ASUS P5W DH - Problems + Fixes Thread

Printable View

Show 100 post(s) from this thread on one page

12-03-2006, 07:04 AM
Sparx

Quote:

Originally Posted by monza1412

thanks Sparx for your anwser, but here where I live, I can only pick up this model of GSKILL, is so damn hard to find the better ones (HZ) cos nobody sells them.
By the way, can you boot with manual timings or you have to put spd timings?
Is it orthos (large test) stable at 400 mhz fsb?

Thanks in advance

Yeah I have the RAM manual- 4-4-4-12-4.
I am running 1:1 with Ortho stable at 400 x9 for 5.5 hours.
Never had a lock up except when I fiddled with my video card overclock, being a noob at overclocking video cards. :hitself:

Where are you are maybe someone can buy it here and ship it to you?
The RAM you are looking at has good timings but I know nothing about it.
I ship overseas all the time I could help?
sparx
12-03-2006, 07:05 AM
lawrywild

Bios archive added to front page :)
12-03-2006, 07:15 AM
lawrywild

I did some testing today

I got a friend who has some D9 which post on manual timings on both his P5WDH's and got him to send me his spd

I flashed my ADATA with the SPD, everything was fine but still no post on manual timings.

So I edited my own ADATA spd to DDR2-800 5-5-5-15 with tRFC @ 42

didn't do anything to help either so now I'm confident this is not an SPD issue, unless it is an issue within the subtimings that SPDTool cannot read..
12-03-2006, 07:33 AM
DutchBBQ

Quote:

Originally Posted by Jason404

The first post in this thread doesn't mention the well known problem of a lot of PCI/PCIe cards not working with this board when using Windows Vista.

This happens with TV-cards, soundcards as well as the latest graphics cards from NVIDEA as far as I am aware.

I've been waiting weeks now for Asus to release a BIOS update that can sort this out.

Jason, a smart guy has found a solution for this major problem.
Read the post of Wugger:
http://vip.asus.com/forum/view.aspx?...e=en-us&page=5
12-03-2006, 07:44 AM
rpr

Quote:

Originally Posted by rpr

I have a major FSB wall/limitation at 356fsb. I can run my E6600 at 356 x 9 for 3.2 Ghz stable at all default voltages. One Mhz over 356 fsb and I'm not stable, regardless of any increases I try in voltages.

Is there any way to distinguish whether fsb limitations are due to the CPU or the motherboard?

P.S.: I also seem to have the fsb wall at 356 fsb even if I lower the CPU multiplier.

Any help?
12-03-2006, 08:42 AM
Sparx

Quote:

Originally Posted by rpr

Any help?

Why don't you flash up to at least 1503.
Also what power supply are you using?
What are your (critical) BIOS settings?
What are your Vcore and Vdimm and MCH voltages?
Are you using 105 for Pcie?

You got good cooling on the MCH heat sink?

Week 27 CPUS are quite good you have to be able to hit 3.5.

This has been covered on the thread here, but it is overwhelmingly HUGE and has to be hard to sort through.
sparx
12-03-2006, 09:00 AM
lawrywild

Quote:

Originally Posted by rpr

Any help?

have you tried using memset in windows to change tRFC to 42 and then use setfsb or clockgen to raise fsb in windows and see whether its unstable above 356?
12-03-2006, 09:00 AM
STaRGaZeR

FINALLY I GOT HIGH FSB (400) WITH MANUAL TIMMINGS!!! I buy some Team Xtreem D9´s and it just boot fine!!! Now can set DDR2-1000@ 4-3-4-4 with FSB 400Mhz, Conroe 400*9=3600Mhz. I´m using bios 1602, will post my bios settings for u guys :)
12-03-2006, 09:01 AM
rpr

Quote:

Originally Posted by Sparx

Why don't you flash up to at least 1503.
Also what power supply are you using?
What are your (critical) BIOS settings?
What are your Vcore and Vdimm and MCH voltages?
Are you using 105 for Pcie?

You got good cooling on the MCH heat sink?

Week 27 CPUS are quite good you have to be able to hit 3.5.

This has been covered on the thread here, but it is overwhelmingly HUGE and has to be hard to sort through.
sparx

Flashed to 1602 yesterday...no improvement.

Power Supply is Antec 550 W True Power...seems to hold very stable voltages on all rails.

BIOS Settings...followed all suggested settings on this thread. Mostly all default, including SPD on Auto, except for PCIe on 100, on 33.3, and I have disabled Hyperthreading. I have also disabled all settings not used (i.e. Qfan).

Voltages...all Auto. I have tried almost every possible voltage setting independantly and in combination with others. Absolutely no help, other than to perhaps allow me to post but to fail Orthos in 60 seconds or less.

PCIe...tried 105, no help. Currently using 100.

Cooling...upgraded to Thermalright Ultra 120 for CPU, and Thermalright HR-05 (passive) on Northbridge. Again, absolutely no help.

This seems to be a classic case of an FSB wall that has no workaround, but I would like to know if it is my CPU or my motherboard holding me back. I have been following the entire thread...but I don't recall seeing any solutions to finding the root cause between the CPU or the motherboard (of course, other than to replace one of them).
12-03-2006, 09:05 AM
nanoprobe32

Quote:

Originally Posted by lawrywild

I did some testing today

I got a friend who has some D9 which post on manual timings on both his P5WDH's and got him to send me his spd

I flashed my ADATA with the SPD, everything was fine but still no post on manual timings.

So I edited my own ADATA spd to DDR2-800 5-5-5-15 with tRFC @ 42

didn't do anything to help either so now I'm confident this is not an SPD issue, unless it is an issue within the subtimings that SPDTool cannot read..

whats chip IC is on the ADATA ram?
and did you use hexworkshop to edit SPD tables because thats
the only way they can be edited.

i came to the conclusion its the ram ICs Elpida & promos
that are the problem, along with SPD timings set to low values.

the Micron D9GMH & Micron D9GKX dont have the 350 + FSB bug
will run at 400+ 1.1 with manual timings.

so if your wanting hassle free overclocking for these asus 975
chipset boards then buy micron IC's

Gskill
team extreme
cellshock
some crosair not all

iam running cellshock pc8000 at 3.3.3.8 & 400 fsb 1.1 with 2.2 volts
12-03-2006, 09:07 AM
rpr

Quote:

Originally Posted by lawrywild

have you tried using memset in windows to change tRFC to 42 and then use setfsb or clockgen to raise fsb in windows and see whether its unstable above 356?

Well, I thought I tried this fix, but I'm not sure I properly implemented it. I didn't use setfsb or clockgen. I opened memset, set trfc to 42, set memset to reapply it, and then restarted and raised fsb in BIOS. If I recall corectly, I posted and booted, but again failed Orthos in less than 60 secoinds.

Everytime I open memset, trfc shows as 42, so I believe that is default of my memory's SPD.
12-03-2006, 09:07 AM
lawrywild

powerchips and yes, the new timings were read properly by cpu-z and bios
12-03-2006, 09:08 AM
lawrywild

Quote:

Originally Posted by rpr

Well, I thought I tried this fix, but I'm not sure I properly implemented it. I didn't use setfsb or clockgen. I opened memset, set trfc to 42, set memset to reapply it, and then restarted and raised fsb in BIOS. If I recall corectly, I posted and booted, but again failed Orthos in less than 60 secoinds.

Everytime I open memset, trfc shows as 42, so I believe that is default of my memory's SPD.

nah, try using setfsb in windows after memset has applied tRFC 42
12-03-2006, 09:08 AM
nanoprobe32

Quote:

Originally Posted by STaRGaZeR

FINALLY I GOT HIGH FSB (400) WITH MANUAL TIMMINGS!!! I buy some Team Xtreem D9´s and it just boot fine!!! Now can set DDR2-1000@ 4-3-4-4 with FSB 400Mhz, Conroe 400*9=3600Mhz. I´m using bios 1602, will post my bios settings for u guys :)

another 1 who has discovered how to break the 360 bug
buy micron D9 with the SPD table's programmed corectly:toast:

no more auto SPD timmings:woot:

team xtreme :toast:
cellshock :toast:
Gkkill :toast:
12-03-2006, 09:09 AM
lawrywild

Quote:

Originally Posted by rpr

Flashed to 1602 yesterday...no improvement.

Power Supply is Antec 550 W True Power...seems to hold very stable voltages on all rails.

BIOS Settings...followed all suggested settings on this thread. Mostly all default, including SPD on Auto, except for PCIe on 100, on 33.3, and I have disabled Hyperthreading. I have also disabled all settings not used (i.e. Qfan).

Voltages...all Auto. I have tried almost every possible voltage setting independantly and in combination with others. Absolutely no help, other than to perhaps allow me to post but to fail Orthos in 60 seconds or less.

PCIe...tried 105, no help. Currently using 100.

Cooling...upgraded to Thermalright Ultra 120 for CPU, and Thermalright HR-05 (passive) on Northbridge. Again, absolutely no help.

This seems to be a classic case of an FSB wall that has no workaround, but I would like to know if it is my CPU or my motherboard holding me back. I have been following the entire thread...but I don't recall seeing any solutions to finding the root cause between the CPU or the motherboard (of course, other than to replace one of them).

Antec PSU = BIG problem on this board
12-03-2006, 09:11 AM
lawrywild

Quote:

Originally Posted by nanoprobe32

another 1 who has discovered how to break the 360 bug
buy micron D9 with the SPD table's programmed corectly:toast:

no more auto SPD timmings:woot:

team xtreme :toast:
cellshock :toast:
Gkkill :toast:

ahaha you call buying new ram a fix to a bug

yeh right :slap:

oh and it's nothing to do with the SPD

my friend has D9 which is pc2-4300 4-4-4-12 and he boots fine at 400fsb.

it's all about how well your ram can handle tRFC, and it isn't fixed until asus give the option to adjust tRFC
12-03-2006, 09:35 AM
rpr

Quote:

Originally Posted by lawrywild

nah, try using setfsb in windows after memset has applied tRFC 42

Sorry to be so lame, but I just googled setfsb and couldn't find a version for this board's chipset anywhere. Link?
12-03-2006, 09:37 AM
Sparx

Quote:

Originally Posted by lawrywild

ahaha you call buying new ram a fix to a bug

yeh right :slap:

oh and it's nothing to do with the SPD

my friend has D9 which is pc2-4300 4-4-4-12 and he boots fine at 400fsb.

it's all about how well your ram can handle tRFC, and it isn't fixed until asus give the option to adjust tRFC

So Larry,
This is a random thing with the tRFC, that is RAM related, from kit to kit if I follow correctly?

Edit: Also I have G.skill and never had a problem - Is g.skill HZ suppose to have this bug?
sparx
12-03-2006, 09:39 AM
syne_24

Quote:

Originally Posted by rpr

Flashed to 1602 yesterday...no improvement.

BIOS Settings...followed all suggested settings on this thread. Mostly all default, including SPD on Auto, except for PCIe on 100, on 33.3, and I have disabled Hyperthreading. I have also disabled all settings not used (i.e. Qfan).

Voltages...all Auto. I have tried almost every possible voltage setting independantly and in combination with others. Absolutely no help, other than to perhaps allow me to post but to fail Orthos in 60 seconds or less.

PCIe...tried 105, no help. Currently using 100.

what's your current settings now?

vdimm
vcore
vmch
12-03-2006, 09:46 AM
lawrywild

Quote:

Originally Posted by Sparx

So Larry,
This is a random thing with the tRFC, that is RAM related, from kit to kit if I follow correctly?
sparx

That's all I can conclude so far.. we've seen people with the same kits and chips, some with the problem, some without. but one thing is for sure, changing tRFC gets you past this. (memset 3.0) and spd timings which sets tRFC to 42 (as proved by memset again).
12-03-2006, 09:47 AM
lawrywild

Quote:

Originally Posted by rpr

Sorry to be so lame, but I just googled setfsb and couldn't find a version for this board's chipset anywhere. Link?

http://www13.plala.or.jp/setfsb/down...15a3_P5WD2.zip

note: you need the ATK1101 driver installed and enabled in device manager (can be found in the acpi/win2000 folder in asus probe II or aibooster etc software installation package)
12-03-2006, 09:51 AM
Sparx

Quote:

Originally Posted by lawrywild

That's all I can conclude so far.. we've seen people with the same kits and chips, some with the problem, some without. but one thing is for sure, changing tRFC gets you past this. (memset 3.0) and spd timings which sets tRFC to 42 (as proved by memset again).

Thanks Larrywild
I guess no need to fiddle with it since I am fine at 400 x 9. But I would like to take a peek at this setting using memset 3.0 once I understand how it works.
Never hurts to know more.
sparx
12-03-2006, 09:59 AM
rpr

Quote:

Originally Posted by lawrywild

http://www13.plala.or.jp/setfsb/down...15a3_P5WD2.zip

note: you need the ATK1101 driver installed and enabled in device manager (can be found in the acpi/win2000 folder in asus probe II or aibooster etc software installation package)

Thanks! I will give it a try and report back with findings.
12-03-2006, 10:22 AM
Gascogne

Im using Super Talent PC-6400 and I can boot fine at 400 fsb with manual timing 4-4-3-8, I have no idea if they are D9 or not but they are good ram atleast. :)

Super Talent PC-6400
12-03-2006, 10:28 AM
lawrywild

Pretty sure they're D9

Show 100 post(s) from this thread on one page