[aag]

I have developed an incredbly annoying problem.


I have a PIII 733, GeForce3. 256MB RAM, Win2K (all SR´s) and I recently acquired the M-Audio DioDelta 24/96 (all digital inputs and outputs). I still have a SbLive inside, which I would like to keep since it provides analog in/outputs.


But now I have inacceptable, unnerving audio hiccups when watching WinDVD2000 3.0. Every 30 seconds or so, sound breaks for some 50-100 milliseconds. This completely spoils the DVD experience. I have a cable modem and access internet from the same PC, but the situation does not improve if I disconnect the cable modem.


I would be extremely grateful for any hint of what might be wrong. Thanks in advance!

 

[jkaiser]

What do your IRQ's look like? I am assuming that the DVD and Harddisk have DMA turned on as it was working before the new audio card.

 

[BHayes]

aag,


I had similar problems as well with my Win2k,M-audio,windvd30 combo. I'm still not exactly sure what fixed it for me, I'm thinking it was a combination of things. I'll give you the list of stuff I tried.


For one thing one of my HD's was starting to fail and it had my TEMP directory on it. So I changed my TEMP/TMP environment variables to a different disk and eventually got a new HD.


I also run the UDAgent( www.ud.com - Cancer research project), and I found that shutting down the agent(not just having it sleeping) seems to help.


Also uninstalled Instant CD/DVD 6. This app installs a service that seems to conflict with WinDVD. I was getting DVD video corruption with this running as well.


Shutting down Norton-Antivirus and not loading the NAV services.


Just as a note, when WinDVD was having problems for me, PowerDVD V3.0 worked fine(cept that I couldn't get DD/DTS output through the m-audio with PowerDVD).


Hope this helps some.

Brian

 

[Aaron Oz]

I have the exact same issue with the DIO 24/96 in ATI DVD 4.1, but it operates fine in WinDVD (all versions). I haven't tried PDVD. Since the ATI beta driver with the overlay adjustments, WinDVD looks at good as the ATI player, so I simply bagged ATI all-together.


Try WinDVD and see if the SPDIF is okay. It probably will be.


Oz

 

[aag]

Maybe my PC is doing things in the background which I am not aware of, and which mess up the timings... Are you aware some utility that logs and reports background activities of the PC over time? simply bringing up the task manager clearly does not cut it.

Also, what is the potential problem with interrupts? I have Win2K and all cards are PCI: I thought that interrupts are a god-dictated given....

 

[aag]

dear jkaiser et al

thank you very much for your support. now I have seen that both sound cards which I have, the sblive and the M-audio 24/96, share IRQ 9. Could this be the problem?If so, how can I change the interrupt line? the option "manual settings" is grayed out. Remember, I use W2K and both cards are PCI. Also, how can I check whether DMA is enabled in the DVD and disk drive. The W2K device manager does not show this, apparently. I hope these questions are not too naiive - it was already once explained to me, on this forum, that I am stupid because I could not get SP/DIF sound out of W2K, and hopefully this time somebody will compassionately bear with me.....

 

[kazushi]

Hi aag,


Sharing IRQ9 may cause the problem, but I'm not sure it will be solved your problem that changing IRQ. Let me explain more details.


The main reason of this problem is somewhere in OS and driver. How they treat interrupt and feed sound stream through SPDIF. If OS become unstable, the device handler for Dio24/96 cannot feed sound stream on time. Then, sound from your receiver is broken. For example, Duell said he has the same problem with ATI player. I guess the reason is that ATI player is more heavier, so it requires critical timing.


BTW, let me describe some possible solutions. First, if you are using Win2k in ACPI mode, there is no way to use other IRQs. All devices share this. However, your system has different INT configuration at BIOS level. So, I suggest you to check this first.


Normally, mother board has 4 interrupt lines and devices use them. If your MB has more than 4 PCI slots, some PCI slots share an interrupt line. You can check this configuration from BIOS. Before your OS starts up, BIOS displays the configuration of PCI slots. You can push pause key to pause boot-up sequence to see the BIOS level configuration. In addition, ESC key is for release the pause. If your cards share the same IRQ at this level, you can change PCI slots for your cards to solve conflicts at BIOS level.


If this doesn't help you, you can install Win2k in not ACPI mode. In this mode, Win2k handle IRQs as what Win98 did. You can push F5 key when Win2k installer ask you to push F6 for additional devices. Some said you can change your machine from ACPI to regular from control panel, but it might cause your system crash.


-- Kazushi

 

[mrdot]

I have the EXACT same problem...Win2k with all the updates, M-Audio and SBLive, WinDVD3.0- I have tried everything to try to fix this problem, to no avail. I run dual boot because winme works fine for dd/dts output...any help would be greatly appreciated


Todd

 

[aag]

maybe there´s a way to increase the priority of the foreground task to such a ridiculous extent that whatever is happening in the background will crawl and, hence, not interfere with the sound... anybody knows of a registry tweak that will effect this? even better would be a tool that logs background activities,and might help identify processes responsible for this mess.... any idea on this???

 

[aag]

well I have nailed down the problem to the M-Audio card. Deactivating it with the Win2K control panel, and running the SPDIF out of the Soundblaster card, takes completely care of the problem. Deactivating the Soundblaster, conversely, does not cancel the stutter from the M-audio, indicating that the problem is intrinsic to the M-audio and not a conflict. Do you believe that I should complain and return the M-audio?

 

[Dean Roddey]

Its obviously not intrinsic to the M-Audio per se, since most people who have them don't have a problem. The problem is that the modern PC is an completely ununderstood collection of compromises flying in tight formation, and any particular combination of compromises might not work correctly. I have a very standard configuration, and I have never gotten rid of my video micro-stuttering, but other people don't seem to have any problem, with the same or even less standard setups. Its all just a crap shoot, and is probably not indicative of any real design flaw in the M-Audio. My 24/96 works wonderfully for instance.

 

[aag]

sure, dean, I understand that, and I agree. However, I am still wondering whether something might be wrong with my particular, individual M_audio card.

best

aag

 

[Dean Roddey]

I doubt it. As I said, its a crap shoot. Some people have problems, others don't. Some people have problems and then they go away for no apparent reason, using the same equipment. I would imagine its just some subtle issue with your particular combination of cards, MB, order you installed them, IRQs they are on, drivers you are using, order you installed them, etc... This is the joy of HTPCs at this point I guess, you always have something to keep you busy.


The only way you could prove your theory, or at least get closer to it, would be to waste your machine, pull out all the cards but the M-Audio and set up nothing but that and see what results you get. If it works, you start putting other stuff back in, but you might get right back to the same place again. I've done it many times (about 50 by now) to try to get rid of my micro-stuttering video, and haven't managed to yet, with many variations.


But I don't think its my Geforce card, because others have the same problems as me, and have had it go away with some change in configuration. And I have reduced it considerable via just (mostly randomly) changing the configuration.


I think that until the computer OEMs understand that we are using their products for this kind of work, and put their much greater resources into figuring out how to make it work consistently, we are going to be in this kind of situation. I assume it would require some serious bus spelunking to figure it all out. Then they would have to update PC specs to require that they meet whatever improvements are required to keep HTPC usage from being a quantum mechanical operation.


Then again, I could be completely wrong...