|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
65.19.73.221
In Reply to: RE: Jriver Internal Volume transparency posted by FlyCast on September 26, 2015 at 05:51:08
"The Jriver wiki page says their 64 bit Internal Volume is the highest possible quality software volume."
A potentially misleading claim.
The issue isn't the quality of the software volume control. It's what happens when the reduced signal exits the computer and hits the DAC. Even if the software volume control were "perfect" if the resulting signal is too low, then the DAC will lose resolution. This will not be an issue if the DAC accepts 32 bit input and has low noise, but it may be an issue if the DAC has 24 bit input or is noisy. It will certainly be an issue with a 16 bit DAC.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Follow Ups:
At least the dither can be improved upon
http://yabb.jriver.com/interact/index.php?topic=100095.0
It helps if a user called Bob Katz ask this type of questions :)
The Well Tempered Computer
Thanks for the link. Bob Katz knows what he's talking about. I read and studied his book on mastering some years back when I started to do audio restoration work, and noted his suspicion and need for quality control over digital signal paths to ensure that bits weren't being unnecessarily changed and that DSP was of the highest quality. (And it wasn't in the case of some early versions of SoundForge that I was playing with at the time.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Last time I used JRivers, it inatslled as a 32 bit program in the 64 bit OS. If this contains double precision code, then is means an additional layer of internal processing. Whether this is good, bad or neutral, I wouldn't know, but I would rather have had a true 64 bit package.
I would also have preferred an audio only, not AV package without loads of caches and registry entries.
Fred,You are confusing two separate issues in computer architecture:
1. Data formats, e.g. 32 bit floating point vs. 64 bit floating point.
2. Memory addressing: 32 bit processor addresses vs. 64 bit processor addresses.These are different parts of the processor architecture. Large (e.g. 64 bit) data formats have been around for decades and work perfectly well with 32 bit memory addressing, providing that the file sizes involved are below 2 GB (or possibly 4 GB). People were using large floating point numbers back in the 1960's, but there was no need for 64 bit addressing until computers starting having more than 4 GB of RAM.
About the only connection between the two items comes with large audio files, where a high sampling rate and a long word length might make the file too large to be handled efficiently by a 32 bit program that was written to use memory mapped file access or to preload an entire track into RAM. You can tell if this is applicable to an audio program by looking at its memory utilization with a task monitor. If the 64 bit data path is just used for on the fly signal processing, then there will be no significant increase in memory utilization, and hence no connection to 32/64 application / OS question.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Edits: 09/27/15
Digital volume control are still best avoided though, and they are not as easy as a knob or encoder:- a remote can actually be more sluggsh.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: