Easy Question - Volume

[taojonz]

Maybe...

I've tracked, I've mixed, but I've never mastered. I've always let an experienced mastering engineer take care of it.

I have one song mixed, and I'm waiting on four more before I send the entire E.P. off to master. In the meantime, I'd just like to share this song on Myspace, etc...

What's the best and most simple/bare-bones way of upping the volume without distortion? Is it as simple as loading the stereo file into a session and pushing the master fader up above zero, until I reach the desired level? (Surely not

I've working with PT LE. I've got the rack 002 bundle with all the included PT plug-ins etc.

[jdg]

grab a limiter, put it on the master bus.
drop the threshold on it till its LOUD as you want.
set the ceiling to -.2dB or something. export as 24bit.
make the mp3 from the 24bit wav (LAME for example works with 24bit input sources)

for PT, everyone loves this limiter:
High-end plug-ins for Pro Tools

good luck

[gwell1]

Do you have Maxim?If you do, set your ceiling to about -.3 db and then lower your threshold until you reach your desired volume. If you have to lower your threshold to more than -6 db, try using two of them, and even the load so that the plugin doesnt overwork itself.

[bob katz]

Originally Posted by taojonz

What's the best and most simple/bare-bones way of upping the volume without distortion?

You can't beat the laws of physics! If it were that easy then there would be a lot fewer mastering engineers. But remember that mastering is NOT all about "raising the level". There's a lot more to mastering, for example, regulating the levels and equalization from song to song in an album, so that they work together. Also, did you know that the word "volume" measures the capacity of a container? "Volume" means quarts, pints, liters, and cubic meters, to me. There is no official definition for the term "volume" in acoustics. Just be aware that, in audio, "volume" is a colloquial usage with an ambiguous and confusing (double or triple) meaning.

[dcollins]

Originally Posted by bob katz

Y Also, did you know that the word "volume" measures the capacity of a container? "Volume" means quarts, pints, liters, and cubic meters, to me. There is no official definition for the term "volume" in acoustics. Just be aware that, in audio, "volume" is a colloquial usage with an ambiguous and confusing (double or triple) meaning.

Although everyone just says "volume," the correct term is "loudness."

"Intensity" is the measured loudness.


DC

[okydoky]

hi,

what i know about this would fill volumes.

yes, i have voluminous data on this subject.

and of course the work bob and dave have done in this area speaks volumes.

but i digress.



right.

[glassmaster]

Originally Posted by bob katz

You can't beat the laws of physics! Just be aware that, in audio, "volume" is a colloquial usage with an ambiguous and confusing (double or triple) meaning.

Well then what about that doohickie with the little pointer- Oh yeah, it's a VU Meter, also referred to as a 'Volume Indicator'. (go ahead Bob; I'll bet you can quote the MRL test tape by heart!)

[bob katz]

Originally Posted by glassmaster

Well then what about that doohickie with the little pointer- Oh yeah, it's a VU Meter, also referred to as a 'Volume Indicator'. (go ahead Bob; I'll bet you can quote the MRL test tape by heart!)

Hi, Dave. Yes, in the original announcer's voice! Here's a trivia piece: Do you remember the STL test tapes, you know that they scored the same guy who was the announcer for the Ampex test tapes?

Further muddying the waters:

An upcoming standard by the EBU will define "loudness" in relation to 0 dBFS, the scale is called "LKFS" (no relation to Mr. K. over here). Thus muddying the waters that the definition of "loudness" relates to the listener's perception since a digital signal has no defined physical presence.

My main gripe with the word "volume" is that the colloquial usage mixes up the terms "gain", "loudness" and "level", which creates ambiguity. Pro Tools' so-called "volume line" does not represent the OUTPUT, but rather the GAIN of the circuit. GAIN is the correct word (Pyramix, Sonic and SADiE get that right). Consumers get confused enough by that water faucet called a "volume control"; the higher you push it the more comes out, but what about what you put in it...

BK

[It'sJoeAgain]

Originally Posted by bob katz

My main gripe with the word "volume" is that the colloquial usage mixes up the terms "gain", "loudness" and "level", which creates ambiguity.

do u also find the same to be true for "resolution"??

[bob katz]

Originally Posted by It'sJoeAgain

do u also find the same to be true for "resolution"??

You're right. That's an even squeakier word. I had to define the term as used in my book to keep from ambiguity! Different people have different ideas of what "resolution" means and I won't get into it here.

BK

[dcollins]

Originally Posted by It'sJoeAgain

do u also find the same to be true for "resolution"??

For our audio purposes, when you see the word "Resolution" just think "Noise Floor."

Just like how the word "you" is spelled "you."


DC

[okydoky]

Originally Posted by dcollins

Just like how the word "you" is spelled "you."


DC

hi,

thannnnnk you.

do any of you guys know how many consecutive samples are required for the "statistical linearization" to be established with d*thering [assuming tpd f, properly applied]? is there a formula or anything?

i have some question as to whether the ear would actually perceive the averaging over time as one specific amplitude. is some of that calculated / realized at the dac or something?



right.

[dcollins]

Originally Posted by oky****

do any of you guys know how many consecutive samples are required for the "statistical linearization" to be established with dithering [assuming tpdf, properly applied]? is there a formula or anything?

There are lots of formulas.

I think it's safe to say that d*ther works as advertised, and outside of the Internets is not even remotely controversial.

i have some question as to whether the ear would actually perceive the averaging over time as one specific harmonic. is some of that calculated / realized at the dac or something?

Despite their uselessness for audio, your ear works much like an FFT spectrum analyzer. Continuous signals are detected below the noise with averaging.

The DAC just tries to keep up.


DC

[okydoky]

Originally Posted by dcollins

There are lots of formulas.

I think it's safe to say that d*ther works as advertised, and outside of the Internets is not even remotely controversial.

hi,

yeah, its not really controversial with me, honestly. i just kind of see it for what it is. i'll use d*ther. its not my favorite thing about digital audio, but its something. i'm just interested in understanding it well, especially since there seems to be a lot of room for it to be misused, or poorly implemented.

are any of the formulas seen as accurate?

Originally Posted by dcollins

Despite their uselessness for audio, your ear works much like an FFT spectrum analyzer. Continuous signals are detected below the noise with averaging.

hmmm. are you saying that my ears are useless for audio, or that fft sprectrum analysers are useless for audio. now i'm confused again?

Originally Posted by dcollins

The DAC just tries to keep up.

and how is it doing with that?

thanks.



right.

[dcollins]

Originally Posted by oky****

hi,
yeah, its not really controversial with me, honestly. i just kind of see it for what it is. i'll use dither. its not my favorite thing about digital audio, but its something. i'm just interested in understanding it well, especially since there seems to be a lot of room for it to be misused, or poorly implemented.

Whew. I thought you were going to frame the question like what's the best d*ther?

are any of the formulas seen as accurate?

No fair comparing audible with visual.

hmmm. are you saying that my ears are useless for audio, or that fft sprectrum analysers are useless for audio. now i'm confused again?

Ears are go.

Does this help?


DC

[okydoky]

Originally Posted by dcollins

Whew. I thought you were going to frame the question like what's the best d*ther?

hi,

Originally Posted by dcollins

No fair comparing audible with visual.

mmm hmmm, i see. i thought there might be some sort of analysis that was agreed upon as "reliable basis", but now that i think about it, i don't know if such a thing could exist in the pro audio community.

Originally Posted by dcollins

Ears are go.

Does this help?

yes, i think so.

i did just realize that i posted in a thread that was not about this topic. probably because i saw you guys posting in this thread. i didn't mean to derail. i'll make some edits.


right.

[IIIrd]

don't fight in the loudness war and you wont be a victim

[+6/185]

Originally Posted by taojonz

Maybe...

I've tracked, I've mixed, but I've never mastered. I've always let an experienced mastering engineer take care of it.

I have one song mixed, and I'm waiting on four more before I send the entire E.P. off to master. In the meantime, I'd just like to share this song on Myspace, etc...

What's the best and most simple/bare-bones way of upping the volume without distortion? Is it as simple as loading the stereo file into a session and pushing the master fader up above zero, until I reach the desired level? (Surely not

I've working with PT LE. I've got the rack 002 bundle with all the included PT plug-ins etc.

Welcome to the loudness war

[Adam Dempsey]

Originally Posted by oky****

do any of you guys know how many consecutive samples are required for the "statistical linearization" to be established with d*thering [assuming tpd f, properly applied]? is there a formula or anything?

Maybe start with the formulas on triangular probability density function.

Originally Posted by oky****

i have some question as to whether the ear would actually perceive the averaging over time as one specific amplitude.

Listening to tpdf (or high pass tpdf as used in some DAWs) at 16 bit should give you the answer.
Personally, I stick with the hearing side of things rather than mathematical analysis.

Back on topic:

Originally Posted by taojonz

What's the best and most simple/bare-bones way of upping the volume without distortion?

Digital gain, just enough so peak level = 0dBFS, or a fraction under.

[okydoky]

Originally Posted by Adam Dempsey

Maybe start with the formulas on triangular probability density function.

hi,

seems like someone would have doped it out for audio.

Originally Posted by adam dempsey

Listening to tpdf (or high pass tpdf as used in some DAWs) at 16 bit should give you the answer.

what is this "high pass tpdf", and who is using it? you could respond in one of the current dither threads if you like, so this thread doesn't totally derail.

Originally Posted by adam dempsey

Personally, I stick with the hearing side of things rather than mathematical analysis.

i know, and hearing is good, but the truth of the matter is that people have different levels of sophistication. one person is almost tone deaf, another hears a singer singing a single tone, another hears the degree of vibrato. so i think it is reasonable to think that different people will perceive "statistical averaging" differently.

by adopting an "its good enough to fool the hearing mechanism" approach, people are not exactly helping the quality of hearing to evolve. to some extent all these "psycoacoustic manipulations" can be see as a subtle dumbing down process.

not trying to be a wet blanket, just trying to stay objective.


right.

[dcollins]

Originally Posted by oky****

hi,
seems like someone would have doped it out for audio.

Have they ever:

Resolution Below the Least Significant Bit in Digital Systems with Dither
Paper Number: 1930 Convention: 72 (October 1982)
Authors: Vanderkooy, John; Lipshitz, Stanley P.

what is this "high pass tpdf", and who is using it? you could respond in one of the current dither threads if you like, so this thread doesn't totally derail.

There are papers at lavryengineering.com that should help here.


DC

[okydoky]

Originally Posted by dcollins

Have they ever:

Resolution Below the Least Significant Bit in Digital Systems with Dither
Paper Number: 1930 Convention: 72 (October 1982)
Authors: Vanderkooy, John; Lipshitz, Stanley P.



There are papers at lavryengineering.com that should help here.


DC

hi,

thanks dave. now that you mention it, i think i've read something mr. lavry wrote about his acoustic bit correction, if that is what you are referring to.

i want to find out about whatever digidesign is using in their "dithered mixer". and there seems to be a reluctance to talk about how many samples have to be averaged for dither to help represent those "in between amplitudes". i know faster sampling also reduces quantization error.


right.

[masterizer]

Originally Posted by taojonz

Maybe...

I've tracked, I've mixed, but I've never mastered. I've always let an experienced mastering engineer take care of it.

I have one song mixed, and I'm waiting on four more before I send the entire E.P. off to master. In the meantime, I'd just like to share this song on Myspace, etc...

What's the best and most simple/bare-bones way of upping the volume without distortion? Is it as simple as loading the stereo file into a session and pushing the master fader up above zero, until I reach the desired level? (Surely not

I've working with PT LE. I've got the rack 002 bundle with all the included PT plug-ins etc.

The way most mix engineers I know demo their mixes is to throw on a limiter with a ceiling of 0.1db, release set to fast or 0, threshold set to where its limiting 2-3db. This should be fine to demo your songs quickly.

Maxim, waves limiters are common amongst mixers. If you are limited to bomb factory plugins you can turn the compressor into a limiter by increasing the ratio.

If you can attend your mastering session defintely do.

[minister]

Originally Posted by oky****

i want to find out about whatever digidesign is using in their "dithered mixer".

Stan Cotey says it is TPDF.

[Waltz Mastering]

Originally Posted by masterizer

The way most mix engineers I know demo their mixes is to throw on a limiter with a ceiling of 0.1db, release set to fast or 0, threshold set to where its limiting 2-3db. This should be fine to demo your songs quickly.

So you set the ceiling at 0.1? Above digital zero. Really? Just sayin...

Sometimes it better to read than type,innit?

[okydoky]

Originally Posted by minister

Stan Cotey says it is TPDF.

hi,

thanks!

i figured it was tpdf, but i am wondering if it is plain old tpdf or high pass tpdf. the reference guides say "noise shaped dither", but i think they might mean high pass tpdf [or its a mistake].

is there some sort of discussion about this somewhere that you could post the link for?

i was going to call them on this new thing i have discovered, called a telephone. apparently it comes included with your internet service. who knew?


right.

[bob katz]

Originally Posted by oky****

hi,

thanks!

i figured it was tpdf, but i am wondering if it is plain old tpdf or high pass tpdf. the reference guides say "noise shaped dither", but i think they might mean high pass tpdf [or its a mistake].

is there some sort of discussion about this somewhere that you could post the link for?

i was going to call them on this new thing i have discovered, called a telephone. apparently it comes included with your internet service. who knew?


right.

A good FFT analyzer (Spectrafoo suggested) will answer all your questions, even the ones you haven't asked, like:

---is the dither working properly?

---what is the noise shaping (if any)

If you're that serious about how your DAW works internally, then get the test tools to prove that it is working properly.

[okydoky]

Originally Posted by bob katz

A good FFT analyzer (Spectrafoo suggested) will answer all your questions, even the ones you haven't asked, like:

---is the dither working properly?

---what is the noise shaping (if any)

If you're that serious about how your DAW works internally, then get the test tools to prove that it is working properly.

hi,

thanks, i heard that program is probably a good one.

but i actually think it is a simple question, especially with the amount of people working with pro tools, and / or mastering stuff that has been produced in pro tools [including redithering files that would already have dither present due to the somewhat omni-recommended use of the "dithered mixer"].

its weird that the information isn't common knowledge. sounds like everyone is guessing.

[edit] i would hope they have it working properly, whatever it is, although i hear you on that, and it probably wouldn't hurt to verify it. it seems to me that the "noise shaped dither" denotation in the manuals could very well mean high pass tpdf. but if that's true, it may have a bearing on what dither would be appropriate for a subsequent final word length reduction. and if they are literally using some sort of actual noise shaped dither, then it might be better to just mix without it unless maybe you're mixing and mastering somehow all in one go "in the pro tools box" [not generally the plan].

i know you like to stay on top of that kind of thing. [edit].

i should be able to get the information.


right.

[Bob Olhsson]

Originally Posted by oky****

..its weird that the information isn't common knowledge. sounds like everyone is guessing...

The information IS common knowledge but the quantity of BS about the subject floating around the net and printed in software manuals is astounding.

[okydoky]

Originally Posted by Bob Olhsson

The information IS common knowledge but the quantity of BS about the subject floating around the net and printed in software manuals is astounding.

hi,

the pro tools manuals are usually pretty straight up, although i do think there are a few issues that they tap dance around, and sometimes that appears to be for fear of liability.

i do not think this issue is really common knowledge. you said it was uncorrelated dither earlier. fine. but exactly what kind of uncorrelated dither? saying its tpdf does not exclude the possibility that it is high pass tpdf. and that is a distinction that could be signficant.

i think its probably tpdf, maybe high pass tpdf.

if you know for sure exactly what it is then you may want to say so, and how you determined it.


right.