Re: Quality of CoreAudio SRC
Re: Quality of CoreAudio SRC
- Subject: Re: Quality of CoreAudio SRC
- From: "Mikael Hakman" <email@hidden>
- Date: Fri, 1 Feb 2008 13:38:02 +0100
- Organization: Datakonsulten AB
With all due respect Jeff:
None of the users I know agree with this. This includes people with personal
and in many cases also professional interests in high quality audio such as
artists, music and studio related people, broadcast folks, audio hard- and
software vendors, and many others. Many of these people work with and
personally enjoy high quality music and other media, which is the reason for
their occupation. I'll guess you could call them studio level audiophiles.
This list includes my customers, suppliers, acquaintances, friends, family,
and me. We do not want to run to the computer every other track to change
sample rate in Preferences and in some cases also in hardware control panel.
We do not even know what sample rate each of our 10 000 tracks is in. So
yes, the hardware setting should be changed by application based on the
actual sample rate of media being reproduced. Then of course there should be
an option to set fix rate and have SRC done. We do not understand why e.g.
iTunes doesn't work this way.
Yes, there are situations when the system has to mix audio from a number of
sources. Then someone has to decide what sample rate to use. This should be
the user. However, there are many situations when only one source is
reproduced, in fact, when the user does not wish to have anything else
reproduced via that particular interface, under any circumstances
whatsoever. When e.g. monitoring a live broadcast, listening to a final
version of some production, or simply listening to e.g. iTunes at 80 - 90 dB
you don't want anything else on that channel. Therefore there should be an
option, which for each interface would allow the user to decide whether to
apply current hardware sample rate or let application to control the
hardware, based on current media sample rate. This cannot be that difficult
to realize in the software. Having an option would satisfy requirements in
both situations and in both user categories.
We want a car with automatic gear box that could be switched to manual when
the need arise, so to speak.
On February 01, 2008 12:01 AM, Jeff Moore wrote:
Basically, what Stephen is saying is something we've been saying since
Day 0 of Core Audio and OS X. The hardware settings belong to the user.
They should not be changed by applications. Rather, applications should
be adapting to the settings of the hardware including responding to
dynamic state changes.
The basic reason why is that the audio system on OS X is designed to be
shared amongst all applications on the system. No application knows
enough to say with any certainty what is going in other applications.
Doing something like changing the sample rate can have serious
consequences to other applications on the system ranging from causing a
glitch in playback/capture or outright failure of some process due to the
interruption or temporary loss of synch. In a badly coded application, it
can cause a crash (see the recent thread about the bug in HALLab for an
example). In fact, I could go on about the bad things that can happen to
other applications when one app thinks it owns the hardware and makes
unwarranted changes to settings.
On Jan 31, 2008, at 2:30 PM, Mikael Hakman wrote:
On January 31, 2008 Stephen Davis wrote:
IMHO, it is bad form to be changing the audio hardware's sample rate
for each track. If you're gunning for the absolute highest quality
output then maybe it's okay but it's pretty obnoxious to the rest of
the system and should be clearly communicated to the user.
Would you or other knowledgeable member on this thread kindly explain
why rate-following would be bad, in absence of other rate- locking
signals such as digital input or word clock etc., please?
Why couldn't e.g. iTunes send the actual sample rate information down
the chain of software layers until it arrives to the hardware (DAC)? The
sample rate would need to be changed only when the rate used by the
actual track is different from previous. I have been discussing this
with few audio interface vendors. While they agree that not doing SRC
would be the best, they blame on application and operating system
vendors for not providing sample rate information to their
drivers/hardware.
For a user it is difficult if not impossible to keep track of what
sample rate a selected media is in, then use a control panel or
Preferences or both to change the rate, and first then to play the
track. Many users want to simply select a track, perhaps in Front Row
using a remote control from the couch, and play it, or even play a
number of randomly selected tracks. Why can't we give even these users
the best audio quality available on the system?
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden