• SONAR
  • [Answered] Question about bit depth... (p.2)
2014/10/22 15:40:04
johnnyV
oranges said: 
"The sampling rate , if not handled correctly (and I don't think Sonar is capable of converting it)"
 
I'm surprised no one commented on this. 
I myself have used both 44.1 and 48. I am unaware that my 48 exports when set to export 44.1/16 bit were any different than my 44.1 exports ?? I didn't use 48 for very long as I saw no benefit for what I was doing. My 01V mixer is 44.1 so I'm stuck for now anyhow. 
 
2014/10/22 15:59:29
dantarbill
johnnyV
oranges said: 
"The sampling rate , if not handled correctly (and I don't think Sonar is capable of converting it)"
 
I am unaware that my 48 exports when set to export 44.1/16 bit were any different than my 44.1 exports ??




I know nothing about the accuracy of the SONAR sample rate conversions.  I have however, written sample rate conversion algorithms myself...and have inspected the results of some other commercial conversion products.
 
Converting from 48 to 44.1 involves a lot of interpolation to do correctly.  (It's a pain in the...rear.)  One software vendor I checked did a very clean sounding conversion, but they also used a mathematical shortcut that caused the resulting pitches to be a few cents off.  While it's not distortion in the "my wave is clipped" sense...it's still distortion.
Consequently, I don't trust tricky math.  If your target is CD (rather than video) I see no reason to tempt the math gods with a 48 to 44.1 conversion.  A better case can be made for 96 kHz, because you can push the brickwall filter farther up into the "now dogs can't hear it" range, which trumps my "tricky math" arguments.  This is also why I remain dumbfounded that 88.2 kHz recording (or 176.4 kHz if you must) isn't standard studio practice.  The conversion math is dirt simple.
2014/10/22 17:22:18
Anderton
orangesporanges
The sampling rate, if not handled correctly (and I don't think Sonar is capable of converting it)

 
If you want a real shock, go here where there's comparative, in-depth analysis spectra of various sample rate conversion programs and algorithms. Sonar has one of the very best - better than Pro Tools HD, Cubase, Reaper, FL Studio, Samplitude, Sequoia, Reason, even Wavelab and Sound Forge Pro. Sonar, Digital Performer, Ableton Live, and Audition all did really well. Most improved: Logic Pro X. Logic used to be horrible, now has one of the best.
2014/10/22 19:48:29
orangesporanges
Good to know, Craig. I'm not 100% sure I knew what I was looking at, so I defer to you. My mistake in saying it can't convert it, I have been using 44.1 forever, so I have never had a need to convert it. I guess my impression was that it should be a multiple of what you expect to end up with, and 88.2 was tough on my old system. Maybe I should try higher sampling rates and see what happens.
2014/10/22 20:33:27
Anderton
orangesporanges
Good to know, Craig. I'm not 100% sure I knew what I was looking at, so I defer to you.

 
Basically anything other than a single, bright curved line is a problem. More other stuff = more problems.
 
I have been using 44.1 forever, so I have never had a need to convert it. I guess my impression was that it should be a multiple of what you expect to end up with, and 88.2 was tough on my old system. Maybe I should try higher sampling rates and see what happens.



The idea that you can't do decent sample rate conversion from, say, 96 to 44.1 was true at one point. However with 64-bit processing, there's plenty of math that can be applied to solving the problem. Any errors are so far to the right of the decimal point they might as well not exist at all.
12
© 2026 APG vNext Commercial Version 5.1

Use My Existing Forum Account

Use My Social Media Account