reamp started to get off time...and finding a new accurate latency offset value didn't seem to work. (One reamp would be off a certain number of samples, setting the offset to correct for that amount would still result in an inaccurate reamp). What's worse is that removing the PDC plugins did not return the offset to its original state (my original setting no longer worked and I could no longer get the timing offset to "hit").
Click to expand...
Surely a product this sophisticated and expensive (at least here in Australia) should not require this much work. I would rather have this sorted by fractal than millions of different delays and so on. I know someone will jump down my neck for saying so but just my thoughts.
 
Oshin said:
Surely a product this sophisticated and expensive (at least here in Australia) should not require this much work. I would rather have this sorted by fractal than millions of different delays and so on. I know someone will jump down my neck for saying so but just my thoughts.
Click to expand...
I actually very much agree. Still - doesn't take away from the fact that the AxeFx3 with Cygnus firmware, is the only digital modeller that satisfyingly actually can replace a real tube-amp...at least in my book. Which again is it's main purpose after all. But I would love to have Fractal address this issue with a little more severity, though I am sure they will fix it eventually.
 
Oshin said:
Surely a product this sophisticated and expensive (at least here in Australia) should not require this much work. I would rather have this sorted by fractal than millions of different delays and so on. I know someone will jump down my neck for saying so but just my thoughts.
Click to expand...
It was mentioned earlier in the thread by @Admin M@ that one of the engineers will be back to looking at the issue once another project is done.


As an aside to that, I've been trying to gather useful information from other users' comments as well as from my own tests/troubleshooting to help facilitate finding a potential fix. A lot of threads on this same topic die off because there is very little input from users who actually experience the unstable offset issue.

Based on other user reports, this is not a DAW specific problem (I've tested in Reaper, and have seen reports on Logic and Cubase as well). It's not an AMD vs Intel processor problem (I've tried on multiple computers). The problem exists for both MAC and Windows users (this based on the fact that I test in Windows and Logic users have reported the problem).

I noticed that the offset would be fine when I tested it in a blank project, but opening existing projects would cause the issue. Even deleting all the tracks/plugins off the project would not re-stabilize the offset. From there I assumed that something in existing projects seemed to be triggering it. So, I tried to recreate the problem by using the steps I mentioned above (https://forum.fractalaudio.com/threads/latency-reporting-to-daw.177908/post-2181398).

In that particular series of tests, the trigger for the offset instability seems to be projects with PDC plugins. But that is based on a series of tests done only by me. It would probably be a big step towards finding a solution if someone else confirmed this behavior...especially in DAWs other than Reaper.
 
I think it's not related to any plugins, the latency instability can be measured with a standalone software like RTLUtility, or using an empty project in Reaper or Cubase (I'm using Reaper and Cubase Pro 11).

Try this :
  • Measure the latency
  • Change the USB buffer size to another value then go back to the previous one
  • Measure again

On my unit every time I do this, I get a different value.

Like you did, based on others reports and my own tests, I found this 4 issues
  • When the unit start, the USB Buffer is reset to 128, no matters what the unit is showing.
  • Latency added by the USB buffer is not reported
  • Latency caused by the USB buffer change over time
  • On my computer (Intel, Win10), I can use the AxeFX ASIO driver in multiple program simultaneously.
 
Rasmus Toftlund said:
I actually very much agree. Still - doesn't take away from the fact that the AxeFx3 with Cygnus firmware, is the only digital modeller that satisfyingly actually can replace a real tube-amp...at least in my book. Which again is it's main purpose after all. But I would love to have Fractal address this issue with a little more severity, though I am sure they will fix it eventually.
Click to expand...
Yeah mate, I also agree with what you have said.
 
sixtystring said:
It was mentioned earlier in the thread by @Admin M@ that one of the engineers will be back to looking at the issue once another project is done.


As an aside to that, I've been trying to gather useful information from other users' comments as well as from my own tests/troubleshooting to help facilitate finding a potential fix. A lot of threads on this same topic die off because there is very little input from users who actually experience the unstable offset issue.

Based on other user reports, this is not a DAW specific problem (I've tested in Reaper, and have seen reports on Logic and Cubase as well). It's not an AMD vs Intel processor problem (I've tried on multiple computers). The problem exists for both MAC and Windows users (this based on the fact that I test in Windows and Logic users have reported the problem).

I noticed that the offset would be fine when I tested it in a blank project, but opening existing projects would cause the issue. Even deleting all the tracks/plugins off the project would not re-stabilize the offset. From there I assumed that something in existing projects seemed to be triggering it. So, I tried to recreate the problem by using the steps I mentioned above (https://forum.fractalaudio.com/threads/latency-reporting-to-daw.177908/post-2181398).

In that particular series of tests, the trigger for the offset instability seems to be projects with PDC plugins. But that is based on a series of tests done only by me. It would probably be a big step towards finding a solution if someone else confirmed this behavior...especially in DAWs other than Reaper.
Click to expand...
I did read the post stating that they would be looking into it and as with six string reply, I agree with your comments. To be honest a lot of this goes over my head therefore my comment was born from frustration.
 
Evohe said:
I think it's not related to any plugins, the latency instability can be measured with a standalone software like RTLUtility, or using an empty project in Reaper or Cubase (I'm using Reaper and Cubase Pro 11).

Try this :
  • Measure the latency
  • Change the USB buffer size to another value then go back to the previous one
  • Measure again

On my unit every time I do this, I get a different value.

Like you did, based on others reports and my own tests, I found this 4 issues
  • When the unit start, the USB Buffer is reset to 128, no matters what the unit is showing.
  • Latency added by the USB buffer is not reported
  • Latency caused by the USB buffer change over time
  • On my computer (Intel, Win10), I can use the AxeFX ASIO driver in multiple program simultaneously.
Click to expand...
Nice. Thanks for providing this info. The USB buffer resetting to 128 after reboot is mentioned in the link below. I checked the firmare release notes for "USB", but it is last mentioned in firmware 15. (I also checked the pinned driver & USB threads, but they seem prior to this report as well.)
https://forum.fractalaudio.com/threads/usb-buffer-size-is-ignored-after-booting-axe-fx-iii.176573/


Also, for clarification, my assumption that plugins have something to do with it came from the steps I provide in the video below. (I rarely ever turn off my Axe-FX.) Unfortunately I did not screen capture the PDC plugin test, but it can easily be replicated.

 
You must log in or register to reply here.
Back
Top Bottom