What does 32 bit float format do for you?

  • Thread starter Thread starter Bigus Dickus
  • Start date Start date
B

Bigus Dickus

New member
I've been playing around with the Cubase demo, and reading through the help/tutorial stuff. I found that resolution input modes are the standard 16/20/24 but also include this "32 bit float" format. AFAIK, there aren't any 32 bit A/D's on the planet, and suggesting that one could output in a floating point format is ridiculous.

So it means it's something internal then, right? Just what is this doing for you? Can you select 32 bit float with a 16 bit incoming signal? 24 bit?
 
It's for internal digital signal processing... less round-off error gets into the resultant 24-bit waveform.......
 
Then it would only make sense to use 32 bit float format for a 24 bit incoming signal, right?

What is the downside? Diskspace? Would that cause a problem if you later wanted to mix 24 bit tracks into the project, or are there plugins that won't function with the 32 float format?
 
No.. it's a good idea to use as high a resolution as possible for any digital processing.... regardless of the "starting" resolution...!

Compare it to a graphic image... if you start with lo-res, start editing it with lo-res, and render it at lo res, you end up with a big fuzzy mess...

If you start with lo-res, edit at hi-res, then render it back down to lo-res, you retain all the original detail plus your edits.

If you start with hi-res, edit at lo-res, then render to hi-res, you end up with a significant loss of detail, because of the lo-res editing....

If you start with hi-res and stay hi-res all the way through - your end-result is clean detailed.
 
:)

Yeah, that's what I had assumed, but I am still a little curious if it causes any compatibility problems with plugins. I'm new at this, so I don't understand the details of plugins, midi, etc.

Are there plugins that, for example, only work with 16 bit files and not 24 bit? Are there any (useful, good, etc.) that will work with 24 bit but not 32 bit float files? In general, will using 32 bit float in any way "handicap" me later with the choice of plugins, midi samples, VST instruments, etc.?
 
BD -- I haven't run into any problems using plugins with the various different bit depths. The downside of 32-bit would be performance issues, of course, since loading a 32-bit wave would take twice the memory of its 16-bit counterpart.

Though I haven't personally used it yet, 32-bit IIRC is for "Tape Emulation".


Chad
 
ok, cool, thanks for the info. I guess I'll use it then unless my computer chokes on the file.
 
Your bitrate shouldn't affect any plug-in's, vsti's and the like. As Blue Bear said (and was kind enough to explain to me) it is basically taking 24-bit info and internally giving it some more definition.

My DAW uses Cubase SX (and VST32 which I don't use anymore) to track, and I record at 32 bit floating...I then do my editing and mastering in SoundForge 6.0 which also accepts and works with 32-bit floating format, fly it back to SX as 32-bit and apply the final dithering in SX to get my final 16/44.1 product. By using compatible software it enables you to keep your projects at the higher resolution.
 
You must log in or register to reply here.
Back
Top