Ignoring 0VU.... urgh...

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

pucho812

Well-known member
Joined
Oct 4, 2004
Messages
14,831
Location
third stone from the sun
With the Modern daw  the in thing, I have noticed that with most session I am getting for mixing and or to edit, that the tracks are being recorded way louder then they should be.


in a daw, depending on calibration, 0VU is usually between -14dBFS to -21dBFS.  SO WTF? If all the tracks are at +3VU and higher how is one supposed to mix that in the box.... it makes it more of a hassle.

I am wondering if this is just because people do not look at a VU scale when recording in the box... can't Manufactures allow you to change the metering between peek and average without any real hassle.

at any rate it is annoying and makes it harder to mix in the box.
 
That seems to be the norm on things I get in from home studio. 

I usually end up (in PT) pushing the faders around for a quick rough, inserting the trim/gain plug into the top slot and copying the fader levels into that plug so all my faders can go back to Zero.  Then I actually start working again....  everything has some headroom again.


My two cents.
 
I try to be around -18. It's funny though when looking at the "low" meter levels. Your instinct is like, gotta get higher above the noise floor! I agree it would be cool to be able to change the appearance. Put -18 at like where -6 is on a dBFS scale... When you think you're gonna clip you'll have 12dB headroom. But then again, on tracks pro tools appears to have some headroom beyond the top of the scale.
 
I'm sure it comes from the idea that louder = more bits used, so "more resolution".
It's an intuitive thing. You boost the gain on your preamp (or whatever's feeding the dac), the meter goes up, and as more level always sounds better, you automatically draw the conclusion that you should get all the way up to 0dBFS, because it 'sounds' better.

When your slo-tools monitor is 4 meter away from your 'gain knob' it's pretty hard to actually see 0VU (-18dB or whatever is calibrated). Digi (and other software manufacturers) should indeed allow you to clearly mark the 0VU point.
 
VU.

...In digital...

Doesn't stand for what you think it does.

It NOW stands for...


"Virtually"

"Useless"


Seriously, there's no law that says things have to be recorded at zero VU.

The most important laws say that the noise floor of the source should be above the noise floor of the converter, and that the PEAK should be below the converter's maximum.

Once you've satisfied those basic requirements, more or less anything goes.

(and a VU can't tell you much about peak...)

Keith
 
Yeah, Virtually Useless, but still known well by enough people to warrant greater inclusion in DAW's.  I'm only used to seeing peak, unless I turn on Spectrafoo as outboard software.  There I can pick from a list, and trim response too.    It wouldn't be that hard to make DAW's show peak with an RMS or VU line also.  That would be useful. 
 
Back in the ADAT and DA-88 days you certainly had to walk a tightrope along the zero border - a dangerous game to hit the sweet spot without clipping. The difference there was really noticeable.

I found the "using up bit rate" idea/approach fairly meaningless in PT and Logic 8 - even at 16 bit. As Keith said, once your off the converter noise floor you've got a pretty large safe zone for most sources.  I stayed as low as possible for med to loud sources. A little higher for low sources like finger picked acoustic - sound was as good as it could be. Only thing I saw with trying to use up the extra level space was less useable room for plug-in processing during mixdown.
 
SSLtech said:
VU.

...In digital...

Doesn't stand for what you think it does.

It NOW stands for...


"Virtually"

"Useless"


Seriously, there's no law that says things have to be recorded at zero VU.

The most important laws say that the noise floor of the source should be above the noise floor of the converter, and that the PEAK should be below the converter's maximum.

Once you've satisfied those basic requirements, more or less anything goes.

(and a VU can't tell you much about peak...)

Keith

True, keith there is no law about how hot tracks should be or are in digital or analog. My point was 2 things.

1. why can't daw manufactures include a VU in their softwarewithout any extra plug in's or whatever.  It could be like a dorrugh meter showing average and peak. If fact I have been toying with the waves dorrough meter and I like it but that is just fluff that should already be in the daw. It doesn't need to be fancy like a dorrough but work in similar fashion.

2. Most tracks I am seeing now days are just too loud. If I have, say, 24 tracks and all the levels are loud then the in the box mix buss will just square off.  I use the vu reference and mention calibration levels to help prove my point that the average level of audio per track is much louder now and that this creates problems when mixing in the box.



 
Display is something that I think is lacking in multitrack daws. Id like to see something like the visualizations in the mastering programs on multiple tracks and master buss etc... Id actually like to see that type of stuff on lcds on each track on say 72ch large frame console. my ears are working full time. It'd be nice to have some other data input (visual)real time. kinda my 10 year plan......
 
pucho812 said:
Most tracks I am seeing now days are just too loud. If I have, say, 24 tracks and all the levels are loud then the in the box mix buss will just square off. 
No it won't. PT uses a 48bit mixing engine, of which 9 are dedicated to providing enough headroom for 128 tracks all at 0dBfs. Most other DAWs use floating-point 32 bit mixing engines, offering 390dB+ of headroom, enough for 10e+15 tracks at 0dBfs.
 
128 tracks of a sine wave VS real music wave forms are 2 different things. I know what digi says it can handle... I know what my ears are telling me, Squared off wave forms and digital clipping...
 
It's an endless battle of visual feed back, you look at the console meters and they're pegged. So that is wrong. But then you look at your PT session and you've got "good strong signal" on the dBfs meters. If you track using console meters everything looks great on the desk, then you look at your PT session and all the dBfs meters are "weak signal".

Studio engineers that came from analog and went to digital usually have a good understanding of what "level" is. They can look at a digital meter and see -16db signal and not think twice about it because 0vu is -16 (or -18 or whatever), but an engineer that grew up on PT wants to get that signal as close to digital 0 as possible because that is what their visual feedback is telling them to do. They never had to align a tape machine and truly understand what "level" is.

A great engineer once told me "as long as the needles are moving, your fine" referring to VU meters when tracking.

The same engineer sat me down and explained to me the whole "using all the bits" thing. He broke it down into math, I don't remember the exact numbers (so correct me if I'm wrong) but it's something like on a 24bit recording you have to get 128dB below digital "0" before you can hear the low resolution breakup of "the bits" on 16bits is like 96dB below "0". So you don't need to get up to 0dBfs to "use the bits".

good quote "you can have your 24bits, my tape machine has billions of bits"  ;D

Ultimately it comes down to gain staging and fader resolution. If your going to stay in the box then yeah, you can track all the way up to digital 0 and start pulling down your faders to keep the mix buss from clipping your converters. I know the software can handle hundreds of dB of headroom, but when it comes back out to the analog world, your converters are still abiding by analog rules.

I personally just listen, if I'm tracking drums I put all the faders at "0" and adjust my input levels until I have a drum mix that sounds good. If the hi hat is peaking at -30dBfs, I don't care. It sits well in the mix with my fader at unity gain. I'm not going to crank my preamp to get level up to digital 0 so that I can just pull down the fader to -30, it makes no sense to me.

I want to hand a PT session to someone and when they open it up all the faders can be at unity and they already have a basic "mix" going with levels. I print the level that sits well in the mix, not the level that visually looks the best. Is that bad? Well, I've NEVER had anyone complain about signal not begin "strong enough" in my sessions. So.... I guess it's ok.

JUST USE YOUR EARS!!! If it sounds good it is good, reguardless of what level it is!  ;)




 
pucho812 said:
128 tracks of a sine wave VS real music wave forms are 2 different things. I know what digi says it can handle... I know what my ears are telling me, Squared off wave forms and digital clipping...
I didn't say you don't hear a thing; I said it's not the mix engine clipping you're hearing. It's something else. Look in the direction of analog.
 
abbey road d enfer said:
pucho812 said:
128 tracks of a sine wave VS real music wave forms are 2 different things. I know what digi says it can handle... I know what my ears are telling me, Squared off wave forms and digital clipping...
I didn't say you don't hear a thing; I said it's not the mix engine clipping you're hearing. It's something else. Look in the direction of analog.

I would agree. a 48bit mixer downsized into a 24 bit converter.
 
Digital summing is no different to analog summing in that you can clip.

If the original contributing signals are un-clipped, there is NO reason for the sum to EVER clip, unless the multipliers (Faders & pan pots) are set too high.

Shrinking down to a 24-bit window is a red herring also. Studer use a summing engine which has something like a 1,500dB range. -It still has to fit through the same window.

It's always down to the user in the end.

Keith
 
For most of my career I have paid attention to audio myths, mainly to search out the tiny morsel of truth behind most of them that show any persistence... I continue to search for a smoking gun behind the many claims that digital combining is somehow inferior to analog. Boy do I wish this was true as I have some of the best technology around for analog combining (current source summing), and I would love to have the world beat a path to my door, but alas, I can't find anything really wrong inherently with digital combining. In theory unlike analog combining you end up with more signal resolution, not more noise, more distortion, more phase shift, etc. 

In practice there are always ways to mess up any implementation analog or digital from things as simple as gain structure, to mistakes in software coding. Properly executed typical floating point hardware has more than enough spot resolution and many times more total dynamic range than the best analog. Even fixed point summing is hard to find obvious fault with, in a world of <3dB final mix crest factor even a -96 dB quantization floor is not much of an issue, and even modest modern hardware is well better than that.

Any real smoking guns out there?

JR
 
In theory unlike analog combining you end up with more signal resolution, not more noise, more distortion, more phase shift, etc. 
If the distortion added by analogue summing sounds more pleasing on a particular recording, that's the method I'll use in preference to the floating point summing in my DAW. For rock stuff I almost always mix in analogue. It's an aesthetic decision. There's nothing wrong with FP ITB summing, but a small amount of analogue "imperfection" can make a mix sound more solid and well integrated. I haven't heard any DSP that can reproduce the effect convincingly, although I've no doubt that improvements in understanding and processing power will eventually narrow the gap to the point where it's near enough.
 
Ok - I normally stay out of these 'religious' debates (and I've just been lurking in Studio A lately since most of my free time is going to recording instead of building gear), but I have to jump in because I want to understand these topics with clarity.  With all the knowledge on this forum we should be able to forge a consensus at least on the objective aspects.

Would someone be willing to expand on the notion that the input level DOESN'T matter in _fixed_ point digital systems?  It would seem that in a 16-bit system (as an example for 'simplicity') each 16-bit sample is a signed integer between -32768 and 32767.  If I input a waveform that is half the maximum level of the converter the result will be a digital representation that ranges between -16384 to 16383, or the range of a 15-bit signed integer.  i.e. for each -6dB below FS one loses a single bit of resolution.  Correct me where/if I am wrong, please!  ;D  I'm not saying its audibly significant in all contexts, but it seems unavoidable due to the math and signal representations.  Certainly this effect becomes less relevant at 24bit resolution, as you can go down to -48dBFS and still have the equivalent of a 16-bit resolution signal.

I can imagine that a floating point system could be different because the magnitude of the represented signal is independent of the precision of the value.  As a result a signal a fraction of full scale can still have the full resolution of a FS signal.  I work on Protools so I am principally concerned with the math behind fixed point systems however.

Philosophically I am 147% behind the notion that your EARS and what you hear are the most important metric.  On the flip-side, much of this debate is based on math and there should be objective answers to many of these questions.  When in doubt in such cases, I bounce two versions, invert one and sum.  If the bits are the same and you think it sounds different, the price you pay for something probably determines how much you like it.  :D

To the original point about tracks recorded "too loud" for ITB mixing - The only way it seems this could matter is the sonic impact of digital attenuation (distinct from summing).  Summing is addition and therefore causes ZERO degradation until clipping.  Any time you sum in protools you have a AUX or MASTER track that represents the buss output - if it clips, you're clipping (bad) - if not you have lost ZERO information.  On the other hand attenuation/gain is a multiplication, therefore the result of the operation exceeds the input resolution (hence the use of the internal 48-bit buss).  Whether anyone can actually hear the degradation caused by this multiplication and eventual truncation is another story however...  :-\

Ok - Now one of you veterans school me on this stuff!!!  :D

jt



 
jtoole said:
Ok - I normally stay out of these 'religious' debates (and I've just been lurking in Studio A lately since most of my free time is going to recording instead of building gear), but I have to jump in because I want to understand these topics with clarity.  With all the knowledge on this forum we should be able to forge a consensus at least on the objective aspects.

Would someone be willing to expand on the notion that the input level DOESN'T matter in _fixed_ point digital systems?  It would seem that in a 16-bit system (as an example for 'simplicity') each 16-bit sample is a signed integer between -32768 and 32767.  If I input a waveform that is half the maximum level of the converter the result will be a digital representation that ranges between -16384 to 16383, or the range of a 15-bit signed integer.  i.e. for each -6dB below FS one loses a single bit of resolution.  Correct me where/if I am wrong, please!  ;D  I'm not saying its audibly significant in all contexts, but it seems unavoidable due to the math and signal representations.  Certainly this effect becomes less relevant at 24bit resolution, as you can go down to -48dBFS and still have the equivalent of a 16-bit resolution signal.
Math doesn't lie... and in the early days of low bit digital, it was a constant struggle to keep the crude digital convertors from rolling over to zero again, past full scale on the high end, and stay out of the quantization grunge at the low end.

Modern convertors clip like good old school analog paths, and have bit resolution below their own (analog sounding) noise floor, so level setting is just like level setting for any analog path... keep it out of the hiss and below clipping.

Of course level matters, but not more or less than any other audio path.
I can imagine that a floating point system could be different because the magnitude of the represented signal is independent of the precision of the value.  As a result a signal a fraction of full scale can still have the full resolution of a FS signal.  I work on Protools so I am principally concerned with the math behind fixed point systems however.

Philosophically I am 147% behind the notion that your EARS and what you hear are the most important metric.  On the flip-side, much of this debate is based on math and there should be objective answers to many of these questions.  When in doubt in such cases, I bounce two versions, invert one and sum.  If the bits are the same and you think it sounds different, the price you pay for something probably determines how much you like it.  :D
Ears matter, but not individual sets but the mass of all human ears collectively... They seem to be voting in favor digital last I checked.
To the original point about tracks recorded "too loud" for ITB mixing - The only way it seems this could matter is the sonic impact of digital attenuation (distinct from summing).  Summing is addition and therefore causes ZERO degradation until clipping.   Any time you sum in protools you have a AUX or MASTER track that represents the buss output - if it clips, you're clipping (bad) - if not you have lost ZERO information.  On the other hand attenuation/gain is a multiplication, therefore the result of the operation exceeds the input resolution (hence the use of the internal 48-bit buss).  Whether anyone can actually hear the degradation caused by this multiplication and eventual truncation is another story however...  :-\

Ok - Now one of you veterans school me on this stuff!!!  :D

jt

bus... ;D

JR
 
Isn't the most important part of recording around 0vu the fact that most analog gear is designed to be hanging out around there?
 

Latest posts

Back
Top