Volume varies with overclock - is this even possible? :confused:

manveruppd · 5 Aug 2010 at 22:12

After the trials and tribulations with my system not POSTing, I left my CPU at stock until I get round to testing my components. However, the volume being output by my graphics card (PCI-E X-Fi Fatal1ty Titanium) is noticeably lower! I checked that all the volume sliders in the X-fi's drivers and in the Windows control panel were turned all the way up, and nothing had changed since before my crashes - nothing except that my CPU speed and voltages are back at stock...

Is there any logical explanation for this, or am I going mad? :eek:

egnshane · 6 Aug 2010 at 00:01

The PCI-E speed shouldn't alter with your cpu mhz. What Mobo is it?

egnshane · 6 Aug 2010 at 00:04

Just looked at your link and you are using a Gigabyte P35-DS3
You should be able to lock the speed of the pci-e if its not already locked

Amonlym · 6 Aug 2010 at 02:16

possibly the supply voltage on the pcie slots is drifting a little bit with increasing loading of the +12 rail and this in turn is altering the overall amplification on the soundcard.. but that would require some terrible design work in a lot of areas (starting psu, visiting mobo ending with soundcard amps)

(group regulated psu's will tend to show an elevated +5v (and sometimes +3.3v) rail as 12V rail loading increases.
if there is no regulation of these voltages on the motherboard (i dont know if this is commonly done or not) then the voltage arriving at the soundcard and available for use would be higher with a greater cpu overclock, with no regulation here then this could result in changing output levels though i would expect it to be a tiny effect

your sure you haven't just changed a speaker volume or partially dislodged a 3.5mm jack while fiddling? - a half inserted 3.5mm jack tends to give a muted, flat, monaural output with reduced overall volume.

manveruppd · 6 Aug 2010 at 14:33

Yes, sorry, forgot to mention my motherboard. (It's actually the DS3R, but I think the only difference is the northbridge heatsink.) I was thinking along the same lines as Amonlym, that as I ratched up my Vcore, the PCI-E voltage maybe also rose a bit, so now that vcore is back at stock the Sound Blaster's getting a tad bit less juice. @egnshane, I'm curious why you think the PCI-E speed would affect volume - the X-fi is just an amplifier, so surely it's a matter of how much voltage it's getting? I'm gonna try going into CMOS and fixing PCI-E speed at 100, then overclocking again - if you're right, then the volume will remain at its current low level.

My audio cables are all definitely plugged in properly, they were the first thing I checked!

I have actually noticed this problem before: a few months ago I tried pushing my CPU to 3.4,and of course it wouldn't even POST and I had to reset the CMOS settings. I left it at stock for a couple of days before I got round to pushing it back to my 3GHz overclock, and in those 2 days I noticed that the volume was a bit low, but in the beginning I thought it was just in my head. But then I overclocked again, and I noticed that everything was NOTICEABLY louder, so I knew something was up. I thought "nah, it can't be" and put it out of my head, but now that it's back at stock and the volume dropped again I'm SURE this is it!

Now comes the hard question: is it possible that this might damage something? Specifically, my X-Fi? Someone in the thread I linked to said that my motherboard might be dying (and I think he's right), so it's possible its current regulation is going senile and pumping more volts through the PCI-E bus. I don't think my PSU is the problem: it's old, but it's a good, solid piece of kit (Antec NeoHD 500), and anyway the X-Fi doesn't have a direct connection to it, it's getting its juice straight through the PCI-E slot. My motherboard, otoh... ok, Gigabyte is a decent brand, but I have the rev. 1.0 of this board, and they've been known to have other problems too (which is whythey made 2 more revisions of it). And the X-Fi IS, after all, an amplifier, and I know enough about amps to know that smoke tends to come out if too much current goes through them...