Gigabyte GA-EP45-EXTREME Living Review

[geo69]

Originally Posted by supershanks

You need to run prime small ffts to stress test the cpu.
No not that I know of. The nearest you get is Dual Bios Recovery source this was included on the EP45T Extreme bios , but I don't think features on my F8 bios i'm currently running on the ddr2 Extreme.

I guess thats pretty much what i did, bugger to get the battery out, what were they thinking to put heat pipes over the battery. I still think somthing is still not quite right as the CPU phase leds still come on when i do a cold boot then they shut off part way through the post. This all went for a shit while running OCCT the system overheated and boom. I will try to uninstall/reinstall Crysis see if that fixes my problem then backmthe the over clock.

[supershanks]

Yep you've got it right re the battery
If you mean the Phase leds next to the ram slots then yes I was surprised they seem to go out shortly after post, so don't perform as stated

Originally Posted by P36 Manual

indicates the CPU loading. The higher the CPU loading, the more the
number of lighted LEDs.

Mine always go out about then . which makes them pretty pointless

[geo69]

Originally Posted by supershanks

Mine always go out about then . which makes them pretty pointless

so your LEDs all come on during post? well then I might be ok then

Thanks

[supershanks]

yes the blurb in the manual & on rthe box reckon that the Phase leds are supposed to show load as the pc is running , however the come on then go out as the baoerd posts, so you bsically see bugger all from then on. So sounds like all is normal with you

[glowball]

I though that in Windows DES enables them? Won't see them otherwise.

[supershanks]

I though that in Windows DES enables them? Won't see them otherwise.

probably does if you install it

[geo69]

Originally Posted by supershanks

probably does if you install it

What's DES "do everything stupid" Just kidding!!!!!

[supershanks]

What's DES "do everything stupid" Just kidding!!!!!

that's great , it's energy saving , if your overclocking it may conflict.
Have had too much fun doing other things to try it

[Raybo58]

Hello. I'm back!

Patience being absent from my virtues, I sprung for an E8600. I just hope I can get close to what other people are getting to make it worth the meals I'll be skipping.

From what I've seen, to get 4.5ghz or better, people have been pumping scary amounts of juice into them. I see others saying that it's overkill, if you know how to tickle the sub-voltages.

But so far none of the people who have claimed to get 4.5 have provided numbers.

Still being in a 65nm frame of mind I went in agressively. I was able to get a working deskop at 4.75mhz, but she failed the stress tests instantly.

So I figured 4.5ghz must be near the sweet spot. But so far no luck. I tried 4.3, 4.25 and 4.0 and I'm getting the same problem with all of them. I can run the Small FFT test in Prime95 for hours but it instantly reports hardware errors whenever I try the Blended test. It won't pass a SuperPI run either.

So I started reading again and came back to one of the first articles I read, which I had ignored the first time...

http://www.xbitlabs.com/articles/cpu...o-e8600_3.html

It says that the E8600 is sensitive to much smaller changes in voltage. After a few more experiments I can now confirm that's true. But in this article they were able to get a stable 4.5ghz without increasing anything but Vcore from their defaults.

Apparently my Mainboard or CPU is saddly lacking because it certainly hasn't been that easy for me. I've been at it 3 days now and I just keep getting the same results.

In the 4th pic on the page they show getting 1.488v in Windows after putting 1.5v in BIOS for Vcore. When I put 1.5v into mine, a reading from Windows reports 1.440v. It appears what I'm missing in the way of power I'm getting back in heat, because at that level my load temps are in the 70's C. I've got pretty beefy cooling, but maybe I'm expecting too much.

I'm starting to wonder if I don't have weak power from the source. I'm in an old building. Or perhaps my cooling is drawing too much Juice. I've read from some people who own Coolit Systems units that they recommend putting them on a seperate power supply because they're so hungry. But my 750w PSU should be overkill for my rig. I don't know. I'm in over my head with anything more complicated than E over I times R in the electricty department.

Or maybe this is what I get for cheaping out and getting a "Moderate" OC mobo. Although some have said the only difference between mine and the Extreme versions of the MB are the cooling pipes. Which are vastly inferior to cooling solutions I have employed.

Anyay, the numbers....

GIGABYTE GA-EP45_DS3R
[u]MB Intelligent Tweaker(M.I.T.)
Robust Graphics Booster .............: Fast
CPU Clock Ratio .....................: 8.5
CPU Frequency .......................: 4.3 ghz
Clock Chip Control 
Standard Clock Control
CPU Host Clock Control...............: Enabled
CPU Host Frequency (Mhz) ............: 500Mhz
PCI Express Frequency (Mhz) .........: 100Mhz
C.I.A. 2.............................: Disabled
Advanced Clock Control [Press Enter]
CPU Clock Drive...............800....: 800mv
PCI Express Clock Drive.......900....: 900mv
CPU Clock Skew (ps)...........0......: 0ps
MCH Clock Skew (ps)...........0......: 0ps
DRAM Performance Control
Performance Enhance..................: Standard
Extreme Memory Profile (X.M.P.)......: Disabled
(G)MCH Frequency Latch...............: 333mhz
System Memory Multiplier ............: 2.40b
Memory Frequency (Mhz) ..............: 1200Mhz
DRAM Timing Selectable ..............: Manual
Standard Timing Control
CAS Latency Time..............5......: 5
tRCD .........................7......: 5
tRP'..........................7......: 5
tRAS..........................20.....: 14
Advanced Timing Control [Press Enter]
tRRD..........................4......: Auto
tWTR..........................4......: Auto
tWR...........................8......: 9
tRFC..........................4......: Auto
tRTP..........................68.....: 77
Command Rate (CMD) ...........0......: 2T

Channel A

Channel A Timings Settings [Press Enter] 
Static tRead Value............7......: 8
tRD Phase0 Adjustmen..........0......: Auto
tRD Phase1 Adjustment.........0......: Auto
tRD Phase2 Adjustment ........0......: Auto
tRD Phase3 Adjustment.........0......: Auto

Trd2rd(Different Rank)........8......: 8 
Twr2wr(Different Rank)........8......: 8
Twr2rd(Different Rank)........7......: 7 
Trd2wr(Same/Diff Rank)........9......: 9 

Dimm1 Clock Skew Control.............: Auto
Dimm2 Clock Skew Control.............: Auto
DDR3 Write Training .................: Enabled
Channel A Driving Settings [Press Enter] 
Driving Strength Profile:............: OC-1200
Data Driving Pull-Up Level...........: Auto 
Cmd Driving Pull-Up Level............: Auto 
Ctrl Driving Pull-Up Level...........: Auto 
Clk Driving Pull-Up Level............: Auto 

Data Driving Pull-Down Level.........: Auto 
Cmd Driving Pull-Down Level..........: Auto
Ctrl Driving Pull-Down Level.........: Auto 
Clk Driving Pull-Down Level..........: Auto 

Channel B

Channel B Timings Settings [Press Enter] 
Static tRead Value............7......: 9
tRD Phase0 Adjustment.........0......: 1-Advance
tRD Phase1 Adjustment.........0......: Auto
tRD Phase2 Adjustment ........0......: 1-Advance
tRD Phase3 Adjustment.........0......: 1-Advance

Trd2rd(Different Rank)........6......: 8 
Twr2wr(Different Rank)........6......: 8
Twr2rd(Different Rank)........5......: 7 
Trd2wr(Same/Diff Rank)........8......: 9 

Dimm1 Clock Skew Control.............: Auto
Dimm2 Clock Skew Control.............: Auto
DDR3 Write Training .................: Enabled

Channel B Driving Settings [Press Enter] 
Driving Strength Profile:............: OC-1200

Data Driving Pull-Up Level...........: Auto 
Cmd Driving Pull-Up Level............: Auto 
Ctrl Driving Pull-Up Level...........: Auto 
Clk Driving Pull-Up Level............: Auto 

Data Driving Pull-Down Level.........: Auto 
Cmd Driving Pull-Down Level..........: Auto 
Ctrl Driving Pull-Down Level.........: Auto 
Clk Driving Pull-Down Level..........: Auto 

{U]Motherboard Voltage Control{/U]
Voltage Type........ Normal 
CPU
CPU Vcore...................1.325v...: 1.48750v
CPU Termination.............1.200V...: 1.300v
CPU PLL.....................1.500V...: 1.500v
CPU Reference...............0.76V....: 0.823v
MCH/ICH
MCH Core....................1.100V...: 1.460v
MCH Reference...............0.76V....: 0.823v
MCH/DRAM Ref................0.900V...: 0.970v
ICH I/O.....................1.500V...: 1.500v
DRAM
DRAM Voltage................1.800V...: 2.3200v
DRAM Termination............0.900V...: 1.0100
Channel A Reference.........0.900V...: 1.0100
Channel B Reference.........0.900V...: 1.0100
[b]{U]Advanced Settings[b]{U]
Limit CPUID Max. to 3................: Disabled
No-Execute Memory Protect............: Disabled
CPU Enhanced Halt (C1E)..............: Disabled
CPU Thermal Monitor(TM) .............: Disabled
CPU EIST Function....................: Disabled

Above are the higher voltages that produce the results I descrbied.
Below are the low end.

CPU Vcore...................1.325v...: 1.4375v
CPU Termination.............1.200V...: 1.260v
CPU PLL.....................1.500V...: 1.500v
CPU Reference...............0.76V....: 0.798v
MCH/ICH
MCH Core....................1.100V...: 1.360v
MCH Reference...............0.76V....: 0.823v
MCH/DRAM Ref................0.900V...: 0.910v
ICH I/O.....................1.500V...: 1.500v

I've tried just about everything in between. I've gone as high as 1.7v on the core voltages, but of course temps become a problem.

There aslo seems to be a strident debate over raising the PLL. Many say that increasing it at all on the 45nm's is very hazardous. Others laugh while they put as puch as 1.8v into it (with LN cooling of course). I'm tending to leave it alone until I can be convinced one way or the other.

I did leave the PLL on AUTO once with a high Vcore and after about an hour of testing I looked at the EasyTune tuner and it showed the PLL as being 1.85v!!! After that I noticed that my idle temps on the 2 cores were radically different. Core 1 reads 12c and Core 2 reads 25c.

I thought maybe I'd done some damage but then I read that Intel admits this is a problem. The temps even out as soon as it's under load. I can't remember for sure, but I think they were even with each other when I was first playing with it.

I should also mention that she's Memtest86 stable, so hopefully I can rule memory out as an influence.

Any ideas?

[supershanks]

I can run the Small FFT test in Prime95 for hours but it instantly reports hardware errors whenever

Cut you short there, if that's the case the implication is that your memory is the unstable part. Blend consists of only 2 tests small ffts - stress CPU & large which stress ram.

Try dropping your ram down to 2.00 ie 1000mhz and retest.
that

Static tRead Value............7......: 8

might ber the cause try it on 9 the same as channel b.