> > Exactly. Microsoft as a company would surely benefit from
> > fixing this bug. The cost of offering solutions to all PCI card systems
> > that can not be moved to a modern computer (under Windows) should be
> > a concern to technically oriented (and responsible) people within
> > Microsoft.
> Hmmm, not exactly.... as far as I understand, the problem happens on
> modern motherboards
> (those with the PCI Express architecture), but that still have some old
> PCI sockets. If all the
> sockets were PCI Express, the problem would not be there. Please correct
> me if I am wrong.
This is correct.
> If this is true, the number of such motherboards is a vanishing
> quantity, tending rapidly to zero...
Are you sure?
I thought there are many industrial/commercial systems that
have a PC as a part of the system and that such systems would
use specialized PCI cards of various kinds that would not
be directly compatible with something that would fit into
a PCI Express slot. (Or USB)
> So it understandable that from a purely financial point of view (costs
> versus benefits) Microsoft
> is not much willing to devote resources to fix a problem that will
> eventually disappear by itself
> in a short period of time...
By stirring up some noise on this matter I now have
some better knowledge:-)
Microsoft will not have any interest in what I might try
to tell them. I will have to send a bug report to
M-Audio and Creative Labs about the misbehaviour of
their products. They may have used the same prototype
for their drivers so they made the same error. In any
case they should be able to see the error with their
own recording software and then they should be able
to fix it themselves or if it is not their drivers
they should be able to file a bug report to
Microsoft that the developers would take seriously.
Problem: Is it likely that M-Audio or Creative Labs
will have any interest in fixing problems with old
> You must be a paying member of the developers community of Microsoft to
> have even the slightest
> chance to report a problem.... admittedly, that's a bad thing.
> > There is a fundamental difference between operating systems as I
> > experience them.......
> From this point of view I completely agree with you. Being Linux
> developed by hobbyists and enthusiasts,
> you have here much more chances to be able to talk with somebody who can
> potentially fix a problem.
OK. I think I have a better understanding now;-)
> > In Linrad that is now through Portaudio. Maybe it will be good,
> > (That is what the portaudio site says) maybe not. In any case it
> > will be possible to replace MME by WASAPI if necessary.
> When initially designing Winrad, I considered the use of Portaudio. As a
> matter of fact, if you browse
> its source code you will find a comment that says that a certain error
> was corrected by some Alberto
> di Bene, I2PHD.... but then I decided to write my own audio routines,
> because in my eyes (but of course
> I may be wrong, sometimes it happens...:-) Portaudio tries to be
> everything for everybody, and probably
> three quarters of its code is a bit redundant, being meant for general
> purpose use. But maybe the
> added overhead is minimal, and I erred in being too cautious...
Well, Portaudio is easy to use. With WMME drivers it is very
inefficient compared to my own drive routines. Direct Sound
seems to work fine with Portaudio, but maybe one could do
Direct Sound even better without Portaudio?