How do you guys deal with latency when reamping?
- Thread starter Loren Littlejohn
- Start date
Glenn Fricker
Very Metal &Very Bad News
Polpotkin
Hit'er in the shitter
It's not a question if the DAW itself compensates for latency, of course it does, we live in modern times.
But it takes time for the signal to leave the output, go through the reamp box in to the amplifier, through all circuits, to the cab. Then it will, no matter how good your daw is, take some time for the soundwaves to travel one inch (or whatever) to your mic and all the way back.
Result = You Will have some amount of latency, or else the physical laws (as i know them) have changed. The latency is outside the box.
But it takes time for the signal to leave the output, go through the reamp box in to the amplifier, through all circuits, to the cab. Then it will, no matter how good your daw is, take some time for the soundwaves to travel one inch (or whatever) to your mic and all the way back.
Result = You Will have some amount of latency, or else the physical laws (as i know them) have changed. The latency is outside the box.
STINNETT
Bad / Nationwide
That latency will be no different than the latency you experience while actually playing the guitar.
Polpotkin
Hit'er in the shitter
Yes that's correct, but that doesn't change the fact that it will take time for the signal to go through the whole system, and come Back. And when recorded again it will by comparsion be delayed, compared to the original track.
Let's say it takes 1ms from when your pic hits the string to till the signal ends up in your computer. Ok, so the original signal is per default delayed by 1ms. Now you send the signal out once again, and that would by my empirical conclusion mean that the reamped signal is 2ms late from "point zero" (when you first hit the string), or 1ms late from the original track. Right?:Smug:
Let's say it takes 1ms from when your pic hits the string to till the signal ends up in your computer. Ok, so the original signal is per default delayed by 1ms. Now you send the signal out once again, and that would by my empirical conclusion mean that the reamped signal is 2ms late from "point zero" (when you first hit the string), or 1ms late from the original track. Right?:Smug:
STINNETT
Bad / Nationwide
Yes that's correct, but that doesn't change the fact that it will take time for the signal to go through the whole system, and come Back. And when recorded again it will by comparsion be delayed, compared to the original track.
Let's say it takes 1ms from when your pic hits the string to till the signal ends up in your computer. Ok, so the original signal is per default delayed by 1ms. Now you send the signal out once again, and that would by my empirical conclusion mean that the reamped signal is 2ms late from "point zero" (when you first hit the string), or 1ms late from the original track. Right?:Smug:
Okay, but I think you're talking about imperceptible latency.
If you were layering your reamped tracks over the original DI tracks, you might need to adjust them to avoid phasing, but if you're just using the reamped tracks you wouldn't need to worry about it.
Polpotkin
Hit'er in the shitter
Yep, that's what i'm talking about. So we're both right
But let's say that you reamp bass and whishes to keep the DI with the reamped signal, or reamp through several amps or different mic positions, you will still need to adjust at least one of the tracks. (even if you're not micing a cab, e.g. DI -> Sansamp)
But let's say that you reamp bass and whishes to keep the DI with the reamped signal, or reamp through several amps or different mic positions, you will still need to adjust at least one of the tracks. (even if you're not micing a cab, e.g. DI -> Sansamp)
i have a c2d 2ghz processor and 2gs of ram and latency is NEVER an issue when reamping as long as i don`t have many plugs in the session
I diden't tought about that, i'll have to do additional testing, if your right, i might use a median value so i won't need to change it everytime i switch buffer size/latency..
You say that latency is only outside the box, but i get about 9 samples of latency doing a loopback test within my soundcard without going through the DA-AD conversion, and about 50 samples of latency doing a loopback test going out of the D-A/output then in the preamp/A-D back to the DAW.
Might be a driver or software issue and might not be the case for everyone, but it's better to fix at least a part of the problem IMO.
As for the additional latency that is occuring in the cables, electronics or in the air (sound waves travel slower than signal) i added 1 sample of latency compensation as a placebo effect. Then i can sleep well thinking that i did something to help against the random hardware/sound latency... Well anyway it's unpredictable and negligible in most of the cases.
ahjteam
Anssi Tenhunen
In case of changing sample rate median value can be enough, but for latency (buffer size) change it is not (at least im my case). Overall delay will be like buffer size(latency value) x 2 + converters latency. So when buffer doubles record placement offset also doubles (aproximately). But I also encounter some strange values at certain buffer sizes on ESI Juli@.
MetalWorks
Member
Ding, Ding....
I have never even stopped to think about the consideration of a delay and realigning the track.
I just reamped and to me it sounded no different than the original track.
In fact, I have even used the scratch Line 6 guitar track along with the reamped track and they sounded fine together.
It has to be extremely minor of an adjustment and I wouldn't have even called it latency.
I'll have to pay more attention next time for shits and giggles just to see if it makes even 1% difference to me to realign it.
If its really that big of a problem, maybe it just varies depending on the recording setup that already has latency issues.
In Nuendo with my Nuendo 9652 & 96K 8 I/O setup and Pro Tools HD setup I have had no latency concerns when reamping.
yeah, i usually align by hand but don't you think a permanent fix woulden't be better ?
True, you shouldn't really use any plugins while reamping with an application which uses your computers host processor.
The reason for this is that an ATA hardrive latencey is about 7 - 11 milleseconds and this becomes worse if you add in plugins.
The plugin may cause the App such as protools to access the swap file on the hardrive and this will increase the latencey of the hardrive.
It would also be adviseable to limit the tracks being played back at the same time when reamping as this will also have the same effect on latencey.
