Higher Vcore=Lower overclock???

[Dacs]

I tried overclocking my main rig (see specs below) and though it was a nice overclock, I feel that the CPU can do better (Tbred B JIUHB).
Strange thing though, I raised the voltage to 1.70 V (current is 1.65 V) and it won't overclock past 1866 MHz! It locks up just right past POST! How come?
I might be thinking that the CPU sucks too much power to the point that my crappy PSU can't handle it. Is that the case?
Thanks for any insights

EDIT: Corrected typos

[F_A_L_C_O_N]

possible, another culprit coul be heat

in first case check voltage reding at your last possible voltage where CPU still posts. Anything closing in on 5% deviation is bad especially when CPU is not even under stress

in second just check temp readings I remember I had Athlon once that craped at something like 52C and I could push the one I have now to 60C and beyond (during summer heat under high voltage) without any problems

[Dacs]

Too bad, my +5V is dipping 4.625 V. Maybe the PSU has crapped out. I should be investing on a quality PSU then
Well, I have a spare 200 W AT PSU here so for the meantime, I might use this to power up my peripherals leaving the 300 W solely for my mobos + fans.
I doubt the heat is the culprit. I'm running my CPU at 57°C max load stable without problems (damn this hot moist tropical weather!). And besides, I ran the PC past cold boot at night when it crapped out so I can rule out heat problem.
Any suggestions to make this CPU past 2400?

[drzaius2]

new power supply definately.. like right now.. NOW!!!..

you can cause damage to your components when the voltage is that low.. mainly to the power supply.. but it could affect other things.. like your CPU..

you want to get it past 2400mhz.. good luck.. get the best cooling available.. and up to voltage as high as you need to.

[drzaius2]

looking over your specs for your

System 1:
Athlon XP 1800+@2280MHz(15x152 MHz)
Jetway V266B mobo
256 MB DDR333 Kingston
yadda
GeForce 4 MX440 64MB DDR
yadda yadda

i assume this is the system you're trying to get up to 2400mhz..

you'd see a much bigger increase in overall speed if you upped the RAM to 512mb or higher.. and changed the motherboard to a NForce2 400 ultra based board.. and a new video card if you want games to run better..

or if you're just doing this to see how high your system can go without buying anything new.. then go right at it.. although as i said in my last post.. your power supply will die on you if keep using it.

[Dacs]

It's not that I need that extra MHz, I'd just like to see how far this thing would overclock. Anyways, I'd upgrade this Christmas season for new parts so by then I can use all the MHz I can squeeze from this CPU.
Thanks for all inputs Strange though, I've been using this PSU for almost 2 years and it didn't gave me any problems whatsoever.
Back to the original problem, should I expect this kind of behavior at all from a crappy PSU?

[drzaius2]

the low voltages are most likely because of the PS.. but also the motherboard could be drawing more then it should..

i had a MSI board and now an Abit, with the same PS.. on the MSI my +5 volt dropped to 4.8v sometimes.. while on the Abit it's always above 5v.. (usually 5.05v)..

[wrathchild_67]

Modern PSU's should have a line tolerance of no greater than 5%. Your PSU is dipping close to 10% from normal. Your mobo isn't an especially great OC'r mainly due to age and chipset. So even if you get a new PSU, your OC could be limited by the mobo simply because it isn't capable of handling a high frequency power hungry monster. Your processor may also be of the rare 1.5v default vcore variety (I know you said it started at 1.65v, but that may be the lowest that the mobo supports). If you do have a 1.5v TBred-B, they are notorious for not liking high voltages, which might explain your instability when upping the vcore. Another thing to keep in mind when overclocking is that increases in vcore may help to boost overall processor operating frequerncy, but is more prone to cause problems with the northbridge and RAM reads/writes and hence make your system even more unstable.

[Dacs]

Yes, in fact this is a 1.5V Tbred B variety. I'm just using 1.65 V because it's overclocked. And you're right, this CPU doesn't like very high voltages
Maybe I should be heading for a mobo upgrade then.
I think it would be in my best interest to leave this CPU as is, so no more overclocking for me now... not until I get a new mobo anyway
Thanks for all guys!

EDIT: I used my spare AT PSU to power up my HD/CDRW/fans leaving the ATX to the motherboard and yet my +5V is way under spec at 4.75V. I have no idea how much of a crap this PSU is
Anyone has an idea on how to mod the ATX with this AT to make a dual PSU?

[wrathchild_67]

Are your harddrives working correctly on the AT PSU? A couple years ago when I tried dual PSU's (ATX/AT like you), I found that I had to connect the harddrives to the ATX PSU because they wouldn't initialize properly on the AT.

[Dacs]

Well, it works along with the fans and the CDRW. As a matter of fact, I'm currently online typing this with that setup.
Is my PSU dying on me? I mean 4.75V without load from other devices aside from the mobo?

[ol' man]

Originally posted by Dacs
Well, it works along with the fans and the CDRW. As a matter of fact, I'm currently online typing this with that setup.
Is my PSU dying on me? I mean 4.75V without load from other devices aside from the mobo?

How do yopu know it is dipping down this low? Are you testing it with a multimeter at the mobo atx connector junction. I have had a few motherboards that told me lies in the bios. i was supposed to get another GA-8IPE1000-L but instead they RMA'd me a MSI NEO i865. It say the 12v rail is at 11.81v but when I test it ion the mobo it is much higher than that at 12.04v. Testing it at the end of a molex tells the same tale. I think my bios is BS'n me. Are you sure yours is not?

i had a MSI board and now an Abit, with the same PS.. on the MSI my +5 volt dropped to 4.8v sometimes.. while on the Abit it's always above 5v.. (usually 5.05v)..

Were you testing it with a multimeter or in the bios? A multimeter will tell the true voltage either at the end of a molex or more accuratly at the atx connector on the mobo. Simply ground the black on a black pin for that voltage or case screw and test the yellow = 12v, red = 5v, orange = 3.3v.

[drzaius2]

no i didn't test with a multimeter... but i had 2 of the same boards at one time.. Abit NF7-S.. and i found that the voltages reported by the bios (motherboard monitor 5.3.3.0 actually) were different.. i dont' see how 2 identical boards with the same bios could report different voltages unless they indeed were different..

i'm not really sure what i'm getting at here.. but i think that measuring the "incoming" voltage is a good idea.. as it's a good way to determine if the PS is not outputting enough or not.. but it can't measure the voltages going through the board.. and eventually what voltages are going to the CPU, etc.. and i would think that these are the voltages the bios measures.. (i could be wrong.. i have no idea where the bios measures the voltage)

even the best motherboard isn't a perfect conductor ..

[Dacs]

I wish that my mobo is lying about my voltages but the reading gives a reading of +4.92V if the rig is not overclocked and using default voltage.
I'm thinking of modding my ATX with my spare AT to be more powerful. Anyone can help me on this?

[NewBlackDak]

Originally posted by drzaius2
no i didn't test with a multimeter... but i had 2 of the same boards at one time.. Abit NF7-S.. and i found that the voltages reported by the bios (motherboard monitor 5.3.3.0 actually) were different.. i dont' see how 2 identical boards with the same bios could report different voltages unless they indeed were different..

i'm not really sure what i'm getting at here.. but i think that measuring the "incoming" voltage is a good idea.. as it's a good way to determine if the PS is not outputting enough or not.. but it can't measure the voltages going through the board.. and eventually what voltages are going to the CPU, etc.. and i would think that these are the voltages the bios measures.. (i could be wrong.. i have no idea where the bios measures the voltage)

even the best motherboard isn't a perfect conductor ..

I have a v2, v1,2, and 1.0. They all read different on the voltages give all the exact same hardware. I can read the rails at the molex though, and it's the same for all 3. Crappy sensors!!