Standard Recording Procedure Q
-
- KVRer
- 3 posts since 4 Dec, 2005
(this was originally posted on Mackie's forum; I thought KVR might get more traffic)
I've looked in the User Guide and Reference Guide, and poked around the FAQs, and haven't found the answer to this question.
I have an Onyx 400F firewire recording interface. I used Tracktion's built-in latency detection scheme and it discovered a consistent 41.06 ms latency from out to in. That's fine, I'm all about consistency and reliability. However, it makes end-to-end monitoring in real-time impossible due to the delay.
No problem! I just make sure that I turn up my input gain on my monitor mix (hardware mixer on the 400F control panel) and I'm fine, I hear myself in real-time.
Well, not really. If I assign, for instance, input 1 to track 1, and track 1's output destination is the monitor mix, then I get myself with zero latency from the 400F, summed with me at 41.06 ms of delay. No dice.
So what do I do? I assign track 1 to outputs 3/4 (for instance). Now, when I practice/record, I get the zero latency hardware mix and nothing else. Fine.
But now, if I want to hear what I just recorded, I have to go and re-assign track 1 to outputs 1/2 (my monitor outs). You can see that this becomes a tedious flip-flop of changing output channels depending on whether I'm listening to the recorded track to see how it was, or whether I'm recording a new clip onto the track.
I've tried a bunch of different things, like muting the track while I'm recording to it, or changing the recording mode to "replace old clips in the edit with new ones." Nothing I've done gets around this.
This seems to me to be a common enough problem that someone must have a solution. Any takers?
Thanks in advance,
Mike
I've looked in the User Guide and Reference Guide, and poked around the FAQs, and haven't found the answer to this question.
I have an Onyx 400F firewire recording interface. I used Tracktion's built-in latency detection scheme and it discovered a consistent 41.06 ms latency from out to in. That's fine, I'm all about consistency and reliability. However, it makes end-to-end monitoring in real-time impossible due to the delay.
No problem! I just make sure that I turn up my input gain on my monitor mix (hardware mixer on the 400F control panel) and I'm fine, I hear myself in real-time.
Well, not really. If I assign, for instance, input 1 to track 1, and track 1's output destination is the monitor mix, then I get myself with zero latency from the 400F, summed with me at 41.06 ms of delay. No dice.
So what do I do? I assign track 1 to outputs 3/4 (for instance). Now, when I practice/record, I get the zero latency hardware mix and nothing else. Fine.
But now, if I want to hear what I just recorded, I have to go and re-assign track 1 to outputs 1/2 (my monitor outs). You can see that this becomes a tedious flip-flop of changing output channels depending on whether I'm listening to the recorded track to see how it was, or whether I'm recording a new clip onto the track.
I've tried a bunch of different things, like muting the track while I'm recording to it, or changing the recording mode to "replace old clips in the edit with new ones." Nothing I've done gets around this.
This seems to me to be a common enough problem that someone must have a solution. Any takers?
Thanks in advance,
Mike
-
- KVRAF
- 12977 posts since 29 Sep, 2003 from Ottawa, Canada
-
- KVRist
- 123 posts since 13 Feb, 2004 from Brookline, MA
My first advice is to lower that latency! 41 ms is way too long for a snazzy new interface and a reasonable computer. My four year-old budget card on a five year-old budget computer gets by well with only 11 ms latency.
Are you getting the 41.06 ms reading by clicking the auto-detect button (then clicking apply) in the setup window? I'm not quite sure what this does--the biggest value I get with this test is .20 ms, and I can't imagine my system can do anything meaningful with audio in just .20 ms. Anyone know what this means?
Are you getting the 41.06 ms reading by clicking the auto-detect button (then clicking apply) in the setup window? I'm not quite sure what this does--the biggest value I get with this test is .20 ms, and I can't imagine my system can do anything meaningful with audio in just .20 ms. Anyone know what this means?
-
- KVRian
- 779 posts since 3 Apr, 2003 from UK
It's the time difference between the output and input stages of the audio device for events that occur at the same moment and so is the amount by which newly recorded audio is adjusted to make sure it lines up with existing audio.
-
- KVRAF
- 2009 posts since 9 Apr, 2003 from Cornwall, UK
-
- KVRAF
- 10815 posts since 26 Nov, 2004 from UK
al the info you need is in this thread!!
A) disable e to e on the input device (thats the things down the left hand side of the screen, if you highlight the input that comes from your desk in the propties pannel there will be a e to e button)
B) try a different driver because that latincey is masive!
thease 2 together should help you out
Subz
A) disable e to e on the input device (thats the things down the left hand side of the screen, if you highlight the input that comes from your desk in the propties pannel there will be a e to e button)
B) try a different driver because that latincey is masive!
thease 2 together should help you out
Subz
-
- KVRer
- Topic Starter
- 3 posts since 4 Dec, 2005
Thanks to all who have replied so far.
I believe the "end-to-end" in the device settings is what I was after. I thought I was accomplishing this by turning end-to-end processing off in the transport controls (the yellow button left of "scroll") but apparently I need to disable end-to-end at the device level. That worked perfectly.
I'm not sure how I can change my latency. I'm using Mackie's ASIO driver for the 400F, my buffer size is large enough that I don't get any pops or skips (512 samples), and I used the "auto-detect" function to determine my latency. And to top everything off, to my ears (and eyes, zooming in), everything is perfectly aligned when I use this latency value. (I.e. I record audio listening to other audio AND MIDI tracks with VSTi instruments [i.e. no latency], and it's all lined up).
So I'm not sure how to combat my latency issue. But since I don't need end-to-end on anyway, I think I can live with it at the (high) value of 41.06.
Thanks again for the answers...
Mike
I believe the "end-to-end" in the device settings is what I was after. I thought I was accomplishing this by turning end-to-end processing off in the transport controls (the yellow button left of "scroll") but apparently I need to disable end-to-end at the device level. That worked perfectly.
I'm not sure how I can change my latency. I'm using Mackie's ASIO driver for the 400F, my buffer size is large enough that I don't get any pops or skips (512 samples), and I used the "auto-detect" function to determine my latency. And to top everything off, to my ears (and eyes, zooming in), everything is perfectly aligned when I use this latency value. (I.e. I record audio listening to other audio AND MIDI tracks with VSTi instruments [i.e. no latency], and it's all lined up).
So I'm not sure how to combat my latency issue. But since I don't need end-to-end on anyway, I think I can live with it at the (high) value of 41.06.
Thanks again for the answers...
Mike
-
- KVRAF
- 6740 posts since 25 Mar, 2002 from sheffield, england
512 samples at 44.1KHz should only give you about 11ms latency
unless firewire adds some extra..?
-
- KVRer
- 16 posts since 30 Sep, 2004
Maybe 3-10 ms,but 30ms ???IIRs wrote: unless firewire adds some extra..?
http://www.m-audio.com/index.php?do=sup ... 3d662fd729
-
- KVRAF
- 12977 posts since 29 Sep, 2003 from Ottawa, Canada
-
- KVRAF
- 4908 posts since 10 Aug, 2004 from Colorado Springs
My Soundblaster Live is 5.3 mS with just a few soft instruments going. Laugh all you want.
-Scott
-Scott
-
- KVRer
- Topic Starter
- 3 posts since 4 Dec, 2005
Ok, thanks for the pointers. I'll try to recalibrate and see why it's so high.
(I run a 512 byte buffer at 44.1 KHz, 24 bit. My system is a 1.2 GHz P3m laptop with 512 MB RAM. Maybe this is why it's a little low?)
Mike
(I run a 512 byte buffer at 44.1 KHz, 24 bit. My system is a 1.2 GHz P3m laptop with 512 MB RAM. Maybe this is why it's a little low?)
Mike
-
- KVRAF
- 2009 posts since 9 Apr, 2003 from Cornwall, UK
personaly I would not use the tracktion built in latency detection scheme. It comes up with zero whenever I've used it, and my setup has worked fine so far by just ignoring it.
I just set my latency on the slider (make sure the external sound card control panel has the same latency, don't worry if you don't have one) restart the driver and your done.
I just set my latency on the slider (make sure the external sound card control panel has the same latency, don't worry if you don't have one) restart the driver and your done.



