Seti-client (CLI) interferes with my soundcard driver in Win2k.

[Elledan]

I've got an SoundBlaster 128 PCI soundcard and am running the v3.03 CLI seti-client.

Now, audio, whether it's from an mp3 file, a video or something else, is being disrupted whenever the Seti-client is running.

Weird thing, however, is that sometimes no such interference is noticeable, then suddenly it kicks in.

Any ideas?

 

[J3anyus]

Make sure you've got your SETI client set to low priority. That's my only guess of what's going on...

 

[Elledan]

It's set to low priority

BTW, I don't notice anything happening to the sound whenever I'm doing some video-editing or 3d-rendering or something similar. It's just with the Seti-client

 

[Maggotry]

My wife's PC does the same thing in The Sims. She'll be playing along just fine, then all of a sudden, the audio gets very choppy. Restarting the game will cure it for a while, then it starts doing it again. It only does it when SETI is running. She has a SBLive. Win 98SE.

Edit: The Sims is the only game that does it. UT and other stuff is fine, just The Sims.

 

[Poof]

That's an old known problem with SB cards & SETI. Not much you can do about it until Creative decides to fix the drivers.

 

[Robor]

What CPU/Chipset and OS are you running? I'm curious as I now have problems with the video intro on some games under XP and 2000 on my AMD dualie but not in Win9x or Win2K on my Intel. The games play fine, it's just choppy sound on some intros. Also, playing video on the desktop is fine - just happens in some game intros. Weird. I haven't tried shutting down the client and launching.

 

[Elledan]

<< What CPU/Chipset and OS are you running? I'm curious as I now have problems with the video intro on some games under XP and 2000 on my AMD dualie but not in Win9x or Win2K on my Intel. The games play fine, it's just choppy sound on some intros. Also, playing video on the desktop is fine - just happens in some game intros. Weird. I haven't tried shutting down the client and launching. >>


Well, it's not like the sound gets choppy, or something, but it's just interference, similar to 'snow' with TV.

Duron 600/KT133, Win2k Pro (no SP's installed).

 

[Elledan]

<< That's an old known problem with SB cards & SETI. Not much you can do about it until Creative decides to fix the drivers. >>


I assume it's a Windows-only issue? I've namely not noticed the same thing with any of my other systems, which run Linux. I only use SB 128 soundcards for my systems, because they're among the most compatible ones.

 

[Robor]

Ah, mine's a different sort of sound problem. It's probably unrelated to SETI as I didn't have the problem with SETI running under Win98 or Win2K when this stuff was in my Intel MB/Chipset combo.

You're a Linux user? I think Poof has found another friend!

 

[Elledan]

<< Ah, mine's a different sort of sound problem. It's probably unrelated to SETI as I didn't have the problem with SETI running under Win98 or Win2K when this stuff was in my Intel MB/Chipset combo. >>

I see



<< You're a Linux user? I think Poof has found another friend! >>


*pats Poof on his back*

Hey buddy, hey pal, how yer doin'?

 

[Poof]

LOL Elledan!

Of the 14 machines here in the livingroom, 9 run Linux (several are dual or quad boot with windoze), 1 runs NetBSD, 1 runs MacOS 7.6, and 3 run windoze.(98, 98SE, 2K AS)

And the sound card issue does occassionally manifest itself with UT running on my Red Hat machine (RH 6.2, 2.4.0-test10 kernel). I use SB Live! cards, where the problem is most prevalent.

I haven't tried running the SETI/UT combo on my SuSE 7.2 though... SInce that machine is a T-bird 1.2@1.4 w/DDR, it might be moot.

 

[Elledan]

Ah, well. Thanks anyway, Poof

Hmm.. perhaps I should try harassing Creative support a bit. I could dump Windows completely as well

 

[Poof]

I could dump Windows completely as well



Go right ahead. I won't stop you.

 

[Elledan]

Nay, I'll first concentrate on getting my other main system running and of course, building my Beowulf-cluster

 

[Poof]

Before attempting an actual Beowulf cluster, you might want to consider a Mosix one. A number of people run this (including here at Anandtech) and have been very happy with it - especially its ease of installation and configuration as compared to the Beowulf hassle.

I'll post the link here but not using the link function, since it doesn't seem to work for me...

http://www.mosix.cs.huji.ac.il/

 

[Engineer]

Via Chipset PCI latency patch link (and article)

Apparantly, VIA chipsets have a problem with the PCI bus which has been discovered. An UNOFFICIAL patch has been developed that has not only cured problems with certain PCI cards (In particular, Radeon Cards and PCI hard drive controllers), the speed is increased dramatically on some of these cards. I'm running it and I can't tell any difference in it, but it is still stable...it may help you.

Good Luck.....

 

[Poof]

Not sure what chipset Elledan has but my UT/SETI problems occurred on the good ole standby BX chipset, not VIA (on my P3 600@972). I haven't tried gaming + SETI on my VIA.

But... as an interesting aside to this (since I just thought of it), I have played DVD movies (software decoder) on my T-bird with SETI running in the background, and have not experienced any issues with the sound stuttering (from the SB Live) or with the video playback.

 

[Elledan]

<< Before attempting an actual Beowulf cluster, you might want to consider a Mosix one. A number of people run this (including here at Anandtech) and have been very happy with it - especially its ease of installation and configuration as compared to the Beowulf hassle.

I'll post the link here but not using the link function, since it doesn't seem to work for me...

http://www.mosix.cs.huji.ac.il/ >>

I actually thought that MOSIX was much harder than any other cluster-methods, but it appears to be easier. Ah well...

BTW, you can create a link using the (L=text)url(/L), with the ( )'s being [ ]'s.

 

[Poof]

I have a co-worker who has used Mosix and he basically said that all you need to do is replace the machine's kernel with the Mosix one and tweek a config file on each machine. If you look at say the Beowulf How-to, it looks rather depressing! :Q

With respect to the links, I can put the link there but something in the forum is adding extra junk at the end of it when you click on it (although the junk isn't visible when you edit the link).

I would normally say it was just me but I've seen a few others mention a similar problem recently.

 

[Sukhoi]

That's odd, I have Sound Blaster cards in my P133 and C433 and have never had a problem.

 

[Elledan]

<< That's odd, I have Sound Blaster cards in my P133 and C433 and have never had a problem. >>

Which SB cards? 128, 512, Live! ?

 

[Elledan]

<< I have a co-worker who has used Mosix and he basically said that all you need to do is replace the machine's kernel with the Mosix one and tweek a config file on each machine. If you look at say the Beowulf How-to, it looks rather depressing! :Q >>

You've to replace the Linux kernel with a Mosix compatible kernel?

To be honest, I found the FAQ for Mosix to be rather unclear on many points.



<< With respect to the links, I can put the link there but something in the forum is adding extra junk at the end of it when you click on it (although the junk isn't visible when you edit the link).

I would normally say it was just me but I've seen a few others mention a similar problem recently. >>

I've never heard anyone complain about it before... for now I'll blame it on you

Okay, off to learn some parallel programming.... *sighs*

 

[Elledan]

<< Via Chipset PCI latency patch link (and article)

Apparantly, VIA chipsets have a problem with the PCI bus which has been discovered. An UNOFFICIAL patch has been developed that has not only cured problems with certain PCI cards (In particular, Radeon Cards and PCI hard drive controllers), the speed is increased dramatically on some of these cards. I'm running it and I can't tell any difference in it, but it is still stable...it may help you.

Good Luck..... >>


Even though the page clearly said that this patch is for KT133A and some other chipsets only, I thought it wouldn't hurt to try it with my KT133 chipset.

I was wrong. Once I shut down the system, I get a nice, fresh BSOD: DRIVER_POWER_STATE_FAILURE.

Well, thanks anyway

 

[Poof]

You've to replace the Linux kernel with a Mosix compatible kernel?

Yep. I expect you bring their kernel sources over, compile, and boot. I believe that it will only work with newer libraries (ie., gnulibc 2.x) and at one time, only with Intel-based processors (although it might just work with the T-birds now - the issue was originally with K6x).

To be honest, I found the FAQ for Mosix to be rather unclear on many points.

Supposedly there's a config file that would go on each machine where you let it know about the IPs of the other nodes in the cluster. Then you can tweek the config files to either allow processes to "float" across the cluster or stay with a specific machine. One machine would act as your cluster controller. My buddy found that if he let the SETI process "float", he would sometimes find it migrating to a slower machine temporarily whenever he opened up something like Netscape! :Q LOL!

<< With respect to the links, I can put the link there but something in the forum is adding extra junk at the end of it when you click on it (although the junk isn't visible when you edit the link).

I would normally say it was just me but I've seen a few others mention a similar problem recently. >>

I've never heard anyone complain about it before... for now I'll blame it on you

PHhhhhhhhhtttttttttttttttttttt!!!!!!!!!!!!!!!!!!!!!!

AND... ugh on the VIA driver mistake there.... Doesn't take much for windoze to puke on a driver.

 

[Elledan]

Thanks for your help, Poof



<< AND... ugh on the VIA driver mistake there.... Doesn't take much for windoze to puke on a driver. >>

*nod*

It's one of the reasons why I prefer Linux over Windows. Windows is just too fragile sometimes.. lol