Hi,

I'm about to replace my socekt 478 motherboard with another socket 478 motherboard that hopefully won't have this problem: when I turn up the volume of my amp I can hear, the mouse moving, disk activity, network activity, 2D accelerated movement of desktop GUI elements (using the scrollbar, moving/maximizing/minimizing windows etc).

I've tested disabled USB, the on-board network adapter, replacing my AGP gfx card with a PCI gfx card, I've also tried an old PCI soundcard and disabled the on-board sound... nothing helped, it's the motherboard itself.

I got a MSI 865PE Neo2 LS aka MS-6728.

So I'm looking for your help. In case you have a socket 478, FSB 800MHz, Intel 465/475 chipset motherboard. Can you do the following test for me: make sure there aren't any possible sound sources like web pages with flash, audio players or system sounds active, then turn up the volume of your amplifier or external boxes (I mean really turn them up to the max if can't hear something at lower volumes) and listen. Then tell me what you hear or don't hear and also tell me the manufacturer and model of your motherboard.

I'm looking forward to your replies. It would be awesome if you could help me!

Hi, I hear no noise with Intel® Desktop Board D865GLC onboard sound (it is Analog Devices AD1985) and Panasonic RP-HJE50 headphones. The PSU is Inwin Powerman IW-ISP300A2-0.

Read this. Its very important :

http://www.pcmus.com/power-grounding.htm

I assume you didn't use a headphone amp, because you didn't mention it. When I directly connect my headphones to the line out/headphone out, I can't hear any noise either. EDIT: To be more precise, I assume that it's not that the noise isn't there, I can't verify that because the signal of my headphone out is simply to low: when I connect my amp between the line out and the headphone and set the volume at the amp so that music would be as loud as when the headphones are connected to the PC directly I can't hear the noise either. So directly connecting the headphones to the PC or using an amp makes no difference for me since I can't hear the noise at those low volumes that are possible with directly connected headphones and thus it is not verfiable whether it is the amp or not. Which makes your test invalid and also won't tell me if grounding issues between the PC and the amp are a problem here.

I have to point this out again: the noise is only hearable at VERY high volumes.



Looks interesting, although it's not very technical.

Sooo, I can record silence from line-in (without jack connections), compress with lossless, and you then may analyse the noise level and spectrum or even amplify/normalize to enjoy the buzzing .

That's a good idea. I did the same, but while it was very white noisy, I could still make out the mouse noise (like geiger ticks) in the background. "Normally" I hear the mouse "geiger" ticks loud and "clear" with very little white noise.

Are you sure the problem is in the motherboard? There are so many other factors involved. The PSU, sound card, cabling, ... even the case could be a problem.

Also, I wonder how much below the normal signal level this noise sits.

Sample file (intentionally slightly over 30 seconds to break copyrights on silence records, ho-ho-ho):
D865GLC-Silenzio.flac
Track Gain : +49.66 dB
Track Peak : 0.010742

Edit: Removed the link.

Egor, did you record that with no connections to the input? I'm asking because an open-ended input may pick up a lot more noise than a connected input.

Yes, line-in is not connected - blue jack is free/empty, (and "autosensing" is not used).

I was hearing the spinning CD-ROM drive in my headphones. I disconnected the cable connecting the drive and audio chip and the problem went away. Even on max volume everything was clear. But recently a heavy noise appeared and when i set the windows master volume the sound is stuttering. Guess that's a PSU problem.

(It's an ancient intel D815 board with AD1885 audio chip)

According to "Amir" at the AVS forum - one of the guys working on audio at Microsoft:

Should be a good thing for the average computer user.

Sorry for replying so late...

I've checked your sample Egor and it indeed includes electro-magnetic noises.

I still like to "hear" other people PCs... but I'm afraid most of them are not clean.

Well, I think an external USB/Firewire sound-adapter would be free from this problem, some models (like Live24bit USB) may be even cheaper than the replacement of the mainboard.

If the PC has grounding issues (like many PC's do) then what difference would an USB external unit make ?

I have corrected this 'chatter' problem by removing washers between the M/B and case.

http://www.pcmus.com/power-grounding.htm

I believe the noise is caused by PCI bus. And in my case there is no washers, Inwin produces top quality computer cases and PSUs and Intel always follows their own design recommendations, but still there is noise.
BTW, what else besides the 50/60 Hz hum can grounding problems cause?

I suppose its easy to try a different sound card or even power supply. If that don't work , I would reseat the motherboard again and re-tighten screws on everything. Any cable may also be a culprit and turn into an 'antenna'.

Hm, I'm still waiting for a guy with no background crackle at all to post here. Because I'm curious if it's actually possible to have a PC with no background noises at all. Another reason why I'm asking for people to test this is because I don't have the budget to go out and buy stacks of new motherboards, sound cards, PSUs and PC cases to do my own tests. And so far my survey's not been a success, except for Egor to whom I'm very thankful.



I was about to reassemble my PC today anyway, maybe it will give me some clues. Unfortunately I don't have a spare ATX-PSU with a P4 power connector.

Well you shouldn't expect quality from onboard sound. When thes WLP 3 comes things will get somewhat better but if you are interested in quality issues you need a discrete sound card. Maybe there are exceptions which provide basic quality but in general that is the case.

I have gutted out my PC now. It's running outside the case with no shielding and grounding at all... and the background crackled hasn't changed a bit. Which means it didn't get worse. So I doubt better groundig is likely to help, if eliminating all extra grounding doesn't make the noise worse.

CSMR: Do you have a discrete sound card? And if yes, can you do the test please?

Here's a tip you might want to try if you're using a sound card with an amp, which worked very well with my Audigy 4 + Pioneer SA-5300 integrated amp (low end from the 70s). If you are using a headphone jack to RCA cable, you can isolate the grounds on the RCA end by using tape, or pliers and tape if you have spare cables. Then use a cable of some sort to connect a screw on the computer case to a screw on the amp (or a proper ground terminal if the amp is equipped with one). After doing both of these I have found a huge reduction in noise generated be the amp, and it's only audible if I increase the volume to unreasonable levels.

Edit:
Essentially you have a cheap ground loop isolator, and the addition of the ground cable ensures no stray noise from the power point (I think) get picked up.

Edit 2:
One last thing - note that I have minimal electronics/electrical experience so perhaps someone else would be able to explain this better.

I have no background crackle, hdd / screen chatter etc. I am running a Turtle Beach Santa Cruz PCI on a P3 550mhz. There is an occasional 'pop' but I have reduced that too by using less aggressive hardware accelleration in windows. Those onboard cards worry me a bit as I've heard this chatter in new PC as well - they all had onboard or cheap cards.



It won't make it worse but won't help either. Try putting it back together, maybe replace m/b standoffs in the case too. It could be a bad m/b design and there is not way to test an onboard soundcard in another system. I am sure it can be solved by some process of isolation.

Thanks, I'm definitely going to try that as well. Isolating the grounds was also one of the "workarounds" mentioned in the article shadowking posted, btw.




What I find more interesting is that you have a P3 550MHz system, I suspect the high data transfer rates of newer to be the real source of increased background crackle and noise. As I said before I did a test with a PCI sound card and the difference was minimal: the noise was only a little bit quieter, but it had exactly the same characteristics like when listening to it through the on-board sound chip.

Also Sebastian Mares mentioned in another thread that he had background noise with his on-board sound and also after he bought and used an Hercules Fortissimo 4. Maybe there are very high-end sound cards out there that provide extra shielding... but those are certainly beyong my budget (and even if I had the money I'd rather think twice before paying some hundred Euros extra, because the cost-value ratio is too bad in this case).

Rather than buying a new motherboard I'll simply buy a new soundboard, since initially I wanted to record some old tapes and old LPs and apart from the bus/background noise... my on-board sounds recording capabilities suck anyway. And I've noticed that when recording from line-in with no cable connected even the on-board sound picks up far less background noise than I actually hear "live" from out of the line out jacks.




Yep, I'll try to experiment a little more...

But nevertheless I find it interesting to "pin point" the root of this problem in a general way. So far I suspect this problem to exist in many if not most "modern" PC systems, and less to non-existing in older ones. Based on what people have said here about their systems. Yet I'd like to hear from many more people, of course.

EDIT: sorry for quoting mess...

I have one on my main setup which is unfortunately in another country. I am not keen on doing tests in any case unless it is really important. Some motherboards will be better than others. Time=money which can be money for a soundcard so rather than doing all this debating IMO you may as well get one.

Hi

Just to let you know I was having this same problem with noises in the audio from the HDD, CD drives, mouse etc. Muting the CD audio input or disconnecting the CD audio cable coming from the CD drive to the motherboard/sound card solved the problem.

I guess the cable was acting as an "antenna" and was picking up EMI noise from devices in the system. Only explanation I can come up with anyway. I thought this particular cable was supposed to be shielded to stop this kind of problem but obviously not.

Thanks.

While muting all possible input channels of the soundchip may help making the noise less noticable by lowering its intensity, it doesn't really eliminate it completely at least not on my motherboard. I've tried that already.

Update: disconnecting my low-budget turntable from my amp (Yamaha AX-396) worked wonders. Now there's mainly homogenous "noise" (nothing to worry about with an analog amp and an analog connection from the PC to it, I guess) and only very faint crackling that's related to mouse movements, hdd/network activity and so on... at full volume (amp) you must know! It's a big difference compared to what it was before.

The turntable was connected with a RCA cable and a grounding wire to the amp and an "Euro"-type power plug to the amp's own power outlets. These kind of "Euro" plugs have no grounding only phase and neutral conductors. I don't know maybe this was a tragic combination that resulted in the noise.