MSI 845 ultra restart problem...

[Duvie]

I pretty much found no issues if you don't change the vcore voltage...The second you do you have a 50/50 chance of a failed restart....

I can defintely run 122 at default so far, which gives me my best memory scores...I can post at 133 with a higher vcore but restart issues abound...


Any mention of bios upgrade? I clearly think this board is not setup for northwoods as bundled software does not work with it correctly....

So much for features of ocing!!!!!

 

[Duvie]

I have managed to get a restart issue even at default...The voltage actual is reading 1.44v while it is supposed to be 1.5v.....Is it my ps??? It is p4 ready as advertised and runs athlon on picky ecs board no problems and all the rail are tight...


PLease help....intel setup makes my 10 plus athlon setups look like a breeze...even with via hassles...

 

[THUGSROOK]

typical lack of voltage during reboots & cold boots...
not to mention youre not even getting full voltage to begin with. (that low voltage reading was from bios right? dont trust software)

bite the bullet and get an asus or abit SiS645

would be nice if some other ppl would chime in here.....

 

[Ark]

Have 2 PCs with 845Ultra-ARU/ 1.6A (from 2 separate sources) - same 1.44V readings in BIOS and in Windows. Did not try higher voltage yet. Running BIOS V3.1.
Known problems - BIOS related:
- Both systems sometimes freese on reboot.
- All red all one green+3red LEDs on initial power up (after AC unplugged),
but systems run fine and LEDs will come to All-Green on reboot (both stystems)
- One system shows CPU temp at 41C, then CPU is absolutely cold (18-19C).

Time for BIOS upgrade, anyone tried v3.2?

 

[Duvie]

I am running version 3.1 according to sandra which is dated 12/27/01....According to msi's site that was a bios upgrade to fix temperature readings WTF!!!!


Now I am concerned I will waste money by getting another PS cause this is likley mobo....

What are your +12V READINGS like??? What is the brand and spec of your board?

 

[Ark]

I just checked voltage with digital meter on HD connector:
+11.94V and +4.99V
Same time Board reported 11.84V - I am sure this is reporting problem, not actual voltage.
Touched HS absolutely cold, but reported 42C.

 

[jonny13]

I get the restart problem on different settings, but found that with the FSB at 125, I reboots just fine. I got about 5-6 reboots without an issue. But, then the problem was that the RAM ratio was automatically set to 1:1, so the RAM was running at 125. This is terrible and I got worse performance than at 117. The spot where it switches is somewhere around 117, so now I just leave it there and it reboots fine. It wouldn't reboot at 119, 120, 121 or 130. Those are the other settings I have played around with. Also, my voltage at 125 and 117 is 1.60, so that doesn't seem to be the issue. Very strange all together. I am looking for a new board.

Jonny

 

[Draco]

I have this board, using bios 3.2. I've tried as high as 115FSB, and it ran games fine, but Prime95 failed after a little while, and I want 100% stablity. So I knocked it down to 110Mhz, prime 95 ran for 24 hours without any problems, 3Dmark, and RTCW all worked great but it still hangs on reboot sometimes and resets the bios back to 100MHz. This is right off a warm reset or a cold boot.

Must be a bios issue... hopefully MSI will release a new bios that addresses this, or maybe this board just doesn't overclock very well.

I have a quality power supply, so I'm not real quick to start blaming that.


Also, what are everyone's voltages, cpu, and more importanly ram. IS anyone adjusting the ram voltage?

 

[Pauli]

I've had my MSI 845 Ultra ARU running for a few days now at 133FSB, default vcore voltage, and no reboot problems whatsoever. Any attempt to go to 135 or higher has not worked yet - I've only tried increasing the vcore to 1.65 though.

 

[Draco]

<< I've had my MSI 845 Ultra ARU running for a few days now at 133FSB, default vcore voltage, and no reboot problems whatsoever. Any attempt to go to 135 or higher has not worked yet - I've only tried increasing the vcore to 1.65 though. >>




Pauli: What are your memory timings set to, and dram voltage. Defaults?

 

[Duvie]

I don't have the board aymore...I updated to the asus p4s333 sis board...

I did it to get past the restart problem, eventhough I felt confident I could have done the wire trick...I did it ultimately to achieve the higher memory bandwidth. I feel some of the argument of an oc'd northwood not competeing is true when coupled with slower pc2100 ram...I couldn't get any better then a 1:1 ratio and was getting about same mem scores I had when tweaking on my ecs athlon board...Now I get 500mb/s more which puts it just past rdram scores on a i850....Asus has already had a bios upgrade to give 4:5 support to ver 133fsb...damn i845 chipset will only do 1:1 and sometimes that is the only option after 125mhz...

I am hoping for a 3:4 atio offering at 133, it should deliver bandwidth score of 2700's....

 

[Skooch]

I've had an MSI 845 ARU for the last couple days, and it's been running my 1.6a at 16X133 (bios set for 1.55 volts) just fine since day one. I ran prime95 torture test for a few hours, and I've been playing SeriousSam 2 and MoH constantly, so Im sure it's 100% stable to my liking. After reading about all these issues getting the board to boot at any bus speed above 133 I decided to test it out. I was able to boot up into Win2k at 140, but Prime95 failed almost immediately. I will probably increase the voltage in the next couple days and see if I can get the cpu stable at 16X140. I think I'm running the 3.0 bios, but not positive, using one stick crucial 512 ram with Infineon chips with all ram timings at default.

The main reason I went with this motherboard is I had bought an ASUS P4S333, but the onboard voice/troubleshooting system was constantly saying "cpu not intstalled". I couldn't even get the board to boot up in jumperless mode, I had to set it to manual to get it to boot, and I was able to install, and run Win2k even though I apparently had "no cpu installed", that must have been a pretty powerful motherboard chipset to run a computer without a cpu lol! I even attempted to clear the bios using the cmos clear jumper, and even successfully updated the bios to 1004c, but the stupid onboard hardware monitor wouldn't shutup about "no cpu installed". I also attempted to disable the onboard hardware monitor voice setting in the bios, but it didn't work, so I gave up on the motherboard.

Forgot to mention everything running at default voltage except the CPU, but even with the bios setting at 1.55, the bios reads as 1.48 volts. I've also rebooted 15-20 times, and had zero problems getting the machine to boot.

 

[Duvie]

Yeh, but the problem for me occurred when the actual voltage was above 1.5v at a higher speed....since you are actually running 1.48v (seems like all msi boards under deliver or have voltage reading issues) that could be one reason...

This reboot issue generally occurs when you reach the point of the cpu needing above default voltage to boost...It is not a specific fsb..The reason the wire trick works is the fact the cpu tells or "tricks" the mobo into telling it needs more voltage, and the board delivers it immediately....Whereas I here the problem without the wire trick is initially the board still reads it as 1.5v and starts with that and obviously fails before it gets to the point to boost it to the desired voltage as set manually in the bios....

Now the asus sis chipset board has a jumper to plus the voltage from default out of the gate....

 

[Pauli]

Draco- yes, RAM timing and voltage is set to default. I was running the 3.0 BIOS, but I flashed last night to 3.2 and noticed that the CPU temp and Fan speed on PCAlert now seem to be reporting more accurate numbers.

Duvie- I'm going to experiment with the voltage now with the 3.2 BIOS and try to achieve higher OC as well. I'll post with the results.