I'm using win2K with the delta1010 and n-track. When i'm dubbing over audio tracks I set my ASIO buffer size down pretty low to the point where I'm probably getting around 10 ms of latency. My question is - how does one eliminate latency entirely from the final recording. It seems to me that if the dubbed over track is consistantly 10ms behind the original, it could just be shifted over by 10ms - then the 2 tracks would be perfectly in time. Seems as if this is an option audio software should have. Does anyone out there know anything about this?
latency compensation
- Thread starter bdemenil
- Start date
Slackmaster2K
Gone
The tracks aren't 10ms behind. Latency is only an issue when mixing or using live input processing. When you move a slider in n-track, for instance, it takes 10ms before you hear the results. Fortunately, 10ms is so fast that it'll seem to be instantaneous.
When you monitor with the delta, it's routing the inputs directly to the outputs. In other words, there's no lag. You're hearing the input before it's processed by the machine.
Now, if you use Live Input Processing to run effects on a track as you record it, then the signal is digitized, processed, and then spit back out. In your case this will take 10ms. If you don't use Live Input Processing (not many people do), then you don't have to worry.
People often confuse "lag" with "latency". Prove this to yourself. Run an ouput to an input and re-record a track. The new track will be in perfect time with the original.
Slackmaster 2000
When you monitor with the delta, it's routing the inputs directly to the outputs. In other words, there's no lag. You're hearing the input before it's processed by the machine.
Now, if you use Live Input Processing to run effects on a track as you record it, then the signal is digitized, processed, and then spit back out. In your case this will take 10ms. If you don't use Live Input Processing (not many people do), then you don't have to worry.
People often confuse "lag" with "latency". Prove this to yourself. Run an ouput to an input and re-record a track. The new track will be in perfect time with the original.
Slackmaster 2000
Slackmaster2K
Gone
Yeah I'm sure.
I did the test on my own again to make sure. I ran ashort TRS cable from output 1 to input 1 on the delta. In the control panel I have HW Out 1/2 set to Monitor Mixer.
I then set the ASIO buffer to 100 samples which corresponds to a little less than 2ms at 48khz. I opened n-Track with a 48khz wave of a high hat ticking and hit record, re-recording the track. Here's the result:
http://www.imt.net/~blarson/100sample.gif
The resulting wave is approximately 1 millisecond behind the original. You might be thinking that's lag, but I don't think so.
I closed n-Track, and set the ASIO buffers to 2700 samples, which corresponds to around 60ms at 48khz. I opened n-Track and repeated the above procedure.
http://www.imt.net/~blarson/2700sample.gif
The resulting wave is still approximately 1 millisecond behind the original. I attribute this to the hardware mixer itself. But clearly, there is no noticable monitoring lag when using the delta, regardless of your buffer settings.
Slackmaster 2000
I did the test on my own again to make sure. I ran ashort TRS cable from output 1 to input 1 on the delta. In the control panel I have HW Out 1/2 set to Monitor Mixer.
I then set the ASIO buffer to 100 samples which corresponds to a little less than 2ms at 48khz. I opened n-Track with a 48khz wave of a high hat ticking and hit record, re-recording the track. Here's the result:
http://www.imt.net/~blarson/100sample.gif
The resulting wave is approximately 1 millisecond behind the original. You might be thinking that's lag, but I don't think so.
I closed n-Track, and set the ASIO buffers to 2700 samples, which corresponds to around 60ms at 48khz. I opened n-Track and repeated the above procedure.
http://www.imt.net/~blarson/2700sample.gif
The resulting wave is still approximately 1 millisecond behind the original. I attribute this to the hardware mixer itself. But clearly, there is no noticable monitoring lag when using the delta, regardless of your buffer settings.
Slackmaster 2000
Last edited: