CPU voltages => is this TRUE ??

[BEIF]

Hi all - BEIF here again !

As part of my ongoing search to fix my problem, I recived a reply post about my P4B-266 & 1.6A o/c issue that went roughly as follows:-

"..........on boot-up, the system delivers 1.5V to the CPU because that is its default value and *THEN*, once the BIOS paramenters are read into the boot-up process, only then is 1.65V is applied to the CPU - so in effect, the system will only use 1.5V to intitially boot up for the first couple of seconds and **THEN** it will apply the CPU voltage figure that is selected and stored in the BIOS.......this is what is commonly known as the cold-boot problem........"

- Is this actually TRUE ??

- I thought that once you saved your higher voltage in the BIOS, that was what was used to boot up the system from the instant you hit the power button ?

- do the P4B-266 series of boards suffer from the "cold-boot" problem ?

- if this is right, why doesnt everybody have to do the wire trick when they O/C ?

- I can upgrade to a P4B-533 for $50US - will I solve the problem by upgrading to the P4B-533 - does the P4B-533 also suffer from the "cold-boot" problem - at this stage the extra $50 would save me a lot of heartaches ?

All advice greatfully accepted.

Benny [BEIF] Ifin

 

[Duvie]

Well I think I mentioned this as well but it doens't appear to be my quote....

I had this issue with my msi board a long time ago before the wire mod...I ran into the problem once I got the system up to 133fsb and I needed 1.65v to be stable....then I had restart issues about 30 percent of time....

I got rid of board....

Next I got asus mobo ran same mobo at 133fsb and it still took 1.65v yet no problems...

DIFFERENCE!!!

The msi board undervolted severly at startup and idle running as low as .05-.06 from what I have set in bios...Asus mobo actually overvolted .03-.04....


I think undervolting of mobos can contribute to this issue....

Bios bugs????

Also since the wire trick tells the mobo that its default is 1.7v and not 1.5v when processor initializes at boot-up the board gives it that much...The fact this (wire mod) has helped so many gives obvious evidence that the board is is too blame for this....Even when boards go up to 1.7v in bios they would have issues. Yet a 1.7v wire trick works fine!!! I think the proof is in the pudding!!!!

 

[paralazarguer]

that is completely false information. Your system will power on even with no CPU installed and give you a beep code or voice warning if you have that kind of board. The system initializes the BIOS without a CPU installed and will boot the system with the vcore set in bios. The information you were given is completely wrong.

 

[paralazarguer]

If you would like more information about how the motherboard doesn't power the processor at first and must wait for the "good power" signal visit this link.

 

[Duvie]

That is full of crap!!! PS...I don't think so....

How can I use same PS all the time and have no issues, huh???? If power supply wont send good power signal that will happen whether one sets the cpu to 1.7v or leaves it at 1.5v....

I think you need to come up with a better explanation and some experience to back that claim up!!!!


Undervolting of boards could be a big culprit as well...either way the mobo and power distribution is likely the problem...


PS alone...ROFLMAO.....c'mon....

Same ps that has restart issues on one board racks up another 200mhz and higher vcore on another...It wuld have had problems in that board as well....

 

[THUGSROOK]

i agree with Duvie 100%.
if you have the problem the 1.7v wire mod fixes it 99.9% of the time.

the problem does seem to be manufacturer related. but at some point it will happen on any board if you overclock it enough.

example: a p4 1.6a / p4b266-C can run 171fsb all day long without any problems. it can even pass prime95 for hours on end. but it has intermittant cold boot and reboot problems. why?

it seems to have everything todo with how the bios handles vcore voltage.
some boards dont apply the boosted voltage until it detects the cpu - and it aint gonna detect the cpu if the cpu doesnt have enough voltage

anyways, the 1.7v wire mod fixed it up in no time and was now able to cold boot upto 188fsb.

check out this thread for more wire mod info.

HTH

 

[BEIF]

Hi Thugsrook & Duvie and others !

Thankyou for the advice - seems that the pin mod **IS** the way to go.

Just wondering Thugsrook, you wrote on another thread:-

".....06/19/2002 3:48 PM.....FYI: the new Asus P4B533 and P4B533-V boards do not need a wire mod...."

Could you please explain ?

=> (a) do you mean that the board allows you to go higher than 1.75V without a PIN MOD ?

or

=> (b) that the bios is written & setup such that it boots from power-button-on-stage-onwards at the CPU voltage that is selected and stored in the BIOS - ie: which is what the P4B-266 does *NOT* do ? ie:- so if I need 1.65V to run at 2.4gig, then when I cold boot on a P4B-533, the CPU will get 1.65V from the instant I hit the power on switch instaed of first giving it 1.5/v and then 1.65V - therefore causing my cold-warm boot poblem ?

If it is (b), is this definelty %101 the case, or is it just "net-rumour" ?

Reason I ask is that I *trust* you guy's alot Thugsrook, so if you say it is abolsute %101 fact that the P4B533 does boot-up from scrtatch at the user defined CPU voltage, then I will happily trade-up to the 533.

Can you please advise / confirm ?

Thanks again,
Benny [BEIF] Ifin

 

[paralazarguer]

I'm not familiar with that board in particular and it may require some sort of hack to get it to work right but the CPU should power on with the set voltage. I suggest that you read that PS guide and then look at a few others for info about the "good power" signal and the BIOS loading by battery power milliseconds before the system is powered on.

 

[Duvie]

Get over the PS thing....It isn't the PSU...I didn't have issue at all with the asus at higher speed and higher vcore....Also the vcore mod would be ineffective if what you say is true and I have seen it work for me and about a couple dozen ppl....


It is mobo related and some manufacturers likley have fixed issues...it happened more so with earlier boards...


The ps link you gave before told me nothing and did not apply to my case and many others...

 

[Mikki]

I'm not familiar with the wire mods and such, and I don't have any experience with the 266 board, but I can tell you for sure that the P4B533-E (and probably the P4B533) has an overvolt jumper giving you a range of 1.5v to 1.8v for the Northwood. Of course many folks will tell you that you don't want more than 1.75v going to your cpu. Hope this helps.......

 

[paralazarguer]

Duvie, I suggest that you read the article again about the "good power" signal. Without that signal, the system will not start no matter how many times you press the power button. As I said in my previous post, I no longer have any idea if it applies to this problem since i"m not familiar with this board and, as the article says, with crappy supplies, this signal is spoofed but it's there none the less. The whole process takes miliseconds but the bios settings are in place before the CPU is initialized. Power supplies 101. http://www.pcguide.com/ref/power/sup/

 

[Duvie]

I am done discussing this.....

I can succeed good power supplies are a must...PS 101!!!


However if I suffered from a bad power signal I would still be dealing with it today and it would happen to anyone regardless if they did a wire trick to tell the bios the northwood defaults at a different vcore then 1.5v....I have seen it help to many ppl with just that simple thing....Nothing is modded to the PSU....So nothing change bad PS will still be a bad PS, yet problem seizes....


I am done!!!

 

[paralazarguer]

As I said twice, I wasn't sure if it applied to this problem since I"m not familiar with this board. I just thought that technical tidbit would be useful. There's something wrong with the board and needs a hack. That's cool. I didn't know.

 

[oldfart]

Jeez. This cold boot thing is not exactly breaking news. This has been around since 815 chipset socket 370 PIII boards came out. This is the reason for the "wire mod" trick. Its a well known issue that affects many boards. The Abit BD7 had a pretty serious problem with it.

As I said in your other post, a 103 FSB o'clock should not exhibit this problem though. 150 FSB I could see, but not 103.

 

[WarCon]

Its not just an Intel thing either. Many people use to have to manually adjust their vcores by connecting the L7 bridges on T-birds when they overclocked to beat the boot up at default voltage problem that many boards had. Not a new issue definitely. Its an overclocker issue.....

You would think the mobo manufacturers would of clued in by now though. That is slightly disappointing to know that we are dealing with similar/same issues with new mobo designs.

 

[THUGSROOK]

Originally posted by: oldfart
Jeez. This cold boot thing is not exactly breaking news. This has been around since 815 chipset.....

do you have any idea how many ppl skipped that chipset and stayed on a BX? :Q lol

 

[BEIF]

Hi Thugsrook !

Dont know about you, but I *am* one of the ones that stuck with my BX [BF6] and skipped the 815 altogether - I guess that is why this is all so foreign to me...

I've checked all the links I know of and can find Thugsrook, and cant find any of them to explain how to PIN for a 1.65V default - instead of a 1.70V default ?

-> Do you know where I can find this info ?

-> Also, are you able to advise yet whether or not the P4B-533 definetly **DOES** boot-up at the higher selected CPU voltages [as oppossed to booting at 1.5v and then upping it during reading the bios] ?

As always Thugsrook, thanks, take care - yours and the advice of others here is truely invaluable !!!

Benny [BEIF] Ifin

 

[THUGSROOK]

1.7v is as low as you can go.
it will give you 1.76v viewed in windows.
do not use the overvolt jumper with the wire mod.

P4B533 does not need the wire mod.

 

[oldfart]

I had a couple 815 setups. They were both pretty good. Last one was a PIII-S 1.26 512K Tually @ 158 FSB, 1.5 GHz. Pretty fast and super quiet system. At least as fast and a heck of a lot cooler than a Woodie P4 at the time. I was not at all interested in P4 until the NW and DDR setups came out.

 

[THUGSROOK]

oldfart ~ i held onto my BX 750@1050 RAID system for as long as i could.
i made a sad mistake when i went to a P4 willy.

 

[BEIF]

Hi Thugsrook

Your wrote again:- on 07/10/2002 12:14 AM ".......the P4B533 does not need the wire mod....."

Please *dont* get mad at me for asking the same thing 3 times, but can you please explain what you mean by this statement and how you found out ?

=> Do you mean that the P4B-533 bios is written & implemented such that it boots-up from scratch instantaneously at the higher CPU voltage that is selected and stored in the BIOS & *not* the defaualt 1.50V ??

I've been searching everywhere to try and confirm this info, but so far can find nothing.

As I said before, I'm quite happy to exchange and upgrade my board [will cost me $50US] if I **KNOW** that I will definetly **NOT** have this same cold/warm booting problem with the P4B-533.

Please Thugsrook, can you confirm / advise ?

Thanks again.
Benny [BEIF] Ifin

 

[THUGSROOK]

whatever Asus did different - it works, thats all we really need to know

(yes i would assume that the bios sets voltage instantly - or it doesnt have to because it doesnt revert it back to default during reboot of shutoff)