12 Feb
2004
12 Feb
'04
6:27 p.m.
Greetings:
I've narrowed down some parts of my current problems with sound. The
mouse is definitely a noise-maker: I move, I get a not-nice ripping
noise that tracks along with the mouse movement. I had hoped that the
problem would be the nVidia binary video driver, but the problem occurs
with the kernel sources driver as well. I've appended some background
info in case anyone sees an obvious flaw in the system.
A couple of things: I have no USB devices on this machine, so how can
I get rid of the usb-uhci stuff hanging off the same interrupt as the
SBLive ? Also, would it do any good to switch slots ?
I've tried a number of different changes to XF86Config, got no joy
from any. :( I've been working through the various suggestions from the
Audio Quality HOWTO, nothing has worked yet. I'm open to any other
non-$$$ suggested solutions...
Best regards,
dp
800 MHz Duron CPU
512 MB RAM
SBLive Value
kernel 2.4.24-ll (Planet CCRMA)
ALSA 1.0.1
[dlphilp@localhost dlphilp]$ lspci
00:00.0 Host bridge: VIA Technologies, Inc. VT8363/8365 [KT133/KM133]
(rev 02)
00:01.0 PCI bridge: VIA Technologies, Inc. VT8363/8365 [KT133/KM133 AGP]
00:07.0 ISA bridge: VIA Technologies, Inc. VT82C686 [Apollo Super South]
(rev 22)
00:07.1 IDE interface: VIA Technologies, Inc. VT82C586/B/686A/B PIPC Bus
Master IDE (rev 10)
00:07.2 USB Controller: VIA Technologies, Inc. USB (rev 10)
00:07.3 USB Controller: VIA Technologies, Inc. USB (rev 10)
00:07.4 Bridge: VIA Technologies, Inc. VT82C686 [Apollo Super ACPI] (rev 30)
00:10.0 Ethernet controller: Lite-On Communications Inc LNE100TX
[Linksys EtherFast 10/100] (rev 25)
00:12.0 Multimedia audio controller: Ensoniq ES1370 [AudioPCI] (rev 01)
00:14.0 Multimedia audio controller: Creative Labs SB Live! EMU10k1 (rev 08)
00:14.1 Input device controller: Creative Labs SB Live! MIDI/Game Port
(rev 08)
01:00.0 VGA compatible controller: nVidia Corporation NV11 [GeForce2
MX/MX 400] (rev b2)
[dlphilp@localhost dlphilp]$ cat /proc/interrupts
CPU0
0: 11070087 XT-PIC timer
1: 16145 XT-PIC keyboard
2: 0 XT-PIC cascade
3: 175283 XT-PIC usb-uhci, usb-uhci, EMU10K1
8: 1 XT-PIC rtc
9: 0 XT-PIC Ensoniq AudioPCI
10: 976593 XT-PIC nvidia
11: 447118 XT-PIC eth0
12: 575359 XT-PIC PS/2 Mouse
14: 27715 XT-PIC ide0
15: 1676 XT-PIC ide1
NMI: 0
LOC: 11070994
ERR: 45523
MIS: 0
12 Feb
12 Feb
6:59 p.m.
On Thu, 12 Feb 2004 17:47:37 -0500
Dave Phillips <dlphilp(a)bright.net> wrote:
Greetings:
I've narrowed down some parts of my current problems with sound. The
mouse is definitely a noise-maker: I move, I get a not-nice ripping
noise that tracks along with the mouse movement. I had hoped that the
problem would be the nVidia binary video driver, but the problem
occurs with the kernel sources driver as well. I've appended some
background info in case anyone sees an obvious flaw in the system.
A couple of things: I have no USB devices on this machine, so how
can I get rid of the usb-uhci stuff hanging off the same interrupt
as the SBLive ? Also, would it do any good to switch slots ?
Maybe your BIOS allows you to disable the USB controller ?
I've tried a number of different changes to
XF86Config, got no joy
from any. :( I've been working through the various suggestions from
the Audio Quality HOWTO, nothing has worked yet. I'm open to any other
non-$$$ suggested solutions...
Since your are using a kt133 based motherboard, may I suggest a BIOS
upgrade ? I had noise in my captured sounds for months until I dared
to flash my kt133 motherboard using a boot floppy image from the FreeDOS
project (be sure to chose "clean boot" or something like that).
Your problem obviously isn't the same as mine but this might help
though.
Best regards,
dp
[...]
--
David
7:16 p.m.
Hallo,
Dave Phillips hat gesagt: // Dave Phillips wrote:
A couple of things: I have no USB devices on this
machine, so how can
I get rid of the usb-uhci stuff hanging off the same interrupt as the
SBLive ?
You could either remove the usb-uhci.o module file, or even usbcore,
or do something like "alias usb-uhci off" to switch it off.
Also, would it do any good to switch slots ?
May well be...
ciao
--
Frank Barknecht _ ______footils.org__
10:23 p.m.
On Thursday 12 February 2004 10:47 pm, Dave Phillips wrote:
The
mouse is definitely a noise-maker: I move, I get a not-nice ripping
noise that tracks along with the mouse movement.
I boot (using lilo & Debian) with the
append="no hlt"
option. Have you tried this?
It helped me with graphics-related noise.
I don't know whether this works on a RH based system.
tim hall
13 Feb
13 Feb
8:49 a.m.
hello, i second the no-hlt (i think it needs the dash) kernel option.
i have also have a kt133 motherboard at home, with an AMD tbird. i had
the same noise problems when moving the mouse. adding the no-hlt option
made it dissapear for me (YMMV)
so i'd say reboot and find out ;-)
good luck, ALEX
--- tim hall <tech(a)glastonburymusic.org.uk> wrote: > On Thursday 12
February 2004 10:47 pm, Dave Phillips wrote:
___________________________________________________________
BT Yahoo! Broadband - Free modem offer, sign up online today and save £80
http://btyahoo.yahoo.co.uk
The
mouse is definitely a noise-maker: I move, I get a not-nice ripping
noise that tracks along with the mouse movement.
I boot (using lilo & Debian) with the
append="no hlt"
option. Have you tried this?
It helped me with graphics-related noise.
I don't know whether this works on a RH based system.
tim hall
8:52 a.m.
On Feb 12, 2004, at 6:47 PM, Dave Phillips wrote:
Greetings:
I've narrowed down some parts of my current problems with sound. The
mouse is definitely a noise-maker: I move, I get a not-nice ripping
noise that tracks along with the mouse movement. I had hoped that the
problem would be the nVidia binary video driver, but the problem
occurs with the kernel sources driver as well. I've appended some
background info in case anyone sees an obvious flaw in the system.
I had this exact problem on an older system, a an AMD K6-2 with a via
chipset. (also NVidia card, btw) The only way I could solve it was to
add a sound card and stop using the onboard sound.
9:41 a.m.
Greetings:
Thank you to all who replied with suggestions for resolving my sound
problem. After working all last night on it I've come to some new
possible conclusions.
First, sound is fine outside of X. I tested realtime output from three
long Csound compositions (Face On Mars, Incantations, and Trapped In
Convert), they all played fine, even when I moved the console mouse.
Second, sound is fine in X with *some* applications. XMMS plays things
perfectly, no glitches while I move the mouse or switch workspaces. The
major offender has been Csound when I run an app (ImproSculpt) utilizing
the FLTK opcodes. The audio distortion had become intolerable, making
the app nearly useless. Btw, the distortion wasn't just from the mouse,
it was bad even without mouse movement, full of pops and crackles. The
mouse really zippered when I moved it across the display, and moving
windows chopped things up more.
Following advice found in an article by Daniel Robbins
(http://www-106.ibm.com/developerworks/library/l-hw2.html) I adjusted
the PCI latency timer (with setpci) for my sound card. At last I
eliminated *most* of the trouble with ImproSculpt. The mouse still zips
but the overall distortion is gone. I thought that perhaps FLTK had been
compiled with GL dependencies, but 'ldd lbfltk.so' showed none.
I do have the no-hlt option appended to lilo.conf, but I think that
the nVidia driver's DRI overrides the option. Bummer. So I'm still
trying to eliminate the mouse zip in ImproSculpt, but overall sound is
much improved.
Btw, here's what I have in lilo.conf, I presume I have the syntax
correct :
append="hdc=ide-scsi no-hlt root=LABEL=/"
Lilo doesn't complain about it, so I figure it's right, but maybe not... ?
Back to the hack...
Best,
dp
11:53 a.m.
On Fri, 2004-02-13 at 06:01, Dave Phillips wrote:
Greetings:
and to you too...
Following advice found in an article by Daniel
Robbins
(http://www-106.ibm.com/developerworks/library/l-hw2.html) I adjusted
the PCI latency timer (with setpci) for my sound card. At last I
eliminated *most* of the trouble with ImproSculpt. The mouse still zips
but the overall distortion is gone. I thought that perhaps FLTK had been
compiled with GL dependencies, but 'ldd lbfltk.so' showed none.
Since you're getting into this, have you checked the MIN_GNT and MAX_LAT
values in the PCI config space of your sound card? Setting the PCI
Latency timer to a higher value only keeps your card on the bus longer
when it finally gets the bus. It doesn't do anything to help it gain
access to the bus sooner. It's probably that some *other* PCI device is
holding the bus and starving you out. Since it appears to be related to
your mouse and hard drives, it would seem to be your South Bridge since
that's what likely controls both of them.
Anyway you can look at a few options:
1) Get a good IRQ. I have an older Dell P3 500 that has this same 'mouse
causes noise on AP2496 under Windows ME' sort of problem when the sound
card is on IRQ 3,4,5,6 or 7. (Assuming traditional IRQs and not
APIC/IO-APIC IRQs.) By moving the card to IRQ9 the problem goes away. (I
know Clemens said the other day that this doesn't matter, but on my
systems it has and continues to matter. I acknowledge the physics in my
universe are somewhat cartoonish at times...) ;-)
2) If it's not an IRQ issue, then maybe you need to get the *other*
devices *off* the PCI bus faster. To do this you would *lower* their
latency timer values. However, you should really look at the MIN_GNT and
MAX_LAT values in all of their PCI config spaces. Lower their latency
timers (or try it) but don't lower them further than their MAX_LAT
values. (IMPORTANT FYI - MIN_GNT and MAX_LAT are specified in terms of
1/4 microsecond units, not clock cycles while Latency Timer is specified
in terms of PCI clocks. You need to do the conversions! Ask questions if
you need to.)
IMPORTANT FYI #2: Messing with any of these values on your disk drives
may lead to corruption.
Just some idea.
Good luck,
Mark
9:19 a.m.
hello again,
--- Dave Phillips <dlphilp(a)bright.net> wrote: > Greetings:
I've narrowed down some parts of my current problems with sound.
The
mouse is definitely a noise-maker: I move, I get a not-nice ripping
noise that tracks along with the mouse movement. I had hoped that the
problem would be the nVidia binary video driver, but the problem
occurs with the kernel sources driver as well. I've appended some
background info in case anyone sees an obvious flaw in the system.
A couple of things: I have no USB devices on this machine, so how
can I get rid of the usb-uhci stuff hanging off the same interrupt
as the SBLive ? Also, would it do any good to switch slots ?
i already replied with the no-hlt suggestion, but still want to add...
my system is actually very similar to yours in that i also have a kt133
chipseet, Nvidia videocard, SBLive card... i even run the same CCRMA
kernel ;-)
about irqs : irq 3 is *very* low priority so definately worth changing.
i would try one of following:
1) see if you can change the irq in the bios. my bios allows to change
certain irq's of pci slots and even onboard devices.
2) move the SBLive to another slot.
if moving to another slot does not change the irq order after you boot
linux, you might have to turn acpi off. you can boot with something
like pci=biosirq or pci=noapic or acpi=off. i need that to avoid that
linux reshuffles my 'perfect' bios irq settings. (i am at work now,
this is from memory 8)
good luck, let us know how you get along...
ALEX
I've tried a number of different changes to
XF86Config, got no joy
from any. :( I've been working through the various suggestions from
the
Audio Quality HOWTO, nothing has worked yet. I'm open to any other
non-$$$ suggested solutions...
Best regards,
dp
800 MHz Duron CPU
512 MB RAM
SBLive Value
kernel 2.4.24-ll (Planet CCRMA)
ALSA 1.0.1
[dlphilp@localhost dlphilp]$ lspci
<SNIP>
[dlphilp@localhost dlphilp]$ cat /proc/interrupts
CPU0
0: 11070087 XT-PIC timer
1: 16145 XT-PIC keyboard
2: 0 XT-PIC cascade
3: 175283 XT-PIC usb-uhci, usb-uhci, EMU10K1
8: 1 XT-PIC rtc
9: 0 XT-PIC Ensoniq AudioPCI
10: 976593 XT-PIC nvidia
11: 447118 XT-PIC eth0
12: 575359 XT-PIC PS/2 Mouse
14: 27715 XT-PIC ide0
15: 1676 XT-PIC ide1
NMI: 0
LOC: 11070994
ERR: 45523
MIS: 0
___________________________________________________________
BT Yahoo! Broadband - Free modem offer, sign up online today and save £80
http://btyahoo.yahoo.co.uk
14 Feb
14 Feb
4:20 p.m.
Dave Phillips wrote:
Greetings:
I've narrowed down some parts of my current problems with sound. The
mouse is definitely a noise-maker: I move, I get a not-nice ripping
noise that tracks along with the mouse movement. I had hoped that the
problem would be the nVidia binary video driver, but the problem occurs
with the kernel sources driver as well. I've appended some background
info in case anyone sees an obvious flaw in the system.
A couple of things: I have no USB devices on this machine, so how can I
get rid of the usb-uhci stuff hanging off the same interrupt as the
SBLive ? Also, would it do any good to switch slots ?
as frank said, either turn off usb in your bios, or if that's not
possible, move the card around.
--
The handles of a craftsman's tools bespeak an absolute simplicity,
the plainest forms affording the greatest range of possibilities for
the user's hand.
That which is overdesigned, too highly specific, anticipates
outcome; the anticipation of outcome guarantees, if not failure, the
absence of grace.
- William Gibson, "All Tomorrow's Parties"
Jörn Nettingsmeier
Kurfürstenstr 49, 45138 Essen, Germany
http://spunk.dnsalias.org (my server)
http://www.linuxaudiodev.org (Linux Audio Developers)
7590
days inactive
7592
days old
linux-audio-user@lists.linuxaudio.org
10 comments
9 participants
participants (9)
-
Alex Timmer -
Dave Phillips -
David -
Fernando Pablo Lopez-Lezcano -
Frank Barknecht -
Joern Nettingsmeier -
Joey Reid -
Mark Knecht -
tim hall