Another sample rate/bit depth conversion question
Old 7th February 2013
  #1
Gear addict
 

Thread Starter
Another sample rate/bit depth conversion question

The post below about sample rate/bit depth conversions really opened my eyes. I've been doing both as a one-step export right from my DAW - I didn't know there were BETTER converters for this.

OK then, are most of you doing sample rate and bit depth reduction (say from 96K/24bit to 44.1K/16bit) using two separate tools/steps? If so, which process do you do first?

Secondly, what is the advantage of working at 96KHz vs 88.2KHz? It seems the resampling artifacts would be less using an even multiple of the final intended rate, yes/no?
Old 7th February 2013
  #2
Lives for gear
 
Alexey Lukin's Avatar
 

Verified Member
SRC goes first, then goes a peak level check, then bit depth reduction. Most SRCs and dithers are good enough, so there's no need to always reach out for external solutions.

If SRC is good, there's no quality difference between 88 and 96 kHz. If it is bad, there may be some.
Quote
1
Old 7th February 2013
  #3
Gear addict
 

Thread Starter
Quote:
Originally Posted by Alexey Lukin View Post
SRC goes first, then goes a peak level check, then bit depth reduction. Most SRCs and dithers are good enough, so there's no need to always reach out for external solutions.

If SRC is good, there's no quality difference between 88 and 96 kHz. If it is bad, there may be some.
OK, thank you.

Now, assuming all the mastering is completed at 96/24, how do you maintain the same levels, and assure no ISPs, once converted? If you have to add a "safety" limiter in the chain somewhere downstream of the conversion doesn't that defeat the purpose of working at 96K?
Old 7th February 2013
  #4
Lives for gear
 
fastlanestoner's Avatar
 

I do it when rendering the stereo master of the track in Reaper.
Old 7th February 2013
  #5
Lives for gear
 
Alexey Lukin's Avatar
 

Verified Member
Quote:
Originally Posted by Nonlinear View Post
Now, assuming all the mastering is completed at 96/24, how do you maintain the same levels, and assure no ISPs, once converted? If you have to add a "safety" limiter in the chain somewhere downstream of the conversion doesn't that defeat the purpose of working at 96K?
Limit with ISP protection at 96 kHz, leave 0.1 dB headroom — and you should be fine after conversion.
Old 8th February 2013
  #6
Some SRCs have a setting to make sure they look at the file and if it clips after conversion, they will automatically bring it down until it is not clipping. I know both R8BrainPro and RX2 have this.
Old 8th February 2013
  #7
Moderator
 
jayfrigo's Avatar
 

Verified Member
In my typical workflow I capture the analog chain at 88.2. I do the peak limiting while at high sample rate because nonlinear processing tends to be cleaner, leaving a few tenths of head room, then synchronous SRC on the way into Sonic where editing, fades, and prep happens, and finally dither to reduce word length as the very last step.

Good software SRC is also used when needed, but you need to know your tools; not all are created equal. Weiss Saracon and iZotope have excellent offerings (especially all the options and control offered in RX Advanced), as is the old Sonic Solutions HD SRC. However, the SRC built into the early versions of soundBlade is not the same and is not great. There was talk of adding the original Sonic HD SRC and/or iZotope to soundBlade, but I'm not sure if this has happened yet. I haven't checked since I don't really need it.
Topic:
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
6777 / High end
69
dantheman360 / High end
20
R'shonn / Geekslutz forum
3
checkers1811 / Music Computers
1

Forum Jump
 
Register FAQ Search Today's Posts Mark Forums Read

SEO by vBSEO ©2011, Crawlability, Inc.