** Official ** Gigabyte UD3 P45 Series -- EP45-UD3 / EP45-UD3R / EP45-UD3P

[pmacdonald]

Thanks for the suggestion radaja.

I'll start this post off by saying that I'm overclocking an E8500 and not an E8400. I copied the code templates and must've forgot to change that at the top.

I tried the template you listed (radaja) and the system failed to boot up. Once I put the VTT to auto the system would start to boot up and then restart itself. The only way the system would fully boot up is if I put the MCH reference setting on auto. It seems that with the settings above or my previous settings the MCH reference and VTT are untouchable or else my system won't boot.

In EasyTune6 with the above my VTT is still shown at a whopping 1.6 which isn't right. I know this is due to my RAM but I'm not sure why. My MCH reference shows around 0.67 in EasyTune with those settings.

One thing I'd like to note is that when I turn XMP to disable my stock timings listed for the RAM change from 4-4-4-12 to 5-5-5-15.

After all this testing I was pretty confident to assume that my problems all reside between the RAM, the high FSB, the VTT, and the MCH reference.

I decided to change my multiplier to 9.5x and to lower the FSB to 422 (4GHz). I used the settings below:

E8500 422 X 9.5 4000MHz F10 BIOS

Robust Graphics Booster.........................[ Auto ] 
CPU Clock Ratio.................................[ 9 ] 
Fine CPU Clock Ratio............................[ 0.5 ] 
CPU Frequency 4.0GHz...........................[ 422 x 9.5 ]

******Clock Chip Control******

CPU Host Clock Control..........................[ Enabled ] 
CPU Host Frequency (Mhz)........................[ 422 ] 
PCI Express Frequency (Mhz).....................[ 101 ]
C.I.A 2 ..................................................[Disabled] 


******DRAM Performance Control******

Performance Enhance .............................[Turbo] 
Extreme Memory Profile (XMP)....................[ Auto ] 
(G) MCH Frequency Latch.........................[ 400MHz ] 
System Memory Multipler.........................[ 2.00D ] 
Memory Frequency 800...........................[ 844 ] 
DRAM Timing Selectable..........................[ Manual ] 

CAS Latency Time 4................................[ 5 ] 
tRCD 4............................................[ 5 ] 
tRP 4.............................................[ 5 ] 
tRAS 12............................................[ 15 ] 


******Motherboard Voltage Control******

Load-Line Calibration [Enabled] 
CPU Vcore [ 1.25 ]...................[ 1.400 ]
CPU Termination [ 1.20 ]...................[ Auto ]
CPU PLL [ 1.50 ]...................[ 1.57 ]
CPU Reference [ 0.760 ]...................[ Normal ]  

MCH Core [1.1 ]...................[ 1.380 ] 
MCH Reference [0.760 ]...................[ Auto ] 
MCH/DRAM Refernce [0.900 ]...................[ Normal ] 
ICH I/O [1.500 ]...................[ Normal] 
ICH Core [1.100 ]...................[ Normal ] 

DRAM Voltage [1.800 ]...................[ 1.90 ] whatever your kit needs for 1000MHz
DRAM Termination [0.900 ]...................[ Normal] 
Channel A Ref [0.900 ]...................[ Normal ] 
Channel B Ref [0.900 ]...................[ Normal ]

With the above settings the computer booted up and didn't seem sluggish at all. I immediately went into EasyTune and saw that the VTT was at the default 1.2V and the MCH reference was at 0.760. I decided to limit the both the MCH reference and VTT to those settings in the BIOS to see if the computer would boot and it did.

So the high FSB and overclocked RAM is what is causing my system to shoot the VTT up to 1.6V. I want to push this system to the max and only be limited by my temps but the high VTT is standing in the way.

How do I keep the VTT down but still increase my RAM speed? The minimum I need my RAM running at is 1000MHz for a 500FSB. I'm sure a new set of DDR3 RAM would solve the problem but frankly I want to figure out why my current set is causing such a high VTT.

[radaja]

dont use auto when OCing.i dont think vtt is your problem.
it really has nothing to do with ram.its voltage for higher fsb(fsb term voltage)
with the voltages i gave mch ref should be showing up as 0.70 in ET6.but maybe
your board dont like mch ref at 0.70v so set my voltages again but set mch ref
one notch lower so it reads 0.730 with vtt at 1.28v.
also you might not be giving your ram enough vdimm for 1000MHz.these boards undervolt
so 1.98v is actually 1.96v

E8500 500x8@ 4000MHz F10 BIOS

Robust Graphics Booster.........................[ Auto ] 
CPU Clock Ratio.................................[ 8 ] 
Fine CPU Clock Ratio............................[ 0 ] 
CPU Frequency 4.0GHz...........................[ 500 x 8 ]@4000

******Clock Chip Control******

CPU Host Clock Control..........................[ Enabled ] 
CPU Host Frequency (Mhz)........................[ 500] 
PCI Express Frequency (Mhz).....................[ 101 ]
C.I.A 2 ..................................................[Disabled] 


******DRAM Performance Control******

Performance Enhance .............................[Turbo] 
Extreme Memory Profile (XMP)....................[ disable ] disable this before you try setting stuff,otherwise it wil reset things
(G) MCH Frequency Latch.........................[ 400MHz ] 
System Memory Multipler.........................[ 2.00D ] 
Memory Frequency 800...........................[ 1000 ] 
DRAM Timing Selectable..........................[ Manual ] 

CAS Latency Time 4................................[ 6 ] 
tRCD 4............................................[ 6 ] 
tRP 4.............................................[ 6 ] 
tRAS 12............................................[ 15 ] 


******Motherboard Voltage Control******

Load-Line Calibration [Enabled] 
CPU Vcore [ 1.25 ]...................[ 1.400 ]
CPU Termination [ 1.20 ]...................[ 1.28v ]
CPU PLL [ 1.50 ]...................[ 1.57 ]
CPU Reference [ 0.760 ]...................[ Normal ]  

MCH Core [1.1 ]...................[ 1.380 ] 
MCH Reference [0.760 ]...................[ 0.73v] 
MCH/DRAM Refernce [0.900 ]...................[ Normal ] 
ICH I/O [1.500 ]...................[ Normal] 
ICH Core [1.100 ]...................[ Normal ] 

DRAM Voltage [1.800 ]...................[ 2.00v] try 1.98v to 2.00v for 1000MHz
DRAM Termination [0.900 ]...................[ Normal] 
Channel A Ref [0.900 ]...................[ Normal ] 
Channel B Ref [0.900 ]...................[ Normal ]

[pmacdonald]

Thanks a ton radaja. I've found out why my previous settings weren't working.

I loaded up your settings exactly as you had them and my system booted fine. I then started to set the values back to the one's I originally had one by one to find out where I went wrong. Each of the settings worked at my previous values until I got down to the final two that were left to change.

The last two I was testing was the CPU term and the MCH reference. The problem I was having is because of the approach I tried to take. I set everything on auto and adjusted each of the settings one by one.

This is where I realized what had went wrong. I set the CPU term to 1.28 and the MCH ref to 0.75. I then set the CPU term to auto and the MCH ref immediately changed to 0.953 which coincidently is where the MCH ref is when the CPU term is at 1.6 which is what EasyTune was originally showing me.

The problem I had when I tried to set the MCH ref to 0.750V and leave the CPU term on auto is that the CPU term would want to shoot itself up to 1.6V but the MCH ref would be stuck at 0.750V so the system wouldn't boot.

Now I just need to optimize the system by running it through Prime95 for awhile and then I'll attempt to lower my voltages. Then I'll try to overclock my system to its limits.

[radaja]

great to hear.keep us posted and if you need any help were here
ive used this template for all my 10+ UD3P's ans many e8400 and e8500's
and it always seems to work.except one board that had issues and i RMA'ed
it and the new one worked great.
so like you said now start working the voltages down.

[pmacdonald]

Well I'm back already... The 4.0GHz was stable for 3 hours so I attempted to bump the multiplier up 0.5x to 8.5x500 / 4.25GHz. I've had no luck so far.

Here's the log that I kept: (hopefully you guys can follow it)
0.4 = FATAL ERROR: Rounding was 0.49609375, expected less than 0.4
0.5 = FATAL ERROR: Rounding was 0.5, expected less than 0.4.

Changed MCH Core to 1.4 and got the BSOD
Changed above back to 1.38 and PLL to 1.5 and got another BSOD
Changed PLL to 1.57 and all the refs back to normal.  BSOD
DRAM Voltage to 2V, VTT 1.3, MCH ref 0.743 booted up
		Prime95 Results:  Worker 2 failed immediately - Rounding was 0.49609375, expected less than 0.4
Restarted with above settings and got BSOD
Standard, 6-6-6-15, VCore 1.4
		Prime95 Results:  Worker 2 failed immediately - Rounding was 0.5
MCH ref 0.763
		Prime95 Results:  Same
MCH core 1.4, DRAM V 2.020V
		Prime95 Results:  Same
PCI E 101, MCH Ref 0.783V
		Prime95 Results:  Same
MCH Ref 0.803 - BSOD
2V DRAM MCH Ref 0.723V
		Prime95 Results:  Worker 2 failed immediately - Rounding was 0.49609375
MCH Ref 0.703V
		Prime95 Results:  Worker 2 failed 0.5
MCH Ref 0.743V PLL 1.50
		Prime95 Results:  Same
CPU Term 1.26 MCH Core 1.3 MCH Ref 0.758
		Prime95 Results:  Same
1.28/1.3/0.750
		Same
1.24/1.3/0.745
		Same
1.28/1.28/0.750
		0.4
1.28/1.26/0.750
		0.5
1.28/1.34/0.750
		0.5
1.28/1.4/0.750 - Ch A&B 1.060V, DRAM Term 1V, MCH/DRAM 1.060V
		0.5
1.30/1.4/0.743 - VCore 1.42500V
		0.5
1.4/1.4/0.806
		0.5
1.5/1.4/0.870
		0.5

Worker two fails every single time immediately.

I'm not sure what settings I need to change. Any help would be appreciated.

Note: The original template used was the one posted above by radaja.

[radaja]

Well I'm back already... The 4.0GHz was stable for 3 hours so I attempted to bump the multiplier up 0.5x to 8.5x500 / 4.25GHz. I've had no luck so far.

Here's the log that I kept: (hopefully you guys can follow it)
0.4 = FATAL ERROR: Rounding was 0.49609375, expected less than 0.4
0.5 = FATAL ERROR: Rounding was 0.5, expected less than 0.4.

Changed MCH Core to 1.4 and got the BSOD
Changed above back to 1.38 and PLL to 1.5 and got another BSOD
Changed PLL to 1.57 and all the refs back to normal.  BSOD
DRAM Voltage to 2V, VTT 1.3, MCH ref 0.743 booted up
		Prime95 Results:  Worker 2 failed immediately - Rounding was 0.49609375, expected less than 0.4
Restarted with above settings and got BSOD
Standard, 6-6-6-15, VCore 1.4
		Prime95 Results:  Worker 2 failed immediately - Rounding was 0.5
MCH ref 0.763
		Prime95 Results:  Same
MCH core 1.4, DRAM V 2.020V
		Prime95 Results:  Same
PCI E 101, MCH Ref 0.783V
		Prime95 Results:  Same
MCH Ref 0.803 - BSOD
2V DRAM MCH Ref 0.723V
		Prime95 Results:  Worker 2 failed immediately - Rounding was 0.49609375
MCH Ref 0.703V
		Prime95 Results:  Worker 2 failed 0.5
MCH Ref 0.743V PLL 1.50
		Prime95 Results:  Same
CPU Term 1.26 MCH Core 1.3 MCH Ref 0.758
		Prime95 Results:  Same
1.28/1.3/0.750
		Same
1.24/1.3/0.745
		Same
1.28/1.28/0.750
		0.4
1.28/1.26/0.750
		0.5
1.28/1.34/0.750
		0.5
1.28/1.4/0.750 - Ch A&B 1.060V, DRAM Term 1V, MCH/DRAM 1.060V
		0.5
1.30/1.4/0.743 - VCore 1.42500V
		0.5
1.4/1.4/0.806
		0.5
1.5/1.4/0.870
		0.5

Worker two fails every single time immediately.

I'm not sure what settings I need to change. Any help would be appreciated.

Note: The original template used was the one posted above by radaja.

well if the first template at 8x500 worked for 3hrs then its just a matter of vcore for 8.5x500@4250,you will just need to increase until its stable.it might be more than your willing to do or temps might get out of control.but by upping the multi its just vcore you need to worry about

[pmacdonald]

That's what I thought at first. I went up to 1.425v and thought that at 4.25GHz that should be plenty. http://www.xtremesystems.org/forums/...05132&page=119 pangingIII hit 8.5x500 with only 1.375v here...hmm

EDIT: I just raised the VCore to 1.45v and still got the same error almost immediately. I'm not sure what to make of this since like you said, simply increasing the VCore should be enough.

[radaja]

every cpu is different.they all have different stock vcore(vid) you can see what yours is by running coretemp

[pmacdonald]

Below is what I'm currently running in BIOS and it's been stable thus far (Prime95 blend for one hour). I'm working on optimizing some settings and am looking for some advice.

When I try to put the vcore at 1.35 the system starts up and it even lets me into windows but I get an error in prime95 after about 2min. (rounding error)

Is there a way to adjust my CPU term and MCH core so that I can lower my vcore? I've seen people in the e8400/e8500 thread run their chips at 4GHz with a vcore of 1.27v.

Something tells me that my CPU term and MCH core aren't optimally set because when I just bump up the multiplier 0.5x my system isn't stable even with an increase in vcore to 1.425v. That seems excessive for 4.25GHz.

E8500 500 X 8 4000MHz F10 BIOS

Robust Graphics Booster.........................[ Auto ] 
CPU Clock Ratio.................................[ 8 ] 
Fine CPU Clock Ratio............................[ 0.0 ] 
CPU Frequency 4.0GHz...........................[ 500 x 8.0 ]

******Clock Chip Control******

CPU Host Clock Control..........................[ Enabled ] 
CPU Host Frequency (Mhz)........................[ 500 ] 
PCI Express Frequency (Mhz).....................[ 100 ]
C.I.A 2 ..................................................[Disabled] 


******DRAM Performance Control******

Performance Enhance .............................[Standard] 
Extreme Memory Profile (XMP)....................[ Disabled ] 
(G) MCH Frequency Latch.........................[ 400MHz ] 
System Memory Multipler.........................[ 2.00D ] 
Memory Frequency 800...........................[ 1000 ] 
DRAM Timing Selectable..........................[ Manual ] 

CAS Latency Time................................[ 5 ] 
tRCD............................................[ 5 ] 
tRP.............................................[ 5 ] 
tRAS............................................[ 15 ] 


******Motherboard Voltage Control******

Load-Line Calibration [Enabled] 
CPU Vcore [ 1.25 ]...................[ 1.35625 ]  1.33 actual
CPU Termination [ 1.20 ]...................[ 1.28 ]
CPU PLL [ 1.50 ]...................[ 1.57 ]
CPU Reference [ 0.760 ]...................[ Normal ]  

MCH Core [1.1 ]...................[ 1.380 ] 
MCH Reference [0.760 ]...................[ 0.750 ] 0.70 in ET6
MCH/DRAM Refernce [0.900 ]...................[ Normal ] 
ICH I/O [1.500 ]...................[ 1.500 ] 
ICH Core [1.100 ]...................[ 1.100 ] 

DRAM Voltage [1.800 ]...................[ 2.040 ] 2.02-2.03 actual
DRAM Termination [0.900 ]...................[ Normal ] 
Channel A Ref [0.900 ]...................[ Normal ] 
Channel B Ref [0.900 ]...................[ Normal ]

[radaja]

run coretemp and tell us what the vid is.sometime cpu ref can help with a slight drop in vcore
but like i said i have one E8400 that does 4.5GHz w/1.29v load and then one that needed 1.46v for 4.5GHz
each cpu is made differently and have different vids.1.33v load is about what all average vid E8400/E8500's
need.also when running lower multi's more vcore is required than running the stock vcore.
example my E8400 only needs 1.20v for 4.0GHz at 9x445@4005 but it needed 1.22v for 8x500@4000

you can try this to see what it takes for 4.0GHz using the 9.5x422@4009

E8500 9.5x422 4009MHz F10 BIOS

Robust Graphics Booster.........................[ Auto ] 
CPU Clock Ratio.................................[ 9] 
Fine CPU Clock Ratio............................[ 0.5] 
CPU Frequency 4.0GHz...........................[ 422x9.5]

******Clock Chip Control******

CPU Host Clock Control..........................[ Enabled ] 
CPU Host Frequency (Mhz)........................[ 422 ] 
PCI Express Frequency (Mhz).....................[ 101 ]
C.I.A 2 ..................................................[Disabled] 


******DRAM Performance Control******

Performance Enhance .............................[Standard] 
Extreme Memory Profile (XMP)....................[ Disabled ] 
(G) MCH Frequency Latch.........................[ 400] 
System Memory Multipler.........................[ 2.00D ] 
Memory Frequency 800...........................[ 844 ] 
DRAM Timing Selectable..........................[ Manual ] 

CAS Latency Time................................[ 5 ] 
tRCD............................................[ 5 ] 
tRP.............................................[ 5 ] 
tRAS............................................[ 15 ] 


******Motherboard Voltage Control******

Load-Line Calibration [Enabled] 
CPU Vcore [ 1.25 ]...................[ 1.350v ] try 1.350v and then try lowering it
CPU Termination [ 1.20 ]...................[ Normal ]
CPU PLL [ 1.50 ]...................[ Normal]
CPU Reference [ 0.760 ]...................[ Normal]  

MCH Core [1.1 ]...................[ Normal ] 
MCH Reference [0.760 ]...................[ Normal ] 
MCH/DRAM Refernce [0.900 ]...................[ Normal ] 
ICH I/O [1.500 ]...................[ Normal] 
ICH Core [1.100 ]...................[ Normal] 

DRAM Voltage [1.800 ]...................[ 2.040 ] 2.02-2.03 actual
DRAM Termination [0.900 ]...................[ Normal ] 
Channel A Ref [0.900 ]...................[ Normal ] 
Channel B Ref [0.900 ]...................[ Normal ]

[pmacdonald]

Thanks, I'll give that template a try tomorrow.

My VID in coretemp is 1.25v.

I bumped up my MCH core to 1.42v to see if it would give me a stable oc at 1.35v. It was prime95 blend stable for 45min then I stopped the test.

I decided that 1.42v was too high at this speed so I set the MCH core back to 1.38v and adjusted my MCH ref to 0.770 in bios / 0.720 ET6. I'm running prime95 right now and it's past where it had failed previously. I'll give an update once it hits 45min (if it gets that far).

After this test I'm going to keep adjusting the MCH ref, MCH core, and VTT in an attempt to lower the vcore.

radaja, can you PM me some of your stable 8x500 templates so I can compare numbers?

EDIT: prime95 blend stable for 53min.

[radaja]

i can send you some templates for sure.
yea thats a high vid chip so 4250 might need more vcore?
what i recommend trying is 7x500@3500 with vcore set to 1.35v
in bios to rule out vcore from causing any problems.and just focus
on making sure 500fsb is stable.meaning you ref's and ram and Nb
are stable.because i just start with vtt at 1.28v and mch at 1.38v
when i get a new cpu or board.but after finding what my new cpu
likes i then start dropping vtt and mch.with 2x2GB ram i usually get
away with 1.24v vtt and 1.32v to 1.34v mch voltage.
with 2x1GB kits i only need 1.26v to 1.30v mch voltage at 500fsb

[pmacdonald]

Is 1.25v really that high of a VID?

I'm finally starting to see some greater potential from this chip. After the above settings worked out I decided to change my MCH core to 1.36 (from 1.38) and MCH reference to 0.750/0.70 in ET6 (from 0.770/0.720 in ET6). I just finished running prime95 on blend for one hour and 3min with no errors. My vcore is still at 1.35v but under HWMonitor it showed only 1.31v which is a greater drop than it has been showing. I'm about to lower the vcore and run prime95 for around 45min. I still feel that I can lower both the VTT and MCH as long as I find the right ratio of MCH to MCH ref. I'll post my updated results tomorrow.

[radaja]

yea the good E8400/8500's with low vids are like 1.12v to 1.20v
but even my good E8400 had a vid of 1.23v so its not a rule but more a guidline
as to how high they will go on acceptable volts.but ive had a 1.25v vid E8500
that did 4.5GHz w/1.44v

[pmacdonald]

Update:

I lowered my vcore to 1.325v in bios (1.30v actual) and wasn't able to get my system stable. Here is some of what I tried:
VTT: 1.24v / 1.26v / 1.24v / 1.26v / 1.26v / 1.28v
MCH: 1.32v / 1.32v / 1.34v / 1.34v / 1.36v / 1.38v

I tried the first 4 combinations with every reasonable MCH reference setting. It seems like this chip won't do 8x500 at 1.3v.

[systemviper]

love the 775's keep on rocking!

[pmacdonald]

radaja, I set my multiplier down to 7x500 to find my lowest mch / vtt setting possible like you suggested but my results don't seem right.

I ran MCH 1.30v with VTT 1.24v and MCH ref at 0.70 (ET6) for 12hrs stable in prime95 (blend).
I kept lowering the settings but everything seemed stable so I decided to turn each setting down to where I thought it would fail immediately and work my way up.
When I set my MCH to 1.10v the system set its resolution to the lowest which I was able to change back but it appeared stable.
I then set the VTT to 1.20 and the only noticeable change was that my Aero transparency settings were no longer able to be activated but it still appears stable.

I'm running Prime95 right now and I'll update later with the results but what is allowing me to have such a low VTT and MCH core?
Also, is the VTT / MCH affected only by the FSB and not the multiplier?

I'm pretty confused with these recent results as I was expecting the MCH to be able to go no lower than 1.32v and VTT 1.24v.

EDIT:

So I got these questions answered myself. I decided to go back up to 8x500 and try to optimize my vcore. The above references worked there too but when I tried to lower the vcore the system would fail after 10min. The higher vcore was just compensating for too low of a VTT and MCH value. 1.325vcore (1.3 actual) at 8x500 is not achievable with my chip. I'm still working on getting 1.33750v (1.31v actual) stable at 8x500 which I know is possible. After that I'm going to shoot for a stable OC at 9x445 to see how much the lower FSB affects my vcore. I should be able to get this done within the next two days. When I get that finished I'll try to get a stable OC at 4.25GHz and go from there.

[pmacdonald]

Well I wasn't able to get 8x500 stable at 1.33750v (1.31v actual). I tried a lot of combinations of MCH/VTT to no prevail.

The most success I had was with the VTT at 1.28, MCH at 1.26, and MCH ref at 0.70v which was stable for 2 hours and 10min.
Changing the MCH ref didn't help and raising the MCH didn't help.
Bumping up the Vcore to 1.34375v (still 1.31v actual) seemed to do the trick at 1.28v VTT and 1.36v MCH with the MCH ref at 0.70v. I'm gonna wait until tonight to run a longer prime95 test on it.

It seems I don't have a golden egg here but I still feel it's safe to run this chip at 4GHz 24/7. Now it's time to push up the vcore and get this baby to 4.25GHz!

[acebmxer]

guys make sure when adjusting MCH Ref or cpu ref to do it in bios and make sure you set vtt to default 1.2 before adjusting any ref voltage. when change has been made then bring vtt back to where you had it set.

[pmacdonald]

Good advice acebmxer but I don't bother to do that myself. Thanks to constant tinkering in the BIOS and the charts radaja sent me I know what the reference will be in ET6 no matter what the VTT is set on.

Update:
I was able to get what seemed to be a stable OC at 4.25GHz (8.5x500) but the voltage required was too high for my liking.
The vcore needed to be set at 1.5v (1.47v actual) which made temps shoot up to 72-73*C after one minute in prime95.
My MCH was set at 1.38v, VTT at 1.28v, and MCH reference at 0.70v. This is by no means optimized because I was unable to run prime95 for an extended period of time.

Before even trying to OC my cpu to 4.25GHz I attempted to OC my GTX 280.
This didn't take long at all because my temps in Furmark were extremely high with the GPU on stock settings.
I decided not even to attempt an OC because the temps were so high (91*C under load in Furmark stability test).
I never hear my fan turn up to 100% in game so within the next week I'm going to run Crysis or CoDMW2 and take temp. readings.
Since I don't game much anymore this isn't that important to me.

The last task I'm attempting is to get a stable OC at 9x445 (4GHz) to see what difference there is between my values at 8x500.

[OCBalmang]

Good advice acebmxer but I don't bother to do that myself. Thanks to constant tinkering in the BIOS and the charts radaja sent me I know what the reference will be in ET6 no matter what the VTT is set on.

Update:
I was able to get what seemed to be a stable OC at 4.25GHz (8.5x500) but the voltage required was too high for my liking.
The vcore needed to be set at 1.5v (1.47v actual) which made temps shoot up to 72-73*C after one minute in prime95.
My MCH was set at 1.38v, VTT at 1.28v, and MCH reference at 0.70v. This is by no means optimized because I was unable to run prime95 for an extended period of time.

Before even trying to OC my cpu to 4.25GHz I attempted to OC my GTX 280.
This didn't take long at all because my temps in Furmark were extremely high with the GPU on stock settings.
I decided not even to attempt an OC because the temps were so high (91*C under load in Furmark stability test).
I never hear my fan turn up to 100% in game so within the next week I'm going to run Crysis or CoDMW2 and take temp. readings.
Since I don't game much anymore this isn't that important to me.

The last task I'm attempting is to get a stable OC at 9x445 (4GHz) to see what difference there is between my values at 8x500.

for 4.2 on air thats pretty good 73C on prime95, now if were talking water GTHO lol!!! my Qx9650 does 4.2 stable at about 65C prime 95 full load pass.. ummm i'm on air... but than again mine is an xtreme edition might have more tolerance... but if im mistaken your working with two cores right?

it should do 4.2GHZ at the same temps with more tweaking unless you got a bad seed

[bluestang]

@ pmac...I used to have my E8500 in a GA-EP45-UD3R and my VID is also 1.250.
Here is a screenshot for you that may help out some way. I still think your memory may be limiting you. Good luck

[Effizienz]

A little temporary result. Now i will reduce the vMCH and vDimm a bit.

[fireblast713]

System:
Intel E8500
G.Skill 4GB Ram
EP45-UD3P Rev. 1.0 F10 Bios
WD VelociRaptor 150GB SATA HDD
Seagate 1.5TB SATA HDD
Hitachi 1TB SATA HDD
Hitachi 1TB SATA HDD
Seagate 2.0TB SATA HDD

Having a strange issue, hoping someone could help me out. So my 150gb Raptor which was my OS hard drive died recently, I RMA'ed the drive and installed the replacement they sent me and now my EP45-UD3P doesn't recognize all my SATA drives, it'll consistently recognize the Raptor and the 2 Hitachi hard drives, however it will never recognize the 2TB Seagate hard drive and will only recognize the 1.5TB Seagate if I set the ICH10 controller to ACHI mode. In ACHI mode it recognizes the Raptor, the 1.5TB Seagate, and the 2 Hitachi. In IDE mode it recognizes only the Raptor and 2 Hitachis.

I checked the power and data cables, tried new data cables, switched the ports they were plugged into (which makes it appear like some ports work and others don't, because when I switch the Seagate into the Hitachi ports, the seagates show up). This makes it seem like the SATA ports have gone bad, however I haven't changed anything except the Bios and the new hard drive, I didn't have any issues prior to the failure of my primary drive. I have also tested the 1.5TB drive in an external enclosure and it works fine, so I don't think the drives themselves are bad.

Anyone have any ideas? I was playing around with the ACHI settings if that matters, but since then I have reset my bios to default settings. I also updated from the F4 to the F10 bios prior to installing the replacement hard drive they sent me. If anyone knows what's going on or has any ideas let me know, I'm considering just RMA'ing the board but I'd rather not if I could avoid it. Thanks.

[NoGuru]

System:
Intel E8500
G.Skill 4GB Ram
EP45-UD3P Rev. 1.0 F10 Bios
WD VelociRaptor 150GB SATA HDD
Seagate 1.5TB SATA HDD
Hitachi 1TB SATA HDD
Hitachi 1TB SATA HDD
Seagate 2.0TB SATA HDD

Having a strange issue, hoping someone could help me out. So my 150gb Raptor which was my OS hard drive died recently, I RMA'ed the drive and installed the replacement they sent me and now my EP45-UD3P doesn't recognize all my SATA drives, it'll consistently recognize the Raptor and the 2 Hitachi hard drives, however it will never recognize the 2TB Seagate hard drive and will only recognize the 1.5TB Seagate if I set the ICH10 controller to ACHI mode. In ACHI mode it recognizes the Raptor, the 1.5TB Seagate, and the 2 Hitachi. In IDE mode it recognizes only the Raptor and 2 Hitachis.

I checked the power and data cables, tried new data cables, switched the ports they were plugged into (which makes it appear like some ports work and others don't, because when I switch the Seagate into the Hitachi ports, the seagates show up). This makes it seem like the SATA ports have gone bad, however I haven't changed anything except the Bios and the new hard drive, I didn't have any issues prior to the failure of my primary drive. I have also tested the 1.5TB drive in an external enclosure and it works fine, so I don't think the drives themselves are bad.

Anyone have any ideas? I was playing around with the ACHI settings if that matters, but since then I have reset my bios to default settings. I also updated from the F4 to the F10 bios prior to installing the replacement hard drive they sent me. If anyone knows what's going on or has any ideas let me know, I'm considering just RMA'ing the board but I'd rather not if I could avoid it. Thanks.

Reflash BIOS and bump ICH up to 1.3.