DISSCUSSION: The Nuendo/ProTools Conundrum

  • Thread starter Thread starter LeeRosario
  • Start date Start date
LeeRosario

LeeRosario

New member
A few months back, I posed an issue here that was really messing with me and quite honestly still is. After several extensive comparisons using just my god given ears, I could tell there was just *something* different about mixing in Nuendo and mixing in Pro Tools. There's something about the summing that's different to an extent. I had to really listen close and live with my mixes for a while. Also, I've been everywhere and ask anyone I could, but I got this:rolleyes: look everytime. No responses from steinberg and digidesign.

Anyway, I finally got a chance to use one of my recent home sessions (all permissions where given by the artist) and cranked out quick 2 minute rough mixes as identical as possible in both Nuendo and Protools. To compensate for panning volume between DAWs, I used the Roger Nicols Inspector Plug-In on the digital PPM setting.

The purpose is to open a discussion about stuff like this and if it's possible that digital bussing is not created equal. I'm not a software programmer, so I hope to learn about whats going on under the hood.

Here are the audio files that go with the discussion: http://www.soundclick.com/bands/default.cfm?bandID=648771

The Cycle (Nuendo Mix)
The Cycle (Pro Tools mix)
The Cycle (Final Nuendo Mix)
Burning June (Final Pro Tools mix. Not the same mix, however engineered in the same way I’ve been engineering for the last year)

Listening Notes-

Title: ProTools/Nuendo Comparison
Purpose: to discern any possible differences between daws
Track: The Cycle
Artist: Arthur and the Dent
Bit Depth: 24-bit
Sample Rate: 96,000khz
Location: Lee’s home studio


Nuendo

Center channel material translates clean, undefeated and maintains clarity with increased number of tracks. The overall mix sounds crisper, although just a little shrill at the higher frequencies (about 10khz and above). Stereo imaging seems to maintain very well through different level changes (sounds clean soft as it does loud, all the way through mastering).

Bussing: Bussing within Nuendo dosn't seem to produce any artifacts or changes to the original signal. The master buss tolerates a very high signal before clipping ratio and also dosn't seem to produce any artifacts.

Pushing the mix inside Nuendo eventually produces a "woofing" defeated sound much like over compressing in a limiter.

Processing: most plug ins sound clean and handle well.

Pro Tools

Center channel material becomes muddy and undefined with the addition of more tracks. Material panned wide left and right dosn't seem to produce any changes to the original signal. The mix tends to sound softer and pumpy and tends to be somewhat muffled and with reduced highend. Stereo imaging seems to degrade with the addition of more tracks. The mix seems to vary in quality at different levels. (seems to have a hard time during the mastering stage).

Bussing: Bussing in Nuendo seems to roll off high end frequencies. It seems that you can squeeze only so many tracks into one buss before you start getting clutter. Tolerates high volume to an extent. The master buss seems to become limited with how far you can push the mix.

Pushing the mix in pro tools eventually produces a very packed and punchy sound with signs of clipping. Without the help of a stereo imager of some type, stereo imaging seems to degrade with exessive volume. Sounds like pushing the mix through a compressor.

Processing: certain plug-ins sound clean and handle well.


Conclusions thus far

Nuendo produces a very clean and open sound. However, the mix can get really trebly and shrill unless proper care is taken. Overall, the mix seems to maintain well across the frequency spectrum, imaging, and volume.

Put simply: it just really comes out "wow". The mix *almost* comes out sounding as good as an analog console mix.

Pro Tools seems to do alright towards a certain extent. The mix seems to be iffy at certain levels. Overall, the mix seems to flatten out and clutter if pushed too far. Opposite of Nuendo, it can tend to sound muddy unless proper care is taken. I remember hearing on how Digidesign programmed a "softclip" feature automatically into the masterbuss (it can't be turned off), so I wonder if that has anything to do with it?

Put simple: sometimes (if I really bust my balls) I can get wow. Alot of times, it can get very "eh".




Any thoughts? Thanks for taking the time and interest.

Lee
 
Last edited:
I won't have a chance to give the samples a listen until sometime later, but in the meantime, a few questions/comments:

I don't understand how you are compesating for panning laws by using Inspector. Ae you simply trying to match overall levels? That may not be enough. Also, you might want to document just which pan law settings you are using in each package.

You should note which builds/versions of each software are you using, so we are all on the same page (is there perhaps a difference between Nuendo 3 and 4, for example?)

I'd also highly recommend that you give the clips generic names and mix up their order to try and at least somewhat blind the test for the rest of us. By giving your detailed description of what you ear in each and then telling us which is which, there is going to be a strong tendancy for us to trick ourselves into hearing the same thing you are, even if we don't.

HTH,

G.
 
SouthSIDE Glen said:
I won't have a chance to give the samples a listen until sometime later, but in the meantime, a few questions/comments:

I don't understand how you are compesating for panning laws by using Inspector. Ae you simply trying to match overall levels? That may not be enough. Also, you might want to document just which pan law settings you are using in each package.

You should note which builds/versions of each software are you using, so we are all on the same page (is there perhaps a difference between Nuendo 3 and 4, for example?)

I'd also highly recommend that you give the clips generic names and mix up their order to try and at least somewhat blind the test for the rest of us. By giving your detailed description of what you ear in each and then telling us which is which, there is going to be a strong tendancy for us to trick ourselves into hearing the same thing you are, even if we don't.

HTH,

G.
Click to expand...


Hmm good points. I've never really gone about testing DAW stuff, so it's kind of new to me.

As far as the Roger Nichols plug, that's the idea I had behind it. I realized how different the pan laws can be that I just need something to give me a close enough mix. The nuendo pan is set to -3 pan law.

As for the mixes, I'll be sure to shuffle them and change up the names to generic names.

Anything else I can possibly do to make this as neutral as possible, I'm wide open to suggestions.


Thanks for the heads up on that.
 
LeeRosario said:
Hmm good points. I've never really gone about testing DAW stuff, so it's kind of new to me.

As far as the Roger Nichols plug, that's the idea I had behind it. I realized how different the pan laws can be that I just need something to give me a close enough mix. The nuendo pan is set to -3 pan law.

As for the mixes, I'll be sure to shuffle them and change up the names to generic names.

Anything else I can possibly do to make this as neutral as possible, I'm wide open to suggestions.


Thanks for the heads up on that.
Click to expand...

Once you nail down pan law, you'll also need to make sure everything else is the same :D Like if you're using an RTAS wrapper for plugs in PT. I don't know what effect that has, but if the plugs are at all different, that would be a concern. It would probably be best to render all tracks with plugs in the same app, then compare the two different apps summing. But then you'll have plugs on a subgroup, that will blow it all to heck :eek:

I would think to be fair you'd also have to ensure that you don't clip the master bus, nor between plugs, nor subgroups. Nuendo should pass 32 bit float between all levels, assuming the plugs handle that internally as well you can do what you like. I do not believe that is the case in PT; certainly if their is an internal limiter on the master bus, that's a huge problem. So I think you would need to establish a mix that never clipped anywhere.

That's just for a strict summing engine test. You might argue that makes workflow a PITA, and you might be right. That is a valid selection criteria irrespective of summing engine performance.

One thing to be aware of when considering the psychoacoustic effect of summing lots of tracks digitally vs. analog is noise. Digitally, you will only be summing the noise that's already on the tracks. In analogland, you have the noise of each channel added to the noise printed on tape. That will have a masking effect on high frequencies, such that a digital mix might seem brighter. Hmm, I don't know if I've ever seen that discussed before :confused:
 
Make sure when you burn discs to compare you write along the outside of them with a green magic marker.

Also, throw salt over your left shoulder before touching the volume knob.
 
pipelineaudio said:
Make sure when you burn discs to compare you write along the outside of them with a green magic marker.

Also, throw salt over your left shoulder before touching the volume knob.
Click to expand...

Last time I did that it got all up in the cd burner. The salt is so unholy :cool:

Once you nail down pan law, you'll also need to make sure everything else is the same Like if you're using an RTAS wrapper for plugs in PT. I don't know what effect that has, but if the plugs are at all different, that would be a concern. It would probably be best to render all tracks with plugs in the same app, then compare the two different apps summing. But then you'll have plugs on a subgroup, that will blow it all to heck

I would think to be fair you'd also have to ensure that you don't clip the master bus, nor between plugs, nor subgroups. Nuendo should pass 32 bit float between all levels, assuming the plugs handle that internally as well you can do what you like. I do not believe that is the case in PT; certainly if their is an internal limiter on the master bus, that's a huge problem. So I think you would need to establish a mix that never clipped anywhere.

That's just for a strict summing engine test. You might argue that makes workflow a PITA, and you might be right. That is a valid selection criteria irrespective of summing engine performance.

One thing to be aware of when considering the psychoacoustic effect of summing lots of tracks digitally vs. analog is noise. Digitally, you will only be summing the noise that's already on the tracks. In analogland, you have the noise of each channel added to the noise printed on tape. That will have a masking effect on high frequencies, such that a digital mix might seem brighter. Hmm, I don't know if I've ever seen that discussed before
Click to expand...


Has anybody noticed any major differences between dry mixes?

The vocals where the only thing with processing already printed but that was done completely outside of both protools and nuendo.

I figure everything else I could leave alone to strictly focus on the effects of the bussing on the mix.

I think both definitely should have tons of headroom, which is why they sound so low, but I can always provide mixes that max out the master buss.

Of course, if we can decide that they warrant such a thing. "A testing all possible things" kind of thing.
 
here's a pretty thorough test

http://www.cockos.com/forum/showthread.php?t=13473&highlight=summing

Interestingly, all the daws summed correctly when left alone, but messing with the levels on cubase/nuendo could make some errors, this is already a known and reported issue at the nuendo forum though. Thing was, even this error was so far down that it strains belief it will be heard when compared with correctly summing daws

Another oddball was logic 8, which could summ correctly, but the fader levels were marked incorrectly. Once callibrated to actual levels, the logic mixes would some correctly as well.
 
pipelineaudio said:
here's a pretty thorough test

http://www.cockos.com/forum/showthread.php?t=13473&highlight=summing

Interestingly, all the daws summed correctly when left alone, but messing with the levels on cubase/nuendo could make some errors, this is already a known and reported issue at the nuendo forum though. Thing was, even this error was so far down that it strains belief it will be heard when compared with correctly summing daws

Another oddball was logic 8, which could summ correctly, but the fader levels were marked incorrectly. Once callibrated to actual levels, the logic mixes would some correctly as well.
Click to expand...

A good piece to the puzzle here.

Reading through your report: A little more renegade and aggressive than I would have approached it :D, but then again I think I've finally reached the point where I'm really burnt on it too. I've definitely lost alot of sleep over researching and countless listening tests, so the black rings around my eyes wants to open the blinds and let the bats out. :cool:

I guess the only clear cut, solid case and point I really have at this point is the difference between the final "The Cycle" mix done on Nuendo vs the "Burning June" mix done in PTLE.


Both where done literally months apart from each other, and in the same way I always run my sessions (everything from my mic selection, tracking, to the processing, mixing in the same room and mastering) and yet they came out wildly different to me.

I don't know, maybe I'm going crazy but Nuendo just seems to give me more room to work with in terms of the mixing inside the box realm.
 
pipelineaudio said:
here's a pretty thorough test

http://www.cockos.com/forum/showthread.php?t=13473&highlight=summing

Interestingly, all the daws summed correctly when left alone, but messing with the levels on cubase/nuendo could make some errors, this is already a known and reported issue at the nuendo forum though. Thing was, even this error was so far down that it strains belief it will be heard when compared with correctly summing daws

Another oddball was logic 8, which could summ correctly, but the fader levels were marked incorrectly. Once callibrated to actual levels, the logic mixes would some correctly as well.
Click to expand...

You never miss a chance to take a shot at Steinberg, do you?

"errors . . . known and reported issue" At the end of that thread on the nuendo board, it was demonstrated that entering fixed values of 6.02dB on the bus (closer to 20*log(2) than 6 is) and its negative on the master removed the "error", which would never be an issue in mixing actual music anyway.

Not a Nuendo user myself, and not gonna be a Reaper user either.
 
If you read my post I claimed that the error (yes its an error, don't try to strawman any motives on me) was so far down its not going to make any difference.
 
mshilarious said:
Strawman? Have you read your blurb lately?
Click to expand...

which blurb is that? the one with the pt icon getting sliced?

I can see how that could be misconstrued as a attack on steinberg...no wait

The blurb about how cool vst is in its portability? Or about how much I like ASIo?

which one?
 
pipelineaudio said:
which blurb is that? the one with the pt icon getting sliced?

I can see how that could be misconstrued as a attack on steinberg...no wait

The blurb about how cool vst is in its portability? Or about how much I like ASIo?

which one?
Click to expand...

The one right below your name! Y'know, I'm Jean-Luc Picard, you're "Official Shill" ;)
 
You can use anything Im shilling with cubendo though
 
Daisy said:
I didn't realize there was a conundrum.

Should I be concerned?

:D
Click to expand...


oh totally. defcon 5:cool:

Nah actually, I was thinking I should gracefully bow out of this one. Pipeline's got the testing down to literally a grain of salt. I figure I'll just stick to philosophizing about mics, and placement and that mysterious hot pocket heating technology.
:D
 
Lee,
I like to read discussions about such things. There are differences that some folk hear & other don't. There have to be differences in how things work anyway. It could come down to the inadvertant "colouration" we like/dislike in some mics or the "warmth" we aspire to from analogue equipment exists in their own, as yet not adjective applied form to some degree within the progs we use, prefer, are stuck with, can't abide.
I was impressed with the knowledge, calls for rigour, side bars etc that came up until the bun fight took over. Show that many minds make fairness harder but infinitley fairer.
 
rayc said:
Lee,
I like to read discussions about such things. There are differences that some folk hear & other don't. There have to be differences in how things work anyway. It could come down to the inadvertant "colouration" we like/dislike in some mics or the "warmth" we aspire to from analogue equipment exists in their own, as yet not adjective applied form to some degree within the progs we use, prefer, are stuck with, can't abide.
I was impressed with the knowledge, calls for rigour, side bars etc that came up until the bun fight took over. Show that many minds make fairness harder but infinitley fairer.
Click to expand...



I think thats exactly what it comes down to...what we have available and what we take from it.

Even if at this moment I'm unable to access in depth information on what makes these things tick, I definitely follow my ears, I like what I'm getting, which of course got me in this shin dig in the first place.

But of course, always in good taste. :cool:
 
You must log in or register to reply here.
Back
Top