A64 3000+ & MSI K8N Neo2, Windows Stability Issues

[Clevor]

Its been rock stable at 2-2-2-5 DDR400 2.6v. I even changed the timings to 2.5-3-3-7 DDR400 2.7v. Both times, the memory was fine. I even ran the ram in single and dual channel of and on for testing.

When I was able to boot into windows, the screen would flicker. And when I restarted my pc, it would take 5-10 attempts of restarting again and again, until I could get into windows. So, I know its not the memory. The memory is amazing, its rock stable and everything, $220 well spend right there, great job PDD Patroit.

2-2-2-5 was OK at default clock of 200 fsb, but at around 215+ I started having issues. The PQI though seems stable at 236, 2-3-3-6.

With the right CPU and ram, I've found this MSI board to be surprisingly stable (considering my previous experience with their P4 boards). Never fails to boot with various types of ram. The board doesn't seem to like 2.5-4-4-6 with TCCD, preferring 3-3-3-6, which is faster anyway.

I tried EB3700s last night and they won't run -2-2- whatsoever. However they do 252, 3-2-3-5 @ 2.85 volts which isn't bad, as that is faster than 2.5-3-3-6. Most the TCCD stuff have to do 250+ at 2.5-3-3-6, so from 250-252 anyway, the ram is faster. Clocked 39.5 ns in the Everest latency bench.

 

[CandymanCan]

He hasn't responded in over 4 hours... I blame it on Sententia.

noob, you should blame it on youreself for being Incompetant

 

[Graphicism]

noob, you should blame it on youreself for being Incompetant

It was a joke... joke he told him to flash, he didn't want to flash risky etc.. and than the long silence. Get back in the cheese CandymanCan.

 

[CandymanCan]

Oh i thought he was serius i was like wtf lol ok wel lsorry if i didnt know it was a joke

 

[Suma]

2-2-2-5 was OK at default clock of 200 fsb, but at around 215+ I started having issues. The PQI though seems stable at 236, 2-3-3-6.

With the right CPU and ram, I've found this MSI board to be surprisingly stable (considering my previous experience with their P4 boards). Never fails to boot with various types of ram. The board doesn't seem to like 2.5-4-4-6 with TCCD, preferring 3-3-3-6, which is faster anyway.

I tried EB3700s last night and they won't run -2-2- whatsoever. However they do 252, 3-2-3-5 @ 2.85 volts which isn't bad, as that is faster than 2.5-3-3-6. Most the TCCD stuff have to do 250+ at 2.5-3-3-6, so from 250-252 anyway, the ram is faster. Clocked 39.5 ns in the Everest latency bench.

Hmm..
There's a difference between the PDP stuff and the PQI.
They use the same TCCD chips but the PCB is different.
If I recall the PDP is using the newer BrainPower PCB as the PQI is using the older BrainPower PCB.

I'm using the PQI myself, and I'm blessing it everday, great RAM!

Suma.

 

[Clevor]

Hmm..
There's a difference between the PDP stuff and the PQI.
They use the same TCCD chips but the PCB is different.
If I recall the PDP is using the newer BrainPower PCB as the PQI is using the older BrainPower PCB.

I'm using the PQI myself, and I'm blessing it everday, great RAM!

Suma.

The PQI runs even better on the DFI LanParty UT3, doing 252, 2-3-3-6 @ 2.9 volts. I got 2x512 OEM sticks early on from Newegg. These sticks did 260 or so at 2.5-3-3-6 on an IS7-E at 2.8 volts. I bought a second pair after they sold out. These are the 2x512 DC kit. I tested on the same IS7-E and this pair did 264 @ 2.6 VDIMM! I benched 3D to check stability, and couldn't believe I was waltzin' the 3DMarks that high at that low voltage. These are spares, so I did not test further or different voltage. I can't imagine how these will do on the MSI board .

 

[deception``]

The temps are fine, around 25*C for case, around 40*C for cpu. My rails are around, +3.3 = 3.20, +5.00 = 4.95, +12.00 = 11.67.

Those rails are pretty low, especially the +12 volt. I'm going to foot the blame on the power supply here. My guess is that, as I mentioned earlier, the Powmax simply cannot provide you with a stable system.

-EDIT- Oops, I see that I have responded too little, too late. Sorry for the double post.

deception``

 

[Neo_peter]

11.67V on a 12V line doesn't seem that bad. That's 97+% of the rail. I'm not saying it's not a dead PSU, but i'm just thinking, if the only data I was given were the voltages stated above, that's not enough to convince me.

 

[bobmanfoo]

but those rails are in bios, imagine what they would look like with all his pc at full load, im thinking it would drop below 11 or real close to it

 

[CandymanCan]

yea prob i wonder what the rails are when using a multimeter

 

[deception``]

but those rails are in bios, imagine what they would look like with all his pc at full load, im thinking it would drop below 11 or real close to it

Dropping below 11 is not typically possible. However, a +12 volt that low is a clear indication that the power supply is struggling to feed the system enough juice. Furthermore, that reading is most likely at idle, and is sure to fluctuate even more under more intense conditions.

deception``

 

[icedogg9999]

The lowest it was in windows, using MBM 5, was 11.50-11.60v. 3.0v for +3.3v, and 4.8v for +5.0v.

I'm going to order the Fortron 530w psu tonight from newegg. Hopefully I'll get it friday/saturday with 2nd day shipping. I do believe its my psu though. Everything running at stock, new parts, never any stability issues before, etc.

 

[Neo_peter]

There are voltage regulators in any piece of hardware that are very sensitive to voltage. And because the voltage is regulated, once the load is high enough to pass the threshold of the voltage regulator, you'll see dramatic drops in the voltage as in 20-30+% instead of a mere 2.x%. That's when you know your power supply need more juice. I mean the manufacturing specification all for the PSU to fluctuate up to 5-10% in some lines. For those of you that know motors, those are the things that draw up to 3x their normal load when it starts up, and while you're in bios the hard-drive has already spun up, there's no big motor which will cause a sudden dip in the voltage so if you can get 11.67V on boot, chances are while you're botting, you're not going to have a voltage dip of more then 10%.

And if the lowest in windows using MBM5 is 11.5V (<5% difference), that is good as far as a PSU manufacturer is concerned, cause the product is built to be within that range.

 

[icedogg9999]

Ok, I just ordered a Fortron 530w psu off newegg. It should be here Saturday, and Sentential said it should fix my instability problem. And everyone else has suggested that to me too.

 

[icedogg9999]

Well, I just received my Fortron 530w psu in the mail today. I just installed it, and windows xp still freezes at its plash screen. Right now, i'm re-installing windows again. I'll report back when installtion is complete.

 

[icedogg9999]

Oh, btw, my rails are perfect. So I know thats not my problem right now.

 

[bobmanfoo]

have you tried unplugging the other hd or drives? seems like a hardware conflict, do you have a sound card? maybe try disabling the onbard sound

 

[ewalk123]

I was having the very same problem with the blue screens at the windows loading screen.

Try uninstalling the nVidia IDE drivers. It worked for me.

 

[icedogg9999]

Well, i just finished installing windows a little bit ago. I installed the nforce3 system drivers and ethernet drivers. My system is 100% rock stable at stock speeds. The Fortron has fixed it. My rails are perfect, i've restarted windows about 10-15 times without any problems loading it back up again. Which I had problems doing with my old Powmax 400w psu.

-Icedogg

 

[ewalk123]

Good to hear. If you start having the same problem again, try uninstalling the nforce3 IDE drivers.

You can do this in "add/remove programs" and just select to uninstall the IDE drivers only.

Glad you are stable! GL.