32 bit float

Posted on

Tim the Enchanter
Member Since: Feb 17, 2008

What actually is 32 bit float? Is it a true 32 bit? Is it better quality than recording at 24 bit?

[ Back to Top ]


Member
Since: Jul 02, 2003


Sep 10, 2008 12:22 am

A 32 bit float is a 23 bit mantissa, with an 8 bit exponent, and 1 sign bit.

It's not any better quality as far as recording goes, but when processing the tracks with effects etc, the accumulated rounding errors that occur with integer based operations are greatly reduced, since only the final result needs to be rounded up/down with FP.

Also as far as recording/playback, no audio interfaces I'm aware of actually record/playback in 32bit FP this all takes place in the software you use, so it doesn't really matter whether you select to record in 32bit FP, the conversion can and usually is done by default in most DAW software these days, except ProTools which still works soley in 24/32 bit integer as far as I know.

Dan

I tune down down...
Member
Since: Jun 11, 2007


Sep 10, 2008 11:16 am

Ugh... What'd you just say? Haha.

Administrator
Since: Apr 03, 2002


Sep 10, 2008 11:19 am

yes, it's more gooder in 32 FP, but it may or may not be used in your system at certain points :-)

Ne'er ate 'er
Member
Since: Apr 05, 2006


Sep 10, 2008 12:36 pm

http://www.splodge-plog.com/wp-content/splodge-uploads/2008/02/my_brain_is_full.JPG


Czar of Midi
Administrator
Since: Apr 04, 2002


Sep 10, 2008 06:05 pm

Herb, whered yer post go??

Veni, MIDI, Vici
Member
Since: Jul 02, 2008


Sep 10, 2008 10:25 pm

It's still there, I guess? I can see the picture. Maybe Herb's brain is full too? >:-)

Czar of Midi
Administrator
Since: Apr 04, 2002


Sep 11, 2008 05:37 pm

Ya, I noticed it when I came back later. Funny stuff Herb.

Related Forum Topics:



If you would like to participate in the forum discussions, feel free to register for your free membership.