Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

Steve McCroskey: Johnny, what can you make out of this?

[Hands him the weather briefing]

Johnny: This? Why, I can make a hat or a brooch or a pterodactyl...

My desktop started having some serious stability issues two days ago. The video on the screen would have random errors, games became unplayable, Firefox would crash from time to time, and eventually Windows 7 started BSODing.

First, I suspected videocard drivers, but after updating to the newest nVidia drivers, then even trying the beta drivers, the problems continued.

Second, I suspect the videocard's RAM had gone bad, but after running several test scans everything was in working order.

That left me with a potential RAM issue. At this point I turned to the old standby, Memtest86, and got to work.

My system has an ASUS M4A87TD Evo motherboard, and 2 x Kingston 2GB PC1333 Valuram.

This mobo has 4 DIMM slots, and with only two sticks you are instructed to run them in slots 1 and 3. After repeated testing with Memtest over the past 2 days, I have the following repeatable results:

Status: PASS
Slot 1 - Stick A
Slot 3 - 
Note: Memtest reports this as 666mhz DDR 1333

Status: FAIL on Test 5
Slot 1 -
Slot 3 - Stick A
Note: Memtest reports this as 200mhz DDR 400

Status: PASS
Slot 1 - Stick B
Slot 3 - 
Note: Memtest reports this as 666mhz DDR 1333

Status: PASS
Slot 1 - 
Slot 3 - Stick B
Note: Memtest reports this as 200mhz DDR 400

Status: FAIL on Test 5
Slot 1 - Stick A
Slot 3 - Stick B
Note: Memtest reports this as 666mhz DDR 1333

Status: PASS
Slot 1 - Stick B
Slot 3 - Stick A
Note: Memtest reports this as 666mhz DDR 1333

After Stick A failed alone in Slot 3 and Stick B passed in both slots, I figured this was a simple case of a flaky RAM stick. But Stick A passes in Slot 3 IF Stick B is sitting in Slot 1, which makes me hesitant to blame my troubles solely on Stick A.

Unfortunately, I have neither more RAM to test, nor another desktop to test my existing RAM on.

Suggestions?


Memtest85's explanation of Test 5 follows for the curious. This is the only test my RAM fails:

Test 5 [Block move, 64 moves]

This test stresses memory by using block move (movsl) instructions and is based on Robert Redelmeier's burnBX test. Memory is initialized with shifting patterns that are inverted every 8 bytes. Then 4mb blocks of memory are moved around using the movsl instruction. After the moves are completed the data patterns are checked. Because the data is checked only after the memory moves are completed it is not possible to know where the error occurred. The addresses reported are only for where the bad pattern was found. Since the moves are constrained to a 8mb segment of memory the failing address will always be less than 8mb away from the reported address. Errors from this test are not used to calculate BadRAM patterns.

EDIT - I have not tried the bit fade. I'm running that now. It takes 3 hours per cycle so it'll be awhile before I can post the results...

share|improve this question
    
+1 for Airplane! ehehehe. –  Shinrai Oct 7 '10 at 14:20

2 Answers 2

up vote 1 down vote accepted

Stick A is bad. I suppose it is remoteley possible that the stick component that is bad is not an actual memory chip, but some other supporting chip used for slot addressing. If that is true, you should be able to run fine with Stick A in slot 3. I encourage you to leave stick A in slot 3 and test by seeing if the original issues persist. If they do, buy a replacement for A and make note that Memtest86 has limitations.

share|improve this answer
    
The desktop has been running fine since I posted, with Stick A in Slot 3. Bizarre. As it's under warranty and still dying when tested in Slot 1, I'm going to return it anyway. Thanks for the help. –  Andrew Heath Oct 12 '10 at 6:32
    
Glad to hear you're back in business. –  Flotsam N. Jetsam Oct 12 '10 at 12:27

Just because you're using Memtest86+ doesn't mean that every block of RAM is accessed and tested just most. This means that the test with just the single stick will have a "higher" chance of more of the memory blocks being accessed and read. With the two sticks there's a "lesser" chance. How much higher and lesser they may be, I'm not sure. My best suggestion is to replace memstick A and see if that changes anything. If not then there is a possibility that the motherboards memory socket is failing, which then means replacement of the Mobo (however I doubt this).

Finally, I'm assuming that you've done this but just in case, make sure that the motherboard memory sockets and the actual memory sticks themselves are free of any dust or debris. You can clean them with canned air, and also rubbing alcohol (just make sure the alcohol has completely dried before seating the sticks of RAM).

share|improve this answer
    
Hi KronoS. Since one pass may not test the stick fully, is there a number of passes in Memtest86+ after which one can be reasonably certain that a stick is good? –  Andrew Heath Oct 7 '10 at 2:41
    
I let it run at least overnight on mine which equates to roughly 8 to 10 passes. I would assume that is sufficient. Do it for too long and you're bound to find some sort of error, but too short and that isn't really a good testing. Perhaps this might be helpful. –  KronoS Oct 7 '10 at 4:00

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.