2 Aug
2015
2 Aug
'15
10:10 p.m.
On Sunday 02 August 2015 12:05:02 Len Ovens wrote:
On Sun, 2 Aug 2015, Gene Heskett wrote:
I agree Len, and we do not have that at the present.
The reasons s/b pretty obvious, given 15 seconds to think about it. And
not likely to get any better until the gear makers are willing to share
a sample of their product, AND full specs disclosure, to the linux
people who will then write code to make it at least as usefull, if not
moreso, than it is with the winderz drivers they paid good money for
someone to half-arsed write. They would get the driver written for
nothing, and the sale a good driver makes possible.
But they can't begin to see the profit potential of giving away one piece
of it, and writing the interface specs. If its proprietary, get a
non-disclosure, but that will also preclude the driver from ever being
FOSS. Makers need to think about this. There are at times, perfectly
good business reasons for firing the legal councel that objects.
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
I've no clue how universal it could be made
to work, but forget the
chips designation by model/device type unless theres more than one
detected by the system, and come right out and say its the blue jack
that is the mic input, the orange jack is the line output. etc etc.
I've no doubt played mix & match with the color of the jack since I
don't have that memorized.
But you get the point.
I had decided not to say anything more... I have also dealt with too
many people on IRC who don't get it. You can take what Paul deals with
on #ardour and multiply that but every distro that has an IRC channel.
However, I am not sure the colour labels in the GUI would help. The
problem would seem to be identifying internal vs. external cards. A
jack GUI might (with some work) put up a dialog box that says
"external audio card detected, but internal audio is chosen. Is this
what you want?" Then have a checkbox that says "Never show this
again".
The solution many users expect is the Windows/Pulse solution of
showing all audio devices and hiding the fact that sample rate
conversion is happening. The profesional world is not that different,
the number of cards that take a number of AES3 inputs with a resample
stage is not that small (some even take mp3 streams in :P ). Certainly
the whole AoIP world is based on the all i/o are visible and usable
concept. (and in sync even)
How much work should the GUI dev put into making sure the user
understands this not so? Or that this is not a good idea? In fact
maybe the user understands this is not a good idea, but they have a
USB input from their mic/guitar and can only use internal audio output
to hear themselves. (why did they buy such a thing in the fist place?)
There are some very good sample rate converters out there that work
with jack (well one anyway). Maybe it is time to automate connecting
extra i/o through s/r blocks. Certainly the idea of being part of an
AoIP network and having _all_ i/o connected to jack all the time does
not make sense. Some kind of GUI that allows dis/connecting a remote
i/o only when needed just makes sense. Most of us here, would want
know and choose which AI is synced and which is s/r. But a number of
others do not care. They just want to use guitar/mic in through a USB
and audio out through internal or even the video conection.
The world of inexpensive PCI audio cards is gone (though I hope to get
years of service from the one I have). The Linux tools we use need to
reflect that. There is a difference between what should be and what
is. Making the best use of "what is" is the way forward in my mind.
--
Len Ovens
www.ovenwerks.net