Audio is progressively delayed
There have been some recent mentions of problems with the timing of mixed down tracks being incorrect, but mostly regarding projects with tracks that have two or more different sample rates.
I have been doing a series of projects with all tracks in 16-bit, 44.1K as I always have. I have been following a procedure of vocal track comping that I have followed for many versions of n-Track. However, recently with some (not all) of the projects, the audio in a rendered track is progressively delayed about 300 ms per minute of audio. So, with a 3-minute song, by the end of the song the audio in the track made by rendering a comp of existing vocal takes, and loading into the song as the final vocal track, is about a second delayed from the original tracks.
I have tried every iteration of setting:
Use system timer for record, or not
Use system timer for playback, or not
Compensate plug-in latency, or not
I am not using dither nor sample rate converting any track.
My soundcard is a Creative X-Fi in the Music Creation mode set to native 16-bit, 44.1K recording with no sample rate conversion.
In essence, everything is set the same as it was for the past two years when I routinely mixed down a comp of several vocal takes into a final vocal track. And, to make this problem more frustrating, it doesn’t happen with every song file. so I can’t reliably duplicate it.
Is anyone else experiencing this problem with the recent builds? I am about to uninstall and try going back to an earlier n-Track version, but would certainly like to keep up with the latest versions if this is a known problem that can be fixed.
at the moment it looks like a fixed value in N program is being turned into a variable one by something, this would acount for many of the problems we are having - but what causes it we have yet to discover -
Is Flavio aware of this issue yet? It is a severe deficiency that has rendered n-Track unusable for me. I cannot continue my album project until it is fixed. Since I cannot find a setup that will consistently duplicate the problem, I have not logged a support email with Flavio yet.
I wanted to go back a few versions to find one that didn’t have this problem, but I cannot find earlier versions of 5.x on the n-Track website. Does anyone know which version started this problem and where I could get the download for the immediately previous version?
Needing to keep my project alive, I downloaded 5.0.9 (2287) which was the only previous version of 5.x I could find and found that if I load .sng files that were made in build 2312 and rendered a comp of the vocal tracks, the timing problem still exists. But, if I made a new .sng file in build 2287, then added all the vocal tracks from the project and did my comp editing, then rendered a new comp track, that track was in time. It may be too soon to tell, but it appears that whatever is causing the problem is being saved in the .sng file.
I will continue working in build 2287 to see if the problem crops up in that build. Meanwhile, I am going to enter a support email to Flavio and direct him to this thread.
I sent Flavio a support request concerning this and a few other issues.
Hang in there!
I was thinking of updating to build 2312 to see if my problem with 24 bit mono recording under ASIO was fixed - now I think I better wait for a fix of my problem to be announceed by Flavio before doing any updates at all…
Apart from the 24 bit mono issue, n-Track (5.1.1 build 2308) works stable and predictable here in my setup, and if it ain’t broke, there’s no need to fix it…
Patiently awaiting the upgrade of n-Track, the copy of Cubase Studio 4 I ordered the other day, and the next issue of Computer Music - bearing a disc with Samplitude 9 SE on it - to arrive…
Computer Music - bearing a disc with Samplitude 9 SE on it
Woah! Are they doing the SE give-away thing again? Coooolll... I gotta grab that. I have the Sam 8SE they gave away but have not done a full project with it yet.
Woah! Are they doing the SE give-away thing again?
I have done a lot of projects with Samplitude SE 8 since I got it. It is by far the most friendly environment I have ever worked in, VST plug-ins like SIR actually works in an Aux channel insert like it is supposed to, automation has time and level written at all the nodes (all right, n-Track got that a couple of builds ago), and mixdown is a breeze in it. The only two things I dislike about it is the fact that a volume envelope on a track going down below 0 dB shrinks the corresponding wave graph to a line, and that it is difficult to rename wave files once they are recorded. I'd love to see what they conjure up in SE 9...
Been there since The Fart Ones started recording – V3 and maybe before. After a lot of fuddling about we finally started putting click at the beginning of each mix so we could visually keep the tracks in sync.
I’m convinced this is plug-in compensation that the file renderer doesn’t take into consideration. It is more likely happening before the file renderer since we can see evidence of it when using plug-ins that cause a lot of compensation in AUX returns. Sometimes the sound from them can come out before the sound being sent to them.
As discovered and reported by Dr J in another thread, at least my problem was caused by mouse clicks along the right side of the timeline accidentally clicking near the scrub adjustment line and incrementing it to a speed of .99. By setting the playback speed to 1.00, my renders are timed correctly.
I have sent another support form to Flavio alerting him to what was causing my problem, and asking that he relocate the scrubbing tool to make it harder to accidentally click and change the speed. I also suggested that the song playback speed indicator glow red when the song speed is anything other than 1.00.
The speed change would definitely be something new and different. The issue I mentioned is extra samples at the beginning of a mixdown, but once the track is slid to the left a little it should line up to the end. If the speed is off no amount to sliding will compensate. That would indeed make mixdowns useless if they are to be reinserted into the song, which is normal for many of us that want to do hard rendered submixes.