KVR Audio

DWb · Post by **DWb** » Tue Oct 02, 2007 7:04 pm

It'd be nice to get the entry-level DAW's in there too - Tracktion, FL, Reaper etc.

Although it'd be fairly remarkable if the 'pro' apps have super-secret magic summing algorithms to make them sound better and they all just happen to sound exactly the same as each other...

DWb · Post by **DWb** » Tue Oct 02, 2007 7:05 pm

Although I'll definitely be linking to this thread next time the standard argument comes up on another forum.

bduffy · Post by **bduffy** » Tue Oct 02, 2007 7:10 pm

TOTALLY.

Pipelineaudio · Post by **Pipelineaudio** » Tue Oct 02, 2007 7:25 pm

DWb wrote:It'd be nice to get the entry-level DAW's in there too - Tracktion, FL, Reaper etc.

Although it'd be fairly remarkable if the 'pro' apps have super-secret magic summing algorithms to make them sound better and they all just happen to sound exactly the same as each other...

REAPER is in there, it was actually the impetus for the test, and of course it summs fine

PaulG · Post by **PaulG** » Tue Oct 02, 2007 7:28 pm

Pipelineaudio wrote: Im waiting for a certain Nigerian Ambassador to transfer funds into my bank account, since you know, I sent him the transfer fee.....

Pipeline,

This test is bogus! You're a liar and a fraud! That Nigerian Ambassador sent the email to ME!

On the other hand, thanks for doing the hard work in what I couldn't clearly explain in my thinking that a host should only be a transparent pipe between the interface and the speakers.

I don't know why I think of it as such a big deal, but the second I saw the first "my host sounds better" thread, I couldn't help but think "then your host is broken".

Quick question, though...(to everyone)...don't host-related errors just result in noise or distortion? Could it express itself audibly other way?

Thanks for any input or answers,

- Paul

joshhunsaker · Post by **joshhunsaker** » Tue Oct 02, 2007 7:51 pm

Nuendo occasionally introduces a millesecond or two of delay to rendered files. That would probably explain why you are hearing really crappy sounding nulls when you cancel the two files...

Pipelineaudio · Post by **Pipelineaudio** » Tue Oct 02, 2007 7:52 pm

joshhunsaker wrote:Nuendo occasionally introduces a millesecond or two of delay to rendered files. That would probably explain why you are hearing really crappy sounding nulls when you cancel the two files...

We lined up the renders. But feel free to download the files and try it yourself.

joshhunsaker · Post by **joshhunsaker** » Tue Oct 02, 2007 8:07 pm

Interesting.

Something of a mystery here.

stefancrs · Post by **stefancrs** » Tue Oct 02, 2007 8:13 pm

joshhunsaker wrote:Nuendo occasionally introduces a millesecond or two of delay to rendered files. That would probably explain why you are hearing really crappy sounding nulls when you cancel the two files...

When I tested this a few years ago, I had to manually align the render from some version of Cubase on the sample-precision level. After I did that it cancelled out fine, before that it did not. Dunno if that's the case here.

DWb · Post by **DWb** » Tue Oct 02, 2007 8:24 pm

Pipelineaudio wrote:
DWb wrote:It'd be nice to get the entry-level DAW's in there too - Tracktion, FL, Reaper etc.

Although it'd be fairly remarkable if the 'pro' apps have super-secret magic summing algorithms to make them sound better and they all just happen to sound exactly the same as each other...
REAPER is in there, it was actually the impetus for the test, and of course it summs fine

Oh yeah. It's in big capital letters at the beginning of the list, making it a bit hard to spot.

Tallisman · Post by **Tallisman** » Tue Oct 02, 2007 10:47 pm

it would be cool if one of you cats with Nuendo would download the little test files and test it...

Or maybe you have? nudge the rendered files if you need but we'd like to see Cubendo Null or understand why it won't... Lots of speculation in this thread... lots of theory but no results yet. I have tested this to death in SX3 and C4 and it is broken. I tested it with the 3 track test, in 24bit and 32bit float. I tested a 24 track version 24&32bits. I tested a 1 track version using sound 1.wav, in 24 and 32bits... none of them nulled. I have zoomed in to align samples (which were aligned already). no Null.

I have done this in Live 5 (though i know not how to set the panlaw in Live 5) no null, but better results than C4 and SX3. I have done this in Reaper. NULLS!

I have tried to prove the methodology wrong, or at least make Cubase 4 work. I can't.

How does Podium and EXT do with this?

thanks
.t

Art Evans · Post by **Art Evans** » Wed Oct 03, 2007 1:08 am

Try a click file (a single sample with a value of say 1000) and mix it with itself in any DAW that seems suspect. Check the value of the mixed click. Should be a single click of sample value 2000. This eliminates any problem with alignment - you should be able to see very clearly that the clicks line up.

joshhunsaker · Post by **joshhunsaker** » Wed Oct 03, 2007 2:27 am

stefancrs wrote:
joshhunsaker wrote:Nuendo occasionally introduces a millesecond or two of delay to rendered files. That would probably explain why you are hearing really crappy sounding nulls when you cancel the two files...
When I tested this a few years ago, I had to manually align the render from some version of Cubase on the sample-precision level. After I did that it cancelled out fine, before that it did not. Dunno if that's the case here.

It would make sense, as both cubase and nuendo use the same SEQ4 engine...

djanthonyw · Post by **djanthonyw** » Wed Oct 03, 2007 3:19 am

my theory is that the same people that worry about this crap are the same people that never make any music

Cordelia · Post by **Cordelia** » Wed Oct 03, 2007 3:31 am

I think you might be right. I've had too much time on my hands since my computer fried and I'm working to earn another one. I'm frustrated and find myself passing time doing detailed tests comparing various compressors and EQs. It's fun in its own way, but it aint making music.

A REAL DAW summing test