ASUS P5E64 WS Evolution - Tips, Tricks, Mods, BIOS, OCing & More...

[StimulatedBoy]

If you have the Load Line calibration @ Performance,this is the reason for the random bsods

The Load Line calibration is set to Normal as I have experienced random BSOD's setting it to Performance previously. Don't know why it is like this but if anyone can tell me, I would appreciate it. So I guess it's my RAM giving me random BSOD's with 0115...

Probably because the 0401 + 0501 BIOS include optimizations for high speed memory (that was written about the 0401 at least on Asus download). I see you are using OCZ 2x2GB 16000 memory how fast can you get them running stable on the Evolution? Last time I saw your settings you were running 418x11. Is that the case still? I had a hard time getting my 2x2GB Corsair DDR1800 memory to run stable at the rated speed. I'm afraid that my CPU is holding me back to go further than around 450x9, but my memory is not going much further than DDR1800.
Could you please post some settings if you are running your memory at DDR1800+ speed?

Higest I have come on Multiplier 11 is 448 FSB which gives me 4.928MHz (rated FSB 1792) and which I find pretty good given the high multiplier. On multiplier 12 I have managed 418 FSB which gives me 5.021MHz (rated FSB 1674). Guess I could manage higher FSB by lowering multiplier to 8 or 9 but I am mostly looking for highest clock (this is more effective in 3dMark series than in SuperPI), so I usually don't do that... As of BIOS 0201 it is not possible to set divider as in 0115, so that option is gone... Problem with these sticks (believe it is Samsung inside) is that they don't respond very well with higher voltage (recommended 1.9V), so it's kind of hard getting timings real tight...

Settings is more or less the same that I have posted before:

CPU Voltage 1.5400
CPU GTL Voltage Reference 0,62x
CPU PLL Voltage 1,6
FSB Termination Voltage 1,40
DRAM Voltage 1.90
NB Voltage 1,45
NB GTL Voltage Reference 0,67x
SB Voltage 1,2
Clock Over-Charging Voltage 0.90
-------------------------------------
Load-Line Calibration Normal
CPU Spread Spectrum Disabled
PCIE Spread Spectrum Disabled


Managed this yesterday and this is my first 5MHz ever: http://valid.x86-secret.com/show_oc.php?id=375121

The SuperPI 1 mill below is the highest I could get MHz wise but the 32 mill. is only an initial test and I now that I can go higher. Too loose RAM timings does not make these any good though but I'll work on them somewhat... ...when I have the time...

[jabski]

Managed this yesterday and this is my first 5MHz ever: http://valid.x86-secret.com/show_oc.php?id=375121

congrats on the 5Ghz

and thanks for the info

[camouflage]

I think it's not the board, more CPU or/ and RAM. With 0115/0401/0501 I can set LLC to performance without getting random reboots. But set to normal or auto the machine doesn't even boot...........

Another thing is if You go for hi FSB with a Wolfdale You have use 0115 to get all the dividers and to set the booster to lighter and try GTLNB 0.67 / GTLCPU 0.63.

Do not give too much PLL - I ran two chips (E8400 & E8500) on this board - both doin' fine with only 1.52 PLL.

Don't know what is best for Quads - but if You run Dual Core go with the 0115, best performance & stability.

[OverClocker_gr]

i was telling that from the beginning... 0115 is the best bios....

[Holmer]

# Rob Halford, thank you very much. I just thought that with memory rated at DDR2000 you aimed for high FSB OC. My sticks uses the Samsung K4B1G0446D chips (probably the same as yours). Very impressive OC especially with LLC on normal . When I set LLC: Normal and CPU voltage to 1.5 I get 1.408 in CPU-Z under load which is far from enough to run my QX9650 at 4 GHz+. I am not too keen on setting CPU voltage 1.55 in BIOS and LLC Normal to get enough juice when I'm on water and not SS as you are. I have also experienced random BSOD's but I can not tell for sure whether it was caused by LLC Performance or just bad settings for the memory (I am leaning towards the latter).

Could someone please tell me which setting in BIOS corresponds to tRD (Performance level)?

# Sav, I am eagerly awaiting you results and settings with your 2x2GB Corsair. I am still working on getting my 450x9 OC stable in Prime blend. It seems that I gain stability by increasing SB voltage to 1.05 and decreasing NB voltage to 1.53.

[camouflage]

i was telling that from the beginning... 0115 is the best bios....

Subscibed - but I wonder what the ASUS guys are doin' there - 3 later versions after that and no improvement, but broken dividers......

Someone should tell them to avoid alcohol & drugs when they're working...

[Kabauterman]

does anone knows a OVP Mod (Over Voltage Protection) for this board?

Yesterday we bench with LN2 and I can't go over 1,93 V...

second problem is that the board have very strange problems with GraphiX OC...
Every time when the system crashes because we overclock the GFX to much.
The system don't boot and give uns a beeeeeeep beep beep beep ...

after we plug out the GFX and wait some minutes
it runs...

we have this problems with all the 3 GFX Cards that we have testet
GX2
HD2900 PRO
8800 GTS 640

BIOS was 0401

[StimulatedBoy]

With 0115/0401/0501 I can set LLC to performance without getting random reboots. But set to normal or auto the machine doesn't even boot...........

Welcome to the wonderful world of... ...ehrmm... ...consistency? Mine is OK at Auto and Normal but BSOD's at Performance no matter what, sometimes it will not make it beyond the MS stripes rolling over the screen showing Vista loading... Hope that someone will give those guys over at ASUS an intro to the terms validity and realibility in research...

Don't know what is best for Quads either but my testing shows that CPU GTL Voltage Reference 0,62x and NB GTL Voltage Reference 0,67x doesn't give ME any problems. With regard to BIOS, I am pretty sure that my BSOD's with 0115 is related to RAM as these are just out (available here less than 2 weeks ago). According to Tony, 2*2 sets also put a lot of new issues to OCing with a Quad. Have adjusted my PLL to 1.52 and followed your recommendation on that without any hassel from my PC. Thx.

@Holmer, I have the same issue with CPU voltage as you as my 1,54 is real while set to 1,59something in BIOS (LLC set to Normal). So don't you worry about upping that CPU voltage a little bit more. My machine runs 24/7 with 1,5V set in BIOS (while CPU-Z shows 1.440-1.448-1.546 -- dependent on idle / load) on 4.6MHz and is benchable up to 4.7something at that voltage. Must admit that your voltage set to 1.5 in BIOS and CPU-z showing 1.4 is a large gap though compared to mine... 450 FSB should not require more than 1.45V or thereabout if I am not mistaken, so try to loose some more and see if that helps. At the same time Tony recommended NB voltage a little bit higher for Quads and 2*2 sticks, so I don't really know...

- Hell bent for leather

[camouflage]

Now it's gettin' better from day to day - here's 8 x 580, lowered the Vcore down to 1.424 and still running only 1.10 VTT:

[ZoLKoRn]

Now it's gettin' better from day to day - here's 8 x 580, lowered the Vcore down to 1.424 and still running only 1.10 VTT:

Nice to see your results is better day to day

Thanks for sharing it

[jabski]

I had some better results with bios 0115 but still nothing special.

Very nice there Camoflage. I am interested to see what happens when you up the mutli

[jabski]

does anone knows a OVP Mod (Over Voltage Protection) for this board?

[StimulatedBoy]

Came across this post over at the P5E3 Premium review and it provided me with some ideas for further Quad testing. Maybe something for you too Holmer...

Some random observations -

RAM

• In general, a delay skew was needed, with more delay needed on channel A than channel B. Between 400 and 420 FSB best stability comes from Normal skew on both A and B channels. 420-440 needs a 25ps delay on channel A. 440-470 needed a 50ps delay on channel A and a 25ps delay on channel B, my board won't successfully post at higher than 470 so I can't comment above.
• Memtest is useless for testing skew. As long as the RAM will boot, Memtest will run successfully. Best test appears to be a SuperPI run, the longer the better.
• Random reboots at idle is a symptom of bad skew.
• Performance level and clock twist is limited by both FSB and RAM. My RAM (the approved 2*1GB Corsair XMP1600 CAS-7) is happy to do PL-6 at up to 420FSB, PL-7 at up to 440 and PL-8 at up to 470. Clock twist is less easy to narrow down, I seem to be able to run various settings at various FSB and RAM settings. Not much help I realise.
• The P5E3 Prem in general seems happy to run 1T timings at all FSB settings.

Board

• vTT (called FSB Term) and vPLL are needed to push high FSB on a quad, with vTT having a greater effect on stability. Between 400 and 420FSB a vTT of 1.2 and a vPLL of 1.6 were needed. Between 420 and 440FSB a vTT of 1.26 and a vPLL of 1.66 were needed. Up to 470 FSB a vTT of 1.3 and a vPLL of 1.7 were needed. I did not feel comfortable running higher voltages than this on a 45nm chip, especially as Intel lists the max safe vTT as 1.35V and many users report killing 45nm chips with vPLL settings of 1.8V.
• Setting strap to "Auto" is useless, set it manually. The board does not seem to realise that at a higher FSB setting a higher strap is needed, and instead changes strap according to the RAM speed setting you've selected. 333 strap was stable up to about 440FSB, 400 was needed above 440. Be aware that RAM dividers are linked to strap settings so your RAM OCing may suffer somewhat.
• Board is quite heat sensitive. Aftermarket cooling or directed airflow is recommended - the X48 chipset has the highest rated TDP of all Intel chipsets at ~35W.
• NB voltage appeared to have little affect on results after 1.4V was reached running 2*1GB. This may change at 4*1GB or at FSB speeds above 470.

[Holmer]

Thanks a lot Rob Halford. I just found out how to change the tRD (performance level) - I wish I had not done that. Of course I could not resist to play with the settng and now I just blew my BIOS. During POST a message telling that a unknown CPU was installed a BIOS update was needed to unleach the full power and the it stalled, I could not reboot and (made the mistake) to turn off the power to restart. Now POST looks very wrong and it is veeeery slow. S**t. I'm afraid the crash free system can not take care of this (I put my CD in but nothing happens (I see horizontal white lines across the screen below the name and the speed of my CPU). What do I do. Will it help to clear CMOS??

It looks like this now. I see no happy ending of this:


UPDATE. OK I was lucky clearing CMOS did the trick I have contact with BIOS again.

[Kabauterman]

http://img385.imageshack.us/img385/8972/ovpevovs5.jpg

Thanks, but thats not a OVP (Over Voltage Protection) Mod
this is a OV (Over Voltage) Mod.

I have done this mod to get higher Voltages in BIOS
but this mod don't helps me with the Protection...

[StimulatedBoy]

Thanks a lot Rob Halford. I just found out how to change the tRD (performance level) - I wish I had not done that. Of course I could not resist to play with the settng and now I just blew my BIOS. During POST a message telling that a unknown CPU was installed a BIOS update was needed to unleach the full power and the it stalled, I could not reboot and (made the mistake) to turn off the power to restart. Now POST looks very wrong and it is veeeery slow. S**t. I'm afraid the crash free system can not take care of this (I put my CD in but nothing happens (I see horizontal white lines across the screen below the name and the speed of my CPU). What do I do. Will it help to clear CMOS??

It looks like this now. I see no happy ending of this:


UPDATE. OK I was lucky clearing CMOS did the trick I have contact with BIOS again.

Tony has entered the discussion over at the P5E3 Premium and has added the following comments that can be found here... http://www.xtremesystems.org/forums/...&postcount=358 Seems also that Samsung chips like advanced while Micron like delays. As we ever would have figured that out...

Good to hear that you're board is up and running again!

[camouflage]

That's important to all the ppl having random reboots at IDLE........

So maybe You don't have to blame the LLC but set some other skew values.......

[camouflage]

BTW - forgot to ask , what RAM slots are You using guys? 1+3 or 2+4..........

[StimulatedBoy]

BTW - forgot to ask , what RAM slots are You using guys? 1+3 or 2+4..........

2+4

[ToyTen]

1+3

[runmc]

2+4

[Holmer]

Tony has entered the discussion over at the P5E3 Premium and has added the following comments that can be found here... http://www.xtremesystems.org/forums/...&postcount=358 Seems also that Samsung chips like advanced while Micron like delays. As we ever would have figured that out...

Good to hear that you're board is up and running again!

Thanks a lot, I was very happy to enter my BIOS and boot to Windows again Very good find. I'm testing CLK skew A/B: Advance 50ps overnight Prime blend. Yesterday evening I started a Prime95 blend that did not fail untill I turned on the monitor this evening when I came home so I am making progress . The settings used were (lending your template Rob Halford ;-)

CPU Features
Ai Overclock Tuner Manual
CPU Ratio Setting: 9
FSB Strap to North Bridge: 333
FSB Frequency: 450
PCIE Frequency: 100
DRAM Frequency DDR:1800MHz
DRAM Command Rate: 1N
DRAM CMD Skew on Channel A: Auto
DRAM CMD Skew on Channel B: Auto
DRAM CLK Skew on Channel A: Auto
DRAM CLK Skew on Channel B: Auto

DRAM Timings
DRAM Timing Control: Manual
CAS# Latency: 8 DRAM Clocks
RAS# to CAS# Delay: 8 DRAM Clocks
RAS# PRE Time: 8 DRAM Clocks
RAS# ACT Time: 24 DRAM Clocks
RAS# to RAS# Delay: Auto
REF Cycle Time: Auto
WRITE Recovery Time: Auto
READ to PRE Time: Auto

READ to WRITE Delay(S/D): Auto
WRITE to READ Delay(S): Auto
WRITE to READ Delay(D): Auto
READ to READ Delay(S): Auto
READ to READ Delay(D): Auto
WRITE to WRITE Delay(S): Auto
WRITE to WRITE Delay(D): Auto

Write to PRE Delay: Auto
PRE to PRE Delay: Auto
ALL PRE to ACT Delay: Auto
ALL PRE to REF Delay: Auto
DRAM Static Read Control: Disabled
DRAM Dynamic Write Control: Disabled

Ai Clock Twister: Auto
Ai Transaction Booster: Auto

Voltage Settings
CPU Voltage: 1.48125
CPU GTL Voltage Reference: 0.63x
CPU PLL Voltage: 1.52
FSB Termination Voltage: 1.36
DRAM Voltage: 1.82
NB Voltage: 1.51
NB GTL Voltage Reference: 0.67
SB Voltage: 1.05
Clock Over-Charging Voltage: Auto

Load-Line Calibration: Performance
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled

Configure Advanced CPU Settings
CPU Ratio Setting: 9
C1E Support: Disabled
CPU TM Function: Disabled
Vanderpool Technology: Disabled
Max CPUID Value Limit: Disabled

[camouflage]

Now I tested slot 1+3 & 2+4 plus I changed the sticks within the slots....

For me there is no noticable differece........... - so I'll keep 1+3 cause I'm running the Corsair airflow triple fan in front of the sticks and I use SATA harddrives........ - so the space at that region is very less....

[camouflage]

I had some better results with bios 0115 but still nothing special.

Very nice there Camoflage. I am interested to see what happens when you up the mutli

Up the multi, hmmm..... - do You think about 8.5 ?

Let's start with 8.5 x 565:

[ToyTen]

Can't wait to get good memories

Here is what I do at the moment, no setfsb or memset used:
Vista32 Home Basic to give some change so it will be better under XP32 Pro...

BP & spi32m:
spi1m:
Occt priority Mix 1h:
3DM Vantage without OC on cards; DD 8.6beta CCC globaly on performances: