Is my Ballistix failing?

bellylard · 28 Feb 2006 at 02:50

I'm starting to think I got a poor set of Ballistix(2x1gb).....When I first got this set they ran fine at 265fsb then after awhile my system started becoming unstable with random reboots, so I backed it off to 255fsb and over the past coupld weeks I started getting more random reboots. It's now down to 250fsb and BF2 started to close or reboot my system after 5 to 30 minutes of play. And just now BF2 won't get past loading all the map data, it shuts down and I get a window pop up saying bf2 ran into a memory error. I'm running 3-4-8-4 1T @ 2.8v

Does this sound like my Ballistix is starting to crap out? What memory diagnostics can I use to figure out?

edit I'm running the bootable Memtest right now, will it stop at an error? Are there any settings I should use?

NightSt@lk3r · 28 Feb 2006 at 04:34

loop test 5 for 1 hour

bellylard · 28 Feb 2006 at 04:38

I'm just going to let the program do it's thing overnight.

Why test5 btw?

NightSt@lk3r · 28 Feb 2006 at 04:56

it usually shows error beffor any other test

Larok · 28 Feb 2006 at 07:14

mine passed memtest but gave an intermittant boot error after 6 weeks of being good.

a_jam_sandwich · 28 Feb 2006 at 07:24

Mine for the 3rd time is starting to fail. I can't be a company that charges so much for the memory can produce such a steaming pile of poo. Moral of the story, do not buy from them again!

tbz_ck · 28 Feb 2006 at 11:52

I would buy again , if only on account of their customer support

chopchop · 28 Feb 2006 at 11:56

tbz_ck said:
I would buy again , if only on account of their customer support

lol... wait till your ballistix dies a few more times... you'll soon change your mind

tbz_ck · 28 Feb 2006 at 13:33

Its died once already

It did last over a year though, and the new stuff is working fine so far (finds wood and touches it)

Reliability aside, it is excellent performing RAM.

bellylard · 28 Feb 2006 at 14:21

Well I let it memtest run for 11 hours and the error count stayed at 0. I'm not sure if the program ran through every on it's own though?

What could my problem be?

AMD64 3000+ @ 2.3GHz (max load temp is 44c)
9800 pro (395/714 passes artifact test)
Antec PP-412X 400W (12v = 12.1v-12.2v, 3.3v =3.15v-3.2v, 5v = 5.07v)
MSI K8N Neo2 Plat.

bellylard · 28 Feb 2006 at 16:48

*UPDATE*

I can pass many loops of memtest #5 @260fsb(5fsb higher than I was experiencing my recently problems at) and @ 2.6v....But I cannot get into windows with at least 2.8v....(have always used 2.8v)

Here aremy dram settings

cas=3
trcd=3
tras=8
trrd=3
trwt=3
trc=12
trfc=18

btw I'm using SideEffects 1.CB3 bios on my K8N Neo2 Plat

Larok · 28 Feb 2006 at 21:26

Buddy, it's your Crucial ballistix memory. Get rid of it. Mine and others failed in the same way and memtest was fine on mine too.

You can confirm this by trying some other memory in your system, then you'll know for sure.

bellylard · 28 Feb 2006 at 23:48

Larok said:
Buddy, it's your Crucial ballistix memory. Get rid of it. Mine and others failed in the same way and memtest was fine on mine too.

You can confirm this by trying some other memory in your system, then you'll know for sure.

If it passes memtest why would they send you a new set??? Do they have other diagnostics tests they use?

Larok · 1 Mar 2006 at 08:19

Try some different memory in your system, confirm that's the issue and then contact the retailer. My understanding is OCUK use a piece of diagnostic software called microscope.

I am not postive but I presume this is it,

http://www.micro2000.co.uk/products/microscope/microscope.html

Dutch Guy · 1 Mar 2006 at 12:21

Instead of Memtest run Prime95's Blend torture test in Windows next.

I think it's a bit early to blame the Ballistix. it could also be other things.

bellylard · 1 Mar 2006 at 16:58

I'll run prime today at 255ht and see how that does.

Penfold99 · 1 Mar 2006 at 21:05

my second set it failing, and ive sent them 3 emails and not recieving a reply, first time their custmor support was great!

bellylard · 1 Mar 2006 at 21:22

Well after a few prime tests I got some very un-settling results from my ballistix. I just put in my old 512 stick and am running at the same clock speed with less volts and it's gone further than my ballistix has recently. I will post the logs later tonite, hopefully leading to more certain answers.

bellylard · 2 Mar 2006 at 00:15

I think my fears of failing Ballistix are comming true. Here are my Prime logs when I tested with my Ballistix I thought might be failing:

Type: Blend - stress CPU and RAM Min: 8 Max: 4096 InPlace: No Mem: 1951 Time: 15
CPU: 2295MHz FSB: 255MHz [255MHz x 9.0 est.]
3/1/2006 12:15 PM Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
Test 2, 4000 Lucas-Lehmer iterations of M19922943 using 1024K FFT length.
Test 3, 4000 Lucas-Lehmer iterations of M19374367 using 1024K FFT length.
FATAL ERROR: Rounding was 0.4970703125, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Torture Test ran 9 minutes 45 seconds - 1 errors, 0 warnings.
Execution halted.

Type: Blend - stress CPU and RAM Min: 8 Max: 4096 InPlace: No Mem: 1951 Time: 15
CPU: 2249MHz FSB: 250MHz [249MHz x 9.0 est.]
3/1/2006 12:28 PM Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Torture Test ran 3 minutes 24 seconds - 1 errors, 0 warnings.
Execution halted.

Type: Blend - stress CPU and RAM Min: 8 Max: 4096 InPlace: No Mem: 1951 Time: 15
CPU: 2249MHz FSB: 250MHz [249MHz x 9.0 est.]
3/1/2006 3:00 PM Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 4000 Lucas-Lehmer iterations of M19922945 using 1024K FFT length.
Test 2, 4000 Lucas-Lehmer iterations of M19922943 using 1024K FFT length.
Test 3, 4000 Lucas-Lehmer iterations of M19374367 using 1024K FFT length.
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Torture Test ran 10 minutes 45 seconds - 1 errors, 0 warnings.
Execution halted.

I decided to try one of my old 512 PC3200 sticks of Samsung and run the same test at the same speed(I used a 166MHz divider and 3-3-3-8 1T in order to get a stable 250fsb)

I put the Samsung ram into my PC and set the fsb to 250 and also put the Vcore down 0.02v and When I came and checked on my computer next it was over 3 hours later and the prime blend test was still running fine. Does this verify my Ballistix are crapping out???

Dutch Guy · 2 Mar 2006 at 08:32

bellylard said:
I put the Samsung ram into my PC and set the fsb to 250 and also put the Vcore down 0.02v and When I came and checked on my computer next it was over 3 hours later and the prime blend test was still running fine. Does this verify my Ballistix are crapping out???

Yes, if all other factors are the same and it runs fine with another stick of RAM it is the Ballistix that is failing.

I guess a call to Crucial is in order (if you ordered directly) or a call to OcUK