64-Bit Cpu Has Nothing To Do With Audio Precision


(Conner_Bw) #1

References:
[1] http://www.steinberg.net/en/company/press/archive/steinberg_press_room_archiv_2006/steinberg_asio_22_standard.html


(Djeroek) #2

You could have just quoted Taktik, Conner :)

(X64 Windows & Renoise - New Renoise Feature Priorities )


(rhowaldt) #3

i wonder where this post is coming from, all of a sudden…


(kazakore) #4

Wonder if he will believe me yet…


(RANSOM) #5

Yesterday i showed this thread to some professional russian audio engineers (they all work on Protools hardware, Pyramix, Sadie and other exotic platforms), they confirmed this topic - cpu bit and os bit has nothing to do with audio precision.


(rhowaldt) #6

although i appreciate your helpful nature, i was being sarcastic :)


(TiLT) #7

As a professional programmer, I can chime in and add my own voice to the chorus of people saying that the host has nothing to do with dynamic depth. I code 32-bit applications all the time, and there are absolutely NO (practical) limits regarding my use of 64-bit variables or data streams. Bit_Arts’ confusion seems to come from not understanding that 64-bit data types can exist and be tossed around in a 32-bit OS, and that this has been possible for ages.

In other words, a 32-bit VST with 64-bit audio processing will both use 64-bit audio internally, in its communication with the host, and within the host itself (unless the host deliberately downscales the signal to a different bit rate, but that’s a different topic entirely). I guess it’s entirely possible that someone would be able to present to me an application that fails to qualify in that it somehow delivers a 32-bit signal to the host despite using 64-bit processing internally (on a 32-bit OS). That would be the result of a bug within the VST (or host application), or even a design decision. Whether or not the OS is 32- or 64-bit doesn’t even factor into the equation and would have no effect on the process.


(Drew Tweedy Music) #8

so, question, if all of this is true, why don’t my 32-bit versions of Renoise have the option to render in 64-bit wav?


(Drew Tweedy Music) #9

A smiley face doesnt do a good job of answering my question…


(TiLT) #10

Why on Earth would you want to do that? I can’t think of any reason why anyone would find such a feature useful.


(Drew Tweedy Music) #11

not gona lie, im a massive audiophile. Ive mixed certain moggs down to 64 bit wav before and it was like cocaine in audio form. i know it would be completely useless in terms of serious prodction, as when you make it a 16 bit mp3 it gets rid of all the quality anyways, but for personal uses, it would be awesome.
I guess i can just render all the tracks in my song as separate files then render them together in 64-bit wav with Audacity or something.


(vV) #12

Because he has the harddrive space and doesn’t know anything else to do with it.


(kazakore) #13

No, he has a pair of these mythical Golden Ears!


(Djeroek) #14

ps2 = 128 bit > nintendo 64


(kazakore) #15

Sony PS1 (32bit) > Atari Jaguar (64bit)


(fladd) #16

You cannot even tell the difference between listening to a 24bit vs a 32bit recording, so I don’t see the point.


(vV) #17

There is no difference between 32-bit and 24 bit, if there is no clipping in either file. 32-bit = 24 bit with clipping info. (which means that destructive data is stored with a 32-bit file whereas a 24-bit file would require a new remix/remastering if it is rendered with clipping)


(fladd) #18

Interesting, didn’t know that. So what is the difference between 32bit and 64bit audio then?
Anyway, my point was that one is not able to hear the difference at all. I mean really no one will be able to tell in a proper double blind test!


(Drew Tweedy Music) #19

I can tell the difference… i guess im the only one here that has an ear for quality… jeeze…


(Drew Tweedy Music) #20

i think i do actually… i find it surprising that no one can tell the difference… its just so obvious to me…