• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Mem errors when chip OC'ed

D

D3v1an7

[H]ard|Gawd
Joined
Oct 22, 2003
Messages
1,256
OK, the rig is in my sig

I recently bought a gig of Kingston pc3200 with 2-3-2-6 timings (which I did set in bios per Kingston)

When I up the FSB to 200 and drop the multiplier to 10 I get tons of errors in test 5 of memtest. I tried upping the vcore to 2.7 and even 2.8 but that didn't solve the problem. I also dropped the multiplier to 9 to make the chip run at nearly stock speeds but I still got a ton of test 5 errors. The only way I could make the errors dissappear is to set the FSB to 166 and run the chip at stock which leads me to believe that either the chip itself or the Northbridge is the culprit.

What other voltages can I tweek to get the chip and northbridge to take the OC to 400MHz from 333? What is the max for a safe overvolt for those other adjustments?

Oh, my temps are freaking great. At stock speeds, my CPU doesn't go over a 100F (38C).

I'm more than a little disappointed that I can't use this new mem to its full potential. Any advice would be greatly appreciated.

Thanks
 
change the 2-3-2-6 to 2-3-2-10 or 2-3-2-11

works much better with the nf7-s

your current settings are the best for running stock speeds, but overclocking will cause corruption. i just learned this lesson the hard way a few days ago, corrupted my registry entirely.
 
really?

Wouldn't increasing the latency that much offset the performance increase that going to a 400MHz bus gives me?

Isn't there any other way of making my box run stable at a higher FSB without dumbing down the memory below the speeds that it says it is rated for?
 
i didnt notice much of a performance drop. you could try somewhere in the middle like 8 or so, but at 6 you will get some pretty corrput data. everywhere i read says that the board likes 10 or 11 and it solved all my problems

www.lvcoyote.com said:
Row active delay

The delay that results when two different rows in a memory chip are addressed sequentially. Also known as T-ras. Values range from 1 to 15. Most NF7 users report better memory performance with a value of 11 or 10. Setting this value too low can result in data corruption. A good way to think of this setting is the time a page in a book is available to be read, before the book is closed and opened to a new page. If the book is closed before the page is fully read, you have to reopen the book to the same page to read it again, which means you wasted time progressing to the next page. However if you are done reading the page, leaving the book open too long also slows your progress to the next page.
Click to expand...
 
Hmm, that's a good explanation, I'll try it when I get home and see what happeneds.

Thanks
 
On nforce2 boards (and nforce3/4 I believe), setting it to 10/11 is actually better performance than 5/6. Just a slight oddity with amd chips and nvidia chipsets...might even be like that on via boards, not sure tho as I have no personal experience.
 
You must log in or register to reply here.
Back
Top