Are downloads supposed to slow down my computer a lot? Whenever I download anything winamp will stutter here and there and the sound sometimes gets staticky. Is this normal?

What type of soundcard do you have? Have you tried updating the sound card / modem / network card / motherboard drivers? What are your system specs? Thanks.

I have an audiophile 24/96 in a AMD XP 1600+ comp w/ 256 mb ram. I'm using the build in network card on my ABIT KX7 mobo connected to the internet via cable modem.
I have all the latest drivers. The stuttering is especially noticeable when I use p2p programs like emule or WinMX. Also, it is more frequent when more bandwidth is being used.

I would recommend trying to install the latest drivers from VIA for your motherboard. You should also go into the device control manager and make sure that your Audiophile isn't sharing an IRQ with your network card. If they are sharing an IRQ, you should try moving the sound card to another PCI slot. If this doesn't work, you should make sure that you are using the latest BIOS for your motherboard. Lastly, you might want to try playing with the PCI Latency setting in your BIOS if your motherboard allows this. Try a setting of 64 first, and if that does not work, try a setting of 0. If none of this works, I'm out or ideas . Good luck!

If none of ViPER1313:s advice help, you could try the "PCI Latency" patch for VIA chipsets from www.georgebreese.com.

I have not tested this myself, but I heard of people that tried it with success.

Are you trying to play local audio files (on your hard drive) or are they coming off the Internet (streaming, etc.)? If they are coming through the Internet, you might notice these kind of problems when you are sharing your Internet bandwidth between the downloads and the audio streaming (i.e. not enough packets are getting through).

You might find some help at Winamp.com's Technical Support Board

I think that is for the KT133A only.....his particular motherboard uses the KT333

Absolutely wrong!!!!!
I had a VIA KT400 board that was so badly hit by "VIA PCI Latency bug" That all sound stuttered and 20% off all files transfered between PCI devices where massacred (wrong CRC)....
I had to send it back... ALL VIA based motherboards can be hit by this to some extent, depending on yr general hardware config... and George Breeses patch only makes it a little better ... like blowing on a wound -- it doesnt heal nothing -- only getting rid of VIA based motherboard is healing....

Saying this.. my personal experience with Winamp is that it will stutter if CPU load is 100% on yr machine for 30 secs or more --- foobar2000 doesn't stutter if CPU load is 100% for 5 mins or more....

Perhaps I have the right combination of hardware or whatever else, but three out my past five mainboards have used Asus VIA chipsets (A7V/A7V133/A7V333 with last two still in active use) and have never given me any kind of troubles like that at all.

I had a Pentium III 500 MHz machine with VIA chipsets that where stable as nothing else since I had an ATARI machine in early 80's... Most VIA based motherboards are probably safe... but my last 2 attempts have been very very bad, personally I'm off VIA for life... but making a totally unreliable statistic from experiences off myself, family & friends; 80% of VIA based motherboards are probably OK....

Only - if you do get the bug, don't try to patch it - throw the board away - I'm talking from experience here....

I know that some older VIA chipsets had problems (Apollo Pro+ for my Celeron 400 comes to mind......so does the KT133A.) Even so, most new VIA chipsets are great. I have a KT266A right now and couldn’t be happier - works great with my SB-Live Value, which is a very picky card when it comes to PCI Latency issues and sound stuttering.

Out of all motherboards I have ever tried and owned, I have had the most problems using an ECS K7S5A w/ the SIS 735 chipset (it was the motherboards fault....not the chipset.) VIA’s older latency problems shouldn't even be an issue here - that’s why I said to mess with the PCI latency as a last resort. By setting PCI Latency to 0 in the BIOS (being an Abit board, I would be surprised if the BIOS didn’t let you tweak this setting), you are in effect applying George's Latency Patch.

Personally, I have had blue screen startup errors and other problems with this patch, so personally I can't recommend it. If you want to try it, that’s great, but I would definitely try to modify the setting in the BIOS first.

Well. maybe you are right... Personally I've had 2 different (different vendors) boards based on the VIA KT400 chipset now - and they've 'caused me nothing but tears - pure fodder for the trashcan.....

Try an kt133a board from HP. I can't adjust anything in the bios. I may look into some of the stuff on the georgebreese site. I'm due for a format anyway.

Anyways; My personal grief aside and back to parke02's problems, there's really no reason to worry him to much about "VIA PCI Latency Bug" when, I guess, that's not what he's suffering from.... P2P progs running on Windows will periodically cause some heavy CPU load for some minutes at a time (anything from 1-30 mins, don't know why...), Winamp output starts to stutter when machine is under heavy CPU load over some time: At least all versions up 'till 2. 89 (?) did, that's about when I switched to foobar.... foobar don't have this problem, so cure: switch to foobar...

The thing is, i experience the stutters when i d/l using anything. Im currently recieving a DCC send through MIRC and maxing my bandwidth at ~200kb. Im playing winamp and CPU usage fluctuates at ~10-20% in task manager. Every now and then, the cpu usage will shoot up to ~80-90% and my comp will sort of freeze for a split second. I don't necessarily think it is winamp that is stuttering, by my comp in general b/c i still get those jumps in CPU usage regardless.

Ok, i just looked through my device manager. I was mistaken when I said i was using an onboard ethernet adaptor. Under network adaptors I have a "Winbond W89C940-Based Ethernet Adaptor (Generic)" and its on IRQ 10. My soundcard "M-Audio Delta Audiophile" is also on IRQ 10. Both devices say "no conflicts" in the resource tab.
I guess I should try and change the IRQ anyways though. How exactly do i do this? Just move my soundcard or ethernet card over one pci slot?

I have had a very similar problem... The sound would stutter whenever I downloaded something. I finally managed to narrow it down to problems with one of my USB devices, my scanner. When I disconnected it the problem went away completely. Later I discovered that I can have it connected if I connect it to an external USB hub.

Hope this helps!

Legba, the only USB device I have connected is a midi keyboard. Most of the time the power is switched off. I'll try removing the USB plug completely and see if it helps.

Ensure your HDs are set to UDMA mode in Device Manager > IDE ATA/ATAPI Controllers > Primary and Secondary. If you're in PIO mode this slows down HD an awful lot, and also makes more CPU cycles get munched when moving data between the HD and whatever.

Yupp Parke, I think it's best if you move the cards.
Try putting them a bit away from each other. Since Network card generate quite a bit of static, it could make you soundcard hiss.
Try putting the Soundcard in the second lowest PCI-socket, and the NIC a high as you can, without obstructing your graphics card.

This should also fix the IRQ problem.
If they still have the same IRQ, try moving the Soundcard to the very bottom.

I had an old BioStar brand VIA686 (?) Pentium3 board that had a similar problem.

Any significant disk I/O would cause my external serial modem to pause briefly, winamp to pause, etc. So could 'large' amounts of cpu usage, although that wasn't as common. (Meaning I could play a 320k mp3 sometimes and that'd consume enough to cause my modem I/O to pause. Actually, that was the biggest annoyance! Because I did/do a lot of music listening while online.)

The disk I/O had to be more than just a little, but it didn't have to be a lot. A lot of disk I/O would, of course, bring the modem connection & music to a hault until that got done.

The system used to be pretty bare, and the only thing installed was a cheapie pci based nVidia video card. No other cards, no usb devices regularly pluged in. USB, audio, etc. were provided onboard.

I never could figure out whether it was Win98se, the video card, the drive, or the mobo.

I did a lot of playing in the bios settings but never managed to adjust anything that helped.

I checked everything I knew to in Win9se. DMA, drivers, etc. etc.

I didn't have another video card that would have worked in that pc, so I couldn't test the possibility that was the culprit. Nor did I have a spare hard drive to test to see if that might have been it.

I finally concluded that it was probably just the via chipset or mobo itself.

For a major chipset vendor, it seems like a *lot* of people have trouble with Via! Via does have a reputation for the first one or two versions of a chipset to be buggy. I'd guess they don't do enough internal testing and let their customers be their beta testers.

I don't know about other chipsets, but if Via has that many problems, then I can imagine what the rest must be like.

I'm not a fan of the Pentium-4 architecture, but I do have to say that you rarely hear people having these kinds of problems with an Intel built mobo.

I don't know of any AMD based chipset that is as reliable.

nForce? nForce2? nForce3? SIS 735 (ones not made by ECS....)? AMD's old 760 boards were workhorses back in the day as well. Secondly, I will personally vouch for the VIA KT266a as a great chipset. In my opinion, VIA's problems have to do with their older Intel chipsets and the KT133a (both of which VIA still seem to be living down today).

The VIA KT333 was never a widely adopted solution, and was really nothing more than a KT266a w/ support for 333mhz memory IIRC. I never heard much flak about this board except it didn't accept new CPUs based on the 333mhz FSB.

I have built 3 systems using nForce2 boards and have had no problems with any of them. Don't make wide generalizations about how AMD boards are not as "reliable" as Intel boards, because most of them are every bit as "reliable". The only motherboard I have ever had die was an Intel based board. The whole P4 vs AMD thing has everything to do with bias and almost nothing to do with facts.

Viper1313...

I readily admit I haven't been keeping track of all the chipsets and mobos.



I've heard a few reviewers also talk negatively about them. Bugs in the chipsets and some boards seem a bit flaky as well. Later revisions have probably fixed that, but it doesn't help anybody who has one. Still, reviewers talk better about it than thei do for Via.

As for Via in general... you'll hear more complaints about them than any other major chipset vendor. I've had problems. A guy I know in Canada had a Duron and the second IDE's dma was unreliable. And other's who've had via also usually comment negatively about stability.

Even most reviews I've read about new via chipsets will say things like "great chipset, wait for the next revision because Via never gets it right the first time or two" and so on. Even with the chipsets of the past few years. Both for Intel and AMD.

They eventually get it right, but early buyers have to suffer through a lot of buggy chipsets.

I'm not a fan of Intel. I despise the P4 architecture. (Sure, it runs fast, but the architecture is a kludge.) But even most reviewers who test mobo's will say the Intel built ones are more reliable than anybody else's. Bland, but reliable. And reviewers who do additional hardware will often say they prefer to do it on Intel mobo's because if they have a problem, they know it's not with the mobo causing problems.

It's not an Intel vs. AMD. It's just that nobody seems to build mobo's for AMD as reliably and constantly as Intel does for theirs. Sure, there have been a few for AMD processors, but not as consitantly as Intel has done.

Yes, Intel has had a few problems. But not as many as any other manufacturer, regardless of the processor type.

I have next to nothing to add to the resolution of the original poster's problem. That will likely be resolved with what has already been discussed. With that possibly winding down, I have this to add to the discussion of chipsets:

nForce IDE drivers have issues. nVidia has circled the wagons on this, and I am awaiting a newer release.

VIA and SiS are low-cost chipset designers, usually selling to low-cost integrators. That low cost is usually maintained by using low-quality (low-cost) parts. While some mainboard builders use better components, like as not the overall build quality of a board based on a low-cost chip is itself low. You really cannot expect too much from something this complex costing so little; corners had to be cut.

I would prefer to see more chipsets coming from AMD themselves. Who should know better how to wire in an Athlon or Opteron than the designer of the CPU itself? Their reliance on outside developers has been the only real concern of mine, save for the exposed cores (bad, bad design). The processors represent a better value, yet the requisite mainboards leave me unimpressed at best.