Latency Compensation Measurement

[GlennO]

I want to make sure I'm understanding this correctly... I'm using Studio One Prof (Ver 4.6.2) and I'm providing a screen shot of each menu for how it reads latency in the I/O preferences. I'm just not sure how to interpret it for what my manual offset should be to correctly offset the latency. Do I only look at the 'Input Latency' under the 'Audio Device' tab and then enter a negative value of that number (which would be -284 samples for the offset)...? Or do I look at 'Instrument' under Monitoring Latency in the 'Processing' tab and enter that as a negative value (-401 samples)...Orrr am I just retarded and both of my ideas are wrong lol hopefully someone can provide a little clarity... Thanks in advance.

I would recommend using the procedure at the top of this thread. That way you'll get confirmation that the value is correct.

 

[MackieFX]

Used this and ended up with 700 samples when using amp+cab only.

Id imagine adding more will add more latency. Its pretty damn annoying

 

[sixtystring]

Used this and ended up with 700 samples when using amp+cab only.

Id imagine adding more will add more latency. Its pretty damn annoying

See more from the link below, but I don't think the # of blocks affects things. (In my case, the reamp offset jumps around, but it does it even without changing #/type of blocks.)

https://forum.fractalaudio.com/threads/latency-reporting-to-daw.177908/post-2184657

 

[GlennO]

Used this and ended up with 700 samples when using amp+cab only.

Id imagine adding more will add more latency. Its pretty damn annoying

There are a couple of obvious exceptions (lookahead compression and pitch), but generally speaking, no, adding more blocks will not affect the latency.

 

[MackieFX]

There are a couple of obvious exceptions (lookahead compression and pitch), but generally speaking, no, adding more blocks will not affect the latency.

The difference between, amp+cab bypassed and on was about 50-100 samples.

 

[GlennO]

The difference between, amp+cab bypassed and on was about 50-100 samples.

That's correct. However, as I said, adding more blocks to that, with a couple of notable exceptions, will not add further latency.

 

[MackieFX]

That's correct. However, as I said, adding more blocks to that, with a couple of notable exceptions, will not add further latency.

aha! my mistake. Thanks Glenn <3

 

[kerninger]

I just measured the recording latency for my FM3 setup as described in post #8 of this thread. The recorded click comes round about 590 samples behind the click signal. With setting a constant offset in Cubase of 590 samples I get the two clicks completely in sync.
But: When I record my guitar through input 1 with usb in Cubase my guitar signal is recorded before the clicktrack signal. It doesn’t matter if I set the offset to 0 or 590 samples it still comes before although I my playing is quite tight on the click.
I tried this with headphones on the FM3 headphones out while recording and with out1 going into a focusrite interface only for using direct monitoring via speakers. The speakers are connected to the focusrite. FM3 is used as Audio Interface.
I tried different Cubase settings and buffer sizes. I really don’t get it. What am I doing wrong?
Why is my recorded guitar signal before the click?

 

[GlennO]

Those instructions mimic plugging a guitar into an FM3 and playing along with a track. There's no difference in the audio routing when you record a guitar, so something in your configuration must have changed after you did your loopback test. For example, changing a buffer size. Or you're using a preset with significantly different latency than the preset in the loopback test. It's suspicious that setting the offset to zero doesn't matter. That should make a big difference in the alignment. How much early is it?

 

[kerninger]

Those instructions mimic plugging a guitar into an FM3 and playing along with a track. There's no difference in the audio routing when you record a guitar, so something in your configuration must have changed after you did your loopback test. For example, changing a buffer size. Or you're using a preset with significantly different latency than the preset in the loopback test. It's suspicious that setting the offset to zero doesn't matter. That should make a big difference in the alignment. How much early is it?

Thanks for your answer.
The setup didn‘t change regarding buffer size. I just switched input 1 back to analog for my guitar and changed to a simple preset with amp and cab block.
The recorded track was about 5000 samples before the clicktrack when setting the offset in my daw to 0. Setting a positive offset of 590 samples (measured before) results in 5590 samples latency. So I have to set a negative offset to 5000 samples in order to get a useful compensation for the recorded track. The value of 5000 I have to check again but it was significantly larger than the measured offset.

And what I don’t understand is the fact that I have to set a negative offset for recording although I measured a positive offset during the loopback test.

 

[GlennO]

Try this: use the preset described in post 8 to measure your latency error. After you measure it and verify a recording aligns correctly, remove the patch cable. Connect your monitors to output 2. Plug your guitar into input 2. Record yourself playing guitar along with the click track you used to measure the latency error. Does that recording align properly?

 

[kerninger]

Try this: use the preset described in post 8 to measure your latency error. After you measure it and verify a recording aligns correctly, remove the patch cable. Connect your monitors to output 2. Plug your guitar into input 2. Record yourself playing guitar along with the click track you used to measure the latency error. Does that recording align properly?

Good idea. But I get round about 3700 samples offset before the click. In the picture the bottom track is my guitar signal, the first track ist the click track. When I set the offset to 0 I get round about 3200 samples offset. So in order to get a "tight" recording I always have to shift the recorded track a little bit to the right.

 

[GlennO]

I wouldn't call 3000 samples a "little bit" . The only difference between the bottom two tracks is the source of the audio coming into input port 2, right? Correct me if I'm wrong, but I don't see how there could be any explanation other than you're playing 3000 samples early. Can you post a screen shot of your Studio Setup -> Audio System dialog?

 

[kerninger]

I wouldn't call 3000 samples a "little bit" . The only difference between the bottom two tracks is the source of the audio coming into input port 2, right? Correct me if I'm wrong, but I don't see how there could be any explanation other than you're playing 3000 samples early. Can you post a screen shot of your Studio Setup -> Audio System dialog?

I also have no other explanation than my playing but honestly, I can‘t believe that. 3000 samples offset are clearly audible when clicktrack and guitar recordings are played together. When I am playing it seems tight enough to me. These are just downstrokes on each click with the strings muted.
What do you mean with „audio system dialog“?

 

[kerninger]

I did another test. I put the guitar cable direct into the instrument Input of my focusrite scarlett with 128 samples buffer size and recorded to the click. With this setup I had no issues of playing "to early", just the usual human swing. That's weird.

 

[GlennO]

Weird indeed. If I had to guess, I‘d say it’s an issue with how you’re monitoring. Otherwise, your Axe-FX is watching you to see which cable you plug in .

 

[ElKabong]

I'm kinda amazed this hasn't been ironed out yet.

Glenn, can you clarify your instructions for me? You say to "Compare the recording with the click track to find the number of samples you need to put in your DAW latency compensation preference. " I imagine I could figure out how many milliseconds one is ahead of the other, but how do I find the difference in samples?

 

[GlennO]

I'm kinda amazed this hasn't been ironed out yet.

Glenn, can you clarify your instructions for me? You say to "Compare the recording with the click track to find the number of samples you need to put in your DAW latency compensation preference. " I imagine I could figure out how many milliseconds one is ahead of the other, but how do I find the difference in samples?

Most DAWs have the option to show samples as the units for the time line ruler instead of beats or seconds. Also, many DAWs have a tool that shows the number of samples in a range you select with the mouse. Here's a screen shot of Cubase with samples shown in the ruler.

It is puzzling this is still an issue four years after it was first reported. I'm not aware of any competitor's usb-equipped guitar modeler with this problem. Or any audio interface of any kind that has this problem for that matter. This is just speculation on my part, but it could be that bugs related to recording get low priority from FAS. Anyway, if you contact FAS tech support, they may have info on the status of a fix.

 

[ElKabong]

Most DAWs have the option to show samples as the units for the time line ruler instead of beats or seconds. Also, many DAWs have a tool that shows the number of samples in a range you select with the mouse. Here's a screen shot of Cubase with samples shown in the ruler.

It is puzzling this is still an issue four years after it was first reported. I'm not aware of any competitor's usb-equipped guitar modeler with this problem. Or any audio interface of any kind that has this problem for that matter. This is just speculation on my part, but it could be that bugs related to recording get low priority from FAS. Anyway, if you contact FAS tech support, they may have info on the status of a fix.

View attachment 101849

Thank you thank you! Thanks to this post, my Axe-FX records nice and tight with the grid now.

 

[Soundstorm]

Hi, has this issue been addressed yet? Does going S/PDIF into a different interface solve the issue? I'm looking at ordering an FM3 very soon and was hoping to be able to use it as an interface so I can keep it as a minimal travel rig.