|
When does bit loss occur? (pg. 3)
|
View this Thread in Original format
| Eldritch |
| quote: | Originally posted by echosystm
you can actually. :p
|
I meant that it's not a choice. It will use 32-bit float internally regardless if you've chosen 16-bit, 24-bit or 32-bit for your project.
| quote: | Originally posted by echosystm
it is of particular benefit for freezing/rendering parts.
|
Yeah, that's what I said. :conf: |
|
|
| DigiNut |
| quote: | Originally posted by echosystm
you can actually. :p |
I think he meant internally, not on export. As most sequencers are written in C/C++, single-precision float is not only the obvious choice for storing signal information but also the only one.* There is no native fixed-point data type.
* OK, so they could also use double-precision (64-bit) floating point, but there was never any rational reason on 32-bit Windows. Maybe we'll see that as 64-bit Windows becomes more mainstream. |
|
|
| echosystm |
| quote: | Originally posted by Eldritch
| quote: | Originally posted by Magnus
Does this mean I should be writing my tracks in 32bit float? Since you mentioned this, I looked and I see now in Cubase that is an option under project setup. I've always had this set to 24bit.
|
You cannot choose 32-bit floating point
|
That's how i read it! Why don't you be a bit more confusing man... :p
But anyway...
| quote: | Originally posted by Eldritch
Go with 24-bit/44.1kHz or 16-bit/44.1kHz. You only need to use 24-bit if you're mastering the track externally.
|
This is why I said use 32bit float, instead of 24bit, in your Cubase project setup; you get no loss when you render or freeze bits. |
|
|
| Eldritch |
| quote: | Originally posted by echosystm
That's how i read it! Why don't you be a bit more confusing man... :p
|
Why don't you be less of a smart ass, seriously. :rolleyes: |
|
|
| echosystm |
| quote: | Originally posted by Eldritch
Why don't you be less of a smart ass |
ahaha that will be the day ;)
to be honest, i still don't understand what you are talking about. why would you chose 24bit when 32bit is there? |
|
|
| Magnus |
So the programming of the sequencer is 32bit float, regardless of what you set your project settings to. However from what I gather although there seems to be come conflicting points of view on this, is that 32bit float, 44Khz is the ideal choice?
 |
|
|
| Subtle |
You can export at 32 bit float.
If the sequencers internal programming is at 32 bit whats the point in having that as an option at all, unless.. its not true.
Its generally recommended to record at 24 bit, atleast thats what alot of producers say. |
|
|
| Derivative |
| quote: | Originally posted by echosystm
jesus christ we have been over this a million times.
ANYTHING OTHER THAN 44K IS A COMPLETE WASTE OF TIME IF YOU KNOW IT IS ONLY GOING TO END UP IN 44K ANYWAY. INFACT, HIGHER RATES MIGHT ACTUALLY SOUND WORSE AFTER DITHERING.
BIT DEPTH, THE HIGHER THE BETTER. |
Thats wrong though. 24/96 properly dithered down to 16/44.1 will sound better than working straight in 16/44.1.
The same can be seen visually by saving a 16 bit bmp picture to 4 bit bmp and then doing the same with a dither filter before truncating bits. The dithered image always has more detail despite having the same palette of colours because the speckled noise helps to define transitions between light and dark.
All of this is academic if your mix sounds though since you can work at 24/96 and dither it down to 16/44.1 like a pro but it will just be a ty mix that has been properly dithered. |
|
|
| echosystm |
| quote: | Originally posted by Derivative
Thats wrong though. 24/96 properly dithered down to 16/44.1 will sound better than working straight in 16/44.1.
|
well, i said "may" sound worse. :p
this is true though. going from 96 to 44.1 puts a fair bit of faith in the dithering algo, in my oppinion. it may sound more true to the signal, but ends up sounding different from what you expected. this is what worries me and is the reason i use 44.1. going from 88.2 to 44.1 is likely a lot more reliable, since it could involve much more simple and accurate (due to less floating numbers) calculations.
whenever i asked this question, people always said "just use 44.1, because the result will be as predictable as the source". it made sense! we had a big thread on this before. i think diginut (?) went into some kind of mathematics/science trance, explaining why 44.1 is the win. maybe someone can dig it up!
| quote: | Originally posted by Magnus
So the programming of the sequencer is 32bit float, regardless of what you set your project settings to. However from what I gather although there seems to be come conflicting points of view on this, is that 32bit float, 44Khz is the ideal choice? |
i make all my projects 32bit float, like i said above. i don't upconvert samples on import though, because i don't see the benefit. recording into your project, 32bit won't really make any difference because your soundcard is only 24bit anyway. however, when you bounce or freeze parts, you want them to be as high quality as possible. in this case, this is 32bit float. :)
regarding the sample rate, people in this thread are divided. if you use 44.1 in, its going to sound the same as 44.1 out. if you use anything else, you rely on your dither. i'm sure most are robust enough to make the difference negligable, but it's a bit of a moot point, since you won't really get much benefit out of a higher sample rate other than for oversampling effects etc. if you want to use a higher sample rate, i guess a safe midway is 88.2, this way it's almost just a division by two (of course, it is more complex than that lol). |
|
|
| Magnus |
| This is why I love this forum. :) Some great and informative information here. A huge thanks again to all! :) |
|
|
| DigiNut |
| quote: | Originally posted by Subtle
You can export at 32 bit float.
If the sequencers internal programming is at 32 bit whats the point in having that as an option at all, unless.. its not true.
Its generally recommended to record at 24 bit, atleast thats what alot of producers say. |
OK, what? Man, sorry, but you're not making much sense on these forums lately.
How could you possibly export a 32-bit float wave file if the sequencer wasn't already keeping the information in floating-point form? Are you actually suggesting that it upconverts when you bounce?
Export is export. It's not the same as mixing. No matter what you export, you're dithering from 32-bit float (unless you export in that format).
Now if you're talking about the "Record Format" in the Cubase project options, that refers to when you freeze instruments or process audio tracks. Or when you - get this, I know it'll shock some of you that this actually still happens - record a live instrument or voice. And there's a very good reason to use 16-bit or 24-bit instead of 32-bit: space. Most of you probably never touch the audio features in Cubase but some of us do, particularly if doing trendy edits like stutters and glitches. My last project had over a gig of audio files lying around; if the audio had been kept in 32-bit it would have been between 1.5-2x as large. Incidentally, this is with NO freeze files or full bounced tracks; just for effects and edits.
Higher bit depths also place a greater strain on system resources, particularly the hard drive. If you're playing LOTS of raw audio at the same time, then using 16-bit instead of 32-bit tracks effectively doubles the amount your system can handle. In practice, it's not quite double, but it's close. Same with the sampling rate; you're cramming twice as much data into 1 second of audio if you double it.
So the question of "why would they even give you that option" is a rather silly one. A lot of producers don't seem to realize that higher quality isn't free, and if you can barely hear the difference or can't hear it at all, then you've hit the point of diminishing returns and probably ought to lower your bits and/or rates.
Even if you do all your internal recording and final exporting at 16 bits, though, the internal mixing in a software sequencer is still happening at 32-bit float. It just is. EOD. |
|
|
| DJ RANN |
| quote: | Originally posted by DigiNut
Or when you - get this, I know it'll shock some of you that this actually still happens - record a live instrument or voice. |
ing heracy! You should be taken out back and flogged for making such a suggestion. Wash your mouth out.
Seriously though, your post addresses the fact that I think people aren't getting the difference between 32bit Floating operation (for your DAW to run on as an function OS/hardware) and bit depth in terms of audio recording/exporting.
On the other note, there was actually a lot of evidence preseted in that thread (Echo) to support both reasonings of the use 44.1k or 96k, but basically it was all moot cause unless your tracks contains real instruments recorded at that level, and all your kit was pro studio (not stuff you can readily buy from audio retailers), even then you would probably have a very hard job being able to ID which was the higher sample rate.
Anyway, at best/if you're lucky it's going to end up on a ty car radio or an over clipped sound system, so 44.1 already allows for so much quality (more than we can hear), so why bother. |
|
|
|
|
