• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Q6600 G0 OC

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Strange thing happened this morning, I decided to drop my multi to 8x and fsb up to 450 but I completely forgot to change the multiplier and it went into windows. To my suprise when I opened up my system properties its showed 4.05GHz!!! I was really shocked and excited so I decided to orthos test, but only 30 seconds in got a restart 🙁 oh well, pretty good for air cooling lol.
 
I've been having trouble "keeping" the multiplier set for certain settings. I'm wondering if the same thing happened to you, but you interpreted it as "forgetting to set the multiplier."

For instance, if I run 8x360 or 2.88 Ghz 1:1 with proven, tight latencies, it will sometimes boot up to show 3.24 Ghz, and I have to go back into Setup and change it. Resetting CMOS seems to resolve it, but I also notice if I drop the muliplier another notch to 7, it shows no such behavior.

It happened again, and this time, instead of changing the multiplier, I dropped the CPU_FSB from 360 down to 356, and it was fine.

Remember __ I have the B3 stepping.

This is really good news about the G0. But I've got bad news about my utility bill. This thing uses over 400W running ORTHOS x 2, and maybe that's not so bad for shorter periods of time, but 8 or 10-hour tests?

Here's "the project after," thematically following the movie title "The Day After:" A solar-panel array to supplement computer-power-consumption.

 
with my ga-965p-ds3 i'll set my cpu to 7x400 which is 2.8GHz but vista will see it as 3.2GHz. is vista mistaken or is my multiplier also not "sticking"?
 
a q6600 with a tdp of 120w or so, using 400w? When overclocked you might hit 150w or so. Unless you're using a 30% efficient PSU, you can't be using 400w, not when 'only' running orthos.
 
Originally posted by: MarcVenice
a q6600 with a tdp of 120w or so, using 400w? When overclocked you might hit 150w or so. Unless you're using a 30% efficient PSU, you can't be using 400w, not when 'only' running orthos.
Click to expand...

????????? Who mentioned heat? or did I miss something? 😕
 
Originally posted by: solofly
I have a nice Q6600 G0 stepping too... 333 x 9 @ 1.1 volts. That's right, overclocked and undervolted at the same time. That's with Ultra-120 eX and a puny 120mm 1100rpm 43cfm fan.😀
Click to expand...

Hehe same here - except mine needs 1.136V for 334x9 to be stable... I tried 1.12V and failed 18 minutes in 🙁

I'm waiting for a bunch of 40mm fans I ordered off Ebay to setup all over my P35-DQ6 passive cooling system, and then I'll go for 3.6GHz (you can never be to careful).
 
MarcVenice: I'm not talking about thermal power dissipation, but electrical power consumption. Surely, the processor itself (and the nV 680i mobo -- a power-hog) uses some fraction of the total 400W. A single nVidia 8800 card uses a lot of power, although my GTS uses less than a GTX. It is over-clocked. It has four hard disks and a hardware RAID controller. Figure all the overhead shows at the idle power-consumption around 280W-plus.

So what would be using the difference of 400 - 280 = 120+ watts? The processor running at full load.

JawKnee530:

As far as I can tell, Microsoft may assume in their OS components -- per calculation and presentation of CPU speed -- that the user's multiplier is defaulted to 9. I wouldn't know for sure, but I think that information is embedded in the processor itself. For instance, your BIOS should post by first showing the processor's actual speed, and then by identifying it as a quad-core 2.4 Ghz (to use the Q6600 as example).

ORTHOS also makes this error, and so does CoreTemp.

Another thing I may have noticed comes from an exchange I had on another thread with member "tylerdustin." He's going to phase-change-cool his system, shooting for a processor temperature of something like -30C (I could be mistaken, and Intel does have a thermal range spec, but he should be able to adjust the equilibrium temperature to a point where . . . .everything works.)

Tyler told me he enabled Intel SpeedStep with his over-clock setting. [We had already been discussing power-consumption.] I decided to try it.

I observed that all my over-clock settings with the default 9 multiplier are perfectly stable with SpeedStep. And all my over-clock settings with lower multipliers just blow up and reboot as soon as the desktop becomes visible.

So I tentatively conclude that SpeedStep doesn't work, or doesn't work properly -- with other than the stock multiplier.
 
Woodbutcher:

I DID. That is, my entire system consumes 414W OC'd to 3.2 Ghz at full 2xORTHOS load. But you must consider that "system" includes a lot of hardware. And I said, since ORTHOS is mostly "goosing" the processor under "small FFTs" that the load-minus-idle power-consumption difference -- 414W - 292W = 122W -- is attributable to the processor.

That's when I wanted to start a forum on "green computing." I think we should have a sticky that takes posts for different projects, like supplementing PC power-consumption with back-yard solar panels, or using laptop drives in some fast RAID setup that reduces power consumption.

 
I've just been reading an interesting forum post by someone over at overclock.net, discussing the role Northbridge latencies play in overclocking and performance, particularly at FSBs beyond 400MHz. Some of you may find it of use.
Link
 
Originally posted by: jawknee530
with my ga-965p-ds3 i'll set my cpu to 7x400 which is 2.8GHz but vista will see it as 3.2GHz. is vista mistaken or is my multiplier also not "sticking"?
Click to expand...

I've noticed that XP gives a mistaken CPU speed calculation in properties, if you use the mobo to lower the multiplier of the CPU. Guess Vista has the same problem.
 
an 8800gts in idle mode, which it should be when you are running small fft's, shouldn't be using a lot of energy. But if you hooked up a kill a wat meter, or a UPS and can see the energy consumption when running small fft's, who am I say to it isn't so. Perhaps investigate in getting your 8800gts to run into 2d mode, doing so, with Rivatuner I think, should save some energy.

And, HD's dont consume that much energy, only when they spin up from the start will they require a few amps, but after that they just keep spinning spinning spinning, costing relatively low energy.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top