So I overclocked my 2500+... (sigh)

Printable View

Show 25 post(s) from this thread on one page

10-30-2003, 03:16 AM
tituswolf

Quote:

Originally posted by ThAz
My old mobo MSI KT3 Ultra showed lower cpu temps with the same processor than my new EPoX 8RDA3G. The MSI should read the temp from the thermal diode and the EPoX uses the socket diode. I relly don´t know which one to trust, but i seams like the EPoX bord (probably all NF2) reads temps higher that they relly are.

I relly hate the manufacturers for going back to socket diode. Just because a cople of pepole with crappy coolers started to call and whine :(

msi boards are horrible for reading temps, their placement of the system temp chip is horrible. they actually placed it under the agp slot about 5 inches away from the gpu heatsink... was always seeing system temps of over 34c.. i wonder why.. as for msi's processor readings, they are horrible as well, was always reading my 2000+ temps at near 45c, put that chip in this asus and I get temps at less than 35c... i also had a kt3 ultra.. the most stable board i ever owned.. just so annoying little quirks
10-30-2003, 03:16 AM
tituswolf

Quote:

Originally posted by xDUCK
Then you have an A2 NB. I had an A7N8X that would do 190FSB stable for days, but 191FSB was lucky if I could benchmark.

a2 Northbridge? wat dat? alls i know is the board is a revision 2 with the nf2 ultra 400 chipset and 198 is all i can get out of her stable... oh i can increase it to 200 but it errors out of prime95 within a minute... it don't believe it is the chip either.. i have lowered the multiplier to 8 and ran it at 200 it still errors out.. just after 7 minutes. i have had 2 barton 2500s and the board behaved the same with both chips.
10-30-2003, 03:19 AM
mystifmagic

Quote:

Originally posted by tituswolf
a2 Northbridge? wat dat?

That's the stepping of the northbridge chip. Just like a CPU it's an identifier, and some steppings overclock better than others.
10-30-2003, 03:21 AM
thebove

Quote:

Originally posted by tituswolf
a2 Northbridge? wat dat?

the stepping (production cycle) of the northbridge. the first ones, the A2 couldn't easily reach 200fsb. like xDUCK said many of theA2 boards would hit a wall above 190 (mine included, 190 is good, but going further gets a little shaky). the later stepping (A1/A3) were pretty much guaranteed to hit 200fsb and beyond.

*edit* grr, mystif ;)
10-30-2003, 03:21 AM
tituswolf

Quote:

Originally posted by mystifmagic
That's the stepping of the northbridge chip. Just like a CPU it's an identifier, and some steppings overclock better than others.

see above... there really is no reason why it shouldn't be able to do at least 200
10-30-2003, 03:23 AM
mystifmagic

Quote:

Originally posted by tituswolf
see above... there really is no reason why it shouldn't be able to do at least 200

I was just telling you what it was. I don't have very much experience with nForce boards, so I cannot verify whether it should be able to do at least 200 or not.
10-30-2003, 03:25 AM
tituswolf

sorry, i wasn't trying to get huffy.... i was just pointing you to my edited post above your own... thanx for your info btw
10-30-2003, 03:36 AM
mystifmagic

Quote:

Originally posted by tituswolf
a2 Northbridge? wat dat? alls i know is the board is a revision 2 with the nf2 ultra 400 chipset and 198 is all i can get out of her stable... oh i can increase it to 200 but it errors out of prime95 within a minute... it don't believe it is the chip either.. i have lowered the multiplier to 8 and ran it at 200 it still errors out.. just after 7 minutes. i have had 2 barton 2500s and the board behaved the same with both chips.

Are we just talking Prime95? Does windows run fine? How about other benchmarks... how about games? If the only thing that screws up is Prime95, I say forget that and happily move on.

Quote:

Originally posted by thebove
*edit* grr, mystif ;)

;)
10-30-2003, 03:39 AM
tituswolf

nah, it includes 3dmark and pi, both error out pretty quickly... but 198 is rock solid stable.. it befuddles me.
10-30-2003, 03:40 AM
mystifmagic

Quote:

Originally posted by tituswolf
nah, it includes 3dmark and pi, both error out pretty quickly... but 198 is rock solid stable.. it befuddles me.

What voltages are you running? How are your 12v and 5v rails?
10-30-2003, 06:19 AM
Thunderbird1GHz

man ill be disapointed if the nf7 rev 2.0 wont even do 200fsb! some say it supports xp3200 but I doubt that since it was rated for 166fsb anyway. just that the new 2.0 revisions overclock very good, just like the tbred B's often did 2GHz, sometimes even more!
10-30-2003, 09:50 AM
hwmicre

Quote:

Originally posted by tituswolf
see above... there really is no reason why it shouldn't be able to do at least 200

]

I was going thru this and all I saw was that your Vcore was at 1.65V.
That can be too low for overclocking. Maybe I missed something, but have u raised it?

Sorry if this was done.....
10-30-2003, 12:09 PM
xDUCK

Quote:

Originally posted by Thunderbird1GHz
man ill be disapointed if the nf7 rev 2.0 wont even do 200fsb! some say it supports xp3200 but I doubt that since it was rated for 166fsb anyway. just that the new 2.0 revisions overclock very good, just like the tbred B's often did 2GHz, sometimes even more!

You're being ridiculous again, Tbird.
10-30-2003, 05:19 PM
tituswolf

yeah, i have had the vcore all the way up to 1.85, it may give me a few extra minutes of stability in prime95 but it always errors out.
10-30-2003, 05:22 PM
mystifmagic

Quote:

Originally posted by tituswolf
yeah, i have had the vcore all the way up to 1.85, it may give me a few extra minutes of stability in prime95 but it always errors out.

"What voltages are you running? How are your 12v and 5v rails?"

That includes your ram voltage; what is it running at?
10-30-2003, 07:41 PM
hwmicre

Quote:

Originally posted by tituswolf
yeah, i have had the vcore all the way up to 1.85, it may give me a few extra minutes of stability in prime95 but it always errors out.

Is that every test at that voltage, or just Prime - which, to beat a dead horse again, to me just isn't worth it.....
10-30-2003, 08:41 PM
tituswolf

just in prime and pi, 3dmark will run fine at 1.675. as for the vrails, i think the 12v is around 11.45 5v are 4.85, i forget the other one... the rails aren't that true.. as for vdimm i think it is set at 2.6 but i have tried up to 2.8 and it still errors out...
10-30-2003, 09:29 PM
thebove

Quote:

Originally posted by tituswolf
just in prime and pi, 3dmark will run fine at 1.675. as for the vrails, i think the 12v is around 11.45 5v are 4.85, i forget the other one... the rails aren't that true.. as for vdimm i think it is set at 2.6 but i have tried up to 2.8 and it still errors out...

12v rail is awfully low, 5v is kinda low too. what kind of a power supply is it? my first guess, based on that info alone, is that the psu isn't up to the challenge of overclocking that system. also, you could download SpeedFan if you don't already have it, and see if your voltages (core, 5v, 12v, and vdimm particularly) fluctuate a lot.
10-30-2003, 10:12 PM
mystifmagic

Quote:

Originally posted by tituswolf
just in prime and pi, 3dmark will run fine at 1.675. as for the vrails, i think the 12v is around 11.45 5v are 4.85, i forget the other one... the rails aren't that true.. as for vdimm i think it is set at 2.6 but i have tried up to 2.8 and it still errors out...

As thebove said your 12v and 5v rails are a bit low, that could hinder you overclock. Have you tried raising your vdimm higher than 2.8v? You should give everything a shot before giving up.
10-31-2003, 12:44 AM
KaoTiK

Quote:

Originally posted by mystifmagic
As thebove said your 12v and 5v rails are a bit low, that could hinder you overclock. Have you tried raising your vdimm higher than 2.8v? You should give everything a shot before giving up.

I don't think his mobo allows VDimms higher than 2.8v, which really sucks.

The 5v rail isn't too bad, but the 12v one is gettin out there. But it shouldn't be causing instability like that, I wouldn't think.
10-31-2003, 12:55 AM
mystifmagic

Quote:

Originally posted by KaoTiK
The 5v rail isn't too bad, but the 12v one is gettin out there. But it shouldn't be causing instability like that, I wouldn't think.

You would be surprised at how little it can take to make an overclock unstable.
10-31-2003, 01:52 AM
tituswolf

i should say that I am not sure at all what my rails are.. is what i remember it being but i haven't checked in a while.. i don't have access to my computer this weekend. my board doesn't go any higher than 2.8 on the vdimm. but i am still convinced its my board... not being able to do 200 no matter what the vcore or multiplier setting is a little weird...
10-31-2003, 01:56 AM
KaoTiK

Quote:

Originally posted by mystifmagic
You would be surprised at how little it can take to make an overclock unstable.

Saps computer ran with his 5v at 4.33, lol. And my computer is perfectly stable with it as low as 4.75.

But 1 extra MHz on the FSB shouldnt blow everything to hell. It's not like he'd have 4.9v on the 5v and 11.9v on the 12v and then puts 1 more MHz on the FSB and his rails drop to unstable levels.

Sounds like the board to me.
10-31-2003, 02:00 AM
lonewolfroger

Quote:

Originally posted by KaoTiK
Saps computer ran with his 5v at 4.33, lol. And my computer is perfectly stable with it as low as 4.75.

But 1 extra MHz on the FSB shouldnt blow everything to hell. It's not like he'd have 4.9v on the 5v and 11.9v on the 12v and then puts 1 more MHz on the FSB and his rails drop to unstable levels.

Sounds like the board to me.

Or his ram needs more then 2.8v`s vdimm to get higher.
10-31-2003, 02:17 AM
thebove

Quote:

Originally posted by KaoTiK
Saps computer ran with his 5v at 4.33, lol. And my computer is perfectly stable with it as low as 4.75.

however he also had to take a BIG hit in his overclock as the strength of his 5v rail came down, so it's not as if the lower voltage didn't affect him in any way, because it certainly did.

Show 25 post(s) from this thread on one page

All times are GMT -4. The time now is 02:47 AM.