Personal computing discussed

Moderator: Captain Ned

 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

motherboard audio degradation

Thu Nov 10, 2016 12:22 am

I've only intermittently been using my USB DAC lately, and since I last used it, I've been thinking my mobo's own audio has been getting muddier and quieter (have to use ~6 dB more to get it to sound right). I just plugged the amp in again, and wow my mobo's audio is falling apart. I knew I'd get some transient response and a touch of FR back by doing that, but this is ridiculous. The mobo's caps still look ship-shape.

Mainly I'm just posting this to mention it can happen, since I haven't heard of this being a serious issue before. It's an ASRock Z97E-ITX/ac, FWIW. My only issue at present is that my volume control gets a bit wonky when using the USB DAC, but I can deal with that.

and my string of freak hardware failures continues.... :evil:
 
HERETIC
Gerbil XP
Posts: 396
Joined: Sun Aug 24, 2014 4:10 am

Re: motherboard audio degradation

Thu Nov 10, 2016 2:28 am

What's your PSU?
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Thu Nov 10, 2016 3:09 am

I assume that's a reference to the string of freak hardware failures. Those have been across several PSUs, all of them top-quality, and none of the failures trace especially well to PSUs (except the one that was a PSU itself - a badly designed APFC controller resulted in the main APFC inductor whining like mad, but AFAICT that failed PSU is still in fine shape as far as output quality). The current one is a Corsair SF600.
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Thu Nov 10, 2016 7:26 am

Well, the silicon shouldn't degrade since it is not being pushed hard. I suppose ESD damage to the onboard audio's built-in DAC (via static discharge to the headphones) is a possibility. Degradation of the DC blocking capacitors on the output is also a possibility, though IMO unlikely unless they were defective to begin with or the mobo is getting really old.

Some other potential culprits... corrosion of the contacts in the audio jack (try plugging/unplugging the cable a bunch of times to scrape the layer of corrosion off), or a software issue (try booting from a live Linux DVD and test the audio with that).
Nostalgia isn't what it used to be.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Thu Nov 10, 2016 4:53 pm

Now that you mention it, some extra resistance on the output could just about explain this; Grados are easily upset by that. It doesn't seem to be at the plug, though.

I doubt it's a software issue for quite a few reasons, but foremost among them is that if this had happened all at once, I would have noticed much sooner. It's got to have been a more gradual deterioration over the past month or so. (Also I'm already running Linux.)

The front panel output is fine, which might narrow it down a bit, but I don't know how that's wired internally (pulseaudio doesn't make any distinction between the two ports).
 
ludi
Darth Gerbil
Posts: 7449
Joined: Fri Jun 21, 2002 10:47 pm
Location: Sunny Colorado front range

Re: motherboard audio degradation

Thu Nov 10, 2016 5:34 pm

If the onboard audio really is objectively degrading at a fast rate, and the easy/obvious culprits such as the drivers and jack can be eliminated (cracked solder joints, perchance?), you most likely have a VRM failure in progress somewhere on the board.
Abacus Model 2.5 | Quad-Row FX with 256 Cherry Red Slider Beads | Applewood Frame | Water Cooling by Brita Filtration
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Thu Nov 10, 2016 5:35 pm

synthtel2 wrote:
The front panel output is fine, which might narrow it down a bit, but I don't know how that's wired internally (pulseaudio doesn't make any distinction between the two ports).

Oh, that's interesting. I would *think* that it is using the same DAC as the front port, but who knows...

Something else to consider (though a long shot since you seem to think the degradation occurred gradually): If this is a Realtek-based implementation, there's probably a software-switchable integrated headphone amp you can enable. Maybe it was on before, and some update has defaulted it to off instead. Google for a tool called "HDA Analyzer"... IIRC it was available from the ALSA web site. It lets you poke at the internal registers in the Realtek codec via a GUI interface. Browse through the output nodes and try enabling the "HP" option for each in turn (this turns on the integrated headphone amp for the corresponding output jack) to see if that makes a difference.

ludi wrote:
If the onboard audio really is objectively degrading at a fast rate, and the easy/obvious culprits such as the drivers and jack can be eliminated (cracked solder joints, perchance?), you most likely have a VRM failure in progress somewhere on the board.

Buildup of corrosion on the jack contacts could do it though.
Nostalgia isn't what it used to be.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Thu Nov 10, 2016 11:41 pm

VRMs don't seem likely, but this is the same board that I had a CPU mysteriously die in, so who knows. I'd think if something like that were going on with audio, I'd notice some other sound quality issues though.

HDA Analyzer throws a syntax error with python 2.7 (I have that and 3.5, 3.5 breaks a lot quicker):

  File "/dev/shm/hda-analyzer/hda_analyzer.py", line 48
    raise ValueError, "URL %s" % url
                    ^
SyntaxError: invalid syntax

I don't know python and don't feel up to debugging this this evening. Maybe later.

The port itself really seems to be alright. Next time I have the machine apart at all, I'll take a look at the board in more detail and see if anything looks wrong (solder joints or such). It's a bit more of a pain to inspect than it would be for some other machines (the downside of tiny cases), and what I'm doing now is mostly just better anyway.
 
Waco
Gold subscriber
Minister of Gerbil Affairs
Posts: 2033
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: motherboard audio degradation

Fri Nov 11, 2016 12:34 am

My wife managed to kill three...yes, three...boards worth of onboard sound by running SLI configs. The only common link to the failures was running SLI, different speakers and headphones for each iteration.
Z170A Gaming Pro Carbon | 6700K @ 4.5 | 16 GB | GTX Titan X | Seasonix Gold 850 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | Cosmos II | Samsung 4K 40" | 480 + 240 + LSI 9207-8i (128x8) SSDs
 
whm1974
Gerbil Elder
Posts: 5462
Joined: Fri Dec 05, 2014 5:29 am

Re: motherboard audio degradation

Fri Nov 11, 2016 1:40 am

Waco wrote:
My wife managed to kill three...yes, three...boards worth of onboard sound by running SLI configs.  The only common link to the failures was running SLI, different speakers and headphones for each iteration.

Good reason not to use SLI configurations.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Fri Nov 11, 2016 2:16 am

How...? :-?
 
whm1974
Gerbil Elder
Posts: 5462
Joined: Fri Dec 05, 2014 5:29 am

Re: motherboard audio degradation

Fri Nov 11, 2016 2:24 am

synthtel2 wrote:
How...?  :-?

Heat? SLI setups do produce a large amount of heat. If the board manufacturer placed the audio chip in between the PCIe x16 slots, then the chip could be cooked to death.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Fri Nov 11, 2016 2:48 am

A fair point. That is a pretty toasty spot. Now that I think about it, that's an argument for horizonal GPU heatsink fins that I didn't consider.
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Fri Nov 11, 2016 6:28 am

synthtel2 wrote:
HDA Analyzer throws a syntax error with python 2.7 (I have that and 3.5, 3.5 breaks a lot quicker):

  File "/dev/shm/hda-analyzer/hda_analyzer.py", line 48
    raise ValueError, "URL %s" % url
                    ^
SyntaxError: invalid syntax

I don't know python and don't feel up to debugging this this evening. Maybe later.

Yuck. IIRC it tries to check for updates of some part of itself when it runs; maybe that piece is broken? I have a vague recollection of doing something to disable that behavior at some point, but it has literally been years since I looked at the source code, and months since I last used it (I am on a different system with a Xonar card now, so my Realtek is disabled).

I'll see if I can dig up the last (working) version I was using.
Nostalgia isn't what it used to be.
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Fri Nov 11, 2016 6:45 am

You can grab the version of HDA Analyzer I was using here: http://uchima.net/trstuff/hda-analyzer.tar.gz

Unpack the archive, and from within the directory where you unpacked it run "sudo ./hda_analyzer.py". No guarantees it'll work on your distro (last system I used it on was running Kubuntu 12.04), and there may be some library dependencies you need to manually install.

Edit: sha256sum output, for verification purposes:
8e04fc426104c915d03329abcc892ba7363eb757d0dbb3266057ba0df6496e9f  hda-analyzer.tar.gz
Nostalgia isn't what it used to be.
 
Waco
Gold subscriber
Minister of Gerbil Affairs
Posts: 2033
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: motherboard audio degradation

Fri Nov 11, 2016 1:20 pm

whm1974 wrote:
synthtel2 wrote:
How...?  :-?

Heat? SLI setups do produce a large amount of heat. If the board manufacturer placed the audio chip in between the PCIe x16 slots, then the chip could be cooked to death.

I blame power distribution, actually. The slots were more than far enough apart, these were stock blower coolers, and there was a decent amount of airflow front->back in the case.

One one, the onboard sound was near the bottom of the board far away from the GPUs. The only common factor was that there were two power hungry GPUs pulling 100+ watts (combined) from the slots.
Z170A Gaming Pro Carbon | 6700K @ 4.5 | 16 GB | GTX Titan X | Seasonix Gold 850 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | Cosmos II | Samsung 4K 40" | 480 + 240 + LSI 9207-8i (128x8) SSDs
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Fri Nov 11, 2016 1:30 pm

Waco wrote:
whm1974 wrote:
synthtel2 wrote:
How...?  :-?

Heat? SLI setups do produce a large amount of heat. If the board manufacturer placed the audio chip in between the PCIe x16 slots, then the chip could be cooked to death.

I blame power distribution, actually. The slots were more than far enough apart, these were stock blower coolers, and there was a decent amount of airflow front->back in the case.

One one, the onboard sound was near the bottom of the board far away from the GPUs. The only common factor was that there were two power hungry GPUs pulling 100+ watts (combined) from the slots.

Yeah, I guess that's a possibility. Sudden large changes in loading on the power rails could conceivably cause overshoot or ringing that briefly exceeds the input voltage range of other devices connected to that rail. What doesn't make sense, however, is that GPUs would be most likely to cause power problems on the +12V rail, but audio codecs are generally powered off of the lower voltage rails (e.g. Realtek typically uses +5V and/or +3.3V).
Nostalgia isn't what it used to be.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Fri Nov 11, 2016 6:28 pm

That version of HDA Analyzer works, thanks. :) HP is off, and turning it on doesn't seem to do anything.
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Fri Nov 11, 2016 6:40 pm

synthtel2 wrote:
That version of HDA Analyzer works, thanks. :) HP is off, and turning it on doesn't seem to do anything.

You should at least hear a moderate volume increase. Are you sure you tried all the nodes that have an "HP" option? If not, you may have enabled the headphone amp on one of the surround channels instead of the one the headphones are plugged into. (IIRC on my system it was node 14, but this likely varies by motherboard vendor/model.)
Nostalgia isn't what it used to be.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Fri Nov 11, 2016 7:05 pm

Yeah, what it gave me was three checkboxes labelled HP / OUT / IN, and when I disabled OUT on that one it cut things, so it was the right one.
 
Waco
Gold subscriber
Minister of Gerbil Affairs
Posts: 2033
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: motherboard audio degradation

Fri Nov 11, 2016 8:18 pm

just brew it! wrote:
Yeah, I guess that's a possibility. Sudden large changes in loading on the power rails could conceivably cause overshoot or ringing that briefly exceeds the input voltage range of other devices connected to that rail. What doesn't make sense, however, is that GPUs would be most likely to cause power problems on the +12V rail, but audio codecs are generally powered off of the lower voltage rails (e.g. Realtek typically uses +5V and/or +3.3V).

It's a mystery to me. I just know it was repeatable and annoying to the point where I bought her a sound card.
Z170A Gaming Pro Carbon | 6700K @ 4.5 | 16 GB | GTX Titan X | Seasonix Gold 850 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | Cosmos II | Samsung 4K 40" | 480 + 240 + LSI 9207-8i (128x8) SSDs
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Fri Nov 11, 2016 8:31 pm

synthtel2 wrote:
Yeah, what it gave me was three checkboxes labelled HP / OUT / IN, and when I disabled OUT on that one it cut things, so it was the right one.

Oh well, it was worth a try.
Nostalgia isn't what it used to be.
 
HERETIC
Gerbil XP
Posts: 396
Joined: Sun Aug 24, 2014 4:10 am

Re: motherboard audio degradation

Sat Nov 12, 2016 5:50 am

Waco wrote:
My wife managed to kill three...yes, three...boards worth of onboard sound by running SLI configs.  The only common link to the failures was running SLI, different speakers and headphones for each iteration.

Synth,
After reading the above post and considering your using a small case-Heat could be a real killer.
Any idea what the temp's are inside case?????????????
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Sat Nov 12, 2016 8:10 am

just brew it! wrote:
synthtel2 wrote:
Yeah, what it gave me was three checkboxes labelled HP / OUT / IN, and when I disabled OUT on that one it cut things, so it was the right one.

Oh well, it was worth a try.

FWIW it definitely made a difference for me; I suppose it may come down to headphone impedance, and/or the specific audio codec used (Realtek seems to make a bazillion different ones).
Nostalgia isn't what it used to be.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Sat Nov 12, 2016 3:19 pm

HERETIC wrote:
Synth,
After reading the above post and considering your using a small case-Heat could be a real killer.
Any idea what the temp's are inside case?????????????

Frosty like Siberia, largely as a reaction to shenanigans like this. I've got one of these beasties in it (though it's not on too aggressive a curve).

just brew it! wrote:
FWIW it definitely made a difference for me; I suppose it may come down to headphone impedance, and/or the specific audio codec used (Realtek seems to make a bazillion different ones).

These headphones (Grado SR80e) are pretty low impedance, so if there's an amp, it should definitely do something. The audio chip is an ALC1150 (I probably should have mentioned that earlier).
 
just brew it!
Gold subscriber
Administrator
Posts: 49711
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: motherboard audio degradation

Sat Nov 12, 2016 3:32 pm

synthtel2 wrote:
just brew it! wrote:
FWIW it definitely made a difference for me; I suppose it may come down to headphone impedance, and/or the specific audio codec used (Realtek seems to make a bazillion different ones).

These headphones (Grado SR80e) are pretty low impedance, so if there's an amp, it should definitely do something. The audio chip is an ALC1150 (I probably should have mentioned that earlier).

The mobo where it made a noticeable difference (Asus M5A97 R2.0) was using an ALC887, and the headphones were Sennheiser PX-100-IIs (which are 32 ohm, IIRC). For whatever that's worth. :lol:

Without the integrated headphone amp enabled, the onboard sound was muddy and thin, and had a hard time driving the headphones to decent volume.

These days I'm using a discrete soundcard + external headphone amp...

Edit: Current mobo has an ALC892... didn't really do a serious evaluation of it, as I'd already decided to go with a discrete soundcard for this build. Like I said, Realtek has a bazillion different sound chips! I suppose I should compare the onboard to the discrete (and with/without the outboard headphone amp) at some point, if only to satisfy my own curiosity.
Nostalgia isn't what it used to be.
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Thu Nov 17, 2016 10:44 pm

All the solder joints in the area look ship-shape.
 
ludi
Darth Gerbil
Posts: 7449
Joined: Fri Jun 21, 2002 10:47 pm
Location: Sunny Colorado front range

Re: motherboard audio degradation

Fri Nov 18, 2016 12:16 am

synthtel2 wrote:
All the solder joints in the area look ship-shape.

...and may still have hairline cracks.  While these are probably NOT the most likely explanation, the only way to be sure is to reflow them.
Abacus Model 2.5 | Quad-Row FX with 256 Cherry Red Slider Beads | Applewood Frame | Water Cooling by Brita Filtration
 
synthtel2
Gold subscriber
Gerbil Elite
Topic Author
Posts: 609
Joined: Mon Nov 16, 2015 10:30 am

Re: motherboard audio degradation

Fri Nov 18, 2016 12:46 am

I'm not going to bother reflowing them, because I've got that external DAC. :wink: Also, the solder work on this board actually is pretty impressive. After that inspection, solder joints are waaay down my list of probable causes, and I've done enough soldering to know.

The whole thing is kind of moot. Curiosity is the main motivation at this point.

Who is online

Users browsing this forum: No registered users and 3 guests