[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Linrad] Windows PCI bug.
- Subject: [Linrad] Windows PCI bug.
- From: Leif Asbrink <sm5bsz.com; leif@xxxxxxxxxxxxxxxx>
- Date: Sun, 30 Aug 2009 03:16:20 +0200
Lately I have spent quite some time in making a stable multiboot
with all the Windows versions(not easy) and several different
Linux distributions (easy)
Now I can back-up and restore all the operating systems with
cp /dev/hdb /dev/sda which will copy the entire hard disk
with all the partitions and all the boot information:-)
This means I can install and test something and then step back
everything to the previous state without all the problems
of incomplete uninstall programs.
One of the things I had planned to investigate was the drive routines
for a couple of soundcards. Problems have been reported with Vista as
well as with 64 bit systems.
What I have found was most un-expected. There is a bug in all versions
of Windows. Something goes wrong with the PCI bus if one tries to
run a soundcard at 96 kHz. It is not only a problem for Linrad.
The same error is present in Winrad both with WMME drivers and with ASIO
drivers. At 48 kHz everything is OK though.
I have tested two different computers with different chipsets.
Both have PCI Express as well as PCI. One is 2 x Xeon Quad, the
other a Centrino Duo. I do not know whether the problem is
because of the presence of PCI express on the motherboards or
whether the problem is due to multicore CPUs.
The bug causes incorrect data but I have also seen loss of data
manifested as a slightly too low reported value for the sampling speed.
Make sure that your system is ok by testing with an audio generator
if you use a soundcard for SDR on a modern computer under Microsoft
Windows. (This problem does not exist under Linux.)
If there is someone on this list who has a PCI card that can sample
at 96 kHz and a motherboard with two or more CPU cores but without any
PCI Express slots - or if someone has a motherboard with both PCI and
PCI Express, but with a single core CPU I would be interested to know
if you see the Windows bug or not.
The problem is not present on Pentium 4 and older where there is no
PCI Express slot.
Leif / SM5BSZ
You received this message because you are subscribed to the Google Groups "Linrad" group.
To post to this group, send email to linrad@xxxxxxxxxxxxxxxx
To unsubscribe from this group, send email to linrad+unsubscribe@xxxxxxxxxxxxxxxx
For more options, visit this group at http://groups.google.com/group/linrad?hl=en