live is not sample accurate!

[mr.ergonomics]

I don't like lurid topics, but this time I have to do it.

1.
I reported a long time ago that there is a flaw in the engine (via forum and tech support). To sum up, at times it's not sample accurate when you record something.
http://forum.ableton.com/viewtopic.php?f=2&t=103857


2. I found out that this is the case with the new fade-in system too.

When you record an audio file, it's sometimes not identical, not sample accurate to the source. Just do a test like this:

- record an audio sample which has "fade on" to another track
I guess you get what I've done out of the picture




- I checked if the recorded file is identical to the original one with an inverted phase test.
- I don't get silence, therefore it's not the same. When you add or subtract 1 or 2 samples delay (as +1, +2 and -1, -2) you have silence.





I haven't figured out what's exactly the reason for this, because sometimes I don't have this problem. It's a bit strange and I don't know if there are similar problems in other parts from the program.

You may ask why 1 or 2 samples delay are a problem, just hear the sample:

- snare hit 1, 2 are added sample accurate.
- snare hit 3,4 are added with 1 sample delay.

http://yshare.de/index.php?h=a7ec007c31 ... 5f0bf7971f

- snare hit 3,4 are muddy and have a lack of attack when you compare them.

In a serious DAW this kind of problems shouldn't happen.

[Johnisfaster]

20 page thread...

[Tarekith]

Is the original file warped?

[Tone Deft]

to add what Tarekith wrote...

complex warp mode(s?) tweaks the loops even at the original tempo.

[mr.ergonomics]

No, it's not in complex mode. You can read the link I've posted (http://forum.ableton.com/viewtopic.php?f=2&t=103857) with a detailed tutorial on a similar problem in session view.

This problem is certainly NOT the end of the world, but I think it has to be solved. I think it's related to the fade mechanism in live.

[cacti]

your stereo input was panned to the left -!.



just kidding, i just wanted to get in on page 1 ha!

[leedsquietman]

so which warp mode are you using.

[Hermanus]

Just let the fade case unchecked in the preferences>>> this way YOU choose if there are fades or not.

About some delay in the sample>>> it's often related to the audio card.
You can't reasonbly expect no latency at all with external usb audiocards.
get a pc and a good pci audio card if you want no latency.

Yesterday, I recorded old tracks I made about 10years ago.

I did in live for record through optical from my md desk to my usb audiocard
I tapped the tempo manually for some, but for other chillout productions with no beat, it was almost impossible to get the original tempo
so there I didn't worry.

Oce you unwarp the loop it's ok.
You can even launch your recorded sample in winamp to see the sample is accurate.
Avoid complex mode... I prefer the repitch for my own.

Man or you decide to stop and be frustrated by this kind of detail.
Or you go forward and find a way to suits your needs and avoid the problem.

Rendering audio, consolidating clips, even use external recorders... You have a lot of ways to make good music!

[clipperer]

hey mr.ergonomics, in a thread that you posted a long time ago you got 180 hits and no confirms, its because generally people dont give a fuck about things like that and i totally agree that in a serious daw this kind of problems shouldnt happen. yes live is not sample accurate, its good to see others posting about it.
ok, now about the problem, it has nothing to do with warp, live is shifting samples and sometimes it shifts to the wrong direction. to replicate this bug other way, put the same sample to 2 tracks, invert one track and consolidate one sample at different locations, youd get the same bug.

[comaduster]

Did you re-record this track with monitoring on the destination track? From my understanding, when you record this way live records 'what you hear', latency included (is this the case?). If you turn off monitoring it should be sample accurate. In your picture monitoring appears to be set to auto when recording. Try turning it to off and re-do the test. Hopefully I'm not wrong about this... hahaha, I'll test it as well.

[comaduster]

I just did it 10 times with monitoring on (on one recording channel) and monitoring off on one recording channel and was able to cancel out via phase reverse every time. That is scary if it is happening sometimes though.

[slow riot]

I have a friend who is particularly sensitive to rhythm who says that live is not especially tight (I think he said clips mode was a bigger offender).

i have noticed it several times but do not trust myself enough to rule out the placebo effect and have not verified this information with recording/cancellations etc etc.

[Tone Deft]

I have a friend who is particularly sensitive to rhythm who says that live is not especially tight (I think he said clips mode was a bigger offender).

i have noticed it several times but do not trust myself enough to rule out the placebo effect and have not verified this information with recording/cancellations etc etc.

what is 'clips mode'?

[Fizmarble]

I have a friend who is particularly sensitive to rhythm who says that live is not especially tight (I think he said clips mode was a bigger offender).

i have noticed it several times but do not trust myself enough to rule out the placebo effect and have not verified this information with recording/cancellations etc etc.

what is 'clips mode'?

I think it's what those in the underground call session view.

[slow riot]

@ tone deft. when sequencing using clips in session mode rather than on the timeline.