Skip to main content

So, I've been recording this band, and we were after drum takes. Typically we would end up with 5 takes, and then load the 3 best into Pro Tools to edit between. The drums sounded really good through the monitor section of the console we were using (I don't say this to brag, it's actually an important component of the story). The Producer was very happy with the drums off tape.

So we load the drums into Pro Tools from tape, and then the Pro Tools guy takes the files to another computer to edit as we continue recording other songs. After we're done with all the takes, and most of the songs are edited, we are ready to do some over-dubs.

We transfer the edited versions into the Pro Tools rig in the room. I set the faders in Pro Tools to unity level, and listen to the drums through the console again. I was not critically listening for differences between tape, and Pro Tools as that was not a concern of mine. Suffice it to say, the drums still sounded good through the monitor section of the console after the transfer.

So now we're moving at a clip, and we load up another song record some overdubs, and then another and record some over-dubs. On the third song, I'm trying to set the fader levels, and I'm having trouble. All of a sudden, for some inexplicable reason, the drums sound like sh*t. Really bad, and not just to me.

The Producer is looking at me, and asking me what happened to the drum sound. So I start messing with EQ, and levels for a couple of minutes, and the sound is drastically off from what it had been all day long. I was having a very difficult time compensating for this sudden difference in sound quality.

Then, a lightbulb went off in my brain. I knew what it was. In less than 30 seconds, I fixed the problem, took off all the EQ's, and reset the faders. The drums were back to the form they had been all day long. The Producer was relieved. The second engineer plainly heard the difference in sound quality.

What was the problem?

Mixerman

Comments

miketholen Sun, 04/22/2001 - 18:24

So now we're moving at a clip
What? a Clip?!? does that mean fast?
On the third song, I'm trying to set the fader levels, and I'm having trouble
Trouble setting levels? huh?
I don't know why PT would automatically sound like shit, I would imagine it would sound like shit all the time! But you said you liked what you heard from the other songs so maybe they all sound like shit.
:D
Is this some sort of riddle? or an oxymoron?
I use Tape w/Tools all the time and have never come across your situation other than PT alway sounds like shit so I Consider it "Par for the course".
Mike ;)

Mixerman Sun, 04/22/2001 - 19:05

Originally posted by miketholen:

So now we're moving at a clip
What? a Clip?!? does that mean fast?
On the third song, I'm trying to set the fader levels, and I'm having trouble
Trouble setting levels? huh?
I don't know why PT would automatically sound like shit, I would imagine it would sound like shit all the time! But you said you liked what you heard from the other songs so maybe they all sound like shit.
:D
Is this some sort of riddle? or an oxymoron?
I use Tape w/Tools all the time and have never come across your situation other than PT alway sounds like shit so I Consider it "Par for the course".
Mike ;)



Stick with me here. (Clip means fast, yes)

It sounded acceptable off of ProTools for 3 songs. True, I would certainly have preferred it directly off tape, but that was not an option, as it was edited and we needed to stay in ProTools (why is irrelevant).

Same drums, same room, and they sounded good going down. In fact, the drums sounded fine coming off of ProTools. (OK, perhaps not ideal, but nothing that mixing couldn't compensate for so let's not get bogged down by that). There was a drastic change in quality of sound after OKing the sound of the drums off ProTools earlier. The drums sounded good the song before, and the drums sounded good after I realized what the problem was. Everyone in the room heard the difference. It was obvious. If I was the only one that heard it, I wouldn't even bring it up.

What happened?

Mixerman

harveygerst Sun, 04/22/2001 - 19:18

Ok, I've never ever worked with protools, but solving a problem always gets me. Doesn't protools and most of these computer programs save stuff several levels deep so that you can undo easily? I'm guessing that there were so many levels of changes from finished songs being saved, you were running out of horsepower on the next song till you flushed all those temp files (or whatever they're called) out of the system.

If that's not the answer, I'll come up with something else.

Mixerman Sun, 04/22/2001 - 21:29

Originally posted by Jay Kahrs:
Were you mixing drum levels and EQ in Pro Tools and not the monitor section of the console?

There was no EQ in ProTools.

BUT, yes the drum levels were changed in ProTools by the editing dude. The levels were set for a stereo output which he was listening to for editing. As soon as I put the levels at unity gain, the drums magically sounded good again.

Now, question number one. Who thinks I'm full of shit, that there would be absolutely no difference?

Question number two. Assuming you believe me, why does changing output levels within Pro Tools, with individual outputs one per input of the console, drastically change the quality of the sound?

I'm not talking a little bit here, folks. Everyone in the room heard this difference without even listening for it. So don't discount it for sensitive ears. That's not what it was.

Mixerman

Guest Mon, 04/23/2001 - 03:00

Very interesting problem, fascinating solution. My question to you is why didn't you cut the damn 2"? You can always leave it in the box until you mix, but there is no way on God's grey earth that P-T will have the impact of 2" on a drum track...especially one that's been left in the box.

You were cutting together sections from 3 different takes? Unless you had to reassemble the drummer a bar at a time, it shouldn't have taken more than an hour or two per song to do it with a razor blade.

Better your gig than mine...

alphajerk Mon, 04/23/2001 - 05:03

could it be the low level bit dithering to the individual outputs? maybe lower volume resulting in less resolution on the way out the individual outputs? or maybe simply louder=better, increasing the level to unity would obviously yield better results, especially going out to an analog desk where your headroom would be seriously affected.

[and checking the pt faders for unity gain when independantly running them out is a no brainer]

and whats up? cant the band do one good take? they suck so bad you gotta splice three together??? :D

MadMoose Mon, 04/23/2001 - 06:48

Originally posted by Mixerman:
There was no EQ in ProTools.

Question number two. Assuming you believe me, why does changing output levels within Pro Tools, with individual outputs one per input of the console, drastically change the quality of the sound?

Is it a two part? #1 I'm guessing it has to do with the internal math of the Fool Tools tracks being at lower then unity gain.

#2 If the tracks are lower then unity gain wouldn't you be using less available bits from the D/A's and therefore have a drop in resolution? So rather then getting 20 or 21 bit's your getting 12 or 13 bits. Am I close?

anonymous Mon, 04/23/2001 - 07:38

Been there, done that--same results. Direct i/o always sounds better with Pro Tools. Not sure why. I could guess that it was 24bit audio being mixed internally to 24 bit stereo. Nasty. You CAN get better results by leaving faders at zero, placing good plugs like Waves stuff (giving you headroom in 48 bit-world) and using them to set/ride levels--still going straight individual outs. In fact, you should try this MM, I think you'll be pleasantly surprised. If you don't have automation on a board, this is a decent way to go. But, I'm sure you don't ever have that problem, huh? :p

Is the summing math on digital boards better? I've barely used an O2R once, wasn't able to make a real judgement.

Mixerman Mon, 04/23/2001 - 08:19

Originally posted by Fletcher:
Very interesting problem, fascinating solution. My question to you is why didn't you cut the damn 2"? You can always leave it in the box until you mix, but there is no way on God's grey earth that P-T will have the impact of 2" on a drum track...especially one that's been left in the box.

You were cutting together sections from 3 different takes? Unless you had to reassemble the drummer a bar at a time, it shouldn't have taken more than an hour or two per song to do it with a razor blade.

Better your gig than mine...

This gig wasn't in LA, (in fact it was in the middle of nowhere) so for a variety of reasons, including time restraints, artists preferences, the fact that the only thing we were using from the studio was the room itself, we even had to get another console, we had no choice but to put it on Pro Tools (this isn't even the half of it, you're just going to have to take my word on it).

You KNOW I would have preferred to cut tape. And you know I would have preferred to stay analog. It wasn't the way to go in this case.

Mixerman

Greg Malcangi Tue, 04/24/2001 - 01:53

Hi Mixerman,

What version of PT were you using?

If you were using version 5 there was a nasty memory leak problem, where closing and opening different sessions (songs) could sometimes produce weird results like the one you mentioned.

If this is not the problem I don't know what is. In more than 6 years with PT I've never experienced the problem you describe. The only way to recreate the problem that I can think of would be if the tracks were recorded at low level, then placing the faders at low level (somewhere near infinity) and compensating with gain to the monitors. Under these circumstances you could well hear the effects of quantisation errors, especially if you have an older version of an 888/24 which only has 20bit D/A converters.

<< Doesn't protools and most of these computer programs save stuff several levels deep so that you can undo easily? I'm guessing that there were so many levels of changes from finished songs being saved, you were running out of horsepower on the next song till you flushed all those temp files (or whatever they're called) out of the system. >>

I can't speak of other computer programs but PT doesn't do this. That's one of the reasons multi-level undo has always been a problem on PT but also why it is so stable. The original files in PT are never changed. Even applying an audiosuite plug doesn't change an original file, it just writes a new permanent one.

Greg

Rog Tue, 04/24/2001 - 07:07

[QUOTE]Originally posted by Mixerman:
If this was a comprehension test you would have failed. I think you need to re-read my posts. I was taking individual outs from ProTools into the monitor section of an analog console.

You only mention this in your third post. If you're gonna set questions, set them. Once and once only and no updates.

Seems you failed a very basic test yourself.

realdynamix Tue, 04/24/2001 - 10:28

Well, if your were at unity gain, including the PT, than I don't know. If you you were at higher levels on the PT, which is actually recommended, than could your name be Rumpelskiltskin, and could the integration of analog to digital and back have something to do with it? But, why would that take a whole 30 seconds to fix!! Just charge it back to the client. Keep's those producers on their toe's.
--Rick

MadMoose Tue, 04/24/2001 - 10:50

Originally posted by Rog:
[QUOTE]Originally posted by Mixerman:
If this was a comprehension test you would have failed. I think you need to re-read my posts. I was taking individual outs from ProTools into the monitor section of an analog console.

You only mention this in your third post. If you're gonna set questions, set them. Once and once only and no updates.

Seems you failed a very basic test yourself.

Nope dude. You failed. Re-read Mr. Mix's first post. Here I'll quote it for you.

<< So, I've been recording this band, and we were after drum takes. Typically we would end up with 5 takes, and then load the 3 best into Protools to edit between. The drums sounded really good through the monitor section of the console we were using (I don't say this to brag, it's actually an important component of the story). The Producer was very happy with the drums off tape.
So we load the drums into Protools from tape, and then the Protools guy takes the files to another computer to edit as we continue recording other songs. After we're done with all the takes, and most of the songs are edited, we are ready to do some over-dubs.

We transfer the edited versions into the Protools rig in the room. I set the faders in Protools to unity level, and listen to the drums through the console again. I was not critically listening for differences between tape, and ProTools as that was not a concern of mine. Suffice it to say, the drums still sounded good through the monitor section of the console after the transfer. >>

I understood what he was doing. A lot of other people did to. It's all good, but you weren't paying attention.

Mixerman Tue, 04/24/2001 - 14:21

Man, communictaion is a bitch.

OK, I will shoulder the responsibility of confusion, and I will put this as plainly as I can.

There were 8 channels of drums. They were coming out of individual channels from ProTools to an analog console for monitoring. So now the 8 drum tracks were coming up on the analog board 1-8. The drums sounded good coming out of ProTools into the console when the faders were set at unity gain within ProTools. If you were an observer at the session you would notice that all the faders were at the same level in ProTools. The faders on the board were at different levels to create the right 'mix' of the drums.

We recorded over-dubs on 2 songs with no problems, The drums sounded good. On the 3rd song, when the second engineer loaded it into ProTools, unbeknownst to me, the faders were set within ProTools at different levels to create a 'mix' for a stereo output. They were set this way for the editor who was monitoring the drums as a stereo output when he was working on them. Normally, the second would reset the faders to unity, immediately. He forgot.

So now, as I'm listeneing, the balances are all out of whack on the drums. So not thinking, I just go to reset the balances on the analog console. As I am resetting the balances by ear, (kind of wondering why I have to change the fader levels so much) I feel as if the drums don't sound very good. Even though the balances are technically right (i.e. the relationships within the mix of the drums seemed correct, there was the right amount of kick drum in the mix, the right amount of cymbals etc..) the drums sound terrible. The snare sounded bad, the kick sounded bad, the overheads are whacked out sonically, etc... But they sounded good the song before. So what happened?

When I turned around I saw that the fader levels were all over the place on ProTools, so I cliked them all to unity, reset my balances on the analog console to where they were from the last song, and the drums were good again.

The mystery part is over. Jak Kahrs had the correct answer. Again, I obviously wasn't clear.

The purpose for the story was to point out that there is a degradation in sound when you use the faders within ProTools to do rides, or set levels when they are being output individually to an analog console. So if you're going to output ProTools into an analog console, the more faders you can leave static at unity level in ProTools, the better from a pure sonic standpoint.

When I point this out to people they are sometimes very surprised. I know that I'm surprised at how drastic the degradation is.

Mixerman

anonymous Tue, 04/24/2001 - 18:54

Hey Mixerman,

I do love a good mystery.

As you have pointed out, taking a "direct out" from the PT fader means 24 bit processing on any fader moves. Many folks have remarked on the loss of fidelity when doing this. Certianly there is something to it.

I just want to mention, that there are analog consequences to this story too. Obviously the signal to noise specs on the board were not as good when the drums were coming out at the "correct" levels. If you have any dynamics on the analog console they too were working "all wrong" with incorrect levels into and from the board's preamps.

Making fader changes on the board wouldn't compensate for distortion in an overloaded pre, or lack of punch in one coming in not hot enough.

I assume that when you were jockeying the sounds around, you didn't reset your preamps on the console. I suppose the moment you considered that would have been the moment you figured out the PT problem. Who knows? maybe that's just how you found it.

I don't question that using direct outs from PT with fader rides is unsound. You can make rides in the output of several good digital plugins that will be better due to dithered output, and higher resolution in the math doing level changes.

But I just wanted to add that there are some consequences in the analog world for what was wrong in addition to the digital ones?

Kyle

Bob Olhsson Tue, 04/24/2001 - 21:54

The problem is with the Pro Tools volume controls. Depending on the program material, they can really turn things to cardboard. One time working on a scream for a film, I dropped the level a little and my wife asked me if I'd blown out a speaker!

I work around it by making gain changes inplug-insthat I know are really well written like the WAVES stuff. Of course this takes forever... sometimes the PT volume controls can sound just fine too such as when you aren't in a hurry...

Rog Wed, 04/25/2001 - 02:53

It makes me wonder sometimes ... reading Digi's stuff about all the artists using PT (quite a roster) and then listening to their CDs - I can hear a real PT sound. It's very cold, thin, lacking emotion and sterile. Think Lenny Kravitz here - I've heard very few rock records as uninvolving as his. No offense if anybody here produced it, it's mixed great, I'm talking about PT's inherent sound.

Has anybody actually used a DAW and been impressed with the sound? I've used Nuendo running on a hammerfall card and it's OK, SADiE was similarly OK, PT sounds worse - nothing I've used compares to a 2" analog mixed to 1/2" 2 track.

anonymous Wed, 04/25/2001 - 06:40

I think we were just talking about the situation Mixerman described. If we're gonna talk about Pro Tools in general, and not just level changes before the "direct out" method, then I think there may be a lot more going on than meets the eye (ear?).

It really belongs in another thread, but there are a lot of problems with the WAY people are using Pro Tools. People are adding lots of distortion with thier analog gear trying to get "good high levels" and are asking thier analog gear to operate at +14 VU. Routinely.

There are many issues that need addressing before you can blame ALL the sonic problems on Pro Tools itself. I am coming to believe that the techniques for making great sounding records on Pro Tools are just being discovered. The preceeding level suggestion is just one example. I'm doing experiments, technical and musical, and I hope to have some answers for myself in the coming months.

To stay on topic, everyone now agrees that while you get 56 bit math from pro Tools internal mix bus, if you take individual tracks before they hit the mix bus, and make fader moves on them, the level change is done as 24 bit math, and is inferior.

Interestingly, some high quality plug ins do math on output level changes at higher bit levels, so if you must ride the level of a single track which is bypassing the internal mixer, use the output level of one of these plug ins. Some of them also dither the result. This is superior to using the PT fader "direct out" at 24 bits.

The message is - either mix in Pro Tools, or take your tracks out at unity, and make no fader moves in Pro Tools. And if you must make moves on "direct out" tracks, do it with a plug in.

As for the rest of the lore, I think a great deal of it is about technique. Not that people aren't doing what everyone agrees is right - but whether or not we have all been doing it wrong together.

More about this later when I have more info to support it.

anonymous Wed, 04/25/2001 - 07:29

Rog,

Go easy man.

Re-read what I said, I wasn't blasting you for being off topic, but rather I was acknowledging that the subject had changed. This is just to keep from confusing the two issues.

Saying that "direct outs" sound crappy with fader moves is one problem - the "sound of Pro Tools" is another. Lumping them all together is what causes confusion, and myths. So I thought underlining the segue was a good idea.

Hopefully, this is a forum for learning, and debunking common myths. Certainly people have corrected me, and I am grateful for the better info.

So, no friction from me, Rog - just discussion on what's really up with DAWs and Pro Tools.

Personally, I think good recordings and mixes are possible on Pro Tools, but I'm relearning my technique, because this is a very different animal, and clearly requires a different approach.

For what its worth, if the moderators want to erase these last 2 post, since its as much spam as topic, its ok with me.

Kyle

anonymous Mon, 04/30/2001 - 09:19

what ever happened to using your tools the best you could.

being of younger status - I always have to make alot of things work that may not be acceptable to some of you - but these things usually make me stronger in the longrun.

I'm now a firm believer that any decent engineer can make a great sounding recording on any medium. Granted some mediums are easier to work with than others.

If you are going to blame what you are mixing on for your sound, or lack thereof - you really should take some time to learn better technique.

analogy time - throw Bob Vila and yourself into a woodshop, give yourself the best damn carpentry tools that money can buy - and give Mr. Vila K-Mart brand crap. Guess what, Mr. Villa is still gonna make a canoe out of driftwood, and you might make a napkin holder if you are lucky and don't splinter yourself.

I look up to alot of engineers that have been doing what I love much longer than I have - or longer than I have lived - but to hear people blaim the sound of a record purely on the medium they "think" (most of the albums on the pro-tools site were not even mixed in PT), just kinda brings me down.

I must be a DAW guy, since I can't afford to work without automation and precise edits - and I also can't afford a 9000J. :D

take care - Jason C. Crouch

Mixerman Mon, 04/30/2001 - 10:31

Well, if you're referring to me, you're preaching to the choir, Jason. Gear are tools in a mix that make life either easier or more difficult. That's actually a step in my 10 Steps To Better Mixing (which you can see on Harvey's forum).

I'm just trying to illustrate a sonic pitfall. Does that mean you can't make a good record through this problem? No. But if people are aware of this, then maybe they can take steps to get around it.

Mixerman

anonymous Mon, 04/30/2001 - 14:25

Don't fret - I wasn't referring to you, rather the replies that followed your posts.

It just gets old really fast when you have to listen to alot of local people that have no idea what they are talking about, judge pieces of gear so harshly.

I even went as far to making a nasty paris post on RAP, which I kinda feel bad about now. Just that every time the Paris rep flows by, I have alot of my clients coming in and trying to tell me all these horrible things about ProTools and that it sounds like crap - and these are people that have never used either program.

Bottom line - I went thru the whole PT vs Paris vs Nuendo vs Performer deal - and it was a great learning experience. What did I come away with - that they are just tools, and that it helps to work with a tool that others are more familiar with, and dare I say might help you make a few bucks back to pay off the shit.

I guess once you get to the level that you can really choose what you get to work with, you would want to chose some things over others - especially when the mixes are of higher levels of importance.

so man - I totally dig it, for me - I'm all about quality - but at the same time I'm all about doing work so that people can hear what I'm all about, and hire me due to what I can do - not what gear I own.

take care - Jason C. Crouch

bluebass Mon, 04/30/2001 - 14:41

Originally posted by JasonCrouch:


analogy time - throw Bob Vila and yourself into a woodshop, give yourself the best damn carpentry tools that money can buy - and give Mr. Vila K-Mart brand crap. Guess what, Mr. Villa is still gonna make a canoe out of driftwood, and you might make a napkin holder if you are lucky and don't splinter yourself.

I really don't mean to be a smart ass here, but you couldn't have picked a worse analogy! Did you ever wonder why the carpenters or (insert appropriate trade mechanics) take a few steps back when Mr.B picks up a tool....it's not because they're amazed at his talent. :D

Regards,

Tim L

anonymous Tue, 05/01/2001 - 02:01

Hello,

So it was established that it was pro tools chopping off bits from the pt fader moves wasnt it?

What if pro tools is the only automation / mixing possible? Would it make a difference to the sound if the all faders were as close to zero as poss and the master was low down, or the master was at zero and the faders were all over the place?

I would love to have the luxury of a big desk and 2" to compare. But i have to make do.

G Kennedy :)

miketholen Tue, 05/01/2001 - 05:44

When I HAVE to mix in PT, When I track I will track into PT at 0dB VU-Very important. because once you have say 16 -32 tracks running it won't overload the "mix" bus w/in tools. You will end up with you faders hovering around zero, thus less math going on.
I have found this to sound much better than slamming level into PT and then turning all the faders down because the mix bus is being overloaded.
Do you guys slam into PT or do you track in at 0dB VU?
Mike

Mixerman Tue, 05/01/2001 - 10:46

Originally posted by miketholen:
When I HAVE to mix in PT, When I track I will track into PT at 0dB VU-Very important. because once you have say 16 -32 tracks running it won't overload the "mix" bus w/in tools. You will end up with you faders hovering around zero, thus less math going on.
I have found this to sound much better than slamming level into PT and then turning all the faders down because the mix bus is being overloaded.
Do you guys slam into PT or do you track in at 0dB VU?
Mike

This seems to me like a great approach to deal with this problem.

Mixerman

anonymous Tue, 05/01/2001 - 11:38

Yes Mike!

I couldn't agree more.

I'm still doing experiments, so I'm hesitant to come out and start yelling this from the treetops, but I am finding more and more that observing 0VU, and NOT "getting all the bits" is a good way to get a good sound in Pro Tools.

This is why I was originally asking Mixerman if he thought there were any analog effects in what he was hearing (either overdriven pres or underdriven pres).

More generally, you also have to wonder what happens to the analog Master section of a console whenever we monitor a Pro Tools track or mix through it, and Pro Tools is hitting it at 0dBfs (+14VU)? This is like every recording studio!

I also agree with Mike that there is something to allowing greater headroom in Pro Tools, resulting in a mix where the faders are closer to unity (subject to less math scaling, and sounding better?) and the mix bus isn't too crowded and colored.

There are also analog implications to running hot levels in Pro Tools. If you use any analog gear on inserts, and you are recording full levels into pro Tools, you are likely going to overdrive the analog gear, forcing it to work at +14db and worse.

Also, running analog gear like preamps hot enough to make peaks at +10 VU going into the converter is going to cause more color from the preamps! Fine for a choice, but not something you'd usually want on every track, and yet it is being done routinely. As the color adds up in multiple tracks, people complain about the mix bus, which may or may not also be part of the problem, especially at higher levels.

So I can't agree more strongly with observing lower levels in Pro Tools. But it isn't that easy.

IF you observe 0VU in Pro Tools, what happens to the "sound" of plug ins? Do they suffer in quality when they aren't being used at their optimum levels? Also, having a final mix that peaks at -8dbfs isn't going to win you a lot friends if its being pressed directly to CD, whereas if its going to mastering, they may well applaud you.

These issues are complicated, and interrelated, and I doubt there's a right answer for all things. I suspect there is a good zone between running the analog gear too hot, and getting a good level into the converters. Each track will require subjective decisions about what tradeoffs we will make, and how they affect the sound. Hopefully we can take advantage of the colorations caused by certain tradeoffs and make choices that will help the sound we are going for. I am coming to think that there is subjective art in setting levels for digital recording - I need to be open to finding it, regardless of what worked well in the past with a console and a tape machine.

Levels are only one issue.

I have also heard many folks complain about "direct outs" in regards to the way Pro Tools does math. While I don't doubt them, I'm still a little fuzzy about what constitutes a "direct out" since using a bus is using a bus isn't it? Is there something functionally different about one bus or the other? Does the summing math get better when you put a Master Fader on a Bus? Or when two faders are being summed? Would having a second blank channel assigned to the output initilaize a mix engine for 56bit math? We know about using quality plug ins to avoid truncation, is it necessary wehn mixing in Pro Tools, or only when going out to a console?

All these and many other questions will soon be answered by all of us in personal subjective (and possibly some "objective"?) experiments. The current thinking is going to change, there are too many contridictions the way it is.

What it all points to is that despite the plug and play marketing of digital recording, people with technique, creative vision and ears will get superior results. I'm ready to forget what I once knew about recording, and look at this beast as totally new. I do believe we are only scratching the surface of what today's version of Pro Tools is sonically capable of. What's exciting is the improvement doesn't require an external hardware or software upgrade. What has to change is the way we use it.

Kyle