• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Interpreting memtest86 results

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

hey560

Registered
Joined
Mar 28, 2004
I have my system at stock speeds and voltages etc (specs are below). I've been noticing the odd reboot every couple weeks. This lead me to test my RAM/CPU using some benchmarking utilities.

I have not overclocked the system since I built it. I need some help in interpreting results from 2 tests I performed.

My CPU/Case temps with ambient temperature of ~24C are:

Idle: 33/31
Load: 41/32

Test1: Prime95 v24.13 "Blend" Torture Test. Ran for 9 hours 51 minutes until a rounding error occured.

Test2: Memtest86. Default test ran for 24hs, 2 errors at same address.

I am wondering if this sort of error is typically expected in hardware that has never been overclocked? Your help is much appreciated.
 
i'd suspect your RAM may be the culprit. You may need to run looser timing that the chips are designed for, in which case, i'd personaly RMA it (if you buy it and they claim 2.5-3-3-7, but you can only get it stable at 2.5-4-4-8, then in my eyes, its defective).

and you show in your sig you have the Value Select Corsair. I had a couple of those and I returned them. Value series OCZ are 10x better. heck, no-name OEM brand mismatched pair ran better in dual channel mode with tighter timings than those 2 sticks of Corsair Value. Sure, Corsair is a good brand, but it seems they put all the lemon chips in the Value series that fail their speed bin tests. I couldnt even break 210mhz on them! (OEM no-name got up to 218mhz, and OCZ Premier series [very cheap price] got me up to 230mhz).
 
I have personally found on my DFN NF4 Ultra-D that when errors occur on Test #5 that that is the result of not enough voltage and errors on Test #8 result in errors in timings that are often fixed by messing with the Tref value. Could you tell us what test caused your errors?
 
Thanks for the replies. I removed the module memtest86 reported an error on and re-ran prime95. I stopped the test at 13hrs with no errors. I will run prime95 on the bad module as well as memtest86 on both modules starting tonight.

Eldonko said:
What RAM, what timings, what voltage.
CAS# Latency 2.5 clocks
RAS# to CAS# 3 clocks
RAS# Precharge 3 clocks
Cycle Time (TRAS) 8 clocks
Bank Cycle Time (TRC) 11 clocks
DRAM Idle Timer 16 clocks

kukyfrope said:
Could you tell us what test caused your errors?
Test 4 [Moving inversions, random pattern]

Supposedly Test 4 is used to find "difficult to detect data sensitive errors"

I will update when I've completed the tests. I don't think I can RMA anymore since its been over a month. I wonder how good Corsair's warranty is.
 
hey560 said:
. I don't think I can RMA anymore since its been over a month. I wonder how good Corsair's warranty is.

I bet(hope) they have more than a month warranty...
 
Corsairs warrenty is lifetime and has the best replacement policy I have ever been through. Post the problem on the forum and ramguy tells you things to try to fix. If he doesnt he gives you the link to fill out the RMA form and off it goes. You get a replacement after a lil while.

JT
 
Back