P4 1.8A FSB 100->133 O/C on an MSI 645E Max2-LRU

Venom_IL

Distinguished
Jul 21, 2002
20
0
18,510
You know, this was supposed to be a walk in the park

"The Pentium 4-1.8A was a bit different and in addition to be it being a more recent processor, it was able to hit much higher core speeds. Using a SiS 645DX motherboard, we were able to easily reach the magic 2.4 GHz overclock on the standard 133/533 MHz FSB", they said

"Considering performance, stability, and availability factors, along with full support for the 133/533 MHz FSB and DDR333, the natural recommendation is a motherboard based on the SIS 645DX chipset", they said

(sharky extreme, and they even used an almost identical system to mine for their <b>successful</b> OC)
and they werent the only ones - everywere on the net ive seen the 1.8A O/Ced easily to 2.4, especially on 645DX boards, which i remember were calld once "overclockers' heaven".

so i went to the store and got myself a P4 1.8A, a SIS645DX based MSI mobo, 256MB Corsair DDR PC3000 and an MSI Geforce 4 Ti 4200 128MB. after i installed everything and made sure all components were orignial and of good quality (using Sandra and 3dmark) i set the FSB to 133, the FSB: DRAM ratio to 4:5 and the vcore to 1.55 and saved the settings in the BIOS

But NOOOOOOOOOO !!!! my luck has to suck!!! so the system cant boot. after messing around with every possible value in the BIOS, i have come to the conlclusion that regardless of the values of the memory voltage / memory frequency / CAS / timing setting mode / MA 1T\2T / Host to memory latency / graphic winsize / ACPI / Primary Graphic adapter / Spread spectrum / Unused PCI Slot Clk / Whatever - i cant boot my computer with 133 FSB, unless the Vcore is set to 1.6V (which is the maximum on my board) in which case the system boots but the OS (WinXP Pro) wouldnt start - either it starts loading up then gives me an error (which tells me that a driver or a HD is responsible for the problem) or it gives me the error even before the OS starts, or the system simply reboots in the middle of loading the OS, and one time the OS actually loaded up only to give me the error a few seconds later.
the error messages were written in white over a completely blue screen and they all started with "windows has halted your system in order to prevent damage to your hardware" or something like that. i should note that all the drivers on my system are the most updated, even the motherboard drivers.

I have tried maxing the DDR Voltage, and messing around with all possibe BIOS values (latency, host to memory...), to no avail.

Im afraid the only solution is increasing the CPU Vcore, in order to do which ill have to wait to the next BIOS update, that is assuming that such an update ever comes, AND that it will enable more than 1.6 Volts... and MSI arent exactly known as a company about which i can make such assumptions. i really hope hacking the processor (for extra Vcore) won't be a necessity because i have zero experience in such things.

so, what do u say guys? any suggestions ???

Thanks a billion in advance !!! :smile:

Rest easy...
Venom is here
 

Crashman

Polypheme
Former Staff
Read my sig. It could be a power supply giving too low a voltage, so that 1.60v setting is actually lower, like 1.54v. Otherwise, I do have a solution for you: The Williamette processor was also available in S478 packaging, which means that your motherboard does support higher voltages, it just won't LET you select them in BIOS.

Based on a 1.50v core CPU, you can raise detected voltage to 1.60v by connecting Vid2 to VSS. You will have to download the pin charts yourself to find out the locations of these pins. The reason you want to only force it to 1.60v baseline is that you will be able to raise it from there in BIOS. If 1.60v almost works, 1.65v should do the trick!

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>
 

Venom_IL

Distinguished
Jul 21, 2002
20
0
18,510
I have a Tsunami Virgo Blue for P4 350W, and under load (prime95) Sandra says the CPU Vcore is 1.57V (when set in the BIOS to 1.6V). So can the Voltage still be the problem here, considering the fact that most ppl were able to perform such an OC not increasing the voltage at all?
If the answer, as i fear, is yes, then im afraid im pretty much screwed because im not going to connect Vid2 to VSS for the following reasons:

1. I have no idea what a Vid2 and a VSS is :tongue:
2. Even if i do read some guide about the subject, i would probably burn the entire processor trying to hack it :tongue:
3. Intel has no way of knowing i raised the FSB, thus the warranty remains, but if i hack the CPU Intel could tell...

In any case, thanks alot for your help. now i guess i know for sure that a BIOS upgrade is what i need, one that would allow me to raise the voltage to 1.65... or maybe i should get a better Power Supplier.

BTW, do you think that the processor is at fault here too? i mean, could it be that my specific processor needs more voltage than other 1.8A's? so if i switch mine with another 1.8A it would be possible to OC it with lower voltage?

thanks again

-------------------------
<font color=blue>Rest easy...
Venom is here</font color=blue><P ID="edit"><FONT SIZE=-1><EM>Edited by Venom_IL on 07/21/02 05:32 PM.</EM></FONT></P>
 

Crashman

Polypheme
Former Staff
I've seen these things hit 2.4GHz requiring anything from stock voltage to 1.65v.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>
 

Crashman

Polypheme
Former Staff
Oh, hey, you MIGHT want to check out the BIOS hackers sites, they do a lot with the BIOS options!

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>
 

Venom_IL

Distinguished
Jul 21, 2002
20
0
18,510
BIOS hackers eh? sounds interesting...
got any URL? cos i couldnt find anything of interest in google

10x !

BTW, is your signature an exaggeration or is it really like that?

-------------------------
<font color=blue>Rest easy...
Venom is here</font color=blue>
 

Crashman

Polypheme
Former Staff
I'd say half of the hardware problems I've seen in here are caused by power supply issues.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>
 

Venom_IL

Distinguished
Jul 21, 2002
20
0
18,510
wicked, i had no idea power supply causes so many problems
oh, and sorry for stealing your sig color, but thats the color that fits my nick best (cash reward for the first person who understands why :wink: )

-------------------------
<font color=blue>Rest easy...
Venom is here</font color=blue>
 

Venom_IL

Distinguished
Jul 21, 2002
20
0
18,510
I've seen these things hit 2.4GHz requiring anything from stock voltage to 1.65v
Well crashman - you might as well add me to you database....
mangaed to get 1.65V by changing CMOS values (credit goes to ockk) - computer boots fine, OS starts fine, I start prime95 - everythings fine, but after 30 seconds i get "Hardware error, rounded 1.5, expected under 1.4" or something like that. I never knew this was possible, for a 1.8A not to run stable on 1.65V...
my CPU has to be the worst ever to leave Intel's factories :mad:
I can't go over 1.65 through CMOS as far as i know, so that leaves me only with VID pinning or something of the sort (which im really trying to avoid for warranty reasons), but first ill see what MSI support says about the BIOS update... wish me luck

-------------------------
<font color=blue>Rest easy...
Venom is here</font color=blue><P ID="edit"><FONT SIZE=-1><EM>Edited by Venom_IL on 07/29/02 06:18 PM.</EM></FONT></P>
 

Crashman

Polypheme
Former Staff
Wow! Well, you're really close then. So close in fact that I would check the voltage under load to see if it's slightly lower than set, like 1.62v or something!

<font color=blue>By now you're probably wishing you had ask more questions first!</font color=blue>
 

Venom_IL

Distinguished
Jul 21, 2002
20
0
18,510
Yes, it is 1.62 underload...
Problem is, switching my power supplier to a better one costs almost as much as it would to switch my CPU to a 2.26B... and there isnt any other way to change my underload voltage, isnt there?
in any case, i really hope MSI get back to me and release a BIOS update already, for crying out loud !

Oh, and I forgot to say that the problem with the CMOS is that i think it keeps three bits for the voltage (i have been told so in MSI's forum) - 8 combinations - from 1.475V to 1.65V in 0.25V increments, so i dont think i can get a higher voltage without a BIOS update, which will use more bits on the CMOS to store voltage data (using software measures, that is)

-------------------------
<font color=blue>Rest easy...
Venom is here</font color=blue><P ID="edit"><FONT SIZE=-1><EM>Edited by Venom_IL on 07/29/02 09:01 PM.</EM></FONT></P>
 

Venom_IL

Distinguished
Jul 21, 2002
20
0
18,510
Yeah, I guess you're right Crashman...
now all I have to do is persuading my dad to let me... which is unlikely to happen.
You know what happens? he keeps yelling at me saying that the whole purchase was based on that OC, and that after all my blabbering about it, i have failed to do it, and that we shouldve gotten a 2.26B blah blah blah - while he is the one preventing a successful OC by not letting me VID pin ! Bah !!!
But I guess you cant help me with THAT :eek: :tongue: :eek:

-------------------------
<font color=blue>Rest easy...
Venom is here</font color=blue>