P2.8C with 250 FSB max?

[caseman]

Anyone have an idea on what could be keeping my 2.8 from hitting anything over 3.5 (250FSB). It runs like a dream (besides a annoying windows update problem) at 250, but as soon as I increase it to anything above 255 it refuses to post and I have to clear the Bios settings in order to get it up again. Both the processor and Northbridge are running cool (32c at idle, 37ish under load) on a Asus P4c800-E dlx. I doubt it's the ram as I've got matched sticks of OCZ PC4000 gold. The PSU is an Antec true430. Changing the voltages doesn't seem to do a thing.

I should be able to head well north of what I am, but so far nadda, and having to clear the bios over and over again is getting less than fun.

Any suggestions would be most helpful

 

[Vapor]

Maybe get a bios update?? Report your voltages, too. Otherwise I would just say that your motherboard can't go any higher or something is wrong.

BTW, how did you get it so high and SO cool??? Some people have problems running the 2.8C that cool at stock speeds under load...how did you get so high and so cool???

Damn Rambus.

 

[lhgpoobaa]

heh. At such a speed and such a temp i would have to say the reported temps are way lower than what they REALLY are.
(i.e. the temp dumbing down factor).

Leave the side fo the case off, direct extra airflow at the NB/CPU, set ram voltage higher, ram timings to most conservative and try again.

If it doesnt work then your cpu or mobo has hit the golden limit.


Dunno bout you but i would be pretty happy with 250fsb.


<b>Lead me not into temptation.
I know the way myself.
Regards,
Mr no integrity coward.</b>

 

[caseman]

I've tried the CPU voltages from 1.6 to 1.7 (in .025 increments) and the Vdimm is at 2.75 (timings at 2.5-4-4-7 and a burst of 8). The Bios is the latest one, 1011 I think.

It's weird. After setting everything in the bios (at 250) it load windows no problem. As soon as I go to reboot however, it won't post again. Set the options back to what they were.. it posts and loads windows. Reboot, no post. Rinse and repeat. To make it odder, I rebooted a number of times earlier today with no hitches what so ever.

 

[caseman]

lhgpoobaa, it's possible that those temps are right, at least I hope. I'm using a SP-94 along with a 92mm Vantec tornado on the CPU. The case itself is running 5 80mm fans.

 

[lhgpoobaa]

what if you set your ram to be 2:3 or half of fsb? Just to rule it out completely.

Anyway if you look at the overclockers database you will see that most high end (2.6C->3.2C) P4's at the moment run out of steam around 3.4 to 3.6Ghz.
The ones that get higher usually have exotic water/cryogenic cooling.

<b>Lead me not into temptation.
I know the way myself.
Regards,
Mr no integrity coward.</b> <P ID="edit"><FONT SIZE=-1><EM>Edited by lhgpoobaa on 11/11/03 08:40 PM.</EM></FONT></P>

 

[Vapor]

I agree...maybe the temps listed were in C and he converted it thinking it was F into C once again (which would mean his core was at 98.6C...ok, so maybe he didn't do that). Either way, either that temp is miraculously low or the reading is wrong (much more likely to be the latter).

Do everything Poobaa said and then report back with what you got.

BTW, Caseman, what cooling are you using??

I'd be happy with a 250, that's for sure (evidently Case isn't though).

Damn Rambus.

 

[caseman]

The speed is more than fine, it's the odd effects of going over that prompted me to ask. Usually I can ratchet a board up until the OS becomes un-stable. This time it just completely refuses to post. Something I'm not really use to.

I'll try running at a 2:3 divider later tonight and let you know what I find.

 

[lhgpoobaa]

If you had a temp probe lying around i would check the temp of the base of the heatsink, just to be sure.

How hot does the heatsink actually FEEL?



<b>Lead me not into temptation.
I know the way myself.
Regards,
Mr no integrity coward.</b>

 

[Crashman]

As soon as I go to reboot however, it won't post again.

That could be caused by the fact that any voltage adjustments you make in BIOS don't initiate until after BIOS has been initiated. In other words, the board "starts" at stock voltage, then sets the new voltage after looking to see what you set it at. Since this takes a few milliseconds, you're running the stock voltage for a few milliseconds, and that might not be enough to get to the point where BIOS makes the voltage change.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[caseman]

The heat sink at least the part that I can get to feel nice and cool.. about the same temp as my desktop.

 

[caseman]

Vapor, I'm using an Antec 1080amg with 5 fans plus the cpu heat sink+fan. The case fans are 80mm (3 of which are panaflo, the other two whatever came with case). As to the CPU I'm running a Thermalright SP-94 with a 92mm Vantec Tornando sitting on top of it. I try to keep the apt temp between 75-80F.

The really nice thing about this board layout and the Tornado is that it shovels out so much air that after the air stream hits the mobo\processor half of it gets deflected right onto the Northbridge. Two for the price of one

If swifttec ever gets their shipping dept in order I'm adding a mcx159-r to the mix for the Northbridge (although looking at my set up I'm not sure it'll fit).

<P ID="edit"><FONT SIZE=-1><EM>Edited by caseman on 11/11/03 11:32 PM.</EM></FONT></P>

 

[caseman]

Damn Wusy, I was afraid someone was going to tell me that.

 

[caseman]

If heat isn't my issue, how would addional cooling help ( as fun as it sounds)?

The thought of running life-giving water through my nice new shiny machine makes me a tad nervous.

Hell, if I'm going to use water, why not go all the way and get a vapo-chill. The case alone costs over half of my whole PC.

<P ID="edit"><FONT SIZE=-1><EM>Edited by caseman on 11/12/03 08:59 AM.</EM></FONT></P>

 

[SJJM]

I going to say it is your ram that is making so you can not go any higher. try putting the voltage up to 2.8 and see if that helps. I have that ram too and it is what is holding me back from hit my goal 3.5 with my 2.6c. I can only hit 255 stable. Your luck cause you got a 2.8c and you are able to hit 3.5.

P4 2.6c at 3.25, Abit IC7-G Max3, OCZ pc 4000, 2X raptors in a raid, and ATI 9700 pro 390/324.

 

[caseman]

I think I bumped up the Vdimm to 2.85 at one point but I'll give it a try when I get home to make sure. I hope not since I bought those chips with the specific goal of OCing. Right now at 250 they are running right where they should be.

Thanks for the idea.

 

[caseman]

Too rich for my blood

 

[shadus]

Water is getting pretty affordable really these days. Vapo's a bit expensive still though

Shadus

 

[caseman]

Ok, new update. After much testing I am almost sure it is the way the MB handles the Vcore. I can play with every possible setting. Some lock up XP during the boot up process, some get me to the desktop before hanging the computer. All these things I'm use to. However, if I keep all the settings the same, but raise or lower the Vcore (anywhere from Auto, to 1.600, to 1.700 and everything in between) the machine won't post and I have to clear the CMOS. Odd huh?

So, after all that I've eked out another 5Mhz for a grand total of 255. My eyes hurt from finding that damn jumper and pins so I'm quitting for the night. Tomorrow I'll be back at it.

 

[caseman]

You just won't be happy until you get me water-cooling my system will you..

I'll take a look at the article.

 

[caseman]

How much worse could the Volcano 7 be compared to the Vantec Tornado? That thing is loud with a capitol L. At least I don't have to worry about the birds waking me up in the morning. All I can hear is the whirrr of the fans.

 

[HardWareBoss]

So buy an agressive radiator.