Hello Michael,

Was is the difference between double precision (64-bit) and 32-bit?

Regards,

Marc

p.s. Is there one better than the other?

----- Original Message ----- 
From: "Michael Gogins" 
To: ; 
Sent: Thursday, March 23, 2006 10:50 PM
Subject: [Csnd] New Csound 5.01 Windows installer


>I have uploaded to SourceForge a new Windows installer for Csound 5.01.
>
> Note the change in naming convention: Csound5.01-win-d.exe, "d" for 
> "double-precision samples".
>
> I will produce a version with 32-bit samples in the next day or so, 
> Csound5.01-win-f.exe.
>
> The installer has the same programs and plugins and creates the same
> directories as Istvan Varga's zip file.
>
> The installer also includes CsoundVST as both a GUI front-end and a VST
> plugin, tclcsound, csoundapi~, the VST opcodes, the Loris opcodes, the 
> fluid
> (SoundFont) opcodes, and API autodocs in addition to the complete HTML
> manual.
>
> New to this installer are winsound.exe, John ffitch's Csound launcher, and 
> _scoregen.dll, a Python score generator for VST hosts.
>
> This installer does not include sources.
>
> Please let me know of any bugs you find, or any suggestions you may have.
>
> Regards,
> Mike
>
>
>
> -- 
> Send bugs reports to this list.
> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk 

I'm no DSP expert, but... from a listening point of view, using 64bit values
to represent audio data allows for increased dynamic resolution which reduces
quantization noise.  

If you don't know what quantization noise is, an extreme example would be to take
an audio file of yours (say 24bit) and convert it to 8bit format.  Basically,
you'll hear a lot more noise.  Whether or not you can tell the difference between
32bit and 64bit depends mainly on the dynamic range of the material.  32bit is fine
for most amateur purposes with low dynamic range (< 80dB??).  You'll want to use
64bit for high dynamic range or if your mixing a lot of different sound sources
together.  Though your end product may use 24bit or 16bit, working in 64bit
all the way through a project (except for the very end) may have a significant
positive impact.

Davis

--- Marc Demers  wrote:

> Hello Michael,
> 
> Was is the difference between double precision (64-bit) and 32-bit?
> 
> Regards,
> 
> Marc
> 
> p.s. Is there one better than the other?