Tracktion Software

.jon · Post by **.jon** » Wed Apr 27, 2005 6:57 pm

voi paska

Didn't hear the difference when listened at work, but now with my setup

I haven't tried with my plugs or anything, but in the sample provided... well there is a rather nasty problem. Probably not meaningful in a musical context on usual material, and my exports from Tracktion have never had any issues- but I have always listened to rather moderate levels.

I'll try here and report later.

DWS · Post by **DWS** » Wed Apr 27, 2005 7:00 pm

semiquaver wrote:I don't hear dither noise in the 24 bit renders - but the tail is identical either way. MJy guess: perhaps dither is being added to the truncated 16-bit file rather than mixed with the 32-bit file before truncation. Because I don't hear the expected benefit from the dithering.

Would anyone associated with 'Mackie' care to respond to this? Seems kind of important. - s

gavriloP · Post by **gavriloP** » Wed Apr 27, 2005 7:08 pm

I know I'm repeating myself like lousy delay fx but:
If I know anything I'm sure the transfer from 32bit floating-point values to 24bit shouldnt be as noisy as transfers from 32bit to 16bit, I mean WITHOUT any dither...

Mikko

IIRs · Post by **IIRs** » Wed Apr 27, 2005 7:09 pm

I just ran my own test.. it seems that there is indeed a problem with 24-bit files.

My usual working practise seems ok: leave Tracktion's dither turned off, and export at 32-bit. Use your audio editor to dither & truncate..

semiquaver · Post by **semiquaver** » Wed Apr 27, 2005 7:14 pm

thanks plat. (and Andrew for the thread) - tell the authorities !

I hope this is fixed soon...

gavriloP · Post by **gavriloP** » Wed Apr 27, 2005 10:50 pm

platinumears wrote:My usual working practise seems ok: leave Tracktion's dither turned off, and export at 32-bit. Use your audio editor to dither & truncate..

Yes, this seems like decent workaround, even though that too is different than pure TapeIt version at very silent level. For a moment I was really worried about my projects but now it seems it will be ok. Thanks for pointing that out.

pdxindy · Post by **pdxindy** » Thu Apr 28, 2005 1:41 am

gavriloP wrote:Yes, this seems like decent workaround, even though that too is different than pure TapeIt version at very silent level. For a moment I was really worried about my projects but now it seems it will be ok. Thanks for pointing that out.

How is it different?

rpc9943 · Post by **rpc9943** » Thu Apr 28, 2005 3:20 am

what precisely am i trying to hear?

Mr. Slater's Parrot · Post by **Mr. Slater's Parrot** » Thu Apr 28, 2005 4:54 am

rpc9943 wrote:what precisely am i trying to hear?

At the very end of each note in the version rendered by Tracktion there is a little (very tiny) fizzy noise just as the note is fading out. In the other version, each note fades out very smoothly (by comparison).

.jon · Post by **.jon** » Thu Apr 28, 2005 7:41 am

Ron,

You can here it by turning the volume up after the main sound, then turning the volume down again before the next note. I didn't have time to boot up my DAW yesterday, so someone could post a better sample - something with a lower initial attack...

gavriloP · Post by **gavriloP** » Thu Apr 28, 2005 9:52 am

pdxindy wrote:
gavriloP wrote:Yes, this seems like decent workaround, even though that too is different than pure TapeIt version at very silent level. For a moment I was really worried about my projects but now it seems it will be ok. Thanks for pointing that out.
How is it different?

There is same sudden drop into the total zero silence but this occurs on a very silent level (over -105db or so the meters tell me). The TapeIt version fades smoothly till the end... of course it won't really matter because it is so very low. There might be some noise too before the drop, I haven't checked but it is certainly inaudible. It just tells me that signal inside tracktion seems to be different than rendered one.

Mikko

semiquaver · Post by **semiquaver** » Thu Apr 28, 2005 3:46 pm

I want to have confidence in the quality of rendering in T. I need to know that when I render a track I am not introducing any error. So even if things are only wierd when there is very low level it could indicate some basic problems. I would love it if someone from Mackie's digital mixer department could work with Jules to verify the audio chain (including mackie's standard dither and noiseshaping) in T.

pdxindy · Post by **pdxindy** » Thu Apr 28, 2005 4:13 pm

semiquaver wrote:I want to have confidence in the quality of rendering in T. I need to know that when I render a track I am not introducing any error. So even if things are only wierd when there is very low level it could indicate some basic problems. I would love it if someone from Mackie's digital mixer department could work with Jules to verify the audio chain (including mackie's standard dither and noiseshaping) in T.

Yeah, it makes me wonder what other ways the sound may be affected, and sorta undermines a basic feeling of confidence that the sound is as clean and accurate as possible...

diverdee · Post by **diverdee** » Thu Apr 28, 2005 4:23 pm

It is indeed a little worrying, hopefully beta-testers, mackie personnel etc. are testing this with a view to confirming & finding a solution if neccesary.

IIRs · Post by **IIRs** » Thu Apr 28, 2005 4:25 pm

As far as I can ascertain, the problem is restricted to integer file formats: if you export at 32-bit your audio will be identical to the real-time playback. I would stake my reputation on it..

.. yeah i know what you're all thinking: "What does a reputation for smugness and arrogance have to do with this problem?"

Bastards!!

Tracktion Software

KVR Audio

Rendering in T is not as perfect as I thought