A16+ZLINK

Music Manic · Post by **Music Manic** » Fri Jul 25, 2003 8:35 pm

If A16 ultra converts to 24bit and sends it down Zlink,do the DSP's process at 32-bit which would waste the other 8bits.
vdat processes at 32bit but I am just wondering why 32 bits would help if the final file is 24bit.Wouldn't the audio word just be truncated.
Could someone expand on the dither process used by Pulsar DSP,because there's a clear difference in sound when recorded ro different sequencers.
Thx

marcuspocus · Post by **marcuspocus** » Sat Jul 26, 2003 2:30 am

I can't explain the dither process as i don't knownothing abotu this.

One thing i know, is that the extra 8 bits are usefull, for processing, it use 32bits float internally. This leave alot more precision in calculations.

If i remember well, i think Timework devices use 56bits internal processing for this reason.

voidar · Post by **voidar** » Sat Jul 26, 2003 8:35 am

The extra bits can be helpfull in recording smoother sound when for example an input (or vdat out) is processed with an effect (reverb i.e.), due to the internal 32 bit processing that SFP uses. I think

Music Manic · Post by **Music Manic** » Sat Jul 26, 2003 9:29 am

Yes I realize te importance of calculations,but what happens to the final file,when it becomes 24-bit?
Thx

Post by **garyb** » Sat Jul 26, 2003 12:55 pm

nothing bad.changing bit-depth is not especially harmful like resampling.32 bit just has 8 bits of headroom when a 24 bit signal is introduced.when the 24 bit signal leaves the 32 bit realm,any processes done at the higher resolution stay nicer sounding.

it's kinda like zooming in and out with the visual.if you have more lines of resolution,there will be more detail in the picture. and if you modify the picture and then lower the number of lines,the picture is still more accurate than if it had been originally shot at the lower level of resolution and was modified there.try it with your digital camera!