Floating point/24/32 bit processing? SampleRate conversion?

CoreyShay wrote on 8/12/1999, 12:53 AM
I have two questions...

1) Is Vegas able to perform 24 or 32 bit calculations on
16 bit files when using plug ins? For example, if I am
working with 16 bit files and have my Waves Q10 EQ plug-in
on Bus A which then feeds a Waves L1 Ultramaximizer, is the
signal after the Q10 EQ truncated back to 16 bits or does
it feed the L1 a 24 bit or 32 bit floating point signal?

2) Also, does real-time sample rate conversion take place
before or after my Bus A plug-ins? I would prefer it take
place before them.


One more thing... Since Sonic Foundry refuses to enable
standard directx plugins as track inserts, then why can't
it enable sound forge to do processing in 24-bit? It
sounds ugly when I have to truncate back and fourth to 16
bit. C'mon Sonic Foundry, you've got a killer interface on
all programs, but lets enable processing of larger
numbers. It's not much to ask!

Comments

pwppch wrote on 8/12/1999, 2:41 AM
1. All audio is converted to floating point before it is
processed/routed in anyway. How a "filter" deals with audio could
possibly cause a change/conversion between the app and the filters.
We always attempt to connect to a filter using floats and stereo.
However, if a filter does not support this, we will convert on input
and its output. The Waves plugins should never have a conversion.

2. Conversion to the "hardware format" (What your card supports) is
done after all processing. No sample rate conversion is done at the
output stage. If your project is 96 and your media is 44.1, the rate
conversion is done on read, before any processing. All conversion are
done on the raw media before any routing or processing.


Thanks
Peter


Corey Shay wrote:
>>I have two questions...
>>
>>1) Is Vegas able to perform 24 or 32 bit calculations on
>>16 bit files when using plug ins? For example, if I am
>>working with 16 bit files and have my Waves Q10 EQ plug-in
>>on Bus A which then feeds a Waves L1 Ultramaximizer, is the
>>signal after the Q10 EQ truncated back to 16 bits or does
>>it feed the L1 a 24 bit or 32 bit floating point signal?
>>
>>2) Also, does real-time sample rate conversion take place
>>before or after my Bus A plug-ins? I would prefer it take
>>place before them.
>>
>>
>>One more thing... Since Sonic Foundry refuses to enable
>>standard directx plugins as track inserts, then why can't
>>it enable sound forge to do processing in 24-bit? It
>>sounds ugly when I have to truncate back and fourth to 16
>>bit. C'mon Sonic Foundry, you've got a killer interface on
>>all programs, but lets enable processing of larger
>>numbers. It's not much to ask!
CoreyShay wrote on 8/12/1999, 9:01 PM
Hey, Peter, I just want to thank you for understanding my question
and giving me such a straightforward and insightful answer.

Again, thanks.