The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
16bit>32bit float>16bit... Vs 16bit>16bit Dynamics Plugins
Old 25th April 2010
  #31
Gear Maniac
Bitter shows weird things in Bias Peak (looks ok in DP).

Bit Usage (which is a function that scan a selection of the audio) clearly shows the correct measures (16bits-32bits). While at the same time, Bitter shows what's on the pictures I've posted.

I don't use any other plugins while analyzing. Just Bitter, and Bit Usage.
Pow-R is supposed not to be working until the user choose it when exporting, you can't hear it until the file is saved and Bias Peak refresh the entire audio.

I have Jack, but it is not in use.

As I said, I have imported an official track from a CD into Bias Peak and it showed the same weird stuffs (see picture 16bit dither previously posted).

But in Digital Performer, everything seems to be fine for Bitter.
That's why I guess, Bitter and Bias Peak are not friends.

Anyway, tomorrow I'll test another dither...this time in DP. And will see/hear what happens.
Old 26th April 2010
  #32
Gear Maniac
Also,

Just thought about trying POW-R1, or POW-R2 in bias peak.
POW-R3 seems to add a little brightness for what I remember.
So I'm guessing that the original source file coud have been already dithered with such a dither shape. And maybe I should look for a more neutral dither.
Old 26th April 2010
  #33
Here for the gear
 

Quote:
Also,

Just thought about trying POW-R1, or POW-R2 in bias peak.
POW-R3 seems to add a little brightness for what I remember.
So I'm guessing that the original source file coud have been already dithered with such a dither shape. And maybe I should look for a more neutral dither.

there are a couple of issues with what you are asking.

the fact that you are now trying to "re-master" something that has already been "mastered" is an issue, and your predicament illustrates two of the "basic rules".

1. you should only use flat, tpdf dither unless you know for sure that no further processing will ever be done to the file.

2. you never know for sure that no further processing will be done to the file.


the other issue to consider is the fact that one cannot actually dither floating point audio in such a way as to remove all the quantization distortion anyway [this is assuming that processing / mixing is, or has been, done in the floating point format].


you may want to consider simply resampling the results of your floating point processing, using good converters. that will get you around at least some of the problems you are encountering.
Old 26th April 2010
  #34
Gear Maniac
Quote:
the fact that you are now trying to "re-master" something that has already been "mastered" is an issue, and your predicament illustrates two of the "basic rules".
Well... I'm not trying... ;-)
Actually I need to master this audio to get it on the release way.
Unfortunately, the mixes I've been sent are 16bit. And I've been told the recording was 16bit too. The question is: why the hell is there any dither on this?
Now, hard to know if a dither was applied during the mixdown for these mixes, or if any dither has been applied at any stages of mixing/overdub process...Go figure...

I'll try the DP+Ozone Dither this afternoon.
Old 26th April 2010
  #35
Gear Maniac
Some news...

I've just tried to import in DP the 32bit float file I previously exported from.
Bad news here, Digital Performer seems to be unable to import 32bit float file.

Back then in Bias Peak, with my 32bit float file.
When I put Ozone on and Bitter just after, if I dither to 16bit then Bitter is able to show the correct bit withotu any artefact.
But once I play back the dithered file, then Bitter shows the same weirdness.
Bit Usage still shows the correct information: 16bit used.

All I can say is that the sound seems to be less chwin-zing (< kind of metalic stuff you don't like on the high freq...). Kind of softness that the music needed anyhow.
To me it sounds ok like this, I have to check that again to make sure.

Especially the process in Bias Peak, applying the Dither to the 32bit float audio, then saving it at 16bit.
Old 26th April 2010
  #36
Here for the gear
 

Quote:
Originally Posted by _barnee View Post
Well... I'm not trying... ;-)
I need to master this audio to get it on the release way.
what exactly is "on the release way"?


Quote:
Originally Posted by barnee
The question is: why the hell is there any dither on this?
people tend to use dither where it is neither necessary or helpful. or maybe they thought they were finishing the process of word length reduction for cd [mastering].

Quote:
Originally Posted by barnee
Now, hard to know if a dither was applied during the mixdown for these mixes, or if any dither has been applied at any stages of mixing/overdub process...Go figure...
why not ask the people who did it?

Quote:
Originally Posted by barnee
I'll try the DP+Ozone Dither this afternoon.
i already told you the "solution" to your predicament. resample.

i think chasing around trying to find "a better dither" is going to a waste of your time. there is only one type of dither that is even arguably suitable for multiple passes. flat tpdf. and you'd still be better off resampling. unless you just happen to like the sound of dither.
Old 26th April 2010
  #37
Gear Maniac
Quote:
what exactly is "on the release way"?
The master being sent to manufacturing plant, to make CDs.

Quote:
people tend to use dither where it is neither necessary or helpful. or maybe they thought they were finishing the process of word length reduction for cd [mastering].
Why they could have thought to reduce the word lenght for CD, when the recording is already 16bit???

Quote:
why not ask the people who did it?
Because it's not a project done last week or month, but years ago.
And good chance it involved several people in the mixing process...

Quote:
i think chasing around trying to find "a better dither" is going to a waste of your time. there is only one type of dither that is even arguably suitable for multiple passes. flat tpdf. and you'd still be better off resampling. unless you just happen to like the sound of dither.
The Ozone dither sounds ok on this project. I'll have to check back tomorrow, but from what I've heard today it sounds way better than with POW-R3 dither.
As I said, I just need to make sure it's ok to use the Ozone Dither that way in Bias Peak. Other than that, I think it's fine.
Old 26th April 2010
  #38
Here for the gear
 

Quote:
Originally Posted by _barnee View Post

Why they could have thought to reduce the word lenght for CD, when the recording is already 16bit???
maybe they were just stupid. or maybe the project was actually recorded or processed at a higher bit depth. you stated you are not sure.

for some reason you seem to think that there is a dither problem on the files you got, so it makes no difference why.

if someone applied dither to a project without word length reduction, then you are stuck with it, and your best bet is to resample rather than use additional dither.

Quote:
Originally Posted by barnee
Because it's not a project done last week or month, but years ago.
And good chance it involved several people in the mixing process...
and you have no idea, or no way of contacting any of those people? you would rather guess?

Quote:
Originally Posted by barnee
The Ozone dither sounds ok on this project. I'll have to check back tomorrow, but from what I've heard today it sounds way better than with POW-R3 dither.
As I said, I just need to make sure it's ok to use the Ozone Dither that way in Bias Peak. Other than that, I think it's fine.
so you like the way the ozone dither sounds, then?
Old 26th April 2010
  #39
Gear Maniac
Quote:
maybe they were just stupid. or maybe the project was actually recorded or processed at a higher bit depth. you stated you are not sure.
Nope, I've been told the project has been recorded on 16bit. This is something sure.

Quote:
for some reason you seem to think that there is a dither problem on the files you got, so it makes no difference why.
I know how a dither could look like...There is some particular shapes. And the files I've been given to master contain a dither shape.
And for other some good reason, I can tell you it has been recorded (or mixed with external analog gears) with a TV monitor somewhere near.

Quote:
if someone applied dither to a project without word length reduction, then you are stuck with it, and your best bet is to resample rather than use additional dither.
I don't get it sorry... You suggest to make an extra DA/AD and record the result on my DAW, right? OK...Then I'm still with a 32bit float file, BUT with an extra DA/AD conversion. And finally, I will need to dither this file to 16bit.
So? what's the purpose of that extra DA/AD? I'd better find the Dither that will play well with my current file, without adding an extra DA/AD conversion.

Quote:
and you have no idea, or no way of contacting any of those people? you would rather guess?
Because I have already asked several questions, and seen not all getting replies...
On a different story, one day I have requested from a band to send me 24bit mixes. Guess what? They recorded all in 16bits, mixed on 16bit, get the stereo file and put it on a 24bit then sent me that...

Quote:
so you like the way the ozone dither sounds, then?
As for today, and what I heard, yes... The Ozone Dither sounds ok for this particular audio file and these particular circumstances.
;-)
Old 27th April 2010
  #40
Here for the gear
 

Quote:
Originally Posted by _barnee View Post
I don't get it sorry... You suggest to make an extra DA/AD and record the result on my DAW, right? OK...Then I'm still with a 32bit float file, BUT with an extra DA/AD conversion. And finally, I will need to dither this file to 16bit.
So? what's the purpose of that extra DA/AD? I'd better find the Dither that will play well with my current file, without adding an extra DA/AD conversion.
no. i'm suggesting you do whatever processing you are going to do in your daw, with the files you were given. then play the result out analog into a different daw [or any digital recording device]. set the second device to record a 16 bit file. there you go. no extra dither.

the "extra DA/AD conversion" is trivial if it is properly done. the bottom line is that you don't have to dither to go from one bit depth to another if you resample the output of the high bit depth source in the analog domain, and, if you are correct in assuming that it has already been dithered once, you are better off resampling than adding a second pass of dither.
Old 27th April 2010
  #41
Gear Maniac
Well...I don't have extra computers to do that. Got an old DAT machine, but it is out of order for a while now... Anyhow, going DA/AD will add some noise.
So, in my case I'd rather add the Dither noise I've found on Ozone, and which it seems to play well on this audio.
Old 27th April 2010
  #42
Here for the gear
 

Quote:
Originally Posted by _barnee View Post
Well...I don't have extra computers to do that. Got an old DAT machine, but it is out of order for a while now... Anyhow, going DA/AD will add some noise.
So, in my case I'd rather add the Dither noise I've found on Ozone, and which it seems to play well on this audio.
you are not really going to add noise by simply going da-ad, unless you use bad equipment or cables.

a second pass of dither is pretty much universally considered a bad idea. there you are definitely adding noise.

and, judging by your previous comments, you are going to have a file that has been heavily processed in 32-bit floating point, so dithering it is not going to remove all the quantization noise anyway. you have a case where you may actually be better off just truncating it rather than adding a second pass of [at least partially] ineffective dither.

but i would still recommend resampling if you can get your hands on the gear.

anyhow, good luck with it.
Old 27th April 2010
  #43
Motown legend
 
Bob Olhsson's Avatar
 

Verified Member
Quote:
Originally Posted by no.posers View Post
...a second pass of dither is pretty much universally considered a bad idea. there you are definitely adding noise.
..
You are adding noise however your only other choice is allowing considerably louder distortion.

Dither prevents distortion, it doesn't just cover it over. A second pass of dither where there has been absolutely no signal processing or gain changes is the only case where you want to leave it off.

And I realize that a lot of magazine articles and software manuals are dead wrong about what dither does or how to use it.
Old 27th April 2010
  #44
Here for the gear
 

Quote:
Originally Posted by Bob Olhsson View Post
You are adding noise however your only other choice is allowing considerably louder distortion.
in the orig. posters case, he also has an additional [better] choice of simply resampling, which would be better if he could get his hands on the necessary equipment to do it.

also, "considerably louder distortion" is probably a bit of an overstatement, in the orig. poster's case, no?

Quote:
Originally Posted by bob olhsson
]Dither prevents distortion, it doesn't just cover it over.
dither does not completely prevent quantization distortion when going from a file being heavily processed in the 32-bit floating-point environment to a 16-bit fixed point file. it may prevent some of it, but its not going to get it all [unless maybe you use some ridiculous amount of dither].

Quote:
Originally Posted by bobolhsson
A second pass of dither where there has been absolutely no signal processing or gain changes is the only case where you want to leave it off.
like it or not, there are some cases where, usually due to mistakes made earlier in the process, you are better off just truncating. you can't just keep piling on dither. in the orig. poster's case, he says it has already been dithered, although probably unecessarily. that is not his fault, but it does limit what he can do with it.

adding a second pass of dither to a file that the orig. poster says already has a "dither problem" would probably be not be helpful, esp. going from floating point where it is not all that effective anyway.
Old 9th May 2010
  #45
Gear Maniac
I'm back with one question.
Let's say I have 2 wav files, both 16bit.
One file is the song, the other is speech/radio jingles/beep/or whatever...

I import this in my DAW, on a 16bit project.
No fader changes, just both files running at the same time... mixed.

Will you then apply dither on the output when boucing?
Old 9th May 2010
  #46
Motown legend
 
Bob Olhsson's Avatar
 

Verified Member
If there are absolutely no gain changes or other signal processing of any kind, they won't need dither because there is just addition and no multiplication or division although they could clip. That's a pretty unusual case.
Top Mentioned Products
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
zzz / Mastering forum
3
m_gant / Mastering forum
14
Beezoboy / Music Computers
9
A1A2 / So much gear, so little time
7

Forum Jump
Forum Jump