Is your processor Running slower than you think it should be ? Or at a different multiplier than it should be ?

[Gillbot]

Originally posted by: spittledip
I just got a Q9550 last night and assembled my new build last night. Today I was testing for OC - lowered multiplier from 8.5 to 6 and upped the FSB to isolate and test RAM. When I finished, I put the multiplier back to 8.5 and went back into windows- CPU-Z showed my multiplier to be still at 6!!! Long story short, I worked for about an hour trying to fix the issue.. and then I remembered this thread. Turns out that CPU-Z was not reporting my multiplier right- I downloaded crystalcpuid and this showed what I had set the BIOS at. Very strange that the multiplier got stuck in CPU-Z- it was showing correctly when I first installed it. Anyway, thanks very much for this thread!!!

Give yourself a pat on the back, VERY few people read this thread.

[BonzaiDuck]

I'm keeping alive my own thread on the "Motherboards" forum, as I attempt to troubleshoot this Gigabyte GA-EP45-UD3R motherboard.

1) The board seems to be finicky about certain types of memory: I'm not the only person who's had trouble with certain flavors of OCZ RAM.

2) the board auto-configures itself to set the multiplier for the CPU lower than the Intel spec. This may have something to do with the power-management features, but I haven't seen anyone post specifics about how the auto-configuration assists with it. I noticed that the autoconfiguration sets the processor to run at 2.166 Ghz (8 x 266) (see (3) below), but MEMTEST86+ shows it as "Core 2 Duo 2.4 Ghz." When I choose the multiplier setting manually so that the system posts at 2.4 Ghz, MEMTEST86+ shows it at 2.397 Ghz. This (inconclusively) suggests to me that the motherboard is actually running the processor under MEMTEST86+ at 2.4 with the motherboard auto-configuration that posts at 2.166.

3) I'm running what is now a dated C2 an E6600 Conroe. Just off the top of my head, I see people using the motherboard with fewer problems (or none at all) who are configuring it for Wolfdale and Yorkfield processors.

Even with the replacement G.SKILL memory (to replace the OCZ modules), I've found this setup locking up after a few iterations of MEMTEST86+, even though I KNOW the modules are rock-solid good and reliable . .. . let me elaborate . . .

4) I went back into BIOS, and disabled ALL the power-management features. It doesn't seem to be locking up now, even with the CPU VCORE and RAM voltages set to the same values effective when the lockups occurred under MEMTEST86+.

Is there a difference in the power-management features of Conroe and Penryn-type cores?

Is it at all likely that -- even if the mobo-maker says the board will run {Pentium D, C2D, C2Q . . } LGA775 cores -- that the BIOS shipped with it may be intended more for Penryn-type cores?

[DominionSeraph]

Originally posted by: foolio65
I hadn't used this machine in a week or so, and when I turned it on yesterday it was incredible slow.

Like insanely stuttering slow, or seemingly normal for an 800MHz processor? (At 800MHz, Win98 should still be super-fast [I have a PIII 700 that boots in 10 seconds] and XP should run ok.)

If it's crazy slow, it could be thermal throttling. My Inspiron 8000 has a loose heatsink, and even though it reports its full 900MHz (no speedstep on P3s), once that fan turns to emergency-high, it dogs waaay down until reboot.

If that's not it, you might have to try clearing CMOS and reinstalling your OS.

And why on Earth would you use Always On?

[shobuddy]

BonzaiDuck, send you a PM

[Aqeel1]

this one is really good and I like it very much .

[Thuggles]

Hi guys/girls. First post on the forum "Woohoo"! Fortunately i managed to resolve my issue before having to ask a question. I can confirm that my i7 920 does not like cpuz. I OC'd the 920 to 3.7GHz and just scared myself when i checked cpuz and found that the multiplier was not dropping from 20! It was hovering within 10MHz of 3.7GHz and chewing more than it's fair share of power. Fortunately coretemp has shown the true multiplier which drops back down to 12 when it's not being spanked. cpuz 1.5 does not work with i7.

Out of interest does anyone have any info on setting up i7's as i am a first time overclocker and i'm probably a bit far off the mark as my temp is getting right up when i'm at 3.7GHz. I'd like to keep the temp down but don't like the idea of water cooling and know that my bios setting can be refined.

[johnny.dacu]

Ok, this is good to be known, i have this problem with my processor and i thought the solution is to reset it. Your advices are very useful, but i'm also curios..if you have reset your computer and all...do you still need the action in BIOS if the processor is slow? or it may be another problem?

[wpcoe]

I forgot about this stickied thread, and posed this question as part of another post. Maybe it's more appropriate to ask in the context of this thread:

My i5-760 consistently shows a 22x multiplier in CPUID's CPU-Z. Whether overclocking, or with stock settings (everything set at "Auto"), I don't see a 21x multiplier which the specifications for this chip (21 x 133Mhz = 2.793Ghz ~ advertised 2.8Ghz) indicate. Is CPU-Z in error?

FYI, at light load, the chip powers down to a 9x multiplier (and core temps drop) for a result of 1.2Ghz. Just curious: is 9x the proper multiplier for that situation?

Originally Posted by spittledip

Turns out that CPU-Z was not reporting my multiplier right- I downloaded crystalcpuid and this showed what I had set the BIOS at. Very strange that the multiplier got stuck in CPU-Z- it was showing correctly when I first installed it.

Based on that idea, I downloaded CrystalCPUID to verify it against CPU-Z. Are you sure CrystalCPUID is not just reflecting the BIOS settings? On my system it sits there showing 2798.05 Mhz as the current clock regardless if the CPU steps down to 9x at idle, or if TurboBoost kicks in (as verified by Intel's own TurboBoost Monitor). I do keep pressing <F5> and CrystalCPUID keeps reporting 2798.05 when all other indications are that that is *not* the current speed.

[SillyBit]

Ya, when I first started overclocking, I took notice of this all important fact about the auto down-clocking to conserve power and reduce heat. I fist saw it using CPU-Z.

[w1llm]

hi

[speedy504]

lol .

[JBT]

I seem to be having issues where my CPU will downclock when I run something like Prime.
I would like the CPU to downclock when idle but not when full load or gaming.

In the BIOS I have the multi set to 44. I have all the power saving features enabled currently. I have my voltage off set at -.55 which should be about 1.33 This is what CPU-Z and Asrock's tuner display. Core Temp shows the VID at 1.375-1.38.

At any rate with this set up, to start my CPU runs at the correct speed of 4.4Ghz. Eventually as the heat creeps up to about 61C on inplace FFT's. It will drop the multi to 42 or 43 once it hits 62C and the temp will top off at 63C.

Is there any thing I can do to keep power savings on but still receive my full OC?

[muk1604]

thx

[nxttruong]

I don't know much about this aspect. I got more from our website.

Thank you!

[alexanderbwinni]

This thread needs to suggest a way to actually verify multiplier when EIST and C1E are enabled. I'm greedy and would like to have an overclock and power saving, but I just can't tell if it's working for me or if enabling these options is permanently locking me to 6x. I've tried running cpuz while stress testing and it still show 6x.

Are there any more reliable tools for testing? I suspect either a software bug or my motherboard.

[crazymonkeyzero]

Wanted to share something I just found out, which may be worth adding to the sticky, in case less experienced member (like myself) did not know.

I just picked up a 3930k on BF and made my 2011 workstation yesterday, and to my surprise it was just plain slow in running computations for my work. The i7 3770k computer I had was ripping through tasks, finishing jobs in 20mins, while the 3930k was taking 32mins for the exact same job (note, all chips are at stock). Upon installing core temp and cpuz, I noticed that voltage on the my 3930k was 0.8 volts at times, and the chip was down clocking to 1.2ghz during the job. Temperature throttling was not the issue, because I have the h100 and temps were always below 50c. After messing with the bios, to no avail, I was just about to give up, when I saw the option in the control panel in windows 7 for power options and noticed that it defaulted to "power saving" After changing this to "high performance" I noticed a 2x increase in speed of the chip. Now the same exact job finishes in 16mins and the 3930k runs 3.5-3.8ghz all the time.

Very happy now Was just about to return the 3930k lol.

[John Tauwhare]

I have a Lenovo U260 laptop with a Core i5-470UM Nehalem CPU clocked at 1333 with turbo boost to 1867. I ran an Excel benchmark test on it expecting a time comparable to a 2GHz Penryn Core Duo of c.100 seconds. But it averaged 172 seconds and ranged widely from 159 to 184. I figured the turbo boost wasn't triggering, or only randomly, and moved on to other things. A couple of weeks later I had time to investigate, and used Argus Monitor to track the CPU clock speed as it ran the benchmark. It wasn't running with a 1333 floor; it was actually spending alot of its time at 667.

THEN, I finally got to reading this thread. And realised my mistake.

So, I selected the High Performance power plan, and went into advanced setting to make sure the processor state would give 100%. Then I reran the benchmark. Argus showed the clock pressed on the 1867 limit the entire run and the time came down to 87 seconds. Which seems quite impressive for an 18 watt part, and has changed my opinion of ultra low voltage CPUs.

Thanks.

[NathanBrook]

Originally Posted by Markfw900

Well, the way I read the Intel documents, C1E is this:
"C1 - Auto Halt: core clock is off. C1 is a state where the processor is not executing instructions, but can return to an executing state essentially instantaneously. Some processors also support an Enhanced C1 state (C1E) for lower power consumption."

While EIST is enhanced speedstep, changing the multi to 6 instead of 9 (as an example) and lowering voltage. I didn't see anything about cache.

Hey its appreciative answer and grabing few points here too....