Sound Difference Quality In Logic, Cubase and Live.

[olafmol]

this low-end is coming from the soundsources you're using, or the processing you're doing

May well be... what I know is that I'm not making a conscious effort to add it in, so maybe others are experiencing this as well for whatever reason.

i don't see what this has to do with perceived differences in sound-engines? if your soundsource has lots of subsonic information it will be no different in different sound-engines

[Chris Cowie]

I used to use Pro tools HD and Mix systems and I can pick out a Mix recorded on Pro tools nearly everytime. It was always the top end harshness and weirdness at the low end that gave the game away. Perhaps its the DAE engine?

I also find I can get better mixes through SX. Maybe this is an ergonmics issue??.....Longer virtual faders etc have a positive effect on my mixing.

One thing I have noticed with Ableton Live is to many Warped sounds particularly drum loops with a lot of top end information creates an MP3 like quality to the sound. Very harsh, but to be expected when warping I guess. Warped bass loops can mess with the low end a fair bit. I now use as little warped clips as possible and I notice the mixes are a bit better.

Robert mentioned earlier that there is no difference between digital audio engines and Im sure hes right........except I would have to dispute that Pro Tools sounds the same as the rest. I dont like the sound of pro tools thats why I got rid of it.

The real test for me has always been to mix a track through a system. I cant tell any difference between those one stereo files tests that have been done to death.

Cheers

[raapie]

Here's a good test:
http://www.kvraudio.com/forum/viewtopic ... sc&start=0

I did this test with Live4.1.1 and Nuendo 2.2b39. Because Live can't export to 32 float I did a 24 bit export. This surely causes truncate errors in Live as well as Nuendo.

I compared the files in Wavelab and there seems to be a loudness difference and the low and midrange sounds different.

People seem to be right: Nuendo and Live do sound different. Haven't decided which sounded better, but you need different source material for this of course. I have to say I am very happy with the results I get from Live, but I have to admit: Nuendo and Live do sound different.

Setting per clip in Live for performing this test:
HQ on, fade off

[raapie]

mmm. the white noise export in Live sounds worse than the export I made in Nuendo.

The Live version lacks highend even while missing a bit of low and lowmids.

[olafmol]

Here's a good test:
http://www.kvraudio.com/forum/viewtopic ... sc&start=0

I did this test with Live4.1.1 and Nuendo 2.2b39. Because Live can't export to 32 float I did a 24 bit export. This surely causes truncate errors in Live as well as Nuendo.

I compared the files in Wavelab and there seems to be a loudness difference and the low and midrange sounds different.

People seem to be right: Nuendo and Live do sound different. Haven't decided which sounded better, but you need different source material for this of course. I have to say I am very happy with the results I get from Live, but I have to admit: Nuendo and Live do sound different.

Setting per clip in Live for performing this test:
HQ on, fade off

i guess you did disable warping?

Olaf

[elemental]

Well, not that it bothers me a lot, (only a little) but moving from SX to Live, I and others I played my tracks to noticed a 'lo-fi' quality to the sound ... SX seems to make it easier to produce really tight and clear mixes... but Live makes it easier to produce music. I also use less warped sounds, only when needed, esp on drums and bass, and have gone back to using recycle for most of my drums.
I've heard Logic users complain of a certain sound of the mix bus that they dont like, I see PT users here complain ... the grass is always greener, as they say...

Maybe lets all get a dangerous 2-BUS and STFU! (yeah right!)
Or a nice old Neve / SSL ... mmmm

[raapie]

yes, sorry I forgot to say: warp was off.

[raapie]

Today I phase-reversed the first white-noise track in wavelab and opened the original and phase-reversed track in Live 4.1.1. Played them both at the same time with Fade and Warp off. Because of phase canceling no sound can be heard. So summing 2 channels works perfectly.

I checked my test-files (Nuendo and Live) again and it seems that Nuendo and Live are not exactly on the same level. So probably sound difference are caused by level differences and the fact that there seems to be clipping occurring. Didn't change the master level on the original test since on the KVR-site they did the same thing. But I found out that I needed to change the level or else serious clipping would occur. A change of gain of about -20 db by the way.

So what? Well my tests are not correct that's for sure. My listening test proved a sound difference but this was caused by clipping and level differences. In short: I am testing the DA convertors of my soundcard (RME Hammerfall Multiface) instead of the rendering engines of both Live and Nuendo.

Maybe I will put more time into this to really test it properly but I guess summing in Live works as perfectly as in any other DAW, it cancels out any phase reversed track. So I don't need more proof that a non-warped track in Live sounds as good as any other DAW.

I left my original messages intact, so I do hope anyone reads this last reply ... I have always loved the interface of Live and it's workflow which I prefer anytime over Nuendo. The phase-cancel test proofed me Live is sample accurate, asking for more is not possible

One of the features of Live which I've never seen in any other application is the ability to record any source, live-input, effects, vsti's, rewire in real-time as audio without the need to stop the track to make a rendered version.

[globalgoon]

quod erat demonstratum

[3phase]

this low-end is coming from the soundsources you're using, or the processing you're doing

May well be... what I know is that I'm not making a conscious effort to add it in, so maybe others are experiencing this as well for whatever reason.

in a digital recording chain unwanted LF is easyly summing up because the in the digital domain you can mix DC values..equals a frequencie of zero with the amplitude x.... Therefore its very usefull to have lowcuts in most of your tracks during a digital mixdown....

This is by the way usefull in a analog mixdown aswell and therefore all big consoles have tunable lowcuts on each channel... But even without lowcuts, an analog mixingdesk channel acts as a lowcut just by its inbuild coupling capaciators...
So the LF problem is automatically a bigger one in the digital domain.

To have the LF only in the signals where you realy need it seriously clears up the sound...And its usually a smaller problem to add lowend punch during the mastering than removing disharmonic LF clowds.
Like allways in audio do a bit less than your brain tells you to do and you are usually fine...

[supster]

everyone drop the engineering talk and blah blah blah

somebody post the blindtest files if you have the gear and can do it right

lets hear what people are talking about once and for all.

otherwise *yawn* ...

[3phase]

Today I phase-reversed the first white-noise track in wavelab and opened the original and phase-reversed track in Live 4.1.1. Played them both at the same time with Fade and Warp off. Because of phase canceling no sound can be heard. So summing 2 channels works perfectly.

.

That is interesting because it indicates that within lives digital domain everything is allwright while the output to the soundcard is causing troubles.. At least on some systems as this thread shows...
Somebody questioned human perception...maybe...but as an audioengineer that had to measure and allign complete TV stations i can tell you that the ear is most of the time a better trouble shooter than the best pro audio analyzer...
And we have a situation here where some people , like me question the quality of the outputs of Live... Even when i agree with everybody here that Live is the best program to create music its a nightmare for me to mix music with it...
The Question is if rendering might be a trick to overcome this phenomen..
Sofar i rarely used it

All the threads and update reports about errors of the rendering dont brought me in a situation where i liked to try it for mastermixdowns...
I will test it... maybe the rendering sounds better than the master output of Live...but it shouldnt be like this.

[mikemc]

Are 0db reference tones available? Also, 3phase, thanks for the pointer about LF and digital, this is what I'm referring to.

[raapie]

for the test files see one of the message above linking to the kvr-site.

with my equipment when I need to rebalance the levels between 2 tracks I hear differences on white noise material. it's very small, but it's their. I need to rebalance the gain compensation using my main amp (a Rotel HIFI).

I am sure that if you're using the same levels you will get the same results in Nuendo versus Live. It seems that a -20 db in Nuendo or a -20 value in Live are causing different levels when I check my digiCHECK.

I am no longer interested in further testing. phase test is good and my general results with Live and great. but anyone who's willing to do it, please, go ahead.

[supster]

for the test files see one of the message above linking to the kvr-site.

I think what we're looking for are actual short music wavs or aiff (dont need to be more than 10 seconds in length) where:

- all of things being equal

- same file recorded in Cubase, PT, Logic, Acid, Live

- nothing but a code to identify where it came from


No offense to anyone in particular but these online debates often end up more like wank sessions where most posters are trying to show everyone and themselves how much they know about the subject at hand

and nothing ever gets resolved

someone that has access to all of the hosts - again, knows what they are doing - and really cares enough about this should prepare the tests
and distribute them

then we will see if we can hear a difference