Multiplier forced to 6x on EP45-UD3R (should be 9x)

[mauibg]

Re: Multiplier forced to 6x on EP45-UD3R (should be 9x)

You can also try other multipliers, 400x 7.5 also gets you there (if those aren't also being reset to 6).

All multipliers from 6x - 9x are available and selectable in the BIOS. Upon reboot (after saving BIOS changes) the MB reverts back to 6x everytime. Previously everything was fine at 9x333 no problem. Only other thing I can mention is that it coincided with the upgrade of my fanheatsink (Arctic Cooler Pro 7) as best I can remember. Maybe I knocked the CPU power conn from PSU to MB loose. I wil check when I get home from work.

BTW, this same PSU was used in my AMD 64 X2 4200+ which uses more power than the 45nm E8500. Outside of that, and the MB & mem, all other H/W is the same. I did however add an extra PCI/IDE ctlr card as well. I will try removing that as well as the optical drives just to lighten the load on the PSU. I do not game with this PC, it's only used for video encoding which means both cores will be pegged 90 - 100% for sometimes 48 hours straight. So I need it to be fast and stable.

I had this problem earlier. My mobo wouldn't allow me to go passed 6x multiplier. No matter how much I changed it, it would always go back to 6x. This is what I did to fix it. Hopefully it will work for you too.

1) Disabled C1E and EIST
2) Upped my CPU Core voltage from 1.1->1.2
3) Upped my DRAM voltage from 1.8->2.1

By doing this, I'm now able to OC my cpu (E8500) to 3.6 GHz. Ran multiple stress tests and it seems to be stable.

Thanks for the tip. Tried these settings with same result unfortunately. Rebooted and forced to 6x. Very disappointed in this MB so far. Brand new and not very useful to me... Although it seems to run fine/stable at 6x in WindowsXP (@ 2GHz)

Try F3 BIOS and make sure that it does actually flash properly? Does that board have Qflash?

Yes I have flashed and used F4, F5 and F6 BIOS versions (not to be confused with the 'F' keys). Qflash is a very easy program. Works great. No solution to my problem though.

EIST and C1E should be disabled. Has this been corrected?

What BIOS are you using, and have you loaded optimized defaults for this setup/BIOS yet? If not, you must load optimized defaults and save/apply/reboot anytime you install a new CPU or anytime you flash the BIOS

As of now C1E is disabled. Please read original post regarding EIST and BIOS info. I have spent hours playing with the BIOS of course saving everytime. That is what is so frustrating; everytime I make the change, it is forced back to 6x on POST/boot. This did not always happen.

The thing is, I had the same problem. For some reason my BIOS would always show the multiplier at 6. If I went into windows and ran cpu-z, eventually it showed my multiplier at 9. I hope there's nothing wrong w/ my board also. I had to set my cpu voltage at 1.2 (default was at 1.1) and I upped my dram voltage also. The multiplier in BIOS right now still shows at 6 but in windows it goes up to 9. I checked using cpu-z, core temp, real temp, and in Orthos.

I also have C1E and EIST disabled.

The change (you may already know) in CPUZ that you are seeing is the CPU throttling from 2 - 3 GHz. EIST being disabled is supposed to prevent this (Enhanced Intel SpeedStep Technology) curious you are seeing this still.

I wish this was the problem I was having. I am aware of this technology and what it can do, unfortuntely my CPU is forced to 6x without the ability to throttle up. The setting in the BIOS under EIST for the clock ratio defines the maximum multiplier for the bus speed.

It seems I am just resigned to a defective MB. My first Gigabyte. Too bad it is a negative experience. Any other socket 775 M/B recommendations that offer the O/C ability that the Gigabytes do? I really wanted this board to work out for me.

[acarcia81]

Re: Multiplier forced to 6x on EP45-UD3R (should be 9x)

Funny, this started happening to me when I changed my fan also. Anyways, try changing settings in the BIOS (cpu voltage, multiplier, ddr voltage and fsb), don't go back to the BIOS after the change, let it boot all the way to windows. Go to Control Panel -> System and see what it reads the cpu @. Mine shows E8500 @ 3.16 3.81 GHz (Since I OC'd).

Here was my post in the gigabyte forum of toms hardware. Hopefully this will help you as well.

EP45-UD3R Won't Save CPU Clock Ratio and Fine CPU Clock Ratio

[VorLonUK]

Re: Multiplier forced to 6x on EP45-UD3R (should be 9x)

When you say "changed your fan", do you mean changed your CPU cooler from the one that comes in the Retail Intel box with an "aftermarket" cooler?

[acarcia81]

Re: Multiplier forced to 6x on EP45-UD3R (should be 9x)

When you say "changed your fan", do you mean changed your CPU cooler from the one that comes in the Retail Intel box with an "aftermarket" cooler?

Sorry, I meant changing the CPU Cooler.

[SpeedEuphoria]

Re: Multiplier forced to 6x on EP45-UD3R (should be 9x)

What are your temps, its possible that the pushpins are not fully seated so it downclocks to 6x. Please check in pc health in bios to see the idle temps.

Also on the PSU, well its used so its always a consideration. How long have you had it.

[VorLonUK]

Re: Multiplier forced to 6x on EP45-UD3R (should be 9x)

Sorry, I meant changing the CPU Cooler.

Just a thought......

As often aftermarket Cpu Coolers have larger and therefore slower rpm fans (for a given airflow), some bioses may assume the CPU fan isn't running at all, especially if it's a PWM type and setup as "Auto" mode in the bios.

In theory it would be a good idea to lean on the side of caution if the bios interpreted that a CPU fan wasn't running and reduce the CPU multiplier accordingly so that just passive cooling would protect the CPU.
However I DON'T know if that is the case and of course those using Passive or Water Cooling methods, would have to attach a "dummy" fan to "trick" the bios.

One way that may prove/indicate that this scenario does in fact exist, is to set the CPU fan if it's in PWM "Auto" mode to a normal or fixed mode, so that the fan is essentially then running at it's full speed. Likely (even if its a large fan 120mm+) it's full rated speed would easily be enough to satisfy most bioses that it is indeed running (if it didn't before).

But as i said above "its justs a thought", as it was mentioned both posters had changed their Cpu coolers. Of course changing a Cpu Cooler can result in other effects, ie from general damage to Cooler misalignment. So although a multiplier reduction would be a great protection mechanism (especially for low TDP 45nm cpu's) against a CPU Cooler malfunctioning, whilst still letting the PC boot, it likely isn't implemented.

[mauibg]

Re: Multiplier forced to 6x on EP45-UD3R (should be 9x)

Sorry got home late so couldnt test/check some things. Will try again tonight.

What are your temps, its possible that the pushpins are not fully seated so it downclocks to 6x. Please check in pc health in bios to see the idle temps.

Also on the PSU, well its used so its always a consideration. How long have you had it.

Thanks for the tip, pushpins fully seated and locked to MB. CPU temps in the 30's in BIOS. Temps stay in the 30 - 40 range under 100%-both cores (still running @ 2GHz thats why). When everything WAS running fine, it was 58 - 60 under 100%-both cores @3GHz. Using CoreTemp in XP. PSU seems fine and stable; just over a year old. Dont have any laying around campatible with this MB that I can test with at the moment. Still want to test reducing the load on it.

Just a thought......

As often aftermarket Cpu Coolers have larger and therefore slower rpm fans (for a given airflow), some bioses may assume the CPU fan isn't running at all, especially if it's a PWM type and setup as "Auto" mode in the bios.

In theory it would be a good idea to lean on the side of caution if the bios interpreted that a CPU fan wasn't running and reduce the CPU multiplier accordingly so that just passive cooling would protect the CPU.
However I DON'T know if that is the case and of course those using Passive or Water Cooling methods, would have to attach a "dummy" fan to "trick" the bios.

One way that may prove/indicate that this scenario does in fact exist, is to set the CPU fan if it's in PWM "Auto" mode to a normal or fixed mode, so that the fan is essentially then running at it's full speed. Likely (even if its a large fan 120mm+) it's full rated speed would easily be enough to satisfy most bioses that it is indeed running (if it didn't before).

But as i said above "its justs a thought", as it was mentioned both posters had changed their Cpu coolers. Of course changing a Cpu Cooler can result in other effects, ie from general damage to Cooler misalignment. So although a multiplier reduction would be a great protection mechanism (especially for low TDP 45nm cpu's) against a CPU Cooler malfunctioning, whilst still letting the PC boot, it likely isn't implemented.

Tried stock cooler as well with same multiplier result. Arctic Cooler Pro CPU fan ~2400 rpm I think. I dont remember exactly off the top of my head; it was just a littler higher than stock. Using the full speed setting in the BIOS did not work well. I tried that when I orignally installed the components (before any of this multiplier problem) and the RPM took a dive and the CPU alarm I set went off (meaning the full speed was more like a crawl causing the dissapation from the heatsink to be drastically reduced resulting in the core temp increase). Just a reminder this is a brand new RMA'd CPU (meaning 2nd one); right out of the box into the socket. So the CPU doest not appear to be the culprit.

I really appreciate the ideas. I want to know the problem as well wanting to have a fully functioning (at proper specs anyway) computer. Before long I'll get tired if I dont find a solution and just buy another board. D@mn I dont want to spend the money... I JUST BOUGHT A BRAND NEW ONE!!!