I don't know why you think that.
When I listen to my 24Bit project before I bounce and dither it, it sounds like it sounds.
Once it's dithered, with any algo, it sounds different.
Actually, the "whole point" of using dither is not that you "not gonna hear it"!
That's funny to me.
Perhaps read up on why we use dither?
Which is strange, because other things you say do actually make sense generally.

 

No you cant hear it, and that is the whole point. You can mistreat a piece of dithered audio so that the dither noise becomes evident. This is not "hearing dither and which algorithm"

Yes of course different algorithms sound different: they are different flavors of noise! There is no black magic, the concept of dithering is very simple

Theoretical reasons? you reduce bitdepth and you wonder why you cant fit all of the previous information?
Do you reduce the size of your HDD and then wonder why you cant fit all the files?
Oh yes, cos theoretically, it is smaller.

 

Ok, If you hear it like that im pretty shocked. What/how are you dithering your files?.. i dont know which daw

I reduce the bitdepth of one of mine (using mp3 LAME converter = piece of shit dithering options)
then phase cancel it with a floating point wav, i get very very low level noise.

I cannot hear this noise unless i boost the fakk out of it.

How is this not your experience
(to clarify no i cant hear the dither between my mp3 and floating wav, i can on the other hand hear the reduced highend spectrum because ive also reduced the sample rate after export from daw)

Why dont you let izotope/voxengo/weiss products do your converting and you might stop hearing it

 

Last edited: Sep 18, 2020

I don't recall anyone saying they can hear which dither. I'd be sceptical of that! We can just hear it has been dithered.

 

Last edited: Sep 18, 2020

Ah, because as I already said, mastering is not what I do if I can help it, and straight away you sound like you are more skilled than me at that end of the production process!
It's totally possible (in fact, it's fact!) that I need to read up on how to get better with dithering.

Oh gosh, I've tried many different algos on a few apps.
Sometimes I just dither in Logic, or sometimes I may use iZotope's algos.
And I'm never happy with the results, but others are and I can get used to it!
"I'm a musician, not a darn scientist, Jim!" (Star Trek ref!)

edit: just seen your edit.
I forgot about Voxengo and Weiss. And I even have them too.
But if I'm not so successful with Logic or iZotope algos, then I don't know if I will get any better of a result!
It's not that the end result is bad per se, but as I've said, I always prefer my finished, undithered mix.

Maybe some algos are better for certain types of music with certain depths of dynamic range?
Is this possible?
But the fact that when others do it for me and I'm like "Yeah, great!" probably means I just don't like having to witness the actual transition from undithered to dithered, because I've got too close to the track by that point.

 

Last edited: Sep 18, 2020

Mission accomplished

This really cranked me up lol

 

I meant file servers and P2P networks. In streaming audio I met 64 kbps files streamed at 192.

 

Last edited by a moderator: Sep 18, 2020

POW-R (Psychoacoustically Optimized Wordlength Reduction) has typically three basic modes and is considered ‘the most sonically transparent dithering algorithm possible’. It is used by many renown companiies such as Digidesign, Cakewalk Sonar, Weiss, Ableton, Magix / Sequoia / Samplitude.
Since it is licensed, you can rarely find it elsewhere.

 

Así se hace, you do believe that fake videos can solve the mystery of fake files?

 

Last edited by a moderator: Sep 18, 2020

Sincerament, no me'n recordo de què parlàvem.

 

Sobre chicas falsas en videos reales.

 

Yea, its not something i worry about either as i play floating point wavs live, so its only an issue for cd/streaming etc
I dont know what Logic dither does, i assume its fine but Izotope is definitely great, i used voxengo r8brain before sticking with it but dither is included in RX so thats now all i need. You can tune the dither as well and i found it makes a huge difference and some introduce very visible noise in the spectrum.
You say you dont usually master if you can help it, so then you wont be dithering stuff either as its always the last process.

 

And Izotopes MBIT+
http://help.izotope.com/docs/ozone/pages/modules_dithering.htm

Some people say that standard TPDF dither is nicer to use before mp3 conversion or other processes, i havent ever checked as i dont expect to process those files any further.

AirWindows guy is always coding new dithers, heres about his TPDF plugin

 

TPDF 'makes no effort to optimize the sound in any way'...
The role of dither is to randomize quantization error /reduce the quality loss/, meaning that it's job is to reduce the error level, not to optimize the sound.

 

You sure can hear dithering, well, once we're talking about audio with like 8 bits or something like that, amplified.

From 24bit to 16bit though? You'll never be able to. Claim otherwise as much as you like, but there's a reason why dithering algorithms don't do taste tests that don't involve signals which essentially were 7-9 bit signals in 16bit headroom scenario amplified to the max: because it's not audible.

There's also a fun proof that Steven Massey made about this. You see, one of his plugins was a dithering plugin - a very popular one at that. Thing is, there was a very common scenario that could occur where the dithering wouldn't work. Guess how many people noticed it, out of all the professional people who used his plugins?

Nobody. Nada. Not a single person. Not even himself, until he saw the lack of it in a spectral analyzer and realized that it's not there. And he developed the damn thing. Dithering did have its uses back in the days especially in really low bit stuff like, say, tape, which varies from what, like 7 to 9 bit or something? But these days, where 16bit is considered as the bare minimum standard? Nope, not a single benefit from dithering. But the good news is that even if you do dither zealously, it doesn't matter, since there is no audible difference between doing it and not doing it. Kind of cool, huh?

EDIT: Now to recall it, I believe it was one of his plugins causing issue with POW-R rather than some dither that he designed.

 

What I meant was I could possibly hear the dithering noise being emphasized by the compression and limiting of the radio station

 

And possible EQ-ing..?

 

Què dius? No t'entenc noi...

 

Mira el teu vídeo.
1. That girl guessing mp3s is a fake, because they say she can hear 18.1 kHz, but 0.1 kHz difference on such high frequency is too small to be discerned by ear.
2. The highest frequency I measured that someone could hear was 17 kHz, and that happened only once, with a 18 old classical piano student.

 

Last edited by a moderator: Sep 19, 2020

The crappy boom box also has a limited range that may have emphasized the noise used in dithering (because noise is added at in the process) as well as the comp/limit of the station. Or maybe the songs were recorded and mastered so horribly. Probably not.
When CRT's make a high pitched "ring" it was about 17K. I could hear that into my 30's. I haven't been around a CRT since then, so I don't know if I could still hear it. I heard a high pitched sound at work from next door that only the younger(20's) people could hear. Nobody else my age or older could hear it. so my frequency spectrum is still pretty good at my age.