Jump to content
Corsair Community

MSI K8T NEO-FSR & CMX512-3200C2PRO


coolied

Recommended Posts

Ram Guy- I am having a problem with my system which is an Athlon64 3200+ running on an MSI K8T NEO-FSR motherboard. I have 2 sticks of Corsair CMX512-3200C2PRO which do not pass Memtest86+ reliably, one stick is a v1.0 and the other is a v1.2 I am running my RAM in DIMM1 & DIMM3 as recommended, at the recommended manual 2-3-3-6 timings and the recommended 2.7vDimm voltage. I have tested both sticks individually with the same results, which is one time it will run Memtest86+ for 9+ hours with no errors, the next time I run it I will have 200+ errors within 10 minutes. It also exhibits this behavior with 2.5-4-4-8 timings and even 3-5-5-11 timings, so i'm really not sure what is going on. I hesitate to blame the CPU as Windows runs stable, and Prime95 seems to also be stable. The one component I am unsure of is the power supply as it is a 420W Raidmax that came installed in my case and doesn't seem to have very stable voltage, my +12v rail seems to run around +11.56v usually, and according to MSI Corecenter, my CPU voltage is between 1.456v and 1.473v which is also reported in the BIOS and by CPU-Z. Since the core voltage is supposed to be 1.5v, i'm wondering if the proper vDIMM is being provided to the memory, or if that is also low. Please advise, as I don't really want to spend any more time trying to blindly fix this issue. Complete system specs: MSI K8T Neo-Fsr motherboard BIOS v1.20 AMD Athlon64 3200+ @ 200x10.0 Raidmax 420w Power Supply 2 sticks Corsair CMX512-3200C2PRO 2-3-3-6 manual timings @ 2.7vDimm Lite-On 16x DVD-ROM Samsung 48x CD-RW WD Raptor 74GB SATA Hard Drive ALPS 1.44MB Floppy Drive ATI Radeon 9800XT Video Card SoundBlaster Audigy Gamer Sound Card So again, please advise... bad RAM, CPU, Motherboard, or Power Supply? Thank you very much in advance. Sincerely, Dave Russell
Link to comment
Share on other sites

Seems like you have been thorough and it's probably a 60/40 chance. 60% PSU is bad or 40% you got a bad Corsair module. Do you have access to another machine to test the RAM? I'd use another machine and again test each stick individually with both Memtest and that Microsoft memory tester. Also, if there is a shop close, see if they a good PSU tester. You may also want to have a look at the Goopsulator by a guy called goosehump. Do a Google search. It's a good little tool for estimating power supply needs. Mike.
Link to comment
Share on other sites

and noticed that 2-3-3-6 timings are only supported on Intel platforms, and that 2.5-3-3-6 timings are for AMD.... thinking back on it, I believe I was running motherboard default vDimm (2.6) when I was testing with looser timings, i'll have to try 2.5-3-3-6 at 2.7vDimm (I believe it's supposed to be 2.7, or is it 2.75? that shows up in RED if I select it in the BIOS)..... the reason I mention the power supply is alot of people seem to have problems with the Raidmax 420w unit, of course a $60 case isn't gonna include a real great PSU..... Dave Russell
Link to comment
Share on other sites

The Corsair is safe and within warranty up to 2.8v so I'd set it on up there. I have not heard either way about your PSU but, good luck with whatever it is. For the long haul, you might be better off with a new PSU anyways just to prevent it from causing any other issues even if it is not causing this current issue. PSU related problems can be subtle and hard to diagnose. Mike.
Link to comment
Share on other sites

I have 2 of those same Thermaltake units and I am very happy with them so far. They are quiet and stable. If it gets to the point that money is no object, I'd suggest a PC Power and Cooling PSU. And, I hear good things about the Aerocool units also. Let us know if it helps. Mike.
Link to comment
Share on other sites

Mike- Will do, hopefully it will help, as I find it hard to believe that i've gotten 2 seperate modules that are bad, especially from Corsair..... that and i'll try running the recommended 2.5-3-3-6 AMD timings at 2.7v and see if that helps, if not i'll try at 2.8v just in case my current PS is providing under spec vDimm voltage...... Thanks for all your helpful suggestions.... Dave
Link to comment
Share on other sites

Got it running nice & stable, but I had to use very conservative timings on the RAM to get it there, 2.5-5-5-11 with Fast (not Turbo) system performance settings. Running (according to the BIOS) 2.7vDimm, but I seriously doubt it's actually there since my power supply does seem to be whacked...... If I kick the core voltage up to 1.55v in the BIOS, the processor actually runs at its rated 1.50v and the ram seems stable at tighter timings that way. Makes sense I guess, with the A64s on-die memory controller and all, the voltage fluctuations were probably causing the instability..... Just gonna run the timings loose until the new power supply arrives.... of course, one great thing about running conservative like this, it's running a lot cooler, was around 46c at idle, now idles at about 36c (around 10 degrees cooler)..... and that's with the stock cooler, but with a layer of AS5 (yes I applied it correctly, according to AS website)..... these AMDs run a little warm as I also have good case airflow and all, but hey it's never had a heat-related problem...... still seems nice and fast as well, have to benchmark it now..... Dave
Link to comment
Share on other sites

2.5-5-5-11 generated an error in memtest86+ test 7, at 31.8MB or so, said it expected ffffffff and the result was ffffffffe, not sure if this indicates a bad RAM stick or not, but this was with 2.7vDimm, now trying 2-3-3-6 timings at DDR333, but I paid for DDR400, so even if it works at 333 i'm going to have to try and RMA it, if my new PSU doesn't help...... Dave
Link to comment
Share on other sites

i must be the unluckiest guy alive when it comes to RAM, most people follow the suggestions on this board and have success, i'm ready to throw the freakin' computer out the window......... just thought i'd vent...... Dave
Link to comment
Share on other sites

[quote name='specmike']Venting is kool. If you feel the need to throw the computer, throw it onto a FedEx truck. I'll send you my address :p: Mike[/QUOTE] *LOL* Yeah, I guess i'm not really ready to give up yet, however, if I can't get this Corsair RAM to work, I may have to see if I can get an RMA for it from Newegg, perhaps if I call and explain my problem to them they will allow me to exchange it for something else, the only RAM I can find on Newegg that is *APPROVED* by MSI for the K8T NEO is Crucial however, and I really want to get this Corsair working....... I guess it would help me alot if I had another mobo I could try the RAM on..... I still find it hard to belive *both* of my sticks are bad, however they both act the same way, here's hoping it's the PSU..... if not i'll have to get ahold of my brother and see if we can try it in his...... same mobo/cpu in his machine..... Dave
Link to comment
Share on other sites

  • Corsair Employees
First off, I do not quite understand, you posted this; [QUOTE]sorry to shout...... just trying to get your attention, really going nuts with this issue, although this thing is like 24hrs+ Prime95 stable while simultaneously looping 3dmark03 CPU tests, so I’m apt to stop worrying about it.....[/QUOTE] As if to say the system is not stable bellow, however you can "this thing is like 24hrs+ Prime95 stable while simultaneously looping 3dmark03 CPU tests," Which tells me if you are having a problem it's likely not the memory. And with this MB and these modules I would test them one at a time with [url]www.memtest.org[/url] to make sure one is not failing. If both pass, then install them in slots 1-3 and set the Dim Voltage to 2.7 Volts. More over in your first post you stated "I have tested both sticks individually with the same results, which is one time it will run Memtest86+ for 9+ hours with no errors, the next time I run it I will have 200+ errors within 10 minutes. It also exhibits this behavior with 2.5-4-4-8 timings and even 3-5-5-11 timings, so I’m really not sure what is going on" Which would suggest some other problem, and the PSU sounds like a likely culprit. However testing your modules on another MB may help to isolate the problem.
Link to comment
Share on other sites

what I was getting at, is that the memtest86 problems I was having were really bothering me, however over the past few days i've had time to do alot of in-windows stability testing and since it seems to be as close to 100% stable as any PC i've ever had i'm just going to stop worrying and start enjoying my system, I re-did the memtest thing 1 stick at a time and only had like 1 failure on each stick over several hours of testing, and now running both sticks in DIMM1 and DIMM3 at 2.7vDimm, only had 2 errors running a test overnight...... from a post i've read somewhere regarding memtest86 and stability, it said that having less than 15 errors over a period of time may not even be a RAM problem, and could be caused by a voltage fluctuation coming from a sub-par PSU, which from what I understand the Raidmax units are..... i've got a Thermaltake Silent Purepower 480W unit on the way from Newegg, so if it's the PSU that should take care of it....... BTW, when I referred to having hundreds of errors within minutes it turns out that was my fault, I accidentally set my tRP to 2 clocks instead of 3 clocks...... and forgot to bump the vDimm up from 2.6 to 2.7..... sorry for any confusion, I must say the way it is running now, I am extrememly happy..... and thanks for your help and patience, i'll let you know how the new PSU works out.... and BTW, I would have tested the memory on another motherboard, however my brother is the only person I know with a mobo capable of running PC3200 RAM (same board as mine) and he's been hard to get in touch with lately, so I really didn't have any options to test on a different board...... and my DIMM tester at work only does up to PC2700..... Dave
Link to comment
Share on other sites

I'm running the MSI K8T Neo-FSR, version 1.1, K8T Neo (MS-6702) V1.2 BIOS, with an Athlon64 3000+, with a single stick of CMX512-PC3200C2, at 2-3-6-3, DDR Voltage (v) 2.70. No problems. I've got it in an Antec Performance Plus 1080AMG case, with the Antec True Power 430, power supply. ATM, CoreCell says CPU temp is 40C, System temp is 29C, while running the UD Agent, in the background. :D:
Link to comment
Share on other sites

  • Corsair Employees
Well let's do this; you have a new PSU on the way, correct? Let’s see what happens when you get the new PSU, if the results change then I would suggest testing the memory in another system the same way over night and see if you still get the random errors. If so, then I would say let's replace your modules, but if the problem does not show up in another system or they change with a new PSU then that would suggest that the problem is some where else.
Link to comment
Share on other sites

I will do that..... I think i've got it narrowed down to the PSU.... it seems, funny as this may sound, that the more fans I run off of the motherboard fan headers, the more unstable my voltage gets..... my new Antec P160 only uses 2x120mm case fans, my old case had 5x80mm.... and as I say, my voltages as reported by MBM5 are much more stable, although a bit low..... so the Thermaltake unit ought to help with that, plus it has an intake fan on the bottom right over the CPU, my Raidmax has the intake in the front, which I would think would affect the CPU cooling...... not sure on that one, though...... If all else fails and it does turn out to be the RAM, I am confident that all I need to do is let you know, if all else fails I can send them to you guys and have you test them for me....... Again, thank you for all of your help and I will stay in touch re. this issue. Dave
Link to comment
Share on other sites

it was my Sound Blaster Audigy, which was busy dying a slow death..... first the 1394 port on it died, and this morning it totally stopped working..... popped it out, ran Memtest86+ just for fun, and seemed to pass with flying colors.... i'd have never thought it was my sound card..... although i've heard horror stories in the past regarding Audigys and K8T800 boards.... Dave
Link to comment
Share on other sites

Unfortunately the Audigy was no longer under warranty, however, I guess even if it were, i'd be hesitant to put another one in this system after all the stuff i've read regarding problems people have had with them..... I picked up a Maddog Multimedia Entertainer 7.1 yesterday at CompUSA, which is based on the Via Envy24 HT-S, a little more hit on the CPU than the Audigy, but my processor can handle it..... seems to work great, and the new PSU is awesome...... had to do a little creative bending to get the drive-bay fan controller installed since my case uses rails, but it's all in and working great!!! Well, gotta run, getting ready for work.... Have a great day! Dave
Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...