|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
65.19.76.104
In Reply to: RE: PC or...... Pono posted by AbeCollins on April 07, 2014 at 12:36:39
" The player software allows me to disable several services at a push of a button when playing audio only, then restore them back to 'normal' the same way."
Do you know which services are being disabled and what, if anything, task manager shows about their typical usage while they are enabled?
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Follow Ups:
"Do you know which services are being disabled and what, if anything, task manager shows about their typical usage while they are enabled?"
See page 16...
http://www.wiebel.nl/zut/Audirvana%20Plus%20User%20Manual.pdf
Actually does very little. The other settings he uses are from he $5 ITune books that Steve N. always hawks. The info mostly came from the CA forums and at one time was posted for free on a certain website.
the CAD script for MAC is more extensive and more effective.
There is also a CAD PC script.
That CAD script was pretty buggy. I just looked at it again and it appears to be more extensive than previous versions with (hopefully) some of the bugs fixed. In any case, I've tried many of the more relevant script lines manually on my Mac.
Thanks. That's about what I would have expected. I'm not an OSX person, but there are similar items on Windows, e.g. System Restore. I've not seen problems on my system with system restore interfering with audio, because it is tracking file system changes. Normally during playback (assuming no other applications that shouldn't be running or evil background processing that also shouldn't be running) there won't be any file system writes, nor any file system operations at all except reads and writes. The bit about USB scanning might be a good one for other systems as well, as it is working the interrupt system and thereby interfering with latency. But I don't know how often it happens on Windows. In any event, the optimizations on page 16 seemed directed toward avoiding clicks or other buffer over/underruns, and not focusing on subtle sound quality. And not focusing on the potential benefits of small buffer sizes, which of course will create clicks if latency isn't low.
As an explanation for why the changes didn't affect sound quality: if the system is running with ample timing margins so as to avoid any clicks then the background processes will have little affect on sound quality, e.g. essentially none when they aren't running. And if they do run every few seconds they will degrade sound quality ephemerally, making it very hard to hear the effects. (For example, most of the subtle sonic affects work by affecting sound staging, and isolated events are perceived by the mind as acoustic events appearing in the sound stage, e.g. a snake rattling in the grass.) However, when these processes hit they may cause interrupt scheduling overhead (latency) or steal cycles from the audio application (depending on priority, but even if there is priority for the critical audio inner loop it can hit OS critical sections and cause priority inversions).
If you reduce the buffer size and this improves sound quality (when there aren't any clicks) then it may be essential to kill off all these background processes when listening to music. However, if you don't like being annoyed by clicks, etc., and don't want to do your homework cleaning up the system then you will have to run with larger buffers. (And at least on Windows 7, windows intensive programs that scroll large windows (such as Firefox) can still cause clicks with the largest available buffers when I am playing DSD).
So I suggest minimizing buffer sizes as much as possible and comparing the sound quality vs. larger buffer sizes. IMO the only software changes more likely to affect sound quality than buffer sizes are things directly on the audio path, which means the player, the interface from the player to the driver, any OS audio stack interposed between the player and the driver, and the driver that sends the signal to the DAC. At least, this has been my experience.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Tony,
Please read the linked page for an idea of what I have been talking about.
It does make a significant difference.
Yes, the player software settings pane has several check boxes for which services I want disabled when the player is launched, then renabled when the player is closed.
I can also view the impact of those services with various UNIX utilities or the Mac's Activity Monitor, which is similar to the Windows Task Manager.
The resource impact of those services are tiny relative to the capabilities of the music server, and any audible differences are also minute.
I've gone beyond the check boxes in the player software and have manually killed some processes and compared. Again, the sonic differences are small and not necessarily 'better' or 'worse'. I find much bigger sonic differences with components downstream from the computer itself.
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: