Go Back   Cockos Incorporated Forums > REAPER Forums > REAPER Feature Requests

Reply
 
Thread Tools Display Modes
Old 04-18-2007, 01:02 PM   #121
Bubbagump
Human being with feelings
 
Bubbagump's Avatar
 
Join Date: Nov 2006
Location: Columbus, Ohio
Posts: 2,028
Default

I understand everything except LSB... so it is a "dither gain" I assume more dither isn't necessarily better? Is this an adjust to taste sort of thing and why was 2 chosen as the default? (Or more correctly, why did griz_lee suggest 2?)
Bubbagump is offline   Reply With Quote
Old 04-18-2007, 01:11 PM   #122
schwa
Administrator
 
schwa's Avatar
 
Join Date: Mar 2007
Location: NY
Posts: 15,749
Default

People who write about such things think LSB of 2 is the ideal dither bit width. If it's much lower it doesn't completely mask truncation noise, if it's much higher it sounds like a bunch of hiss.
schwa is offline   Reply With Quote
Old 04-18-2007, 01:35 PM   #123
Bubbagump
Human being with feelings
 
Bubbagump's Avatar
 
Join Date: Nov 2006
Location: Columbus, Ohio
Posts: 2,028
Default

Dig. That's pretty sweet then... out of the box and ready to roll with psycho then. Cool!
Bubbagump is offline   Reply With Quote
Old 04-19-2007, 05:18 AM   #124
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default

Thanks for that Schwa! I really don't read the forum often enough.
griz lee is offline   Reply With Quote
Old 04-19-2007, 11:26 AM   #125
schwa
Administrator
 
schwa's Avatar
 
Join Date: Mar 2007
Location: NY
Posts: 15,749
Default breakthrough!

I know people are bored of dithermania, it's so March 07, but there's some new stuff.

Billoon (aka QA Terrier) noticed that the 48k psycho dither generated very very faint psycho shaped noise even with the dither off and no audio signal, which is weird. Without getting too mathematical, I think the cause was summed floating point error, which led to the realization that there are multiple solutions to the noise shaping equations (complex roots). Adding a constraint that forces the floating point error to converge led to what I think is a completely independent solution to the previously-solved 44.1k noise shaping problem, as well as a new solution system for the 48k problem (as well as for any other sample rate, I think).

In any case, blah blah blah, I updated the uber dither with the new coefficients, the 48k noise shaping is clearly better (well, clearly when you isolate the noise and boost it 100 dB, anyway), and the errors converge so Billoon's ghost noise goes away.

In my little world at least, this is pretty exciting.

Last edited by schwa; 04-19-2007 at 11:40 AM.
schwa is offline   Reply With Quote
Old 04-19-2007, 11:39 AM   #126
schwa
Administrator
 
schwa's Avatar
 
Join Date: Mar 2007
Location: NY
Posts: 15,749
Default

Couple of nice pictures courtesy of Voxengo Span...

Truncation noise when rendering 64 bits to 16 with no dither or psycho shaping:


Noise from the same render with dither and psycho shaping turned on:


Human hearing threshold at frequency curve:


... dither "whitens" the ugly digital truncation noise, psycho shaping pushes the noise energy around until it matches the human hearing threshold curve.
schwa is offline   Reply With Quote
Old 04-19-2007, 11:43 AM   #127
Diogenes
Human being with feelings
 
Diogenes's Avatar
 
Join Date: Aug 2006
Location: A place that allows me to protect myself...
Posts: 8,245
Default

Schwa, you are THE man! Good stuff!

Explanations AND pretty pictures. Nice.

D
Diogenes is offline   Reply With Quote
Old 04-19-2007, 12:32 PM   #128
griz lee
Human being with feelings
 
griz lee's Avatar
 
Join Date: Jul 2006
Location: in a hotel room near you
Posts: 1,175
Default

Wow. That's detail, and stuff like that is important in ditherland.

Can this get into the next release, i wonder? Cool work Schwa.
griz lee is offline   Reply With Quote
Old 04-21-2007, 06:29 PM   #129
schwa
Administrator
 
schwa's Avatar
 
Join Date: Mar 2007
Location: NY
Posts: 15,749
Default

Updated the uber dither. Now supports psycho noise shaping at 44.1, 48, 88.2, 96k sample rates.

(All other dither features work at any sample rate, but psycho is only calibrated for those 4.)

... thanks Billoon for given these a test listen, as usual ...

Last edited by schwa; 04-23-2007 at 11:38 AM.
schwa is offline   Reply With Quote
Old 04-23-2007, 06:15 AM   #130
schwa
Administrator
 
schwa's Avatar
 
Join Date: Mar 2007
Location: NY
Posts: 15,749
Default

Minor update to the uber dither: Removed overload disabling. Simply clamping the output prevents feedback divergence (aka brain explosion).

This should also help CPU a little bit.
schwa is offline   Reply With Quote
Old 04-23-2007, 02:23 PM   #131
norbury brook
Human being with feelings
 
norbury brook's Avatar
 
Join Date: Mar 2007
Location: London UK
Posts: 3,378
Default

Quote:
Originally Posted by LOSER View Post

And to sum the above up: I plead for two additional post fader master FX slots (one for a limiter and the other one for a dither plug-in). Maybe include a (flat noise) dither into ReaInsert, too?

Which is exactly what Cubendo has for the above reasons.

MC
norbury brook is offline   Reply With Quote
Old 04-23-2007, 02:27 PM   #132
LOSER
Human being with feelings
 
Join Date: May 2006
Posts: 2,373
Default

Quote:
Originally Posted by norbury brook View Post
Which is exactly what Cubendo has for the above reasons.

MC
Why stop there? - Post fader FX per _track_!!!


Nice work Schwa, you obviously got the math .
LOSER is offline   Reply With Quote
Old 04-23-2007, 02:37 PM   #133
schwa
Administrator
 
schwa's Avatar
 
Join Date: Mar 2007
Location: NY
Posts: 15,749
Default

Quote:
Originally Posted by LOSER View Post
And to sum the above up: [b]I plead for two additional post fader master FX slots (one for a limiter and the other one for a dither plug-in).
That point is worth making extra clear: if you insert dither_psycho anywhere, you can't so much as touch the signal after that or it will get un-bit-reduced and then re-truncated. You can't even move a fader downstream from dither_psycho. It has to be the very last thing that happens to the audio before it's written to disk.

Really it should only be on the master, and only with the master fader at zero.
schwa is offline   Reply With Quote
Old 04-23-2007, 02:47 PM   #134
LOSER
Human being with feelings
 
Join Date: May 2006
Posts: 2,373
Default

Quote:
Originally Posted by schwa View Post
Really it should only be on the master, and only with the master fader at zero.
Exactly and that's why we need post fader FX slots (at least ONE on the master) so you can actually move your master fader, PLUS have the fabulous psycho_dither in the "extra post fader specailly reserved for dither plugs" slot thingy. (And while Justin is at it he can also add unlimited post fader FX on all tracks, just for the sake of it ... j/k ... well if he likes he can though .)
LOSER is offline   Reply With Quote
Old 03-27-2011, 02:51 AM   #135
mahasandi
Human being with feelings
 
mahasandi's Avatar
 
Join Date: Mar 2007
Posts: 478
Default

Hey guys I know I'm resurrecting an old thread but I've noticed that reaper has 2 dithers now js unsupported dither and utility dither psycho?

what are the differences?
mahasandi is offline   Reply With Quote
Old 04-03-2011, 02:07 AM   #136
Billoon
Human being with feelings
 
Join Date: Sep 2006
Location: Arse end of the earth.
Posts: 2,988
Default

Wow, this is an old thread.

They are both different implementations of dithering.

Dither psycho is the one to use if you don't like Reapers built in dithering or you want to dither your playback, etc.

Schwa's Psycho Dither would be a better choice though.
Billoon is offline   Reply With Quote
Old 04-24-2014, 08:13 AM   #137
giro1991
Human being with feelings
 
Join Date: Apr 2014
Posts: 9
Default

Hello!

I'm new on the forum. I'm really here to get this plug in as it appears it is the only 32bit dither in-the-world (or atleast with my google).

However, I notice the thread is titled VST though I cannot figure how to make it into a dll to be read as a vst in programs other than reaper, i.e universal VST.

I would have PM'd the dev directly, but I have too low post count .

Thanks!
giro1991 is offline   Reply With Quote
Old 04-26-2014, 05:13 AM   #138
Billoon
Human being with feelings
 
Join Date: Sep 2006
Location: Arse end of the earth.
Posts: 2,988
Default

Yeah, you can't use it in other programs directly. You may be able to use it with the Reaplugs bundle but i haven't tried it so not sure. You could use Schwas Psycho Dither though.

See here...

http://www.stillwellaudio.com/plugin...psycho-dither/
__________________
Fortune favours the prepared...
Billoon is offline   Reply With Quote
Old 05-08-2014, 09:27 AM   #139
giro1991
Human being with feelings
 
Join Date: Apr 2014
Posts: 9
Default

Hello, sorry for late reply but is schwa also 32bit?

Thanks
giro1991 is offline   Reply With Quote
Old 05-08-2014, 10:31 AM   #140
urednik
Human being with feelings
 
urednik's Avatar
 
Join Date: Apr 2010
Posts: 1,247
Default

schwa is every bit

Schwa is John Schwartz, a developer of Reaper. So could you please ask again and consider this fact?
__________________
W10 (64) Lenovo E540 - SSD; Lenovo B590; W7 (32), Compaq 610 (2.1Ghz core 2 duo, L2 cache, 2GB RAM); DPA 4018, Schoeps MK2, Schoeps MTSC 64, Neumann mk184, AEA Ribbon 88mk, AKG SolidTUBE; Focusrite Scarlett 18i20, recording merely live acoustic music.
urednik is offline   Reply With Quote
Old 05-10-2014, 06:20 AM   #141
Billoon
Human being with feelings
 
Join Date: Sep 2006
Location: Arse end of the earth.
Posts: 2,988
Default

Quote:
Originally Posted by giro1991 View Post
Hello, sorry for late reply but is schwa also 32bit?

Thanks
Yes.
__________________
Fortune favours the prepared...
Billoon is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 10:19 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.