Reasons for variable quality with Transpose?

rossc · Post by **rossc** » Thu Sep 15, 2005 4:45 pm

Something that I'm sure a lot of people must have noticed is that when trying to alter the pitch of a track in Ableton some tracks stand up admirably to it whereas others suffer really badly.

As a general rule of thumb I don't like to transpose a track more than 2 either up or down since the degredation in quality is usually very noticeable beyond that point. However, some tracks seem able to hold up to 3, or at a stretch 4 in either direction, whilst a few others i've played sound woeful even with just a +1 pitch.

Is there any explanation for this? I know that certain sounds, like for example rhodes chords don't fare to well when pitched, but I have heard some tracks with those type of sounds that seem to disprove that. I wonder if it's maybe something to do with the production values of tracks - maybe the amount of EQ or compression or is it something else entirely?

On a similar theme I also notice that tracks when converted to a low quality mp3 (say below 64k) also behave differently, with some retaining some sense of dignity even at very low quality whilst others sound really woeful. Is there a similar reason for this as with pitching in Ableton?

Thanks.

rossc · Post by **rossc** » Fri Sep 16, 2005 9:48 am

Does anyone know? I'm particularly interested to find out as I'd like to maximise the quality of my own tracks when their transposed.

Max Kachanoff · Post by **Max Kachanoff** » Fri Sep 16, 2005 10:28 am

Want to obtain a better quality - render them in SForge or Wavelab.

rossc · Post by **rossc** » Fri Sep 16, 2005 10:37 am

Yep, I guess that could work if I was looking to do a pre-programmed studio DJ mix for example, but if I was doing a live mix i'd ideally need to be transposing quickly within Ableton.

I'd also be curious still to know why certain tracks perform better than others when transposed within Ableton?

Max Kachanoff · Post by **Max Kachanoff** » Fri Sep 16, 2005 10:45 am

Well, I sometimes consider ableton pitch factor relating more to FX than a true shifter. If you need more - do the pre-proccesing.
I guess the difference depends on the clip itself - its type, polyphony, compexity, richness of formant content and quality maybe.

rossc · Post by **rossc** » Sat Sep 17, 2005 9:39 am

Yeah, I had thought that the complexity of a track could be a factor too. Thanks a lot for your replies.

jms5881 · Post by **jms5881** » Sat Sep 17, 2005 10:41 am

rosec:

does it have something to do with your sampling rate and plugins? for instance using a plugin usually decreases your sampling rate by way of changing your buffer size. So tracks with this additional processing wouldn't hold up as well to transposition.

what kinds of tracks are your transposing? it may have to do more with the differenced between the source material

inis · Post by **inis** » Sat Sep 17, 2005 11:32 am

Ask BT. If anyone has seen his video on M-Audio, he talks about how he is pitch shifting songs to make them all in the same key, or at least sound good chromatically, whatever. But as we all know, he uses live. He does a demo if two songs in the video. So, someone out there must have an answer for you.

rossc · Post by **rossc** » Sat Sep 17, 2005 12:59 pm

That's interesting. in terms of my own tracks i'm all software based, writing mainly house music on Cubase with Halion and various other VSTs + several fx plugins (just the usual delays, reverbs, etc). Up 'til now I've always ran my tracks at 16bit, 44.100 sample rate.

So you reckon that upping the sample rate could have a positive effect with this kind of thing?

Thanks for the info.