Question about the old ProTools 'small' sounding mix problem

[gubu]

Originally Posted by Lee Flier

No. Like I say it's been obvious to me for years and it really becomes more apparent the more tracks you have and the more processing is going on. But even unprocessed, there's a difference.



Are you talking about Lynn Fuston's Awesome DAW-Sum test? If so, Lynn is a guy, and there definitely were differences in that test, although all the audio was unprocessed and, IIRC, there weren't many tracks. Plus I don't think 64 bit engines had come along yet then.

Ah, I just assumed that 'Lynn' was a woman.

I was reading up last night on pan laws, DAW engines and the comparative tests that have been done on different DAWs, and a couple of things struck me.

First off, there is a lot of writing done about the fact that null tests 'prove' that all DAW's handle audio identically, and will sound the same.
But, those tests are really quite simple, and don't factor in plugins, aux busses etc.

So the question with regard to null tests is:- would 2 full sessions on a different DAW each, mixed with identical pan laws, pan positions, fader positions, aux send/returns, and plugin parameters produce a null in one of these tests?

I highly doubt it.

The other thing is pan laws. What's the point of a non-user-adjustable pan law in a daw (or indeed any control over pan laws), when the system already gives you control to the nth degree over trims, levels, balance, and everything else that affects the pan position and level of a signal?

I don't see the point of that. It seems like an ok idea (and is used by plenty of engineers, I'm sure) if the law is user-adjustable to any value, but fixed by the system?

In any event, knowing that PTLE 7.4 uses a pan law of -2.5dB might go some way towards explaining why mixes often translate as 'narrow'.

[gubu]

Ok, well #1 is Reaper, and #2 is ProTools.

It just goes to show!! Not that it was any kind of scientific test in the first place.

So, in the meantime, I've been reading mountains of material, including many of the docs here at HC, and apart from the pan laws, there should not be any quantifiable difference between different platforms.

Maybe not Again, the null tests etc. use really quite simple setups to 'prove' that everything is identical, so I guess it's better to just pick the DAW that works for you and use it. The proof is in the final mix, and the truth is that the more time and effort I spend in getting a mix done, the better it sounds in the end.

About Reaper, I just don't like it. Too many reasons to mention, but editing workflow especially is not anywhere near as quick as with PT (even on my older version of PT), and PT is way more stable on my machine (again, old, so YMMV). I thought I could hear some sonic differences for the better between it and PT (the bottom end is different, to my ears), but as I described earlier in the thread, this was nowhere near a scientific test, what with the different reverbs in each example and whatnot.

Anyway, I'll be sticking to PT for now. And will try not to resurrect these 10 year old conundrums again anytime soon!

What can I say? I live in the far northwest of Ireland, at the edge of the known world. News can sometimes take a while to get here

[WRGKMC]

I use several DAWs off an on. When it comes to the finished product I honestly couldnt tell you
which song was recorded on a different DAW program, even if thet the same song, same tracks.
The GUI can fake you into thinking theres a difference but for most of the basic items, panning,
trims volume, there isnt any difference at all. Plugins, sure. Anothther routing options, busses etc,
there may be minor differences there as well as the actual meter responces.

You want a big change of sound quality, its all in the preamps and converters.
A DAW program has zero effect on tracking unless you're preprocessing the signal through the DAW mixer.
Thre signal goes through the converters and straight to the hard drive. The only time the DAW comes into play
is when you're mixing.

I used to think Cubase sounded different than Sonar. Cubase seemed to sound plusher to me and Sonar sounded
more direct when I was mixing. I thought thay might have used different wiring archectecture. (not that wiring applies to software
but you get my meaning) The tracks played back through Cubase sounded like the signal had less processing before the signal was
sent back to the interface and monitors and had a more analog sound quality. Sonar sounded flatter. Every stage weather it passed through
a buss or mains remained the same.

I thought the plushness carried through in cubase to the mixed down file. That was untill I did identical mixdowns with
dry tracks on both DAW programs. Then comparing the two mixdowns both with my ears and overlaying the two in Har-bal,
a frequency analizer that will let you take two recordings and superimpose one over the other, there was virtually no difference
in frequency content.

The rest, how a program feels working with it, ease of editing, stock plugins, GUI, colors etc are what make most people
decide which program they prefer to use regularly. I will say its good to work with several different DAWs, if only to expand
your experteese, but to be efficiant of as many sustems you can. Most have simular tools, and some have cool tools others dont.

[gubu]

Thanks WRG,

Ya, I've used different DAWs on different platforms. Cubase, Adobe Audition etc. on PCs and Macs.

The point of this exercise was to try and gain some sonic advantage from changing to a different DAW. It seems that Reaper may have some sonic advantages on/after mixdown, but the slower workflow and decreased stability on my machine compared to PT, along with some other glitchy behaviour, makes it a non-runner for me.

For now, I could see myself importing 2 track mixes/single tracks into Reaper to take advantage of a few plugins that don't have RTAS support. I used to use Adobe Audition in this way a lot when I had it, particularly for the 'spectral edit' feature that Audition has. But, for the same money, I could get a VST-RTAS wrapper and forget Reaper altogether.

I'll definitely try some other DAWs whenever I get to upgrade my machine. Cubase has always made sense to me, and I like the look of Nuendo, even tho it seems extravagantly priced. So, you'd never know.

Also, your point about preamps and converters is spot on. I often wish that I'd bought a few less high $$$ plugins back when I was rich, and spent the money more wisely on 3 or 4 channels of good preamp with/without S/PDIF out, and a couple of better mics.

[Lee Flier]

Ya, no question that converters and preamps make a bigger difference than DAWs. And yes, I've seen all the evidence that there "should" not be any difference between DAWs but at the end of the day, there is. Not if you're just doing raw summing of tracks with all the faders at 0, but once you start doing processing, there is. So whether you notice the difference depends a lot on how many tracks you have, how much processing you're doing, etc. and of course that's a lot harder to match between DAWs. A null test does show differences once you get into an actual working situation, so no one can really claim that two DAWs do all their DSP in exactly the same way. They don't, so the differences at that point are subjective. For a lot of people it's just going to boil down to workflow and user interface, as it does for you. For some, there are sonic differences that are noticeable. For me personally, I don't really care which DAW I use from a workflow perspective - I don't do a ton of editing, for instance. They all do pretty much the same things as far as I'm concerned, given the way I use DAWs. So it's ALL about the sound for me, and in a real-world work environment I have always preferred the sound of Reaper, Nuendo and others over PT.

[BushmasterM4]

But Lee,,, in this test you thaught 2 was Reaper and it was PT's Just saying And yes you were listening on crappy speakers, but 90% of listeners use, well, crappy speakers. I use 3 DAW's PT and Reaper being two and other than panning, there is no audible difference, well, IMHO.

[UstadKhanAli]

Test 2 sounds more vibrant to me. But it also sounds like a different mix.

[gubu]

Originally Posted by UstadKhanAli

Test 2 sounds more vibrant to me. But it also sounds like a different mix.

Yes, test 2 is a slightly different mix.

Different reverb (I don't have that reverb as a VST), which might explain the vibrancy. But then, listen to the build up of mush as the solo starts, compared to Test 1.
Is this the reverb, or the DAW?



And different processing on the bass guitar, because I hit the limit of my machine with less plugins on Test 1.

So, as I said above, not a scientific test, by any means. More of an attempt to find some sonic advantage from a different DAW, because I know plenty of people who moan about the sound of a bounced mix in the older versions of PTLE, with some justification, possibly.

[Lee Flier]

 

Originally Posted by BushmasterM4

But Lee,,, in this test you thaught 2 was Reaper and it was PT's Just saying And yes you were listening on crappy speakers, but 90% of listeners use, well, crappy speakers. I use 3 DAW's PT and Reaper being two and other than panning, there is no audible difference, well, IMHO.

 

I don't think most people routinely use speakers as crappy as the ones in my laptop. And they are different mixes, so it's really not a good test. I've done much better tests, enough to know there is a difference, and not just on great monitors - I could hear the difference on desktop computer speakers or typical earbuds. It's not that obvious on some types of material, others it is. If you can't tell the difference in your own work, that's cool and I believe you. You might not be pushing the envelope of stuff that highlights the differences. I don't always either.

[CME]

And just want to point out that pro tools has revamped their native mix engine. Since Pro Tools 9 it uses a 64-bit float mix engine. Before that it was a 32-bit float mix engine. Just something to think about.

[gubu]

Originally Posted by CME

And just want to point out that pro tools has revamped their native mix engine. Since Pro Tools 9 it uses a 64-bit float mix engine. Before that it was a 32-bit float mix engine. Just something to think about.

Indeed.

Unfortunately, I'm still stuck with and old Mac, so I won't be upgrading to a 64 bit DAW any time soon.

In fact, I'm sure a lot of what I've posted in this thread is pretty irrelevant to many here, due to the age of my rig

[russrags]

Hi gubu,

Hey neat song, I really like it. Here's what I've figured out, I go with Pro Tools as there are more commercially available Studio's around the World that use PT than any other DAW. If your completely in house use whatever, but if your going to freelance and network, PT and Logic are your main options.

Old Mac's: There have been more HIT records recorded on old macs than any other format in the digital age. I often discourage people from thinking, "if I just had this, or if I just had that," try and make the most out of what you've got, and let that equipment KNOW who's boss. The equipment works for you, not the other way around.

Stereo width: yes the sound stage has gotten smaller in the digital era, but that's mostly due to records being made at home. Back in the days of Analog (Big Machines) they were primarily used in commercial studios with bigger rooms and microphones. You've got to have a reference of that big sound, before your going to be able to duplicate it at home. I do a lot of recording and mixing at home, but at some point on nearly every project, I'm down at a big studio re-amping and using microphones to make things bigger.

Over the years my mixes get better and better, but there is always more and more work involved to get there. You've just got to work it, and then work it some more, and squeeze out every last drop of blood.


Russ
Nashville

[gubu]

Thanks Russ. Words of wisdom there for sure!

I think I'm on the path you describe. Yes, I GAS for this or that preamp, or monitors, or for a more powerful machine. But I know that I've not achieved the very best that can be achieved with what gear I have, so it means having to 'work it, and then work it some more', and finding more inspiration to keep pushing things to get that extra .01% of quality into each finished project.
Maybe I'll never produce top notch commercial material, and deserve the kind of gear that I dream about. But even if it takes the rest of my life, I'm sure going to enjoy trying

Considering that Sgt. Pepper was recorded on a bunch of 4 track machines, it's obvious that the material, and the chops that the personnel bring to a project is far more important than this DAW, or that plugin suite, or preamp, or whatever.

At the same time, it's no bad thing to try and get the most you can out of the gear and chops that you do have, which is sort of what this thread was about, I think!

Anyway, I'm happily working on that track, and some others, at the moment, trying to improve and 'squeeze out every last drop', in the full knowledge that the final results are in my own hands.

[Zooey]

Originally Posted by gubu

Indeed.

Unfortunately, I'm still stuck with and old Mac, so I won't be upgrading to a 64 bit DAW any time soon.

Pro Tools is still a 32 bit application (even version 10). He was talking about the mix engine--totally different issue.

[gubu]

 

Originally Posted by Zooey

Pro Tools is still a 32 bit application (even version 10). He was talking about the mix engine--totally different issue.

 

Yes, we were talking about the mix engine - sorry if I wasn't clear, but I won't be running a 64 bit mix engine on that old Mac any more than a 64 bit DAW

[Zooey]

 

Originally Posted by gubu

Yes, we were talking about the mix engine - sorry if I wasn't clear, but I won't be running a 64 bit mix engine on that old Mac any more than a 64 bit DAW

 

If you ran Reaper on your Mac, then you've run a 64 bit DAW. But I think you said you were running out of processing power on that mix.

[gubu]

Originally Posted by Zooey

If you ran Reaper on your Mac, then you've run a 64 bit DAW. But I think you said you were running out of processing power on that mix.

Well, yes.

I seem to remember that it gave various mix engine options, but it wouldn't even playback in 64bit. I had to switch to 32bit float.

[Mixerman]

Originally Posted by gubu

Running PTLE 7.4 on an old Mac.

First off, I know I'm pretty low on the curve as an engineer compared to most. But this is not a 'blame the tools' thread. Just something I've been scratching my head over.

Lately I mixed some material in which all of the elements were sonically well above average. From the perspective of harmonic content and dynamics, the mix stands up well in comparison to commercial material.

But, it doesn't compare well with commercial material in terms of stereo width, and low-level/high frequency fidelity (as best as I can explain it), making the whole thing sound 'small'. Also, the rough demos (with little or no processing) that I had been exchanging with a collaborator who uses a different DAW (also with little or no processing) don't compare well either. The PT versions were much fuzzier around the edges.

I know at least one other forumite here agrees that PT systems are prone to this problem, and I know a well-respected producer here in Ireland who has abandoned PTLE altogether, in favour of Logic, for exactly the same reason.

Also, I've heard murmurings on these very forums that the mix buss architecture in PT leaves a lot to be desired, in comparison to that of other 'pro' platforms.

So, my questions are these:-

To those of you who have had a similar experience with PT, is this just an issue with PTLE, or is it also a problem on HD systems?

Has this issue been fixed/modified at all on later PTLE releases? It's up to version 10 now, right?

Is there a workaround for this problem in the older versions of PT?

The best workaround I've found is external summing, and it pertains to all DAWs that I've used. Perhaps less on some than others, but it's an issue regardless.

There are all sorts of summing boxes on the market, and there will be many people on the internet who will argue that they hear no improvement between summing ITB and summing OTB. I, on the other hand, have successfully demonstrated this problem to numerous clients, and have never encountered someone who could not readily hear the improvement.

Lastly, this cannot be demonstrated in a results based test. Here is what I wrote about testing your summing box in Zen and the Art of Mixing:

Originally Posted by Mixerman

Testing Your Summing Box

If you

[gubu]

Thanks Mixerman!

A quick question:-

What is the difference between using a summing box and a good mixing desk?
Say you had an interface that gave you 8 or 16 outs from your DAW (and I don't BTW, but I used to regularly use a HD2424), and you run it thru a decent 8 or 16 channel desk, possibly with some outboard. Is this not the same, or at least very similar, to running a summing box?

I ask because I used to mix live recordings from a HD2424 thru a Soundcraft GB2 at one time, only using ProTools as my 2 track master recorder, and the mixes have none of the 'haze' that I just can't seem to get rid of when mixing ITB.
It seems to me that a reasonably good desk could be set up exactly like a summing box, and should give you some of the sonic advantages that you describe above, no?

[Mixerman]

Originally Posted by gubu

Thanks Mixerman!

A quick question:-

It seems to me that a reasonably good desk could be set up exactly like a summing box, and should give you some of the sonic advantages that you describe above, no?

Yes. With an emphasis on "reasonably good desk." Converter quality will make a difference too.

Enjoy,

Mixerman

[gubu]

Ok,
I was chatting with a producer/tech/wizard friend of mine for an hour or so on the phone this afternoon, and he brought up the issue of summing ITB vs OTB.

He says, basically, that the 'haze' that some of us complain about with ITB summing is a processor artefact caused by the fact that the processor has no choice but to drop bits at the summing buss because the system architecture doesn't have enough bit width thru the processor to accurately sum the audio from multiple tracks. And that it's less of a problem in a 64 bit system, but that you're basically screwed in the world of 32 bit floating, if you're running more than a handful of tracks.

He also goes on to say that it's a problem with all DAWs, but that PTLE is either deliberately crippled, or the RTAS format does not use system resources efficiently enough. His example was going from a single 1G machine with 2 or 3G of ram to a dual-core 2.4G machine, and still maxing the CPU with far less plugins in PT than he would in Logic or Sonar, with the attendant loss of bits due to insufficient bit width at the processor.

Like Mixerman above, he says that the only solution is external summing, and that even a 'reasonably decent' mixer is preferable to mixing ITB. From what I've heard with my own ears, I'm inclined to agree. I've heard a lot of his stuff, and the material he produced while mixing ITB on PT has markedly lower sonic quality to everything he ever did on tape in the old days, or more recently using PT, but summing with a desk that he built himself from scratch.
From my own work, I can say that nothing I ever mixed thru a desk has that 'haze' that you get ITB. But as I don't have my friend's knowledge and experience, I'm happy to take his word for what is actually going on under the hood.

So, there you go!