Quote: Originally Posted by
RedGTiVR6 
durwood - great input!
your edit reminds me of an issue I was having trouble working through in my head.
If we do all of the processing on the computer, that means that we will be pretty much stuck with 48kHz.
Most of you know that I have been researching bit-accurate/bit-perfect playback in a bit of detail.
Then it hit me (don't you hate it when you overlook something so incredibly obvious!>)....it's not a bit-perfect stream if you tune on the computer....
but wouldn't it still make sense to strive for bit perfect in just about every aspect so that you know you are getting the true 44.1kHz rate before any processing? then you process the sound, and spit it out, making sure windows doesn't resample it to 48kHz then either...
is any of that coherent? anyone have any input?
Ya I have been tossing it around too and I thats why I want the computer to do all the processing. In the digital domain, its pretty much all mathematics.
With the computer AND a soundcard, there are basically two processors. Your CPU can do what ever it wants as long as the software tells it to do the right thing. Then you have your soundcard with its DSP processor on board that is limited to audio only related processing with limited resources.
In my thoughts, to maintain the ultimate SQ, this would be ideal:
Option 1
CPU
All sound adjusting is done by mathematics before the soundcard ever gets a hold of the signal
- Resampling 44.1kHz audio to 48kHz/96kHz (if needed depending on what the soundcard uses to process)
- EQing (via VST plugins)
- Time delays (Freq Allocator)
- Crossovers (Freq Allocator)
Disadvantages:
-The CPU will probably be doing ALOT of work.
Option 2 (If you have a kx driver compatible card)
CPU
Handles things the soundcard cannot do or do well
-resampling 44.1kHz to 48kHz or 96kHz depending on soundcard used
-EQing (via VST plugins)
KX driver compatible Soundcard
-Time delays (Kx driver w/UFX plugin package)
-Crossover (Kx driver w/UFX plugin package)
Advantage:
The workload is split between the two processors
It's not Bit perfect, but CD audio can be refined or smoothed out by resampling and not really affecting quality. The D/A dacs are really the only thing you need to worry about with SQ with both of the above scenarios. At least this is what I am thinking...
Any thoughts????