Nuendo's Digital Mix Buss

  • Thread starter Thread starter LeeRosario
  • Start date Start date
i dont write code or anything like that so my limited understanding is that the principal diff between nuendo/cubase and PT HD (not le???) is they both do double pricision but steinberg uses a 32bit float and PT a 48 bit fixed....

I always understood LE does 24-bit and HD works on 48-bit. Which sucks, cause I'm not about to spend 10grand on a personal HD system just yet. :D

I'm just confused because you say you don't want answers about summing when you're asking about summing.

Specifically Nuendo's, which they seem to guard that information really well.


BTW, the audio engine in Cubase SX is the same as Nuendo. You can save $1000 or so.

I've heard the difference too, I just figured it was the difference between pan law and the different plugins.

Where I first heard about this Nuendo thing...

I remember a couple of years back, I was working with this producer in tampa, who was an ex-Nuendo licensed dealer and he went on and on about Nuendo and how it sounds. Eventually he ended up mastering all his projects in this thing. In my head I'm like, "man this sounds like a crock of shit, man".

So he gave me a copy, brand new box and everything, that I never gave any thought to.I finally got around to installing it a month ago. That first listen sparked the whole thing.

Hey Farview, did you ever get a chance to read on what Cubase/Nuendo does or if it even does anything different at the summing stage? It'd be cool to find out whats happening under the hood. I understand your bussiness is centered around Nuendo and the Ghost console. I assume you print your tracks from the console's master buss or do you feel that Nuendo sounds about the same? Or maybe it's in the file encoding?

It's one of these rare occasions that I can't seem to really come up with a good conclusion, I guess mostly for personal knowledge.


I even looked up Chuck Ainlay, I know he's an avid Nuendo user, but he pulls his mixes from the SSL. Also, I read the producer for the last Breaking Benjamin album reportedly used Nuendo for most of that project. No luck there.

It'd be nice to have some piece of mind about this, I'm not usually willing to mix inside the box on a paid project like this.
 
Last edited:
I am very reticent to step into this thread; I'm still exhausted from the phase/polarity thread, which is another one of those never-ending topics with stubborn people on both sides. I'll just throw this in to piss everybody off ;):

Make sure when talking about how two different apps "sound different" that you specify in your own head just what you're listening to. I understand completly and don't dispute the arguments and findings of those standing upon summing tests and null tests, or those pointing out the whole panning laws thing.

However, I have heard differences between apps that seem to have nothing to do with either the summing engine or the pan laws. Namely: Take a straight WAV file and just play it back in the different apps. No panning (or panning laws equalized) no summing (in fact, no other tracks even in the editor.) Just a straight playback.

A double blind would be best for truely testing this, but I would swear based upon previous experiences that I've detailed in other threads that one can often hear a subthle but qualitative difference in the sound of the playback. This does not necessarily mean that they are editing or summing anything different, just that the playback sounds different.

Now, as I've said before, I will be the first to admit that what I just described doesn't make much sense to me; I don't understand *how* that could possibly be. But maybe there's something going on outside of the summing engine and the actuall editing engine, in the monitoring/playback that is causing a perceived difference.

The test I'd like to see is not a summing/null test of what comes out of the digital editor, but rather a test of what is coming out of the monitors.

I am 60% dubious as to there actuially being a difference, but the other 40% tells me that it's a difference I have expereinced first-hand that, based upon my experiences, seems unrelated to pan laws, the summing algorithm, or phychological bias.

I have no definitive stand here, just adding another variable to the equation for everybody else to fight and scrap over ;).

G.
 
I am 60% dubious as to there actuially being a difference, but the other 40% tells me that it's a difference I have expereinced first-hand that, based upon my experiences, seems unrelated to pan laws, the summing algorithm, or phychological bias.

I have no definitive stand here, just adding another variable to the equation for everybody else to fight and scrap over ;).

G.

OK, in light of my thread proving no difference between the sound of summing engines (except for the broken cubase/nuendo summing system which really doesnt sound MUCH different if they are nulling to -99, which is lower than 99.99999999999999% playback environments will EVER reproduce)

even still

and you'd think I'd be the LAST person to say this, but:

You arent nuts

Playback and render, with timebased plugins especially, but just disk caching/buffering issues in general can make a difference.

Toss out the timebased effects and the difference is probably too subtle to be played back on 24bit converters, but it CAN be there
 
You arent nuts
Well, I wouldn't go THAT far, pipe ol' boy. :D
Throw out the timebased effects and the difference is probably too subtle to be played back on 24bit converters, but it CAN be there
What I have heard was indeed subtle. Like I explained it once before, it was like the difference between two slightly different converters, or maybe even the difference between bad clocking and quality clocking. I'm not saying that was the CAUSE; it wasn't. The comparisons were on identical systens with identical gear and identical sources all the way through. I'm just saying the the perceived difference (whether it was real or imagined) was in that class of difference; slightly different focus to detail, some seeming a bit more veiled than others. Easy to miss if you're not listening closely.

The kicker for me that keeps me from just dismissing the difference out-of-hand was both another person and I noticed it completly seperately of each other on our own systems in our own homes, and kind of dismissed it until the subject came up in a phone call some time later when we,both basically said, "You noticed it too?" Not exactly a double-blind test by any means, and it still doesn't prove anything. But it is a curious enough of a situation for me personally that I have to wonder if there *is maybe* some real component to the difference in perception that is outside of the pervue of the summing area of the software.

G.
 
Playback and render, with timebased plugins especially, but just disk caching/buffering issues in general can make a difference.

Toss out the timebased effects and the difference is probably too subtle to be played back on 24bit converters, but it CAN be there

I would think all bets would be off as soon as any plugins are used due to delay compensation, right? I have no idea and kind of don't care:) I'm just curious. I'm a Cubase user, so I don't worry about delay compensation at all but I assumed that each DAW would vary slightly in this area.
 
I would think all bets would be off as soon as any plugins are used due to delay compensation, right? I have no idea and kind of don't care:) I'm just curious. I'm a Cubase user, so I don't worry about delay compensation at all but I assumed that each DAW would vary slightly in this area.

that would make an interesting test :)

I THINK, though I cant say for sure, that any not broken would come up the same...but we should certainly try...See if you can think up a freeware plugin with lookahead, that should throw a serious monkeywrench into the loop
 
that would make an interesting test :)

I THINK, though I cant say for sure, that any not broken would come up the same...but we should certainly try...See if you can think up a freeware plugin with lookahead, that should throw a serious monkeywrench into the loop

Maybe download the 15 day demo of Sonnox Limiter?
http://www.sonnoxplugins.com/pub/plugins/support/demos.htm

I think there are other freebies out there, but they're mostly home brewed so maybe not reliable.
 
BTW, the audio engine in Cubase SX is the same as Nuendo. You can save $1000 or so.

I've heard the difference too, I just figured it was the difference between pan law and the different plugins.

BINGO!!!!!!
 
I suppose that there MIGHT be some audible differences in the way the audio is routed to/from plugins....ie...is this routing done 24 bit fixed of 32 bit float? Possibly 48 bit?

Who knows.

The differences I have heard from DAW to DAW are SO insignificant in an overall way that I just can't possibly use that as a basis for choosing the DAW I will use.

Double blind tests are the ONLY way you can for certain say there is a difference. Even the 3daudioinc.com stuff isn't double blind.

Really.
 
I suppose that there MIGHT be some audible differences in the way the audio is routed to/from plugins....ie...is this routing done 24 bit fixed of 32 bit float? Possibly 48 bit?

Who knows.

The differences I have heard from DAW to DAW are SO insignificant in an overall way that I just can't possibly use that as a basis for choosing the DAW I will use.

Double blind tests are the ONLY way you can for certain say there is a difference. Even the 3daudioinc.com stuff isn't double blind.

Really.



Just a theory and strictly going on instinct.

It's definitely rare that I go on a personal crusade to discover what it is about a program that's giving me a different result than what I've experienced before in a DAW I've been using for most of my young career.

Plus of course, you pose a good question. It may be just a 32bit float processing issue, but then again, in lacking that software programming knowlege, I can't figure for sure what happens to the life of that signal once the plugs are inserted.

Like would it truely matter if you take a 24bit track, throw it into a 32bit session, apply processing that should "theoretically" continue to stay 32bit, weighing the plug-ins design integrity, and pump it out in a 32bit float format? I'm wondering if that is truely explaining the improvement in clarity. A big part of me just can't believe that's a complete story. There's gotta be something else going there.

[edit]Another theory I'm comming across is in the way Pro Tools soft clips at the master buss inside the software. I know in LE, you get something like 3db headroom (beyond slamming the master buss) before you really get noticeable clipping. In HD, I can't really compare, because I've only ever run HD through a console completely bypassing the summing algorithms altogether. So I personally can't conclude if it's a bit depth thing or something else.[edit]

Plus I remember reading a comprehensive debate and study on digital summing and analog summing almost 2 years ago which came to a hazy conclusion and left me wanting more, which I think was a mix mag article. Multiple respected engineer's, great listening environments, the works. It never answered differences, if any, between DAWs. Just digital summing vs analog.

So in this specific case, the only conclusion I can draw so far is LE seems to smash much more aggresively when you slam the master.The attack sounds like it rounds off alot more aggressively on top of the imaging closing up the harder you slam that buss. Not that I'm in a habit of doing that. I just personally think it's important to push the limit to see exactly what I can and can't do.

I'm not feeling I get that fatigue as much in Nuendo. Plus I'm finding Steinbergs peak master plug seems to let me cheat a couple of extra db in there if I really need it.

Slamming the master in Nuendo just gives me these totally different results. It'd just be nice to know what exactly the programmers had in mind in that situation from both Digidesign's perspective and Steinberg's perspective. Those being the two DAW in question.

I can't agree that a double blind test is the absolute and only way, but multiple accounts does add great insight to the overall picture. I'm not convinced as of yet though. I still feel I probably will end up going with Nuendo, just for the ease of cross training and willing to bet on something new if anything.

I'll continue to ask around, perhaps something else will play up. Either way, I do have the deadline to meet, so hopefully I can post up some sort of result once the thing is said and done.

Great insights so far.
 
Last edited:
I dont understand the point of going over the master
 
Why would you be slamming the master? Other than to test how gracefully the summing buss fails under poor gain staging
 
Why would you be slamming the master? Other than to test how gracefully the summing buss fails under poor gain staging

Oh of course. No debating that.

I guess it's just more of a personal thing. I think probably because I had done a couple of mixes on LE in the past that sounded kind of pleasing to me when I pushed them a little bit into the master buss. If it sounds like ass, then I definitely don't go that route.

The best thing I can compare it to is like using the softclip feature on steinbergs dynamics plug on the entire mix. That's something I might try on this project. I was curious to see how Nuendo handled that. Thats where I found there where differences.

Of course, I could be full of shit. I might just be hearing something that's not really happening.


The other question I had Farview, since your familiar with Nuendo/Cubase:

Does Nuendo do anything different in monitoring playback that LE dosn't? I mean I wouldn't see why it should, if someone uses the same audio card for all applications. I just still don't get it. I litterally passed another mix from protools into Nuendo 5 minutes ago, no processing, no panning; absolutely nothing, and I hear things better than in LE. For example, the kick and snare tracks retain thier crack and clarity and stay firmly audible behind 40 something other tracks (guitars, synths, vocals, etc).

Does that make any sense?
 
I've kind of noticed the same thing you are describing. I always assumed that it was the Digi interfaces that were doing it. I've never been in a position to use Nuendo with a Digi interface.

I also really don't come anywhere near 0dbfs on the mix buss, so we may not be hearing the same thing.
 
I've kind of noticed the same thing you are describing. I always assumed that it was the Digi interfaces that were doing it. I've never been in a position to use Nuendo with a Digi interface.

I also really don't come anywhere near 0dbfs on the mix buss, so we may not be hearing the same thing.

Definitely understandable. I mean, all I really have here is a shitty 002 rack, so I'm probably not even giving this listening test a fair chance.

I'm still pretty stoked about trying some mixing with Nuendo. So when this thing is done, whenever that is, I'll be sure to pop up in the mixing forum and throw up some old LE mixes with the new ones.

I really appreciate you voicing your opinion. Goes for the rest as well.
 
No pan law, etc etc. What I was getting was better detail throughout the soundscape without hazzyness down the center channels. In fact, especially the center channel material. I even dumbed it down to 24-bit to even it out with LE, which is what I'm used to using at home anyway.
You do realize that Pro Tools LE uses a -2.5dB pan law, so everything you pan center is 2.5 dB down in PTLE compared to Nuendo, where you are using "no pan law"?
 
You do realize that Pro Tools LE uses a -2.5dB pan law, so everything you pan center is 2.5 dB down in PTLE compared to Nuendo, where you are using "no pan law"?
Pan law in Nuendo is user selectable. It's in project setup.
 
It would be interesting to hear the results with the pan law set up the same way in Nuendo as it is in PT, as he said in the post I quoted, he was using no pan law in Nuendo, which would make mixes sound quite different up the middle of the mix.
 
It would be interesting to hear the results with the pan law set up the same way in Nuendo as it is in PT, as he said in the post I quoted, he was using no pan law in Nuendo, which would make mixes sound quite different up the middle of the mix.

Using the same pan law in PT as in nuendo, mixes null 100%, unless you move a fader, at which point they null to like -99, which is still jack diddly nothing
 
Back
Top