Audio sync problem
-
- Posts: 9
- Joined: 15 May 2016
Hi all
I'm sorry in advance as I'm sure this has been answered before. I've searched the web and here to no avail though.
I'm a fairly experienced reason user but have only really started using vocals.
I have noticed that when I record, and then play back the vocal are always a smidgen behind in the whole timing of the sequence.
I have to unsnap the track and drag it a few milliseconds forward so that is syncs again with the midi stuff.
I'm sure there is a setting that effectively does this to compensate.
Any pointers on how to do this.
Thanks
I'm sorry in advance as I'm sure this has been answered before. I've searched the web and here to no avail though.
I'm a fairly experienced reason user but have only really started using vocals.
I have noticed that when I record, and then play back the vocal are always a smidgen behind in the whole timing of the sequence.
I have to unsnap the track and drag it a few milliseconds forward so that is syncs again with the midi stuff.
I'm sure there is a setting that effectively does this to compensate.
Any pointers on how to do this.
Thanks
All DAWs work in such a way that they record the audio to the track *where it is heard* during recording. This means that when you use software monitoring the audio will be recorded to the track a bit later than your actual performance because of the roundtrip delay of your computer. This is exactly what was heard during the recording except that your actual acoustic performance is a bit "sooner". So if you're (acoustically) dead on the beat with the performance it will be a bit late in the recording. As said the DAW kind of expects you to go after what you're hearing.
So solutions? For one you can lower the latency of your system by lowering the buffer size. This will minimize the delay. You can also turn off software monitoring and use the hardware monitoring of your sound card. Here its important to actually turn off software monitoring in the DAW and not just use hardware monitoring and turn down the channel in the DAW so that your DAW knows that you're using hardware monitoring. Then it will put the audio dead where it was actually acousically performed in relation to the backing playback - because that is what is heard during recording in that case.
I hope this was clear enough, I'll explain further if something is unclear.
So solutions? For one you can lower the latency of your system by lowering the buffer size. This will minimize the delay. You can also turn off software monitoring and use the hardware monitoring of your sound card. Here its important to actually turn off software monitoring in the DAW and not just use hardware monitoring and turn down the channel in the DAW so that your DAW knows that you're using hardware monitoring. Then it will put the audio dead where it was actually acousically performed in relation to the backing playback - because that is what is heard during recording in that case.
I hope this was clear enough, I'll explain further if something is unclear.
-
- Posts: 9
- Joined: 15 May 2016
Thanks. I think that kinda makes sense. You would think there would be a setting to simply auto move forward the recording forward a set amount of time after the recording in the sequencer. This seems like a basic idea and would compensate. But alas it looks like it will have to be a manual job after recording each section which will slow down the music making process.
I'll look at what you suggested but logically it will never be no latency and so it'll have to be moved anyhow.
Thanks again
PS What is the 'recording latency compensation' setting in the audio options? I was kinda hoping this was what I was looking for but
A) there is nothing about it in the manual
B) changing it seems to have had no effect
I'll look at what you suggested but logically it will never be no latency and so it'll have to be moved anyhow.
Thanks again
PS What is the 'recording latency compensation' setting in the audio options? I was kinda hoping this was what I was looking for but
A) there is nothing about it in the manual
B) changing it seems to have had no effect
Last edited by pugnacious on 15 May 2016, edited 1 time in total.
Sure you have no latency if you use hardware monitoring or e.g. an external desk for the monitoring. But then as said you have to turn off software monitoring so your DAW knows you have zero latency and puts the track accordingly. Imo this behavior is very logical and consistent although you do have to know it works like this. Most instrumentalists do actually listen to what they hear and play accordingly. Many acoustic instruments also have latency so this is imo the normal way to do this.
-
- Posts: 9
- Joined: 15 May 2016
PS What is the 'recording latency compensation' setting in the audio options? I was kinda hoping this was what I was looking for but
A) there is nothing about it in the manual
B) changing it seems to have had no effect
A) there is nothing about it in the manual
B) changing it seems to have had no effect
It compensates for the slight delay, if monitoring has not been used on that track. When recording audio (and using software monitoring) I always create an extra track just for monitoring. The track that is record enabled will compensate for the delay and you'll still hear what you're doing.
Cheers!
Fredhoven
Fredhoven
No need to be confused. All you need to know is that if you don't want to nudge your recording afterwards, you create an extra track for monitoring only. The record enabled track doesn't monitor the signal and the automatic latency compensation compensates for the latency automatically.
Cheers!
Fredhoven
Fredhoven
-
- Posts: 9
- Joined: 15 May 2016
Great. If that works, I'll be well chuffed. I'll give it a go and let you know
Actually "compensation" is what happens when it shifts the track back If the driver reports the wrong latency the track is not shifted by the actual latency but by what the driver reports. Putting the track dead on (what happens when you have no software monitoring) is "not compensated".Gaja wrote:No need to be confused. All you need to know is that if you don't want to nudge your recording afterwards, you create an extra track for monitoring only. The record enabled track doesn't monitor the signal and the automatic latency compensation compensates for the latency automatically.
-
- Posts: 9
- Joined: 15 May 2016
genius (gaja).
thanks for this. all working now.
i kind of understand now as well.
i still dont know why any musician would want the recording to reflect when the computer played back the sound and not when they performed the sound. This means the musician would have to play/sing ahead of time just to make the recording work.
But anyhow, Im happy now as I know how to deal with it.
Great to find this forum. Since the official forums died, I didnt know there was a place. Sorry, Ill have loads of questions!
thanks for this. all working now.
i kind of understand now as well.
i still dont know why any musician would want the recording to reflect when the computer played back the sound and not when they performed the sound. This means the musician would have to play/sing ahead of time just to make the recording work.
But anyhow, Im happy now as I know how to deal with it.
Great to find this forum. Since the official forums died, I didnt know there was a place. Sorry, Ill have loads of questions!
Because as said most instrumentalists *listen* to what they do and (often unconsciously) adapt to that. Think about it, a violin doesn't instantly produce a sound when you first move the bow, neither does a piano when you hit the key and for synthesizer sounds the core sound also doesn't happen instantly, theres the attack etc. This is actually when you become a good musician, when you *listen* to what you do instead of just doing it so you can adapt and mold the sound you produce.pugnacious wrote:i still dont know why any musician would want the recording to reflect when the computer played back the sound and not when they performed the sound. This means the musician would have to play/sing ahead of time just to make the recording work.
Honestly *I* don't understand how one can ignore whats heard so far that you only realize later that theres a shift happening. Especially if that shift is so large that it becomes a problem. In that case I'd always go for hardware monitoring or low latency DSP processing to actually hear the sound where I would want it, which in turn also makes the DAW record it in that "location".
This is why people want low latency in their DAWs btw.
-
- Posts: 9
- Joined: 15 May 2016
Ok
Steep learning curve
I'm a sax player so get all you said, but adding in 'another delay' with computer makes me have to relearn what has taken years to master ie that gap between starting the note and it sounding and lengthening it. Anyhows, this is all academic now, and I thank you for all your advice.
I have a fairly decent MOTU audio device, so ill def look into hardware monitoring. Given all the knobs and sockets though, that may be another major challenge, perhaps ill just stick with the workaround gaja showed me
Steep learning curve
I'm a sax player so get all you said, but adding in 'another delay' with computer makes me have to relearn what has taken years to master ie that gap between starting the note and it sounding and lengthening it. Anyhows, this is all academic now, and I thank you for all your advice.
I have a fairly decent MOTU audio device, so ill def look into hardware monitoring. Given all the knobs and sockets though, that may be another major challenge, perhaps ill just stick with the workaround gaja showed me
Ok yeah, that's how you would say it I guess, it's a fixed value, not an actual compensation.normen wrote:Actually "compensation" is what happens when it shifts the track back If the driver reports the wrong latency the track is not shifted by the actual latency but by what the driver reports. Putting the track dead on (what happens when you have no software monitoring) is "not compensated".Gaja wrote:No need to be confused. All you need to know is that if you don't want to nudge your recording afterwards, you create an extra track for monitoring only. The record enabled track doesn't monitor the signal and the automatic latency compensation compensates for the latency automatically.
Cheers!
Fredhoven
Fredhoven
- kuhliloach
- Posts: 880
- Joined: 09 Dec 2015
I 2nd the idea above of getting the buffers setting down to 128 or 64 samples for doing the recording, and muting other things if needed to make the recording go smoothly. Just curious if this problem is stemming from a system that is set to 256 samples or higher. I usually monitor the "delayed" signal while recording. Is that a bad idea?
No its not a bad idea. With 10ms or so of delay you'll probably just adapt your playing slightly and thus play in time. As said you always *hear* whats recorded during recording anyway. The only time this becomes really important to know is if you have a high buffer size, actually have software monitoring enabled but don't use it (i.e. have the channel turned down) and play with other means of zero-latency monitoring (e.g. listen to your instrument acoustically, use an external desk or hardware monitoring). Then of course the result will be very unexpected.kuhliloach wrote:I 2nd the idea above of getting the buffers setting down to 128 or 64 samples for doing the recording, and muting other things if needed to make the recording go smoothly. Just curious if this problem is stemming from a system that is set to 256 samples or higher. I usually monitor the "delayed" signal while recording. Is that a bad idea?
-
- Information
-
Who is online
Users browsing this forum: No registered users and 30 guests