RuSo

sorry for the long rather confusing story but here it goes...
I have currently big problems with 2 P5K boards, 1 P5K and 1 P5K/Pro
They wont run 4GB OCZ reaper 1066 5-5-5-15, at their rated speed on stock cpu Q6600 2.4Ghz
I currently own a ATI HD4850 run XP MCE and Vista Ultimate SP1 fresh install on seperate hdd´s

We tryed there in the shop another PSU and replaced the Reaper for 2X2GB corsair XMS2 800
And those gave problems too, at stock cpu/ram speed intel burn test fails within a second.
Same as 4X 1GB Reaper, but no problems in burn test with 2X1GB Reaper)
So the shop offered to replace the board with an P5Q pro as they were puzzled too.
But i dont want the same problems again, the modules were tested in pairs in memtest and gave no problems.
When 4GB in, 4X 1GB or 2X 2GB then most games crash randomly, displaying the vpu recover a loss of signal leaving my monitor in power saving mode or a "0x00000008 memory could not be read" error.
This occurs eventually on 2GB too, but the intel burn test is ok
I can run memtest hours... stable
my previous gfx card 8800GTS suffered driver stopped responding problems randomly and texture problems and the card is fine in another PC

all seperate parts are ok but it all goes wrong when it must work together
Is this issue my mobo or my CPU or My Gfx or my memory i ran everything stock, disconnected all unneeded devices it still makes problems....very strangely

spaceballsrules

This sounds an awful lot like a buggered CPU, as uncommon as that is. Try a different one.

__________________
spaceballsrules EOC Stats Page

Asus Crosshair II Formula Owner's Club

RuSo

tryed the Q6600 with the P5Q no change, then tryed with an old p4 775 3.4EE on The P5Q and makes te same error.....linpack fails after 1 sec with 4GB, all runs stock Reaper 1066 memory at 800 because of P4 FSB speed

Acoma_Andy

Which set are you using now? Your sig says Kingston.

What's the memory voltage and timings at now? Perhaps you can post a Cpu-Z screenshot of the memory tab.

__________________

superk

update to the latest bios for all components, including hdd firmware if possible. ensure all your drivers are up-to-date. Verify with the QVL to see if your ram is compatible or not.

Ensure that NOTHING is overheating.. that intel burn test is insane..

If all of that fails, or even perhaps before hand, try running the ram at 800mhz instead of 1066. I have a P5K, so I know that its supposed to be compatible, but just try it. If it works, it is a good chance the memory controller on the motherboard(s) are shot.

Also, for your memory.. perhaps try to loosen the timings, and verify your voltages.. maybe its not running 1066 stable cuz its not gettin enough juice..

__________________
.: My Case Mods :.

Hades (Antec 900) * Antec 1200

.: My Case Rebirths :.

Agamemnon * Heracles

.: My Lego Cases :.

Lego PC v1 * Lego PC v3 * Lego Xbox

RuSo

i tryed all u suggested above i have tryed 3motherboards all with same issue the P5Q came just out of the box

the intel burn test runs easy 6 passes on my quad with 2 GB and also on the P4
but fails after cold boot in 1 second. the temps are fine of all tested i tested the board bare only on motherboard tray of my Thermaltake shark with distancekeepers of course and all needed for booting my OS

i watched in the QVL and there is HOCZ2RPR10662GK in there.
Mine are OCZRPR10661G
the corsair XMS2 2X2GB 800mhz which i tested with in store are not in there.
I think maybe your right about this, well.....it has to be.... i cant think about anything else anymore.. i will get some modules in QVL to test and report back

also take note that both my OS are 32bit the reason for 4GB 4x1GB is only dualchannnel + max amount
i ran with memory hole on and off which limits the memory amount to 3328MB for 32bit with same results
EDIT: hmm if i run the test with only half memory then no problems, it makes errors on my laptop too and it has also 4GB then its just the program, a memory allocation error or something
But it does not explain the vpu and black screen with high pitch whine out speakers issue or maybe it does or maybe my 4850 was D.O.A.??? because i have run gta almost every night and when i wake up and look there is eiter nothing on screen, or "memory could not be read" error this always happens while gaming, it doesnt overheat it has a bigger fan (glacius) then reference model (ASUS)

superk

Well, what I suggest would be to try a 64bit OS. But I mean.. if memtest passes, then the ram is fine. Does the memory pass memtest at 1066 in the rig that it complains during linpack burn test?

__________________
.: My Case Mods :.

Hades (Antec 900) * Antec 1200

.: My Case Rebirths :.

Agamemnon * Heracles

.: My Lego Cases :.

Lego PC v1 * Lego PC v3 * Lego Xbox

RuSo

yes the memory run ok in memtest decreasing the test to half memory solved the intel burn programs error, but i will test tonight if my games still crash leaving my computer crashed in described ways above
It crashes also with only 2GB and with 2GB kingston in my sig
That was always fine, so i now really suspect the ATi 4850 card

superk

yeah it seems quite possible. Did you update to the latest bios on the video card? Try that if you havent already. System bios as well if possible.

Which slot to you have your 4850 in? I have an 8800gt in mine, and its actually in the black slot.. yeah I know its slower, but its just for video for now... I have a heatsink to remove before I can install it in the PCIEx16 slot...

If all else fails, try it in the x4 slot.

__________________
.: My Case Mods :.

Hades (Antec 900) * Antec 1200

.: My Case Rebirths :.

Agamemnon * Heracles

.: My Lego Cases :.

Lego PC v1 * Lego PC v3 * Lego Xbox

RuSo

i narrowed it down to 2 problems
1: 4x1gb reaper makes a vista display driver recovery this explains why multiple graphics cards gave problem solved by using reaper 2x2gb modules (on 1100 5-5-5-15 2.2v 4hours memtest)
2:i have never ever saw such a bad case of
"how NOT to apply thermal paste" as i removed the stock heatsink (glacius) and looked under it, litterally everywhere around the GPU was ceramic paste.... there was also a hardened out grey layer on he heatsink (original paste) which was damaged resulting in only a contact with 75% of gpu.
In the small store i bought it,i explained and i asked is it possible to exchange for a good card if damaged by this fabrication error? the card makes loud noise under stress (its not the fan) and possibly the memory error comes from overheating/burned memIO on graphics card...
They said that it was not permitted to remove the heatsink???
and will not swap it for a new
only after a week of testing, to make sure i havent damaged the card while repairing it??? (i only cleaned gpu and reapllied paste)
after testing period then they will send to supllier (RMA to asus maybe better option?) these problems got me swapping mobo's cpu's ram reinstalling OS etc. 3 months alltogether,
i will never buy there anymore, cause the last a 7600GT that was faulty too, (heavy artefacting) and a brand new antec 550w trio psu (which looks a bit used) where the seperate fan connector (molex) has no power, for me that is enough

tnx for all the help in here REP+ greetz