- Joined
- Apr 20, 2005
- Location
- Gday from Melbourne Australia!
Hello, i was just wondering, i get a prime 95 hardware error that rounds 0.5 when expected <0.4 does this mean i have too much VCore? or does it mean too little?
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Prime 95 Errors?
- Thread starter -_{MoW}_-Assasi
- Start date
OP
- Joined
- Apr 20, 2005
- Location
- Gday from Melbourne Australia!
- Thread Starter
- #3
yea, i run 2 instances, core 1 seems to be the one that seems to error the most, sometimes core 2
and what does it mean when my pc locks up?
and blue screen of death with 0x000000F9 and 0x00000000A, any translations of those?
and when Verifying DMI pool data is suppost to pop up, but doesnt, and then reboots? what does that mean?
I know this is very basic knowledge, but i dont know em
help very appreciated
and what does it mean when my pc locks up?
and blue screen of death with 0x000000F9 and 0x00000000A, any translations of those?
and when Verifying DMI pool data is suppost to pop up, but doesnt, and then reboots? what does that mean?
I know this is very basic knowledge, but i dont know em
help very appreciated
OP
- Joined
- Apr 20, 2005
- Location
- Gday from Melbourne Australia!
- Thread Starter
- #4
bump to get noticed
- Joined
- Mar 1, 2004
it means you are unstable
could be lack of voltage
mobo errors
cpu errors
memory errors
basically, your puter cant run at those speeds wiith the settings you have chosen
time to tweak or lower your clock speeds
could be lack of voltage
mobo errors
cpu errors
memory errors
basically, your puter cant run at those speeds wiith the settings you have chosen
time to tweak or lower your clock speeds
- Joined
- Jul 29, 2001
I don't see a power supply mentioned anywhere in your sig, what PSU do you have?
- Joined
- Jan 29, 2003
- Location
- New York City
When the system locks up before the DMI pool data it usually indicates a memory error. Loop Memtest test 5 & 8 to see if your getting errors.
OP
- Joined
- Apr 20, 2005
- Location
- Gday from Melbourne Australia!
- Thread Starter
- #8
yea, mem test is fine, its a super flower 500 watt 14cm fan.
I just wanna know these things as a simple guide for latter use.
I pass memtest 3 times, i know my mem can handle it, so it could be my mem controller or something
I just wanna know these things as a simple guide for latter use.
I pass memtest 3 times, i know my mem can handle it, so it could be my mem controller or something
OP
- Joined
- Apr 20, 2005
- Location
- Gday from Melbourne Australia!
- Thread Starter
- #9
help?
- Joined
- Aug 31, 2001
- Location
- New Zealand
Get a digital multimeter and take some readings off a molex at idle and load.
OP
- Joined
- Apr 20, 2005
- Location
- Gday from Melbourne Australia!
- Thread Starter
- #11
where can i buy such a product and how does it work?
Does downclocking your prøcessor frequency make the errors go away?
All of our hardware hits a peak at some point... your processor may simply be peaked out for the temperature and VCore that you are currently running it with.
I've got the same board as you, and the same thing happens to me when I am running a largely unstable processor or memory clockspeed, and trying to coax it into Windows. Downclocking, lowering your temperatures, or increasing your component voltage (to a certain point) will help alleviate this - the issue is a by-product of unstable clockspeeds.
I hate to say it, but 3 passes of memtes86 is not enough to call your memory stable, especially not with these sketchy DFI motherboards. At the very least, run tests #5 and #8 for a few hours each.
With the DFI nF4 motherboards OCCT should also be used for memory stability testing, due to the rather serious VTT droop issue that every single one of these poorly designed boards is prey to (VTT will droop a great deal under certain load conditions, OCCT and memtest86 T#8 being the best examples).
If I were you, I would start by peaking each of your components independantly of the entire system, so as to determine the capabilities and limitations of each part in your machine.
First, drop your LDT multiplier to 2X, run your memory on a divider, and see how fast you can get your processor to run with stability. As you push your processor frequency upwards continue to overvolt without hesitation, so long as you continue to see signifigant stable frequency gains from each overvoltage step. Be thorough in your stability testing - you want to get a good idea of where your chip tops out with stability on your cooling.
Overclocking the processor seperately from your memory removes the memory as a possible source of error - when instability occurs you will know without a doubt which component has peaked. Instability is your means of probing the capabilities of your hardware.
Next, keep your LDT multiplier at 2X, drop your processor multiplier to 8X, set your VCore to 1.45V, and push your memory frequency as high as you can without compromising stability in OCCT and memtest86 tests #5/#8. This will give you an indication of the upwards capabilities of your memory. Your motherboard's HTT peak can be determined in the same manner, with the use of a memory divider.
It is also a good idea to work out the best timing configuration for your unique set of RAM, while you are pushing the memory independant of the rest of the system.
Finally, armed with an idea of where each component peaks out, you can troubleshoot your system overclock and determine which part is holding you back.
All of our hardware hits a peak at some point... your processor may simply be peaked out for the temperature and VCore that you are currently running it with.
I've got the same board as you, and the same thing happens to me when I am running a largely unstable processor or memory clockspeed, and trying to coax it into Windows. Downclocking, lowering your temperatures, or increasing your component voltage (to a certain point) will help alleviate this - the issue is a by-product of unstable clockspeeds.
I hate to say it, but 3 passes of memtes86 is not enough to call your memory stable, especially not with these sketchy DFI motherboards. At the very least, run tests #5 and #8 for a few hours each.
With the DFI nF4 motherboards OCCT should also be used for memory stability testing, due to the rather serious VTT droop issue that every single one of these poorly designed boards is prey to (VTT will droop a great deal under certain load conditions, OCCT and memtest86 T#8 being the best examples).
If I were you, I would start by peaking each of your components independantly of the entire system, so as to determine the capabilities and limitations of each part in your machine.
First, drop your LDT multiplier to 2X, run your memory on a divider, and see how fast you can get your processor to run with stability. As you push your processor frequency upwards continue to overvolt without hesitation, so long as you continue to see signifigant stable frequency gains from each overvoltage step. Be thorough in your stability testing - you want to get a good idea of where your chip tops out with stability on your cooling.
Overclocking the processor seperately from your memory removes the memory as a possible source of error - when instability occurs you will know without a doubt which component has peaked. Instability is your means of probing the capabilities of your hardware.
Next, keep your LDT multiplier at 2X, drop your processor multiplier to 8X, set your VCore to 1.45V, and push your memory frequency as high as you can without compromising stability in OCCT and memtest86 tests #5/#8. This will give you an indication of the upwards capabilities of your memory. Your motherboard's HTT peak can be determined in the same manner, with the use of a memory divider.
It is also a good idea to work out the best timing configuration for your unique set of RAM, while you are pushing the memory independant of the rest of the system.
Finally, armed with an idea of where each component peaks out, you can troubleshoot your system overclock and determine which part is holding you back.
Last edited:
