memory probs?

alright people,

i'm having problems with my RAM, it's prime stable, but i get crashes in counterstrike?

the error is something like 'the memory read from 0x00000000 could not be written to 0x0000000'

any idea's?

settings:-

FSB 200mhz
multiplier @ 11
RAM @ 100% (200mhz)
RAM timings 7 3 3 2.5
CPU 1.650v
AGP 1.5v
DIMM 2.77v

thanks for any help

Is it just CS that crashes? Or games in general? Is your video card overclocked?

~THT

i think it's games related as it freezes abit in battlefield 1942 aswell. if i set everything back to default in the BIOS i dont get the error?

my graphics card is running @ default

thanks

Welcome to the forums!

What motherboard and what is the original speed of your CPU? If you are OC'ing up to a 3200+, then you might want to try move vcore before anything else. Try 1.70, then go up to 1.75.

Remember, prime95 doesn't stress all functions of a CPU. Have you stress tested with 3dMark?

Also, you memory timings should be like this:

-->9<-- 3 3 2.5 where 3+3+2.5 = 8.5, but 9 to round up.

I would actually try 2.3.3.8 as well.

There is a good post about this somewhere in these forums from Mushkin's FAQ/ tech support that dicusses this memory timing.

And lastly, to round out the testing, use memtest86 at your default timings and settings and target (11x200).

Good luck!

EDIT: Ah, I see you have the 8rda+. Your sig is so dark i didn't see it! I had memory errors at CAS 2.5 at 200-220 FSB. Going to CAS 2 eliminated these errors. But of course, if the vcore solves your stability issues, run with that.

thanks for the welcome & info m8, i'll give the memtest, vcore & timings a go

i havent tried a stress test in 3dmark2001, it did make it through 1 test though.

P.S. sig now bright


EDIT: it's booting fine with timings as 8 3 3 2 is it ok to run the memtest with these settings or do you recommend i change back to defaults?

stress said:

it's booting fine with timings as 8 3 3 2 is it ok to run the memtest with these settings or do you recommend i change back to defaults?

Click to expand...

I would try both. It takes about 15mins to run a full memtest. For the sake of solving all the unknowns, I'd give those timings and your defaults a shot.

sig looks great

i've run the memtest on defaults and the over clocked settings and it passed both? but still crashes with the same memory error in game.

could it be the overclocked settings are conflicting with my graphics card?

i've got the AGP at 66mhz i think thats right for a 4x card, think i'll try looping 3dmark when i get back from work.

thanks

stress said:

could it be the overclocked settings are conflicting with my graphics card?

Click to expand...

Possibly. Have you tried the game at CPU defaults as well?

drshivas said:

Possibly. Have you tried the game at CPU defaults as well?

Click to expand...

yeah, with the CPU/RAM set to default i dont get any errors in the same game.

only way i can test it really is when i upgrade my vid card, thats the next on the list

i do sometimes get cold boot problems where the monitor turns off when the desktop is supposed to load, this was happening on my old setup (everything else is different except the vid card)

stress, did you get the error when you had that overclock but without the zalman heatpipe? with faster cpu/ram, you can better take advantage of the videocard and perhaps the card could not handle it with only passive cooling

also, i believe the 8rda+ allows the vdimm up to 2.93 or so, why don't you give that a shot?

drshivas, nice observation on the mushkin latency reccommendations, however i believe the exact formula from mushkin for tRAS translates as this:

CAS + tRCD + 2 cycles = tRAS

(2-3-3...2+3+2=7, tRAS = 7)
(2.5-3-3...2.5+3+2=7.5, tRAS = 8)


from Mushkin:
"the minimum tRAS would be tRCD+CAS latency + 2 cycles (to output the first burst of four and make way for the second burst in the output buffers).
Any tRAS setting lower tRCD + CAS + 2 cycles will allow the memory controller to close the page “in your face!” over and again and that will cause a performance hit because of a truncated transfer that needs to be repeated. Along with those hassles comes the self-explanatory risk for data corruption. That one is not a real problem as long as the system is kept running but in case it is shut down and the memory content is written back to the hard disk drive, the consequences can be catastrophic. For the drive, that is. "

i did have the cold boot problem before i put the heatpipe on, it did the same thing with a crystal orb on there. i think that problem started when the original fan died and i didnt notice for a while

if i can find the crystal orb, i'll give it a try with that on instead of the heatpipe & see what happens

thanks

drshivas said:

I would try both. It takes about 15mins to run a full memtest. For the sake of solving all the unknowns, I'd give those timings and your defaults a shot.

sig looks great

Click to expand...

If I actually run all the tests, it takes me 8 hours or so. What's your definition of full? All the default tests?

You may also want to lower your multiplier when you run memtest, to make sure that you are only testing the staility of your ram.

Thanks for the clarification R4z0r4mu5 Pr|m3. I guess I miss calculated.

Gnufsh: What I mean by a full memtest is whatever it runs by default. When I boot off my memtest CD, it starts running, and takes about 15mins to finish a pass. I assume that 1 pass is a full test of the memory, right? I've been using one pass as a quick litmus test for OC stability. So to answer your question, I have been using the default tests as a "full test."

(Damn, my record in this post is looking like 0-2!)