Corsair 5400UL & Asus P5WD2 DDR860 4-2-2-2

[The Stilt]

I got my Asus P5WD2 Premium board yesterday from one of our distributors. They had received their shipment from Asus just few hours earlier, so this is probably one of the firsts Glenwood boards seen in Finland.

Because my Corsair 5400UL sticks couldn´t meet their limits on any 925XE board, I was quite eager to find out how would they perform on 955X boards.

So I ripped the package open, and bolted my 660 chip and 5400UL sticks on.
I dropped the multiplier to 14x and yanked and yanked the fsb up with SetFSB program...

First time system failed at 14x308 clocks, when memory was operating at DDR820 speed. I was running them at 2.3V, which was the highest voltage that can be selected from the bios.

Naturally I wasn´t satisfied, so I made a DDR VDD mod for the board. Then I raised the voltage to 2.39V and booted up for the second run.

Now I was able to hit 322FSB. 322FSB with 3:4 memorydivider means 430FSB or DDR860 for the ram so here we go :D

:fact:
Corsair 5400UL @ DDR860 4-2-2-2 (http://www.elektroni.fi/Stilt/Corsair_5400UL/5400UL_DDR860.png) (Verified (http://valid.x86-secret.com/show_oc.php?id=17781))

P4 660 @ 4520MHz (1.35V, stock cooling)
Asus P5WD2 Premium (VDDR mod, active cooling on NB)
Corsair 5400UL 1GB Kit (4-2-2-2-5-Turbo-HyperPath3 enabled / 2.39V)
Matrox Millenium II PCI

Thanks for Franck Delattre for CPU-Z (http://www.cpuid.org/) version with P5WD2 support, and abo for SetFSB (http://www13.plala.or.jp/setfsb/) program.

I´m currently held back by cpu, since 4.5GHz is pretty much maximum with stock cooling ;)
I´m going to bolt on my Mach2 and see how these sticks really perform.

[Kunaak]

Get to the good stuff, and show us what them sticks can really do :)

[Twisted-net]

I want to see some sandra membench of that :)

[b|gf|sh]

Dude thats amazing like :cool:

Would love to see some Everest bandwidth screenies at those speeds :eek:

[Supertim0r]

beat this monster :slapass:
:woot:

[The Stilt]

Get to the good stuff, and show us what them sticks can really do :)

I will when someone sends me the bios that was used in the sample boards. It would allow use of 2:3 (1.5x) and 1:2 (2.0x) memory dividers :slobber:

[V4n{}u|sH]

WoW :slobber:

[LarsK]

Awesome stuff, just what I what to read about :clap:

Man now I really can't wait to get the P5WD2, I've got one waiting at the post-office but can't get it before tuesday because monday is a national holiday....

What rev. are your Corsair memory sticks? I've got at dual kit with the same rev. number as the ones Anandtech tested, ie. "XMS5400v1.2 0515017-2". Haven't tested them at all yet, just got them from Germany last week. Had to get the mainboard from Sweden as well.... :rolleyes:

Arrgh now I really can't wait to give the system a go in the Mach II GT...

Is the Vmod fairly easy on this board, and how much extra voltage can you potentially get from a Vmod, and what voltage is still "reasonably" safe for the memory? Hope you don't mind the questions, but that's what you get for running cutting edge hardware... :p:

[The Stilt]

Here are some results: :D



Corsair 5400UL @ DDR900 4-2-2-2 - SuperPI 1M (http://www.elektroni.fi/Stilt/Corsair_5400UL/SuperPI1M-DDR900.png) (Verified (http://valid.x86-secret.com/show_oc?id=17836))

Corsair 5400UL @ DDR906 4-2-2-2 - Everest Write (http://www.elektroni.fi/Stilt/Corsair_5400UL/Everest-WR-DDR906.png) (Verified (http://valid.x86-secret.com/show_oc?id=17827))

Corsair 5400UL @ DDR906 4-2-2-2 - Everest Read (http://www.elektroni.fi/Stilt/Corsair_5400UL/Everest-RD-DDR906.png) (Verified (http://valid.x86-secret.com/show_oc?id=17827))

Corsair 5400UL @ DDR906 4-2-2-2 - Everest Latency (http://www.elektroni.fi/Stilt/Corsair_5400UL/Everest-LC-DDR906DL.png) (Verified (http://valid.x86-secret.com/show_oc?id=17827))

Corsair 5400UL @ DDR906 4-2-2-2 - Sandra 2005 Buffered (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-BF-DDR906DL.png) (Verified (http://valid.x86-secret.com/show_oc?id=17827))

Corsair 5400UL @ DDR906 4-2-2-2 - Sandra 2005 Unbuffered (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-UBF-DDR906DL.png) (Verified (http://valid.x86-secret.com/show_oc?id=17827))

With correct operating system, I would have hit 5000 in Sandra unbuffered with same clocks :( Unfortunately there isn´t SetFSB or Clockgen program that works in this OS.

P4 660 @ 4760MHz (1.43V)
Asus P5WD2 Premium (VDDR mod, active cooling on NB)
Corsair 5400UL 1GB Kit (2.49V / 4-2-2-2-5-Turbo-HyperPath3 on)
Matrox Millenium II PCI
Mach2 R404

These sticks seem to love voltage, so for next run I will bump it up to 2.55+ :D

LarsK

LOT code of my sticks are 017-5 v1.2.

VDDR mod for this board couldn´t be any easier. Simply solder 10k vr between sixth leg of RT9214 chip and ground. 8k resistance will give you stock voltage, and 2.49V will be achieved with 4.4k resistance when 2.3V is selected from the bios.

I assume that ~2.6V is the maximum voltage that should be used with these Micron BT-37E "Texas" chips. Japanese guys have used 2.8V with Samsung CD5 chips for benching.

[Bennah]

This is overclocking at its best :D

Get the board only a short time after Asus start to send them out. Try it out on air. Needs more vdimm so you figure out a vdimm mod. Still not happy so the Mach II gets dusted off and bolted on, then the fun begins. Produce a bunch of benchmarks and verified screenies :D

Great work The Stilt, Im impressed with his Intel stuff :toast:

[saaya]

very nice man! :toast: congrats :D

front page worthy! :D

[Supertim0r]

Do they get hot ?
How is DDR2 compared to DDR in term of heat ?

[5-Clicks]

with lower volts = cooler generally...

nice results Stilt! :clap:

[LarsK]

Here are some results: :D
LarsK

LOT code of my sticks are 017-5 v1.2.

VDDR mod for this board couldn´t be any easier. Simply solder 10k vr between sixth leg of RT9214 chip and ground. 8k resistance will give you stock voltage, and 2.49V will be achieved with 4.4k resistance when 2.3V is selected from the bios.

I assume that ~2.6V is the maximum voltage that should be used with these Micron BT-37E "Texas" chips. Japanese guys have used 2.8V with Samsung CD5 chips for benching.

Excellent, thank you very much :toast: Oh, and congrats on the results as well, simply awesome

[elec999]

You have an amazing system.
Very nice job with the ram
Thanks

[Xassius]

ddr2 = ultra cool compared to ddr.

[robberbaron]

Just imagine what kind of latency that would be getting on an AMD system! DDR2 is pretty nice.

[Astral7]

:eek2: Very nice The Stilt! Looks like DDR2 has finally arrived. Those timings at those speeds are amazing.

:toast:

[IluvIntel]

I got the same sticks as The stilt, but are limited by the 925X chipset and P5AD2 - MoBo.
I'm curious as to how they handle sustained bandwidth transfers at those 900 + speeds and with those timings, I mean its one thing to boot into windows and run Everest and Sandra. But can they handle the "Acid" test, eg: running 2 x distributed computing projects and net surfing and emailing at the same time with 100% cpu load without crashing or lockups? This is doing it for several hours or more.

But its really good to see someone doing great stuff anyway with this combo of MoBo and Corsair. ;)

[FX Freak]

nice overclock, ddr2 forever :D :)

[blackjok3r]

Give us SuperPi 32m please. Stability will be proven then.

[kromosto]

unbuff scores are like buff scores of a 478 on 865 or 875 without pat :toast:

[Cornelious0_0]

unbuff scores are like buff scores of a 478 on 865 or 875 without pat :toast:

I know hey? that's flippin' sick.....congrats Stilt, truely amazing. :toast:

[The Stilt]

DDR960 4-2-2-2 is history :D

1GHz speed wouldn´t have been a problem, if I could get over 1.52V vcore out from the board...

Anyways here we go again:

Corsair 5400UL @ DDR960 4-2-2-2 - Everest Read (http://www.elektroni.fi/Stilt/Corsair_5400UL/Everest-RD-DDR960DL.png) (Verified (http://valid.x86-secret.com/show_oc?id=18003))

Corsair 5400UL @ DDR960 4-2-2-2 - Everest Write (http://www.elektroni.fi/Stilt/Corsair_5400UL/Everest-WR-DDR960DL.png) (Verified (http://valid.x86-secret.com/show_oc?id=18003))

Corsair 5400UL @ DDR960 - Sandra 2005 Buffered (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-BF-DDR960DL.png) (Verified (http://valid.x86-secret.com/show_oc?id=18003))

Corsair 5400UL @ DDR960 4-2-2-2 - Sandra 2005 Unbuffered (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-UBF-DDR960DL.png) (Verified (http://valid.x86-secret.com/show_oc?id=18003))

System:

P4 660 @ 5.05GHz (14x360) 1.52V
Asus P5WD2 Premium (VDDR & Vnorth modded = 1.88V)
Corsair 5400UL 1GB Kit (2.65V VDD, 4-2-2-5-Turbo-HyperPath3 enabled)
Matrox Millenium II PCI
R404 Mach2 for cpu and Swifty MCX-159R for chipset.

Without Vnorth mod and Swifty on chipset the board was pretty much maxed out at 340FSB with 3:4 memory divider. After raising Vnorth voltage from 1.65V -> 1.88V and adding Swifty MCX-159R I was able to reach 360FSB.
Without vnorth mod and better chipset cooling I was limited to 343FSB no matter how much voltage I fed to those sticks.

With 2.65V VDD these sticks don´t even warm up that much, and they still seem to get advantage from higher voltage. Currently I´m limited by cpu, because I can´t get any higher vcore than 1.52V out from the board. If the bios wont solve the problem, atleast vcore mod does. With Abit AA8XE Fatal1ty this chip maxes out at 5.4GHz with same cooling ;)

1GHz DDR with 4-2-2-2 settings is pretty close now ;)

[jVIDIA]

:slobber:

It's so nice to see such a worderfull setup like yours, doing so good :toast:

I'm also waiting for a P5WD2 Premium and 1GB Corsair XMS2-5400UL :D

But I'm on air and still dont know wich cpu to buy : 630 or 640 ?! Any advice ?

I will when someone sends me the bios that was used in the sample boards. It would allow use of 2:3 (1.5x) and 1:2 (2.0x) memory dividers :slobber:

The 2:3 and 1:2 divider don't come in the retail bioses ? :mad:

What do you think of the new OCZ EB PC2-6400 and the OCZ EL PC2-8000 !?? What's the use of the 8000's ?


edit:

In the Anandtech review of the Asus P5WD2 the used the Corsair 5400UL and the say :

We were amazed to find that the Corsair CM2X512A-5400UL could reach DDR2-1066. This yielded a Sandra standard bandwidth of just over 6400 MB/sec. To put this in perspective, this is the first motherboard and memory that we have ever tested capable of running at DDR2-1066. Frankly, 1066 required slower 5-5-5-15 timings at the maximum 2.3V available on the Asus. We also needed to push all chipset and bus voltages to the maximum and set CPU voltage at 1.60V to achieve this milestone.

Could it be the mem's were not common retail ones and where choosed ? :slapass:

[macci]

over 5000MB/s unbuffered is pretty damn insane :toast:

Great overclocking there Tarmo :D

[b|gf|sh]

Dude that ram is nuts! :eek:

So whats tricks you gonna try next to get to 1GHz DDR?

(Also you still using the corsair heatspreaders?)

[The Stilt]

Dude that ram is nuts! :O

So whats tricks you gonna try next to get to 1GHz DDR?

(Also you still using the corsair heatspreaders?)

Get more voltage for cpu. That should be it ;)

It´s almost impossible to remove heatspreaders from this kind of BGA ram.
There is a bigger chance to kill the ram than not to kill it :)

The memory is not the problem, cpu is. Thats why there is no SuperPI at this time.

[Bennah]

Insane to say the least... great work once again :toast:

[The Stilt]

Insane to say the least... great work once again :toast:

Thanks guys :toast:

[jVIDIA]

At 2.3v and using the most relaxed timmings the your ram can hit 1000Mhz ? Like in the Anandtech P5WD2 review ?

[Byron]

The Stilt... this is awesome man... any pics of the setup?
1) are those XMS2 or TWIN2X?
2) can you also run a memtest86 and tell us the benches of L1,L2 and Memory please!!!

[The Stilt]

The Stilt... this is awesome man... any pics of the setup?
1) are those XMS2 or TWIN2X?
2) can you also run a memtest86 and tell us the benches of L1,L2 and Memory please!!!

This is a TwinX kit.
Running memtest won´t help that much since it will be corrupted after ~ 4.3GHz.

[LarsK]

Damn, those results are plain crazy... :slobber:

Thank you for making my last day of wait simply intolerable...... :D

[Byron]

Thanks,
Why memtest86 will be corrupted :confused:

[The Stilt]

Thanks,
Why memtest86 will be corrupted :confused:

32-bit limitation kicks in when cpu frequency passes ~ 4.3GHz.

[Sampsa]

TARMO, I'll send you the Asus P5WD2 Premium sample BIOS later today :cool:

[Rabbi_NZ]

The Stilt,
amazing results... looks like DDR2 is now at the stage it IS better than DDR... at least in your hands it is.

Congrats!

[FireDragon]

MAn you need an ES chip...contact Fugger I know he has one...so you can run 1:1 with a lower multi...that would be CRAZY...GREAT JOB THERE!!!

Dragon

[mike]

Awesome results indeed! Having the same rig - I'll have fun trying to chase your scores...DId you ever get that bios?

[nCrusader]

How high can they go with CAS3?

[cupra]

that is very nice man!

great clocks you have there, congrats :toast:

[Soltani]

Hi

Yes this is a pretty nice board!

I broke 1120MHz with my Centon mem - http://valid.x86-secret.com/show_oc.php?id=16918

When I try using the Setfsb, that you link to, I just get an error?

I have a couple of Bios for the board, and 1 of them supporting 1066Mem freq, let me know if You want one.

[LarsK]

Hi

Yes this is a pretty nice board!

I broke 1120MHz with my Centon mem - http://valid.x86-secret.com/show_oc.php?id=16918

When I try using the Setfsb, that you link to, I just get an error?

I have a couple of Bios for the board, and 1 of them supporting 1066Mem freq, let me know if You want one.

I might just want one later. Maybe I'll catch you on tweak.dk later tonight. So far I'm installing and updating windows.... Just did a quick test to make sure new board, memory and psu was working properly, so I very conservatively ran memtest and installed @ 250 Mhz fsb and DDR675 @ cas 3-2-2-8 and 2.1V in the bios. No errors an the memory isn't even warm at all. Nice. Memtest said 3558 MB/s I think it was, this is with 640 @ 4 Ghz and -29°C in bios.... :D

[The Stilt]

Hi

Yes this is a pretty nice board!

I broke 1120MHz with my Centon mem - http://valid.x86-secret.com/show_oc.php?id=16918

When I try using the Setfsb, that you link to, I just get an error?

I have a couple of Bios for the board, and 1 of them supporting 1066Mem freq, let me know if You want one.

Pm sent :)

[LarsK]

The Stilt > Did you get the Vcore issue sorted? I can't seem to get the Vcore to go over ~1.496 in the bios, regardless of setting from around 1.45V and up it just stays the same....

Thus I'm stranded @ 290 Mhz fsb and 4647 Mhz right now. Could probably go a bit higher as this is completely stable, memtest, prime and everything.

Memory running @ 2.3V in bios @ 4:5 divider 725 Mhz cas 4-2-2-8, memtest86 ver. 3.2 shows 3886 MB/s, no 32 bit bug in this version as far as I can tell. Would be awesome to see what you could score with that insane mem speed.... :slobber: :D

[The Stilt]

Many thanks for the bios Mr. Soltani :toast:

With little eased timings and with silly low voltage (2.38V) I managed to do this (http://valid.x86-secret.com/show_oc.php?id=18325) :D

That was just an appertizer. More to come soon ;)

[b|gf|sh]

Many thanks for the bios Mr. Soltani :toast:

With little eased timings and with silly low voltage (2.38V) I managed to do this (http://valid.x86-secret.com/show_oc.php?id=18325) :D

That was just an appertizer. More to come soon ;)

:eek:

I know what I'll buying when I get payed then

Any plans to try out the new asus intel nf4 board (or any other new intel boards for that matter)?

pld dude :cool:

[Soltani]

Nice clock!

Hehe You are getting close to my 1. place - http://valid.x86-secret.com/records.php

Do You know why I get this error when trying using the setfsb?

[The Stilt]

Nice clock!

Hehe You are getting close to my 1. place - http://valid.x86-secret.com/records.php

Do You know why I get this error when trying using the setfsb?

I´m also getting the same error.
The error has something to do with the different bioses.
Abo designed the program for boards that use retail bios. It think these unofficial bioses change something in the pll interface.

[The Stilt]

:D

Corsair 5400UL @ DDR1134 4-3-3-1 - SuperPI 1M (http://www.elektroni.fi/Stilt/Corsair_5400UL/SuperPI1M-DDR1134.png)

And that is not even the best one ;)

This time I don´t have to offer a verification url, because I screwed up a bit :slapass: When I was submitting the highest result, it accidentally ticked the publish box and the result was not published. Because results cannot be removed or submitted twice, I have to wait untill CPU-Z author removes or makes the verification url public.

I still got the verification files, so do not worry ;)

P4 660 @ 4830MHz (14x345)
Asus P5WD2 Premium (1.85V Vnorth VDD, 0208 bios, thanks to Soltani)
Corsair 5400UL 1GB TwinX-kit, 2.57V 3:5 DDR 4-3-3-1-4-Turbo
Matrox Millenium II PCI
Mach2 R404, Swifty MCX-159R

The new bios works great, and even the vcore adjustments work now..
Unfortunately SetFSB program doesn´t work, so I´m still very cpu and performance limited (no tweaks, due SetFSB) :hrhr:

Tarmo will return.

[Byron]

excuse me, but i'm noob with DDR2, can someone explain something, does this Asus mobo have divider for running the memory faster than its actual speed? (3:5 , 3:4 dividers..) cause i've never seen at any other motherboard a divider to run the mem faster at least in DDR1 mobos... so, thought might that be a function that only DDR2 mobos have ? is that correct?

Cheers,
B

[nCrusader]

3:5 is the ratio between FSB and MEM freq. FSB:MEM
3:5 gives 500MHz on memory if your FSB is 300MHz.

[Byron]

yes i know that, i just didn't knew that DDR2 mobos have this function..., so far with DDR1 you could use divider to run the memory bellow the CPU FSB.

[The Stilt]

excuse me, but i'm noob with DDR2, can someone explain something, does this Asus mobo have divider for running the memory faster than its actual speed? (3:5 , 3:4 dividers..) cause i've never seen at any other motherboard a divider to run the mem faster at least in DDR1 mobos... so, thought might that be a function that only DDR2 mobos have ? is that correct?

Cheers,
B

Thats not a DDR2 feature only, even some old 845PE boards like Abit IT72 had it (IIRC). Canterwood and Springdale (8x5P series) can also do the trick. Motherboards capable to it are Asus P4P800-E, Abit IC7, IS7 and AI7 and also Abit AS8 and Asus P5P800 board.

[mike]

Awesome score! Keep on pushing!

:D

Corsair 5400UL @ DDR1134 4-3-3-1 - SuperPI 1M (http://www.akiba-pc.com/Corsair_5400UL/SuperPI1M-DDR1134.png)

And that is not even the best one ;)

This time I don´t have to offer a verification url, because I screwed up a bit :slapass: When I was submitting the highest result, it accidentally ticked the publish box and the result was not published. Because results cannot be removed or submitted twice, I have to wait untill CPU-Z author removes or makes the verification url public.

I still got the verification files, so do not worry ;)

P4 660 @ 4830MHz (14x345)
Asus P5WD2 Premium (1.85V Vnorth VDD, 0208 bios, thanks to Soltani)
Corsair 5400UL 1GB TwinX-kit, 2.57V 3:5 DDR 4-3-3-1-4-Turbo
Matrox Millenium II PCI
Mach2 R404, Swifty MCX-159R

The new bios works great, and even the vcore adjustments work now..
Unfortunately SetFSB program doesn´t work, so I´m still very cpu and performance limited (no tweaks, due SetFSB) :hrhr:

Tarmo will return.

[Byron]

Thats not a DDR2 feature only, even some old 845PE boards like Abit IT72 had it (IIRC). Canterwood and Springdale (8x5P series) can also do the trick. Motherboards capable to it are Asus P4P800-E, Abit IC7, IS7 and AI7 and also Abit AS8 and Asus P5P800 board.
Cool, thanks for clarifying this ;)

[Sentential]

The Stilt how in the world are you getting CPUz to read your RAM timings correctally? :confused: On my gigabyte they are purely blank :(

[The Stilt]

The Stilt how in the world are you getting CPUz to read your RAM timings correctally? :confused: On my gigabyte they are purely blank :(

Timings should be displayed correctly with last version of CPU-Z (1.28.6).
To make ram speed and ratio visible, I had to get "special" (1.28.7) version from CPU-Z Franck.

[LarsK]

I seem to be hitting a brick wall @ 290 Mhz fsb, cpu is stable well above this, memory also. What settings do you use for pci-e to run above ~300 Mhz fsb, should extra NB cooling be neccessary from 290 and up? Any clues would be more than welcome. Rather stuck @ 290 Mhz fsb and DDR725 atm....

[Sentential]

Timings should be displayed correctly with last version of CPU-Z (1.28.6).
To make ram speed and ratio visible, I had to get "special" (1.28.7) version from CPU-Z Franck.
Could you PM me his email? I really need him to take a look at the giga. I need to get this working ASAP!

[The Stilt]

I seem to be hitting a brick wall @ 290 Mhz fsb, cpu is stable well above this, memory also. What settings do you use for pci-e to run above ~300 Mhz fsb, should extra NB cooling be neccessary from 290 and up? Any clues would be more than welcome. Rather stuck @ 290 Mhz fsb and DDR725 atm....

I set PCI-E to 101MHz from bios.
I used enhanced NB cooling right from the start.
I used the stock NB heatsink with 40mm fan on it.

With that cooling it ran at 340FSB with 3:4 divider.

[Sentential]

I seem to be hitting a brick wall @ 290 Mhz fsb, cpu is stable well above this, memory also. What settings do you use for pci-e to run above ~300 Mhz fsb, should extra NB cooling be neccessary from 290 and up? Any clues would be more than welcome. Rather stuck @ 290 Mhz fsb and DDR725 atm....
I think your issue is Asus related. I am on the stock cooler, running passive, no fans no nothing on it and mine goes past 300 effortlessly. On the gigabyte there is a max of .35v. With .25 I have managed to get as high as 310 100% stable on the chipset.

The only thing I have done to the chipset was pull the shim and add AS5 to the NB

[LarsK]

Thanks guys, I'll try the 101 Mhz thing and then look at the NB cooling solution. I've been through several settings this morning and nothing seemed to matter, it just won't even post beyond 290/291. Really odd though as any benchmark is completely stable @ 290.......

Vcore is also funny. Anything from 1.450V-1.16125V gives me ~1.49V actual Vcore, and 1.625-1.700V gives me an actual Vcore of 1.58V.

AI booster shows I'm only getting 2.12V for the Vddr, I'll probably mod the thing over the weekend.

[The Stilt]

I´m going to receive my MSI P4N board tomorrow, so we will see soon if these sticks will do any better on NF4 IE platform s:D

[Sentential]

I´m going to receive my MSI P4N board tomorrow, so we will see soon if these sticks will do any better on NF4 IE platform s:D
Dont hold your breath. Ive got a friend who sez that it doesnt have a PCI lock :(

[LarsK]

Nice The Stilt, looking forward to that.

Well, I found a setfsb program that allowed up to 307.5 Mhz fsb, and booting at 290 I can get to the max 307.5 and at least run superpi 1M without isues. Stock cooling on NB, no mods on mainboard yet either. Getting a score of 26.297s @ ~4920 Mhz on my 640. Highest I tested 3dmark was @ 300 Mhz fsb (4800 Mhz) and it ran without issues. So I'm sure it's just a bios bug or something.

Screenshot of superpi mod 1.4 uploaded here:http://www.hardware-test.dk/oc/billedbrowser.asp?id=7913&pic=10

I'm not near your results The Stilt, but it's pretty decent for a non modded board and a 640 cpu IMO. I bet I can get the sucker into the 5 Ghz club with a better bios, or another setfsb version :)

[Sentential]

Honest to god guys. I dont see why everyone is buying the P5WD2. The gigabyte blows it away. I dont need to use clockgen at all past 300....yet im the only one useing this board? :stick:

[Mesce]

I got a hell of a lot of problems with my msi p4n diamond. I cant put my finger on it. Wont even run 1:1 at stock.

Not tryin to hijack thread :)

[Sentential]

The Stilts, just out of curosity what volts are you using for DDR800 with 4-3-3? 2v or so?

[The Stilt]

The Stilts, just out of curosity what volts are you using for DDR800 with 4-3-3? 2v or so?

Never tried so low speeds with such settings.
These sticks run fine at DDR800 with 2.2V and 3-2-2-1 settings.

[Sentential]

Never tried so low speeds with such settings.
These sticks run fine at DDR800 with 2.2V and 3-2-2-1 settings.
Fantastic, thankyou much :toast:

[The Stilt]

Honest to god guys. I dont see why everyone is buying the P5WD2. The gigabyte blows it away. I dont need to use clockgen at all past 300....yet im the only one useing this board? :stick:

That wouldn´t be the exact term that I would use to describe that Gigabyte board ;)

I didn´t use clockgen for that 345FSB run with 3:5 divider ;)

[Sentential]

That wouldn´t be the exact term that I would use to describe that Gigabyte board ;)

I didn´t use clockgen for that 345FSB run with 4:5 divider ;)
Ah nvm then :banana: I was under the impression that past 280 they started to get awefully shaky in BIOS. My mistake :toast:

[abo333]

Nice The Stilt, looking forward to that.

Well, I found a setfsb program that allowed up to 307.5 Mhz fsb, and booting at 290 I can get to the max 307.5 and at least run superpi 1M without isues. Stock cooling on NB, no mods on mainboard yet either. Getting a score of 26.297s @ ~4920 Mhz on my 640. Highest I tested 3dmark was @ 300 Mhz fsb (4800 Mhz) and it ran without issues. So I'm sure it's just a bios bug or something.

Screenshot of superpi mod 1.4 uploaded here:http://www.hardware-test.dk/oc/billedbrowser.asp?id=7913&pic=10

I'm not near your results The Stilt, but it's pretty decent for a non modded board and a 640 cpu IMO. I bet I can get the sucker into the 5 Ghz club with a better bios, or another setfsb version :)
Please check "Ultra" before clicking "Get FSB".
You will be able to try 5Ghz or more.

[LarsK]

Please check "Ultra" before clicking "Get FSB".
You will be able to try 5Ghz or more.

Thank you. I figured it out a few minutes after I posted :D I was able to pass the 5 Ghz mark indeed. However I need a bit more Vcore to get the cpu stable past ~4.95 Ghz.

Did you do the programming on this? If so nice work, it works like a charm, however I'm getting som odd sisandra results. They don't increase at all, but 3dmark and superpi increases almost linearly with speed.

[Apoptosis]

Hey guy's this is Apop over at Legit Reviews... Been testing on the same exact setup and found some weird things going on with the 1:2 or 1066Mhz memory divider. It doesn't seem to be working correctly. I have a thread going on here: http://forums.legitreviews.com/about1931.html that goes into further detail on the issue. I'm looking forward to what you guys have seen and can if you can confirm what I have been seeing during testing.

Awesome overclocking with the 3:5 divider... some major bandwidth! :toast:

[The Stilt]

Hey guy's this is Apop over at Legit Reviews... Been testing on the same exact setup and found some weird things going on with the 1:2 or 1066Mhz memory divider. It doesn't seem to be working correctly. I have a thread going on here: http://forums.legitreviews.com/about1931.html that goes into further detail on the issue. I'm looking forward to what you guys have seen and can if you can confirm what I have been seeing during testing.

Awesome overclocking with the 3:5 divider... some major bandwidth! :toast:

Hi Apoptosis,

Unfortunately I have to disagree with you :D
Atleast with retail boards it works perfectly.

Sandra Unbuffered with 3:5 divider (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-UBF-3;5.PNG)

Sandra Unbuffered with 1:2 divider (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-UBF-1;2.PNG)

Asus P5WD2 Premium "54MG0" with 0408 bios.

So it is too early to say if those CPU-Z H.O.F scores are invalid.

[Apoptosis]

Well I can't run 4-4-4 timings at 1066 with my memory and unmodded board so I ran 265FSB @ 1:2 and 3:5 memory ratios with Sandra 2005 SR1 on buffered memory testing and got the following results.

Results with the 1:2 divider (http://forums.legitreviews.com/images/random/asusp5wd2/12divider.jpg) (6700mb/sec)

Results with the 3:5 divider (http://forums.legitreviews.com/images/random/asusp5wd2/35divider.jpg) (6600mb/sec)

I'm also on a retail ASUS P5WD2 Premium Motherboard with the 0408 BIOS.

What are your thoughts? I only ran the tests once, so these are well within the margin of error to be identical results.

[gandalf027]

Can someone please send me the cpuz 1.28.7 please? :)

Thanks

[The Stilt]

Well I can't run 4-4-4 timings at 1066 with my memory and unmodded board so I ran 265FSB @ 1:2 and 3:5 memory ratios with Sandra 2005 SR1 on buffered memory testing and got the following results.

Results with the 1:2 divider (http://forums.legitreviews.com/images/random/asusp5wd2/12divider.jpg) (6700mb/sec)

Results with the 3:5 divider (http://forums.legitreviews.com/images/random/asusp5wd2/35divider.jpg) (6600mb/sec)

I'm also on a retail ASUS P5WD2 Premium Motherboard with the 0408 BIOS.

What are your thoughts? I only ran the tests once, so these are well within the margin of error to be identical results.

Can you run the unbuffered test please?
400MB per sec difference in unbuffered certainly doesn´t fit in the error margin.

[macci]

Been testing on the same exact setup and found some weird things going on with the 1:2 or 1066Mhz memory divider. It doesn't seem to be working correctly.
Like The Stilt already mentioned you really should try sandra unbuffered and/or some benchmarks that actually measure something that has to do with actual performance. Sandra Buffered is one of the worst benchmarks there is.
Its well known that Sandra buffered is very dependant on system FSB on P4 rigs and thats the reason you dont see much of a difference between different dividers - Your FSB is the same!
Try unbuffered or superpi or 3dmark2001 lobby hi and you should see a difference between the dividers.
The real world performance gain might not be too big thou as the system FSB might be the limiting factor there.

[Apoptosis]

Let me run a few more CPU's & benchmarks and see what happens... either way it's a good problem solver...

well i ran some more benchmarks and it looks like the difference between buffered and read scores are pretty much nill between 900 and 1066 memory clocks. It also looks like the dividers are working right and that everest is showing the wrong values. Sad that the only time you'll see the improvment on these overclocks is on certain benchmarks or when testing unbuffered and write tests.

This corsair 5400UL is some pretty amazing stuff. I'm happy at 5-3-3-8 timings at 1051MHz as it's getting Sandra buffered scores of 8,000MB/s and unbuffered scores of 5,000MB/s

Stilt you got some crazy fast Corsair modules on your hands! Congrats and keep those results coming! :woot:

[Byron]

The Stilt, i've looked my bios on my IC7 but there is no option for running the mem higher than the cpu... is there any special option m8?

Cheers,
B

[The Stilt]

Anyone tried Memtest at 320FSB with 3:5 ratio (1.06GHz 4-3-3-4 for ram)?
Memtest v1.55 reports following:

L1 Cache: 31375MB/s
L2 Cache: 24383MB/s
Memory: 5007MB/s :ROTF:

EDIT:

330FSB with 3:5 (1.1GHz 4-3-3-4 ram)

L1 Cache: 32353MB/s
L2 Cache: 25144MB/s
Memory: 5163MB/s

...

[The Stilt]

The Stilt, i've looked my bios on my IC7 but there is no option for running the mem higher than the cpu... is there any special option m8?

Cheers,
B

Select 533PSB NB Strap and it will appear.
Thou there is no way that unmodded IC7 would handle it.

[LarsK]

Damn, that's like A LOT af bandwidth there.... :p:

[The Stilt]

:D

Corsair 5400UL @ DDR1142 4-3-3-1-2 - SuperPI 1M (http://www.elektroni.fi/Stilt/Corsair_5400UL/SuperPI1M-DDR1142.png) (Verified (http://valid.x86-secret.com/show_oc.php?id=18984))

Apoptosis:
I think that 5086MB/s unbuffered score in your article is a bit weird.
Did you deselect all eight boxes in the option menu of Sandra?

I can hit only ~ 4900MB/s with identical setup. I have higher clocks and tighter latencies than you did (320FSB, DDR1070 4-3-3-1).

[The Stilt]

Damn, this DDR2 "war" with Amin Soltani is getting tight :D
Currently we are neck by neck in frequency, but I can´t do any better from this: Corsair 5400UL @ DDR1170 5-3-3-1-2 (http://valid.x86-secret.com/show_oc.php?id=19074)

Getting that far took so high VDD, that I don´t even dare to say it loud ;)

Maybe it is time to move for Corsair 6400 stuff, and see how it does.
DDR1300 - DDR1400 would be quite nice even with CL5 latencies :D

[Byron]

Anyone tried Memtest at 320FSB with 3:5 ratio (1.06GHz 4-3-3-4 for ram)?
Memtest v1.55 reports following:

L1 Cache: 31375MB/s
L2 Cache: 24383MB/s
Memory: 5007MB/s :ROTF:

EDIT:

330FSB with 3:5 (1.1GHz 4-3-3-4 ram)

L1 Cache: 32353MB/s
L2 Cache: 25144MB/s
Memory: 5163MB/s

...
Holly JESUS! 5K memory... drool... that is some serious data transfer m8 :D
Me with 875 chipset and turbo at 261fsb's is 3.8K exactly with DDR1 though.

[Soltani]

Damn, this DDR2 "war" with Amin Soltani is getting tight :D
Currently we are neck by neck in frequency, but I can´t do any better from this: Corsair 5400UL @ DDR1170 5-3-3-1-2 (http://valid.x86-secret.com/show_oc.php?id=19074)

Getting that far took so high VDD, that I don´t even dare to say it loud ;)

Maybe it is time to move for Corsair 6400 stuff, and see how it does.
DDR1300 - DDR1400 would be quite nice even with CL5 latencies :D

Nice overclock, with nice timings! :toast:

I see You got the 1. place again!

I won't try to clock my Centon any further. Ill wait untill I receive my new PC5300 modules from Centon, hope they can do better, with better timings also.

[death metal]

Heya stilt, I've been watching this thread so closely =)...any chance you can post the result for the MSI P4N as well?

I am a bit worried about a post I read before that C19 doesn't fully support dual core though (the system will run fine, just that only one core will be recognized, or something like that).

And btw, nice OC. If you can have the chance to test the 955x vs C19, do advise which one to get =). I am leaning over to C19 but Mesce's experience somewhat dishearten me but I'd really like SLI. And last two questions, what is the highest vcore, vchipset, vdimm on unmodded board does that mobo is capable of? does it need vdroop mod too?

[The Stilt]

Heya stilt, I've been watching this thread so closely =)...any chance you can post the result for the MSI P4N as well?

I am a bit worried about a post I read before that C19 doesn't fully support dual core though (the system will run fine, just that only one core will be recognized, or something like that).

And btw, nice OC. If you can have the chance to test the 955x vs C19, do advise which one to get =). I am leaning over to C19 but Mesce's experience somewhat dishearten me but I'd really like SLI. And last two questions, what is the highest vcore, vchipset, vdimm on unmodded board does that mobo is capable of? does it need vdroop mod too?

I just cancelled my MSI P4N board and ordered P5ND2 SLI-DLX instead ;)
Too bad that Asus still use 3 stage cpu power cirquit in P5ND2 :slapass:

No way it will handle dual core if it´s having hard time with single 5GHz Prescott...

I´ll try to arrange something little, concerning that NF4 IE vs. Glenwood thing ;)

EDIT:

Maximum voltages for P4N in v1.0 bios:

Vcore: VID - 1.55V
VDDR: 1.8V - 2.4V
NB Voltage: 1.xx - 2.xx (?)

[death metal]

Thanks for the quick reply =)....I guess I'm going for the Asus 955x then =)...and I hope you don't mind my queries again: can you post the stock vCore, vAGP, vDIMM for the Asus i955 mobo? Do I have to vdroop mod it to stabilize the vcore or can it hold its own on a simple/basic Swiftech Rev3a cooler kit without any droop mod?

[SteveS]

Maybe it is time to move for Corsair 6400 stuff, and see how it does.


Have you tried the Corsair 6400 RAM? I have ordered a Gb of it to go in my P5WD2 board and was wondering how it went and the best timings for it.

[The Stilt]

Thanks for the quick reply =)....I guess I'm going for the Asus 955x then =)...and I hope you don't mind my queries again: can you post the stock vCore, vAGP, vDIMM for the Asus i955 mobo? Do I have to vdroop mod it to stabilize the vcore or can it hold its own on a simple/basic Swiftech Rev3a cooler kit without any droop mod?

P5ND2:

Vcore: 0.9375V - 1.700V (+ 0.01V overvoltage option available)
DDR VDD: 1.8V - 2.3V
Chipset Core Voltage: 1.4V - 1.5V
South bridge SATA Voltage: 1.5V - 1.6V
CPU Termination Voltage: 1.25V - 1.4V

For basic use, there is no need for Vdroop mod because Asus has used different regulator chip than in their previous boards. P5WD2 board uses Analog Devices 3181 regulator chip instead of older 3180 chip which have almost removed the droop "problem". I think that P5ND2 uses the same chip as P5WD2.

[death metal]

Wow, thanks for the info The_Stilt =)....I didn't know that revision on the vregulator chip can do such enough wonder, at least for non extreme OCing =)...I'm a bit worried about the Chipset Core Voltage though, 1.5v seems a bit low. I am used to pushing about 1.65 to 1.8v during bench run for older chipsets, even with a 1066Mhz 925xe (I use about 1.7v for DDR750+).

I guess vNB modding is called for this time, I am targetting 285FSB to 295FSB with NB1C cooling on the chipset. I haven't used the CPU termination voltage though (I haven't used any P5AD2-E Premium or Fatal1ty board yet). I wonder why it's such a low setting though, vCore, and vDD both seems "within" my expectations.

And hey, am looking forward for that Asus C19 bench run soon :)....(please, don't confuse me which should I get this time: Asus i955 or Asus C19 *lol*, if you came up with great score on C19, I'll be resetting my mindset again hahaha).

[Mesce]

Im watching this thread carefully as well. I have p5nd2-sli and have tested 2 different 5400ul's and the same thing happens. I cannot manage 2:3 or even 1:1 at 4ghz on my 660. The timings of 3-2-2-8 even dont work. I will watch to see how the stilt manages to fix this problem bceause I am all alone in searching for people with p5nd2's

[LarsK]

The 0408 bios really did solve a LOT of problems for me. I couldn't get the memory to post at anything above DDR730 before, or the motherboard above 290 Mhz fsb. Now I'm running 300 Mhz fsb and DDR1000 stably.... :D Only at 4-4-4-12 timings but still, no mods and default chipset voltages

[Tony]

I tested the NF4 asus a while back and its not to good, i also hit 5007MB/s in memtest with the new 800EB which uses the same IC's as the corsair.

DDR2 is actually getting quite nice now, the only issue is the cpu's it uses are poor.

[jVIDIA]

where do you guys get the P5WD2 Premium bios you all are talking about ?!

[Mesce]

I tested the NF4 asus a while back and its not to good, i also hit 5007MB/s in memtest with the new 800EB which uses the same IC's as the corsair.

DDR2 is actually getting quite nice now, the only issue is the cpu's it uses are poor.


Im currently waiting for a bios update but there doesnt seem to be one anywhere. Im glad it isnt just me with this problem and it seems like a NF4 problem in general because I tested the msi p4n and it was worse than this board. At least I manage 1:1 with this board at stock.

Do you have any bioses that I could use for it?

How far did u manage to push the memory?

[LarsK]

where do you guys get the P5WD2 Premium bios you all are talking about ?!

There's a link in the Intel section. Look here http://www.xtremesystems.org/forums/showthread.php?t=63655

[Mesce]

The stilt have you got hold of a p5nd2-sli yet? Im anticipating the arrival because Im sure it has a bad bios causing not low latencies and not 2:3 dividers.

[The Stilt]

It´s getting weird with these Corsair 5400UL sticks...
In legitreviews article (http://www.legitreviews.com/article.php?aid=204&pid=2) there are pictures that show V1.1 Corsair 5400UL sticks. According the article those sticks contained chip labeled as D9CWX which points to part number (MT47H64M8BT-3). After removing the spreader from my V1.2 sticks I found chips labeled as D9DQW which means MT47H64M8BT-37E stepping A chips, which are 3.7ns stuff instead of 3ns...

[The Stilt]

I´m getting there ;)

After little enhancements, I was able to achieve higher overclock with less voltage. At first I removed heatspreader from the sticks, and added a more powerfull fan to delived higher air bypass for the sticks.

Then I soldered one Rubycon 470µF / 6.3V cap to empty solder spot under NB.
This fixed the shutdown problem (system shuts down when it´s supposed to restart).

With 0.05V lower voltage than in previous run I got this (http://valid.x86-secret.com/show_oc.php?id=20117) :D

Going for more.

[Byron]

neat!
where exactly did you solder the cab btw ? pic?

[The Stilt]

neat!
where exactly did you solder the cab btw ? pic?

http://www.akiba-pc.com/Corsair_5400UL/P5WD2_CAPADD.jpg

It´s for the chipset vdd.

[LarsK]

That's awesome news. Guess I'll have to solder one of those cab's on as well, as I'm affraid the shut down for a few seconds and then restart is eventually going to kill my mach II GT cooler.... what bios version do you currently use? Still the 0408? I heard news of the 0413, that's why I ask....

Also, can you actually lock the multiplier at 14x in the bios or are you using a software programme for that? Because I can't do it in bios with my 6640, but I found a programme that will do it in windows. However the setfsb version I used with the 0124 bios won't work with the 0408 bios on this board....

[The Stilt]

That's awesome news. Guess I'll have to solder one of those cab's on as well, as the shut down for a few seconds and then restart is evetually going to kill my mach II cooler.... what bios versione do you currently use? Still the 0408? I heard news of the 0413, that's why I ask....

Also, can you actually lock the multiplier at 14x in the bios or are you using a software programme for that? Because I can't do it in bios with my 6640, but I found a programme that will do it in windows. However the setfsb version I used with the 0124 bios won't work with the 0408 bios on this board....

I use 0408 bios because it´s the newest one.
I use 660 chip which is PRB 04B chip. With 04B chips (TG >= 115W) multiplier can be locked to 14 via Lock Free function. With PRB 04A chips like your 640 multiplier can only be changed via speedstep.

I have tried to help Abo to make a working SetFSB version, but Asus has changed something in SMBus, so after updating the bios from 0124 version SetFSB will stop working.

[LarsK]

I use 0408 bios because it´s the newest one.
I use 660 chip which is PRB 04B chip. With 04B chips (TG >= 115W) multiplier can be locked to 14 via Lock Free function. With PRB 04A chips like your 640 multiplier can only be changed via speedstep.

Well I only asked about the 0413 bios because it's mentioned here (http://www.asus.com.tw/news_show.aspx?id=968) at the bottom of the page.

So if I order a 660 I will be able to set the multiplier @ 14x in bios then I assume. Great I need to get the 640 back in the Shuttle anaway... :p:

I have tried to help Abo to make a working SetFSB version, but Asus has changed something in SMBus, so after updating the bios from 0124 version SetFSB will stop working.

Yes, unfortunately I get an SM Bus I/O error with bios 0408. Tahnks for answering my questions :toast:

[death metal]

What was the specific problem with shutdown? You mean when you chose "Windows shutdown", it will restart instead?

Also, is that new OC on water or just plain air =)...nice tip on adding cap btw...

I´m getting there ;)

After little enhancements, I was able to achieve higher overclock with less voltage. At first I removed heatspreader from the sticks, and added a more powerfull fan to delived higher air bypass for the sticks.

Then I soldered one Rubycon 470µF / 6.3V cap to empty solder spot under NB.
This fixed the shutdown problem (system shuts down when it´s supposed to restart).

With 0.05V lower voltage than in previous run I got this (http://valid.x86-secret.com/show_oc.php?id=20117) :D

Going for more.

[The Stilt]

What was the specific problem with shutdown? You mean when you chose "Windows shutdown", it will restart instead?

Also, is that new OC on water or just plain air =)...nice tip on adding cap btw...

No, the board will shutdown itself when it is supposed to restart.
That´s because the MCH voltage in unstable under load. Additional capasitor will fix the issue.

Cooled with stock heatsink up to 4.55GHz :)

[death metal]

No, the board will shutdown itself when it is supposed to restart.
That´s because the MCH voltage in unstable under load. Additional capasitor will fix the issue.

Cooled with stock heatsink up to 4.55GHz :)

I'm getting excited with this board =)....I'm guessing there is a trace already from the vacant capacitor slot going to the NB.

And woah, that stock heatsink can boot the CPU@4.55? You must have really low ambient temp :p: :p: :p:

[The Stilt]

And woah, that stock heatsink can boot the CPU@4.55? You must have really low ambient temp :p: :p: :p:

Not really, thermometer shows currently 25c in my workshop.
For 4.55GHz I have to set vcore to 1.425V any lower or higher will make system even more unstable that it already is at 4.55GHz :)

[Merkor]

Not really, thermometer shows currently 25c in my workshop.
For 4.55GHz I have to set vcore to 1.425V any lower or higher will make system even more unstable that it already is at 4.55GHz :)

So what are your temps then with the boxed cooler (CPU and MB/idle and load)?

No, the board will shutdown itself when it is supposed to restart.
That´s because the MCH voltage in unstable under load. Additional capasitor will fix the issue.

It is possible to increase the MCH voltage in the BIOS. Does it perhaps solve the problem by just increasing this?

[The Stilt]

Corsair 5400UL - Memtest86+ v1.55 Bandwidth (http://www.elektroni.fi/Stilt/Corsair_5400UL/5400UL_MemtestBW.jpg)

These modules really seem to like of burn-in.
At the start, that kind of speeds required over 2.6V VDD, but now I need only 2.4V to achieve same speeds and settings.

Memtest test #5 seem to be the hardest for overclocked DDR2 ram.
Still I can run it at DDR 1.1GHz speed with 4-3-3-2 settings without a hitch :D

When I have more time I´ll bolt the Mach2 on again, and run some more memtest. Currently I´m cpu limited since I some crappy Artic Cooling heatsink (which is actually worse than the original).

[The Stilt]

It is possible to increase the MCH voltage in the BIOS. Does it perhaps solve the problem by just increasing this?

Really?
Is 2.1V MCH VDD enough?
So far it hasn´t helped ;)
:D

[jVIDIA]

So much fuzz about this mem's on the web and Corsair doesn't have any 5400UL's in stock in Europe !!!! BAHHHH :slapass:

Where did you get yours ?

[SteveS]

Just a question.
I have the same Mboard, diffrent ram though.
But I cant seem to go any higher than the pic below
What do I need to do to get better OC?

Should I consider getting water cooling? I am currently on the stock air as I just could not get the XP-120 working right.
Any advice would be appreciated.

http://hem.bredband.net/stesmi/955.jpg

[The Stilt]

Gotta to love them :D

Corsair 5400UL - Sandra 2005 Unbuffered (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-UBF-DDR960WR.jpg)

Corsair 5400UL - Sandra 2005 Buffered (http://www.elektroni.fi/Stilt/Corsair_5400UL/Sandra-BF-DDR960WR.jpg)

:slapass:

[Mathias-K]

Hehe, Stilt, i used to think BH-5 @ 250MHz 2-2-x was fast.. Now you're running 480MHz 2-2-x :o

Damn, thing are getting xtreme these days :)

[V I P E R]

@ The Stilt - can you run 2 video cards in SLI on this board?
Have you tested more 955 boards? Is the ASUS suitable for not so extreme clocks and what vmods the mobo requires?

[The Stilt]

@ The Stilt - can you run 2 video cards in SLI on this board?
Have you tested more 955 boards? Is the ASUS suitable for not so extreme clocks and what vmods the mobo requires?

I have two Gainward 6800 Ultra cards available, but I don´t have any interests in nVidia currently. Waiting somethin better ;)

P5WD2 is the only 955X board I have currently tested, but I have pretty much used every single Socket 775 board available from Abit, Asus and DFI.
Asus P5WD2 might be the best Socket 775 board ever built.

Asus fits perfectly for daily or extreme use.
The modifications are very simple just like the modifications for their previous boards have been.

[V I P E R]

@ The Stilt - Thanks for the answers man. You rock ... damn nice results. I'll build the same system :). Can you only test is SLI working on that board even in 4x mode for the second card?
Currently I have ASUS P4C800 E Deluxe + P4 3G @ 4G default voltage with silent air cooling. Mobo is vdroop, vdimm modded so I think that I will not have any problems mod the new ASUS model :).

Please advice what DDR2 memory is best to buy. What chips I have to look for - I now Micron, but what revision, stepping or what to search.?
I seen these two models and they seem very good.
Personally I prefer this one because of its price: http://www.newegg.com/Product/Product.asp?Item=N82E16820220056
What you think?
And the next one is like yours:
http://www.newegg.com/Product/Product.asp?Item=N82E16820145564
But it is 45$ more expensive.
What should I buy

[Sentential]

Now that Im on the P5WD2 things are going very well but Im still having issues. Are you guy's FSB fluctuationg at all? Mine still changed by almost 10mhz on the FSB still.

[FUGGER]

I will pop some heat spreaders and let you know what D9's I have.

Thanks for the mod info, it will surely help.

[LarsK]

I just put in a 660 in stead of the 640, and with the 14x multiplier so far I'm running @ 345 Mhz fsb without any problems. My fsb does fluctuate a little bit, but it never goes below 345 Mhz and it only increases 3 Mhz max. Also it doesn't fluctuate under all benchmarks, just some.

I have had the system running at 350 Mhz fsb also, but my memory was more than maxxed out when I did, so I did get some odd results. However I'm still on the standard passive northbridge cooler and I have not done any mods yet, so I'm quite happy. My Corsair PC5400UL does seem to max out at around 4500 MB/s in memtest86, pretty much regaerdless of memory/ fsb Mhz and timings, it doesn't want to go above that. I'll hopefully get the mod done pretty soon. I'm also doing the cap mod (thanks The Stilt for that also). Gonna put some decent cooling on the Northbridge as well I guess :D

[pcdoc1]

Damn, guess my 650 needs to go.......... Hello Ebay!

[The Stilt]

I wrote a little P5WD2 mod guide for Akiba-PC.
Read it here (http://www.akiba-pc.com/article.php?46.0) .

[death metal]

Wow, thanks Stilt =), but where is the "capacitor" addition modification there? hehehe...you missed it ;)... :stick:

[The Stilt]

Wow, thanks Stilt =), but where is the "capacitor" addition modification there? hehehe...you missed it ;)... :stick:

Soon :)

[V I P E R]

The Stilt - can you answer to my question above please! You can send to me a PM if you don't to fill your thread with useless posts.

[The Stilt]

The Stilt - can you answer to my question above please! You can send to me a PM if you don't to fill your thread with useless posts.

I don´t have enough time to test those cards in SLI, sorry :(
There will be so much better stuff available in the markets soon, so I have no interests in any nVidia products anymore.

There is atleast two different Micron D9 stepping that overclock best.
One is MT47H64M8BT-3 and one is MT47H64M8BT-37E.

If you want to get some really good stuff, get Corsair 5400UL ;)

[Sentential]

Are you getting FSB fluctuations? Also could you link me to the BIOS you are currentally using?

[The Stilt]

Are you getting FSB fluctuations? Also could you link me to the BIOS you are currentally using?

I haven´t noticed any FSB fluctuation in "normal" use.
When the motherboard is pushed to it´s limits and FSB passes 360FSB barrier it fluctuates about 6MHz under load. Little higher FSB Termination voltage will fix that thou ;)

[webwilli]

I just put in a 660 in stead of the 640, and with the 14x multiplier so far I'm running @ 345 Mhz fsb without any problems. My fsb does fluctuate a little bit, but it never goes below 345 Mhz and it only increases 3 Mhz max. Also it doesn't fluctuate under all benchmarks, just some.


which tool you are using for read out the fsb?? cpu-z?

cpuclock show me a different fsb and the fsb fluctuate heavy. is cpuclock to old for this new board or show cpu-z the wrong? i have 7200 mb @ sandra mem benchmark with DDR733 and 4-4-4-6. this can´t be right.

[LarsK]

which tool you are using for read out the fsb?? cpu-z?

cpuclock show me a different fsb and the fsb fluctuate heavy. is cpuclock to old for this new board or show cpu-z the wrong? i have 7200 mb @ sandra mem benchmark with DDR733 and 4-4-4-6. this can´t be right.

Yes, that's one of the programs I use to see if the fsb is fluctuating. However at 360 Mhz fsb and beyond, even 3dmark says results cannot be published because of cpu speed variation. Also SiSandra and other benchmarks shows really borked results. I know the cpu is stable at this spedd as I've run benches at almost 200 Mhz more at the defaul 18x multiplier

[V I P E R]

@ The Stilt - If I fill all 4 slots with memory what will happen, because on the 875 mobos if you remember the PAT was turning OFF, and I want to know what is the situation with this board.

[Sentential]

Do you have a link to the BIOS you are using? We all really need a copy of it. My vcore settings do not work at all on this set. PLZ link us.

[jVIDIA]

Can someone tell where are the critical PLL and mosfet chips that need extra cooling ? Like the FSB PLL generator ?

Please put a pic of the P5WD2 with some kind of indication where are those critical chips.

It's because I would like to put some passive cooling sinks on those critical hot spots !

thanks.

[pcdoc1]

Sentential: I'm using 0408 and it works pretty well.....

You can find it here. (ftp://ftp.downloadbase.de/p5wd2_0408.zip) If there's a newer one out there, I'm not aware of it..........

[Sentential]

Thanks

[FUGGER]

New versions 1.29 of CPUID works good for P5WD2. Correct memory speed and core voltage.

[LarsK]

New versions 1.29 of CPUID works good for P5WD2. Correct memory speed and core voltage.

Yes it works fine here as well. It doesn't seem to show any fsb fluctuation under load anymore either.

[nutcase]

Can you do a small experiment for me?

Just trying to figure out why my Bandwidth efficiency is so much higher than others..

I only see 3 things different than yours:

1. I am running with HT enabled (But I already tried this. made no difference)

2. I am running 1:1 on the Ram while yours are at Higher Mem Multi's (my Mem won't run at this FSB with 3:5 or higher)

3. I am running 0124 Bios (will load 0408 later)..

I am starting to suspect Asus's Hyperpath only runs at a 1:1 ratio.

[LarsK]

Loose the 0124 bios it's bad. The motherboard is doing so much better with the 0408 bios IMHO.

No problem here running the 3:5 divider, even without mods I've had it stable @ 300 Mhz fsb and DDR1000 @ cas 4-3-3-8. However with the 660 cpu @ 14x multiplier, the memory gets taxed much harder pr. clock, but still it runs fine at 4:5 till above 350 Mhz fsb. I can do 3:4 at relaxed timings, but 3:5 is currently a no-go @ 350 Mhz fsb regardless of timings. I'm hoping a volt mod will help solve this.

I have both Turbo and Hyperpath 3, as well at Hyper Threading enabled all the time.

[nutcase]

Loose the 0124 bios it's bad. The motherboard is doing so much better with the 0408 bios IMHO.

No problem here running the 3:5 divider, even without mods I've had it stable @ 300 Mhz fsb and DDR1000 @ cas 4-3-3-8. However with the 660 cpu @ 14x multiplier, the memory gets taxed much harder pr. clock, but still it runs fine at 4:5 till above 350 Mhz fsb. I can do 3:4 at relaxed timings, but 3:5 is currently a no-go @ 350 Mhz fsb regardless of timings. I'm hoping a volt mod will help solve this.

I have both Turbo and Hyperpath 3, as well at Hyper Threading enabled all the time.

This is why I asked.. :eek:

Trying to figure out why I am getting a consistent 95-98% efficiency on a P4.. I will be switching to 0408 after I finish playing with 0124..

[The Stilt]

This is why I asked.. :eek:

Trying to figure out why I am getting a consistent 95-98% efficiency on a P4.. I will be switching to 0408 after I finish playing with 0124..

You boot up with 200FSB, right?
Thats it. Again there is ol' good Asus NB Strap trick which forces tightest possible settings on when the system is booted up with 200FSB.

Try to boot up with 266FSB (1066FSB NB Strap) and see how much the results will be decreased ;)

[nutcase]

You boot up with 200FSB, right?
Thats it. Again there is ol' good Asus NB Strap trick which forces tightest possible settings on when the system is booted up with 200FSB.

Try to boot up with 266FSB (1066FSB NB Strap) and see how much the results will be decreased ;)

I am booting up at 340 FSB on the CPU FSB and 1:1 on the Ram..

booted up at 336 FSB and 4:5 ram divider But had to drop to 4-4-4-12 timings to make into windows. But Sandra is still showing 90%+ on the ram. This is the Max the ram will do as it is only PC4200 Ram.

BTW: Nice work on the WR..

Let you know.. I am not using SETFSB right now..

[V I P E R]

I have both Turbo and Hyperpath 3, as well at Hyper Threading enabled all the time.
Can you use 4 sticks of ram and see is there any MEM bandwith decrease?

[jVIDIA]

What software do you guys use to read the P5WD2 temps ?

I'm so sad that MBM has been stoped ....

[webwilli]

i´m use

MBM
Systool
Asus PCProbeII

[jVIDIA]

MBM suports the P5WD2 ?!?

Which sensors do you use ?

[webwilli]

MBM suports the P5WD2 ?!?

Which sensors do you use ?

no special :p:

i have use config wizard.

see my mbm.ini attach.
rename in .ini
copy to DATA dir
if necessary i can upload the hole DATA dir

careful. its german ;)

[gloatlizard]

the stilt,plz tell me what is the max 1m pi running of your memories on 3224,thanks;)

[gandalf027]

no special :p:

i have use config wizard.

see my mbm.ini attach.
rename in .ini
copy to DATA dir
if necessary i can upload the hole DATA dir

careful. its german ;)

Huu cool . .gona try it :)

edit: it works . thanks!!!

[jVIDIA]

Gandalf, apanhas-tes o config para o MBM primeiro que eu ! AI AI :slapass:

Funca bem ?

:D

[cpulloverclock]

I am booting up at 340 FSB on the CPU FSB and 1:1 on the Ram..

booted up at 336 FSB and 4:5 ram divider But had to drop to 4-4-4-12 timings to make into windows. But Sandra is still showing 90%+ on the ram. This is the Max the ram will do as it is only PC4200 Ram.

BTW: Nice work on the WR..

Let you know.. I am not using SETFSB right now..
I'm for a bug

I dont think that's possible to break more 80%+ on an Intel system ;) :)

[gloatlizard]

the stilt plz can you tell me what do you do with your very lucky memories with 3224 timings?i want to compare them to mine

http://img79.echo.cx/img79/390/test24vk.jpg

[Brave]

hehehe.....tive a ler o thread todo... Gandalf e Jvidia...essa cena bule bem?? Gandalf....já tens outro sistema?? lol..vendeste o k eu te vendi? :P

Abraço ..e contem novidades disso :)


Good work people...keep going :P

[jVIDIA]

What version of the CPU-Z do you folks use with tne P5WD2 ?

I tried the last one, 1.29, and it doesn't looks good even with no OC :

- the multiplier is always switching from 14x to 16x
- the CPU speed is always switching from 28xx to 32xx
- the memory fsb speed and the multiplier in the mem tab is wrong

Any ideas what's wrong ? Is this normal ???

[The Stilt]

What version of the CPU-Z do you folks use with tne P5WD2 ?

I tried the last one, 1.29, and it doesn't looks good even with no OC :

- the multiplier is always switching from 14x to 16x
- the CPU speed is always switching from 28xx to 32xx
- the memory fsb speed and the multiplier in the mem tab is wrong

Any ideas what's wrong ? Is this normal ???

You have C1E (Enhaced Halt State) enabled in bios.
CPU-Z version 1.29 shows all the information correctly.

[ztop]

You have C1E (Enhaced Halt State) enabled in bios.
CPU-Z version 1.29 shows all the information correctly.

i have a questin for you...
the p5wd2 premion...non wifi version...is 300 fsb stable from bios... not by set fsb without any kind of mod???i want to order it,,,, :)
thanks...

[The Stilt]

i have a questin for you...
the p5wd2 premion...non wifi version...is 300 fsb stable from bios... not by set fsb without any kind of mod???i want to order it,,,, :)
thanks...

Yep, it can boot up to 350FSB straight from the bios.
Thats pretty stable frequency.

[ztop]

Yep, it can boot up to 350FSB straight from the bios.
Thats pretty stable frequency.
thanks!

[whozyodaddy]

Why would you use SetFSB over your original BIOS to inrease FSB? SetFSB is an executable that you run in Windows, is that why you choose it?

Oh, and here is a newbie question, why do you drop the multiplier to 14x? Can you achieve higher FSB with a lower multiplier?

[jkabaseball]

I use SetFSB for on the fly OC.

[ztop]

Yep, it can boot up to 350FSB straight from the bios.
Thats pretty stable frequency.

with a pci vga and a ata hd??or with a pci\ex vga and a sata hd?
thanks

[The Stilt]

with a pci vga and a ata hd??or with a pci\ex vga and a sata hd?
thanks

With PCI vga and SATA hd.

[jVIDIA]

You have C1E (Enhaced Halt State) enabled in bios.
CPU-Z version 1.29 shows all the information correctly.

Thanks Stilt ... must see that :toast:

One more for you guys ... i've tried the 0408 and 0422 bios and they booth have the "shutdown & restart" feature when exiting from bios @ fsb >= 250 !(the original and the 0208 bios are fine !)

Is it normal ? Can it damage any component ?

Asus must do something about it ... :slapass:

Another problem is that with the original bios (0128) and with the CD bios (0208) I have Windows XP running fine with raid 0, but with 0408 and 0422 the Windows XP doesn't pass the blue bars zone ! It stays there forever :mad:

Is there any known problem between bios when using a raid 0 configuration ?
Must I reinstall XP again ?

It sucks to have some bios booting windows fine and the 04xx don't !

[Sentential]

Where is 0422 BIOS at? Where on earth are you finding these BIOS?

[jVIDIA]

Where is 0422 BIOS at? Where on earth are you finding these BIOS?

:fact: (http://www.stcom.co.kr/dataroom/dataroom_content.asp?no=417&part=)

Enjoy :D

[ztop]

With PCI vga and SATA hd.

and finally....wich bios do you have used for your insane scores??

[jVIDIA]

Folks, I've tried 0408 & 0422 bios without success using Raid0 !!!

Let me explain ...

With my P4 640 + P5WD2 Premium + 0124 bios (original one) / 0208 bios (in suport CD) + 2 x Hitachi 80GB SATAII the Raid0 works like a charm.

I flashed the 0408 and then the 0422 bios but windows XP won't boot with any of those bios !! It stayed forever stuck on the blue bars zone !

Then I tried to install again the XP having the 0422 bios already flashed, but during windows instalation after choosing the ICH7R / Raid option from the "raid floppy disk drivers" the windows setup finishes loading some other drivers and then says "Starting Windows ..." and after some 20 seconds I get a BSOD !!!!! :eek:

Al this @ default bios options (200 fsb).

Any idea what is going on ? :confused:

The 0408/0422 bios also as a bad feature .... the "shutdown & restart" after exiting bios and I have read somewhere that it also occours after a windows restart !

I'm also suspicious about this bioses, since they are not official ones ! :rolleyes:

[MDboyz]

I got my new P5WD2 premium board, and the bios is 0304 version.
Here is a link for anyone wants to try it.

0304 (http://home.comcast.net/~mdboy2/0304.rom)

[jVIDIA]

Hi MDboyz, great offer ;)

Does that bios have the 800 and 1066 mem options ?

Are you using Raid 0 ? And if so is all ok ? Because with me the 04xx bios give'd me BSOD during windows setup :mad:



:toast:

[Sentential]

:fact: (http://www.stcom.co.kr/dataroom/dataroom_content.asp?no=417&part=)

Enjoy :D
Thankyou much

[jVIDIA]

Good news from 0304 or not ?

Is Raid 0 ok with 0304 ?

EDIT:

Did you guys know that Corsair has changed the specs of the 5400UL from 3-2-2-8 to 3-3-2-8 ?!?! :mad:

Read it for your selfs :



Specification Change Notice
(http://www.houseofhelp.com/forums/showthread.php?p=198002#post198002)

Product : CM2X512A-5400UL and TWIN2X1024A-5400UL

Timing Specification:

Previous: 3-2-2-8

Revised to: 3-3-2-8

This change is effective immediately. Parts with the revised specification will have a specification number of XMS5400v1.5.

This change was made due to a subtle change in the performance characteristics of the RAMs used to build these modules. While the XMS5400UL modules typically continue to function at the 3-2-2-8 latency setting, they will no longer pass Corsair’s rigorous qualification with those settings, and may not demonstrate long-term stabilility.



After this I feel a little deceived by Corsair !! :slapass:


EDIT 2:

Well after reading this 'Specification Change Notice' with atention I became happy because I just realized that my 5400UL v1.3 were made with better chips than the current 5400UL's v1.5. !!! "....subtle change in the performance characteristics of the RAMs used to build these modules..."

:D :D :D

In the begining I tought that all versions of the 5400UL where revised to 3-3-2-8 ! ... but it's just the new ones that have chips not so good :banana:

[Dark79]

0304 gives me the same problem as 0204 - if I set FSB higher than 265, board won't POST. I just get a black screen =P No idea about the RAID thing though, sorry

[jkabaseball]

anyone get the 1:2 working yet?

[jVIDIA]

anyone get the 1:2 working yet?


Yep ... I'm with 1:2 @ 250fsb :YIPPIE:

http://jvidia.no.sapo.pt/Images/Benchmarks/cpu-z_memory.jpg

And rising ....

And what about this unbuffered Sandra scores :

http://jvidia.no.sapo.pt/Images/Benchmarks/sandra-mem_unbuff.jpg

Nice or what ?! :banana:

[Brave]

jvidia....what is your specs system ?

[ztop]

0304 gives me the same problem as 0204 - if I set FSB higher than 265, board won't POST. I just get a black screen =P No idea about the RAID thing though, sorry

i hawe the 0304
raid 0 on ich7 ok
an 300 fsb ok....

[jVIDIA]

jvidia....what is your specs system ?

See my sign please ... it's updated ;)

[sluflyer06]

well my board died last night.....i flashed 0422 to try it out and got.
DMA-1 Error
Fatal Error...System Halted

Completely hardlocked during post...no cure...could be related to the scrape in the PCB near mounting hole I did...but worked with other BIOSES...except 0408 didn't fix vcore and didn't cause shutdown-restart problem......BIG QUESTION is was it the BIOS? or the scrape that possibly cut between 0-3 traces but worked with 0304

[Mesce]

I manage to hit 3-2-2-4 at 320 mhz 1:1 but I cannot select cas 2. Is there a bios out that will allow cas 2?

[jVIDIA]

Holly god ! Dead mobo ! :eek:

If you turn the system on what happens ?

Tried CMOS reset with battery off ?

[sluflyer06]

when system is turned on....beeps twice, then reads CPUID and RAM and then
DMA-1 Error
Fatal Error...System Halted

you can't do anything crashfree doesn't load...no bios access i took battery out and left unlplugged for awhile...try no hardware cept VC and ram.

HOW IS 0422....does it fix Vcore adjustments not working? does it fix the overvolting at idle?.....these were I issues I had with 0304 for the ONE HOUR i had it.

[jVIDIA]

Did you take off the battery and short circuit with the CMOS jumper for at least 10 seconds ?

Haven't noticed nothing special .... can you explain those bugs a litle better ?

[sluflyer06]

well in the reatail 0304 bios....if you change the Vcore in the BIOS it doesn't do anything...it just stays the same...and it overvolts during idle...

with 0408 its suppoed to fix those problems.

does 0422 allow you to successfully change the Vcore and then when you read it with say CPU-Z it shows teh values you set.

[ztop]

well in the reatail 0304 bios....if you change the Vcore in the BIOS it doesn't do anything...it just stays the same...and it overvolts during idle...

with 0408 its suppoed to fix those problems.

does 0422 allow you to successfully change the Vcore and then when you read it with say CPU-Z it shows teh values you set.

mine with 0304 in idle overvolt....but i can change the v-core

[sluflyer06]

so when you change the Vcore it shows it properly under load? and also as you increase Vcore does the overvolt @idle increase accordingly?

i hope it was jsut the damage I did and not severe bios issues with the newer batched boards because....at the same Vcore/FSB settings I had 100% stable on my AS8 it wouldn't even superpi....i'm so frustated/angry and hoping it was cut traces because I can't keep switching mobos :::sigh:::

[ztop]

so when you change the Vcore it shows it properly under load? and also as you increase Vcore does the overvolt @idle increase accordingly?

i hope it was jsut the damage I did and not severe bios issues with the newer batched boards because....at the same Vcore/FSB settings I had 100% stable on my AS8 it wouldn't even superpi....i'm so frustated/angry and hoping it was cut traces because I can't keep switching mobos :::sigh:::

now i have set 1,45 v in bios....end bios accept the changement.. and from asus probe in idle i have 1,5v + o -....in full load i have 1.45-1.46 v stable

[sluflyer06]

alright so at least its responding to the changes....though the overolting would be terrible if you actually needed a high Vcore during load.

[LarsK]

Well guys, I haven't had the time to get the voltmod done just yet, but I should shortly as I just started my vacation (finally!).

In the meantime I've gotten myself another pair of TwinX 5400UL sticks, I now have 4x 512 MB in there. So far I've seen no decrease in performance, stability or even overclockability from going from 2x to 4x memorysticks. Quite good i think. 4 sticks (2GB) @ DDR2-954 cas 4-3-3-4-2 turbo and hyperpath3 enabled, 100% memtest stable. Pretty nice I think. Now if only Asus would fix the cold boot problem in the 0422 bios.... :slapass:

[sluflyer06]

how Cold is the Cold boot problem? (haha like I needed another worry)

and does it have the problem with 0408

[LarsK]

how Cold is the Cold boot problem? (haha like I needed another worry)

and does it have the problem with 0408

I don't think it's temperature related, just an issue where it hangs on the post screeen on a "cold" boot. No problems on reboots at all, just a problem when you turn the system on in the morning/afternoon whenever you get up ;)

And no, the 0408 didn't have this problem. I can reach the same oc speeds on cpu and mem on both bios'es, however some memory ratios do NOT work on the 0408 (system will hang and not even post), and those now work on the 0422. In fact they all do now. Ive tried them all.

[sluflyer06]

well thats very good news...and I"m assuming 0422 fixed all the Vcore problems (overvolting and non response to bios settings)

2. What do you do when it hangs at post after being off, just hit reset or CTL ALT DEL?

Haha I feel dumb for assuming it had a cold lock problem because it would have been the first I've heard about on Intel. doh.

[ztop]

alright so at least its responding to the changes....though the overolting would be terrible if you actually needed a high Vcore during load.

i'm 24h 4200 mhz vc def on my 560j....1.4v..1.42 full load..1.27 in idle
no problem at all

[jVIDIA]

Folks .... bad new on 0422. I've been playing wih the bios voltage and overvolting is still there !

The 'non response to bios settings' is gone .... is responds allrigth.

Larsk (and others) , in memtest86+ 1.60 I get no errors but with memtest 3.2 for windows I get errors !!! In which should I believe ?!?!? :confused:


And as for the voltage options in bios, what do you have on those :

FSB Termination Voltage
MCH Chipset Voltage
ICH Chipset Voltage

And which one should be changed from 'Auto' to get a stable OC ?

Do any of those help on getting better memory latencies ?

[IluvIntel]

Greetings All,

Whats the highest FSB you can get with this board using 14 x multi?
But with the system fully stable under full load. I presume that is with max FSB volts using 0422 bios.

Cheers :toast:

[sluflyer06]

according to the stilt I think with 0408 it'll do 350fsb without any vmods....if you search through the thread i believe it in here somewhere.

I've yet to see less than 300 stable except on retail bios (0304 and older)

[LarsK]

Greetings All,

Whats the highest FSB you can get with this board using 14 x multi?
But with the system fully stable under full load. I presume that is with max FSB volts using 0422 bios.

Cheers :toast:

Mine's stable up to around 345-350 Mhz fsb. It does even run everything just fine up to around 360 Mhz fsb, however results show some/a lot of speed variation. I know the cpu isn't the problem, as it will do around 200 Mhz more on the 18x multi.

Highest speed I ever achieved with default cooling and no mods was 370 Mhz fsb. Not particularly stable though.

[IluvIntel]

Mine's stable up to around 345-350 Mhz fsb. It does even run everything just fine up to around 360 Mhz fsb, however results show some/a lot of speed variation. I know the cpu isn't the problem, as it will do around 200 Mhz more on the 18x multi.

Highest speed I ever achieved with default cooling and no mods was 370 Mhz fsb. Not particularly stable though.

OK thanks LarsK,

Nice to see you, you've posted on OCAU forums in Australia, right?

I'm user RodneyJM on those forums. :)

In using the WD SATA II (300mb/s version) I get windows reporting errors around 288mhz FSB. I know its a bit risky using SATA HD for O/C, but thought I'd give it a try on this newer 955X chipset mobo. This is with FSB v on 1.5. PCIe v on 1.55. HD locked at PCIe 120 ;
Both chipsets on Auto voltage, no need for tweaking these I believe because of relatively low bandwidth RAM (I like using 1:1 :D ).

Are going to try ATA133 HD soon, see how high FSB can go - this is with prolonged CPU max load test, not just being able to boot into windows and run SuperPi with CPU-Z screenshot. :slap:

IMHO O/C is useless unless I can use it every day for the things I like to run without lockups or crashes. ;)

[LarsK]

OK thanks LarsK,

Nice to see you, you've posted on OCAU forums in Australia, right?

I'm user RodneyJM on those forums. :)

In using the WD SATA II (300mb/s version) I get windows reporting errors around 288mhz FSB. I know its a bit risky using SATA HD for O/C, but thought I'd give it a try on this newer 955X chipset mobo. This is with FSB v on 1.5. PCIe v on 1.55. HD locked at PCIe 120 ;
Both chipsets on Auto voltage, no need for tweaking these I believe because of relatively low bandwidth RAM (I like using 1:1 :D ).

Are going to try ATA133 HD soon, see how high FSB can go - this is with prolonged CPU max load test, not just being able to boot into windows and run SuperPi with CPU-Z screenshot. :slap:

IMHO O/C is useless unless I can use it every day for the things I like to run without lockups or crashes. ;)

Yeah It's me allright. I agree system should be 24/7 stable, otherwise it's not really useable. At 345 Mhz it actuallt is completely stable without any speed variation on the cpu either.

I actually run my OS on a 2x WD Raptor SATA Raid 0 array, also a couple of Raptors on the SIL SATA controller, an old Maxtor ATA drive and 2 dvd-rw's, so both ATA controllers are in use as well. I've got all like 5 USB 2.0 devices plugged in the mobo as well, 2 of them drawing power from the mobo, both network adapters are being used and I use the on-board sound card as well.

I've got all motherboard voltages maxxed out, 4x Corsair PC5400UL blocks mounted @ 2.3V so considering that everything on the mainboard is used 24/7, I think acheiving a 24/7 stable 345 Mhz fsb speed is pretty darn good for stock cooling too. I'm going to put water cooling on it here pretty soon, as well as volt mod it a bit, at least NB/mem term. and Vddr. I'm hoping to get 360-365 Mhz 24/7 stable without any cpu speed variation.

[IluvIntel]

Yeah It's me allright. I agree system should be 24/7 stable, otherwise it's not really useable. At 345 Mhz it actuallt is completely stable without any speed variation on the cpu either.

I actually run my OS on a 2x WD Raptor SATA Raid 0 array, also a couple of Raptors on the SIL SATA controller, an old Maxtor ATA drive and 2 dvd-rw's, so both ATA controllers are in use as well. I've got all like 5 USB 2.0 devices plugged in the mobo as well, 2 of them drawing power from the mobo, both network adapters are being used and I use the on-board sound card as well.

I've got all motherboard voltages maxxed out, 4x Corsair PC5400UL blocks mounted @ 2.3V so considering that everything on the mainboard is used 24/7, I think acheiving a 24/7 stable 345 Mhz fsb speed is pretty darn good for stock cooling too. I'm going to put water cooling on it here pretty soon, as well as volt mod it a bit, at least NB/mem term. and Vddr. I'm hoping to get 360-365 Mhz 24/7 stable without any cpu speed variation.

Hi Larsk,

So you actually run you O/S on the SATA raptors in raid 0 config, connected to the ICH7R southbridge?
And you get those high FSB O/C with all that other gear attached?
Hmmmm, I should just max the volts on everything in the chipset dept and the PCIe v. It annoys the hell out of me that I get window errors above 287mhz fsb, So it seems using more power to components will help here. But what if the system shuts down and reboots again. Maybe D0 stepping 560 cpu is no good even with 1.51 + vcore.... :(

You getting 345mhz fsb with stock cooling would IMO be easier to do in the climate of your country than down here in warmer Australia, even in our winters. Thats a fantastic achievement considering all the gear thats plugged in on that mobo.

[FUGGER]

My mobo was unmodded and I reached booted at 416FSB and PLL'd to 450FSB.

I have the new setFSB and plan to push it again tonight. Ive had the AC on full blast for a few hours now and plan to breach -130c under load.

My older P5WD2-P maxed out at 365FSB or so.

Mine has silver sinks and the 8 pin power next to CPU.

[LarsK]

My mobo was unmodded and I reached booted at 416FSB and PLL'd to 450FSB.

I have the new setFSB and plan to push it again tonight. Ive had the AC on full blast for a few hours now and plan to breach -130c under load.

My older P5WD2-P maxed out at 365FSB or so.

Mine has silver sinks and the 8 pin power next to CPU.

Damn, that sounds really cool :p: :D I would really like to get mine completely stable beyond 360 with "everything" attached.

Mine has silver heasinks and an 8-pin power connector as well. Also I'm using the PCP&C 850W SSI PSU, I'm quite sure that doesn't hurt my oc ability any either ;)

[sluflyer06]

Got my board up and running: Currently @ 4.410 using 315mhz x14.
7800 is @ 530/1350 (stock cooling until i get my LS then it goes water)
Aquamark 92,133 (http://arc.aquamark3.com/arc/arc_view.php?run=625119469)
http://img310.imageshack.us/img310/173/315memory7yc.th.jpg (http://img310.imageshack.us/my.php?image=315memory7yc.jpg)
http://img349.imageshack.us/img349/6347/superpi26sec8me.th.jpg (http://img349.imageshack.us/my.php?image=superpi26sec8me.jpg)
SuperPI: 26 Seconds
Sandra Memory: 7,393MB/s

[gandalf027]

My mobo was unmodded and I reached booted at 416FSB and PLL'd to 450FSB.

I have the new setFSB and plan to push it again tonight. Ive had the AC on full blast for a few hours now and plan to breach -130c under load.

My older P5WD2-P maxed out at 365FSB or so.

Mine has silver sinks and the 8 pin power next to CPU.

Does the new setfsb work with the new bios like 0422 or 0509?

thanks .. and if yes, where can we get it?
:)

[aristidis]

Just a question about the VDD mod. With the mod on and stock mem volts i get a lot of errors with memtest 1.60 and also with Win memtest. Without mod everything is fine. Strange think. I have also noticed that the broblem is specially with the second dimm slot.

[sluflyer06]

Don't flame me for this but....How do you make Sandra run the bandwidth test unbuffered as opposed to default.

[aristidis]

Don't flame me for this but....How do you make Sandra run the bandwidth test unbuffered as opposed to default.


I din't say anything about Sandra. Just memtest with Spd and 1.8 volts with bios settings

[sluflyer06]

i know you didn't.......but everyone here uses sandra unbuffered as a memory bench and I can only get it to do buffered.

I can read, I didn't exactly see the word sandra in your post........yet my question still stands and thank you for your concern.

[aristidis]

Im' not talking about benching memory, just problem with memory after the VDD mod. It was also imposible to run Sandra with so many memory erors.

[BigStan]

Don't flame me for this but....How do you make Sandra run the bandwidth test unbuffered as opposed to default.
Along the bottom of the Sandra Mem screen on the left side is an icon for settings. Uncheck these and when you run Sandra you should get Unbuffered scores.

[sluflyer06]

Thanks BigStan

[BigStan]

You are welcome. You might be paying me back soon. My P5WD2 came in today. Mem, vid card, etc should all be here tomorrow. Prolly will be needing some help shortly.

[sluflyer06]

I'm always willing to assist in overclocking ventures. :) God knows I've helped about almsot every EOCF member that owns a P5WD2 with one thing or another.

[IluvIntel]

I don't have the option to unlock cpu with 14x multi on this mobo, are using 630 cpu so stuck with 15x multi. :(
I am using 0422 bios, anyone have idea how to overcome this?
Also, other options in CPU config menu are set to auto, and thats the only choice you can take... :confused:

[noosphere]

Any news???
Why i cant pass off 300fsb with multi 14x ou 280 with multi 15x???
Strange! If I put 305fsb (Multi 14x) and increase vcore for 1.50 it continues to be instabel... What could be?

[mozzartusm]

Any news???
Why i cant pass off 300fsb with multi 14x ou 280 with multi 15x???
Strange! If I put 305fsb (Multi 14x) and increase vcore for 1.50 it continues to be instabel... What could be?

I would put your PSU at the top of the suspect list. I can run benchmarks all day long at FSB's alot higher than 300 with Vcore at 1.475 so thats not your problem.

Im running a 3.73EE so if your using Dual core then my previous statement may not apply. I havent had a chance to test a dual core yet so im not up to date on how much voltage the CPU needs.

[noosphere]

I would put your PSU at the top of the suspect list. I can run benchmarks all day long at FSB's alot higher than 300 with Vcore at 1.475 so thats not your problem.

Im running a 3.73EE so if your using Dual core then my previous statement may not apply. I havent had a chance to test a dual core yet so im not up to date on how much voltage the CPU needs.

I allready check with more vcore... and the limitation is the CPU.... i cant push more the cpu :(
I participate in some foruns in allmost the people that have a DC, they just can reach to 4200 :D

Regards :O
Noosphere

[The Stilt]

Bios version 0509 is the worst overclocker.
It isn´t even stable with 2GB RAM at 266FSB 1:1 / 3-2-2-1.

Use version 0408.
It´s most stable and overclockable version.

[noosphere]

Bios version 0509 is the worst overclocker.
It isn´t even stable with 2GB RAM at 266FSB 1:1 / 3-2-2-1.

Use version 0408.
It´s most stable and overclockable version.

:cool: Strange!
But with de 0509 e can push my cpu at 4200mhz (300*14) and the memory at 1000mhz with X14.... on the other hand with 0408 e just can put the cpu at 4200 (280*15) and the memory only reach at 9xx, because i cant pus more than 4200 in my cpu.

Could you explain me please!

Sorry for my english ;)

Noosphere

[BigStan]

Where did 0408 come from. The only one on the ASUS web site is 0422. Do you have a link to 0408?

[noosphere]

Where did 0408 come from. The only one on the ASUS web site is 0422. Do you have a link to 0409?


Take it 0408 (ftp://ftp.downloadbase.de/p5wd2_0408.zip)
I am unaware of 0409! i never see it before. Where did you see someone speaking about 0409?

Regard

Noosphere

[jVIDIA]

Noosp, BigStan was refering to 0509 and not 0409 .... he must have spell'd it wrong ;)

[BigStan]

I did spell it wrong, buit I was referring to the 0408. He posted me a link and that was what I was looking for.

Thanx for the link.

[noosphere]

Noosp, BigStan was refering to 0509 and not 0409 .... he must have spell'd it wrong ;)
Right ;)
Thanks dude :P


I did spell it wrong, buit I was referring to the 0408. He posted me a link and that was what I was looking for.

Thanx for the link.
Bigstan, post news after you do the test to 0408 bios ;)

The Stilt ... can you answer me to my question please ;)

Regards dudes :

Noosphere

[LarsK]

Bios version 0509 is the worst overclocker.
It isn´t even stable with 2GB RAM at 266FSB 1:1 / 3-2-2-1.

Use version 0408.
It´s most stable and overclockable version.

Hmm, ok.... With bios ver 0509 I can run 280 Mhz fsb with 2 GB memory (4x 512mB PC5400UL) @ 3:5 cas 4-3-3-2 I find that pretty good. Also I don't have the cold boot issue I did with 0408

[noosphere]

Hmm, ok.... With bios ver 0509 I can run 280 Mhz fsb with 2 GB memory (4x 512mB PC5400UL) @ 3:5 cas 4-3-3-2 I find that pretty good. Also I don't have the cold boot issue I did with 0408

Did you try run memtest86 (bootablecd)! Do you have some errors with this timings?
What is your multi and cpu?

Noosphere

[LarsK]

Did you try run memtest86 (bootablecd)! Do you have some errors with this timings?
What is your multi and cpu?

Noosphere

Yes I did, from bootable floppy thogh ;) No errors what so ever. P4 660 @ 18x multiplier (default). System as in sig. Turbo and the other memory optimization enabled (can't remember what it's called atm.and I'm @ work) as well.

There has been no difference in maximum memory oc potential between bios 0409 and 0509 for me. However the 4:5 divider that didn't work for me with the 0408 bios now does work with 0509, so for me the 0509 has been the better choice

[jVIDIA]

Yes I did, from bootable floppy thogh ;) No errors what so ever. P4 660 @ 18x multiplier (default). System as in sig. Turbo and the other memory optimization enabled (can't remember what it's called atm.and I'm @ work) as well.

...

Larsk, my 5400UL with 4-3-3-8 @ 1000 mhz with memtest86+ tests ok but with Memtest 3.2 for windows (http://hcidesign.com/memtest/) I get errors !!!

In which program should I belieave ?!

Maybe they both are correct and it must be something in my windows ....

I use windows xp + sp1 update. And you folks ?

[LarsK]

Larsk, my 5400UL with 4-3-3-8 @ 1000 mhz with memtest86+ tests ok but with Memtest 3.2 for windows (http://hcidesign.com/memtest/) I get errors !!!

In which program should I belieave ?!

Maybe they both are correct and it must be something in my windows ....

I use windows xp + sp1 update. And you folks ?

Good question. I'm not really sure, but in my experince if there are memory errors memtest86 is pretty good at finding them. My blocks won't do DDR2-1000 @ cas 4-3-3-8, "only" DDR2-980, but both memtest86 (I believe I'm using ver. 1.55) and memtest 3.2 for windows show no errors for several passes. I'm using WinXP Sp2

[jVIDIA]

My blocks won't do DDR2-1000 @ cas 4-3-3-8, "only" DDR2-980, but both memtest86 (I believe I'm using ver. 1.55) and memtest 3.2 for windows show no errors for several passes. I'm using WinXP Sp2

2 questions about that :

- what do you get @ DDR2-1000 ?!? BSOD or mem errors ?

- when you said "both memtest86 and memtest 3.2 for windows show no errors for several passes", is that @ 980 ?!

And in my case what should I believe memtest86+ or 3.2 for windows ?!

[LarsK]

2 questions about that :
- what do you get @ DDR2-1000 ?!? BSOD or mem errors ?


I get errors in memtest86, and I also have random stability issues. Just as one would expect with faulty memory.


- when you said "both memtest86 and memtest 3.2 for windows show no errors for several passes", is that @ 980 ?!


Yes, that is @ DDR2-980


And in my case what should I believe memtest86+ or 3.2 for windows ?!

Personally, I would probably believe both. However, errors in windows could be related to something other than the physical memory, and if you have more than 10+ error free passes with memtest86 I would personally look elsewhere for the problem

[IluvIntel]

Why are all you people trying to get 100% perfect memory with no errors?
If you don't use ECC ram, then its a waste of time IMHO.
This Corsair 5400UL memory is not designed to be ECC capable, so why bother with perfection at high bandwidth?

you may as well get a XEON board with ECC and then Overclock that.

[jVIDIA]

Why are all you people trying to get 100% perfect memory with no errors?
If you don't use ECC ram, then its a waste of time IMHO.
This Corsair 5400UL memory is not designed to be ECC capable, so why bother with perfection at high bandwidth?

you may as well get a XEON board with ECC and then Overclock that.

IluvIntel, my goal is to have a fast OC'ed system but also a stable one !

I don't want just a system to do benches !

It must also work 100% when for instance I do some media encoding and run 3D aplications !

;)

[IluvIntel]

I am now running DDR2 @ 936mhz with 4-3-3-6 and its Sp2004 stable (12 hour test) running 2 instances of it both on priority 9. This is with 2.3 Vdimm. Northbridge @ 1.65v. = No errors for Ram only test. :toast:

The Corsair 5400UL's are bloody good value for money !! :fact:

I would test for longer, but I have to use the computer for other purposes ! ;)

[noosphere]

I am now running DDR2 @ 936mhz with 4-3-3-6 and its Sp2004 stable (12 hour test) running 2 instances of it both on priority 9. This is with 2.3 Vdimm. Northbridge @ 1.65v. = No errors for Ram only test. :toast:

The Corsair 5400UL's are bloody good value for money !! :fact:

I would test for longer, but I have to use the computer for other purposes ! ;)


I´am with 4.2.3.4 with same memory--- rock stable

[IluvIntel]

I´am with 4.2.3.4 with same memory--- rock stable

What bandwidth is that with? did you test for errors and for how long?

[V I P E R]

I am now running DDR2 @ 936mhz with 4-3-3-6 and its Sp2004 stable (12 hour test) running 2 instances of it both on priority 9. This is with 2.3 Vdimm. Northbridge @ 1.65v. = No errors for Ram only test. :toast:


Can you give a shot of Everest Read?

[IluvIntel]

Can you give a shot of Everest Read?

See post #241. ;)

[V I P E R]

See post #241. ;)
I saw it several times but you didn't :)
You have attached two times the Everest write and latency, but there is no picture of the read test.

[IluvIntel]

@ V I P E R,

I'm having trouble with my screenshot utility - it don't wanna save where I tell it too ! damm it !
Using the freebie - Irfanview v3.97.

[V I P E R]

Push the Print Screen button on your keyboard then open Paint and paste in Paint the taken picture (you can use Ctrl + V) :).

[IluvIntel]

ok, V I P E R I got it togeather now, but not with Everest.
Would you like Sandra? :)

This is with 4-3-3-7 timings @ 15 x 283mhz, 4:5, I have found it to be the most practical upper limits for my system when doing the things with it that I want.
In other words, distributed computing projects with net surfing and occasional gaming.
All this evey day, all day without lockups or crashes. ;)

[V I P E R]

IluvIntel - thank you my man :). Your scores are great, but I want that Everest score because I'm on my way getting 955 because my ASUS P4C800 E Deluxe i rather old, even that it is a rock solid PC - P4 Prescott 3Ghz @ 4100Mhz watercooled with Storm G4, Black Ice Extreme 2 and MCP-655.