Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[djwaudio]

Im wondering if anyone can explain if there is a difference between the upsample process that my M6000 BW2 does, vs a sample rate conversion. Since the 6000 is clocked at 44.1 when I'm running a CD master, I'm thinking the upsampling it does, is somehow more benign than if I were to just run a 88.2 signal and SR convert off-line. What's cooking under the hood?!

[Treelady]

It would be good if someone who worked on that stuff from TC could answer that.   There are alot of really well-implemented things TC does in their pro gear, but they do little to tout their horn about it.

[djwaudio]

Good point, I'll have to corner the engineers from TC at AES this year and get the story. 

I guess the question still stands for other "up-sampling" processors as well.  Just wondering how it's different.  I guess I'll have to dig up a Ken Pohlman book and look it up.

[bradsarno]

I've been using a GW Labs upsampler/jitter-reducer for a long time now. Since I operate on a single-clock system/loop this takes my typically 44.1kHz raw file from my 44.1kHz system clocked by a Burl B2, upsamples to 96kHz on the way to my D/A which feeds the analog gear. I can easily toggle between 44.1kHz and 96kHz, and it seems to do a perfectly acceptable job of upsampling. But for me the big benefit is the gentler slope of the 96kHz anti-alias filtering. Seems to be much cleaner and less phase shifted in the high, audible frequencies. Things like stick attacks and pick attacks become more distinctly separate and more clearly defined. Crisp elements become softer and more detailed, more dimensional. So I don't know if this GW Labs is a stellar upsampler, but the benefit is unquestionable. Feeding the analog chain a D/A signal that's got the gently sloped filters is a big plus.

Brad

[Treelady]

Have you compared the GW vs. upsampling your client source files to 96k and pitching those directly to your D/A?

[bradsarno]

Have you compared the GW vs. upsampling your client source files to 96k and pitching those directly to your D/A?

No, but I should and will. I guess it all comes down to algorithms, right?

B

[Treelady]

Having your files in one session that is 44k makes archiving and session management much easier.   If the hardware you're using sounds the same, then your current method is the best.  If you need to do any plug-in processing of the pitch audio before you DA, doing it at 96k (IMO) sounds better, but that means two sessions and more hard drive space.

[GlensAudio]

Hello Treelady (in Pittsburgh);

I'd go with the best sounding and just forget about the numbers.  A D/A (or A/D ) conversion involves slinging bits around your little box in ways you could only never imagine.  So, IMHO, if it'd digital and you like what you hear than go whit that.  Ears rule!  If 'that' presents challenges in archival so be it. There should be a method for compression in the archival process.

Glen