IBM Thinkpad 600E bios mod for processor update

[Sharedoc]

L2 cache looks ok.

Here are couple of possibilities

(1) Maybe the processor runs too hot? This could explain why it does not boot up to windows but makes safe mode (which is much lighter computing task).
You can check this simply by booting in the ice box (hehe)

(2) Maybe there are some Win98SE drivers or initialization files that cause the problem. Safe mode booting omits these.

Do you have access to Ubuntu Linux DVD? Thinkpad can boot directly from Ubuntu DVD. Would be interesting to know if your system boots to Linux ok?

[x6xdemonx6x]

I know this is the 600 section but no one has responded on the 390 side so I thought I would try here I was wondering if anyone had any input on potential problems of takeing a p3 450mhz 390x to a p3 850mhz cpu? From what I can tell cache, fsb, and form factor are the same

[trillium]

Although there are some 390 (E & X) posts in this thread, I decided to respond here - http://www.wimsbios.com/phpBB2/post42605.html#42605 - where it might be more easily found by someone searching for '390' issues.

The short answer to your question - 'no problem'!

[trillium]

Sharedoc,

In regard to the possibilities you mentioned:
1) I'm sure it's not related to CPU temperature - I'm doing all this with the system partly disassembled and can check the heatsink by touch, and it's not abnormally hot at all.
2) Yes, I'm pretty sure it is, as you suggest, "... some Win98SE drivers or initialization files that cause the problem ...". The question is how to identify them!

As regards Linux. I had not ever tried it, but found and downloaded a version called Puppy Linux v2.12 which runs from a CD. It was only a 70MB download for the ISO image; I burned a CD and gave it a try. It did boot up on the system with the FSB modified to 100MHz and the P-III installed, but it took a very long time to complete the boot - almost 4 1/2 minutes! When I switched to the P-II/300, (450MHz with the 100MHz FSB), with the same CD the boot completed in a little under 2 minutes! What I was timing was the entire process from the disappearance of the IBM ThinkPad logo screen to the appearance of the Linux GUI desktop and intro screen; this required responses to some prompts about testing the video card but I think any delays would have been similar for both trials. I'm not at all familiar with Linux so didn't do much more than just test if it would boot, although I did note that the floppy drive worked OK with the P-II, but didn't work with the P-III. So the P-III system booted, but it seems something's not right!

I've been trying a few other things hoping to get Win 98SE to boot properly with the P-III, but still without success. I also installed the P-III/500 in an unmodified 390E system (66MHz FSB) with the correct '390E' BIOS. The BIOS says it's a "Pentium Pro/333MHz", which is consistent with it reporting "Pentium Pro/500MHz" in the modified, 100MHz FSB, system. And guess what - the boot hangs at exactly the same point! So it seems the problem's not related to the 100MHz FSB mod as I had thought! My experiences suggest that one cannot simply swap the CPU from a P-II to a P-III in a 390E! It may still be possible to get the P-III system to run Win 98SE, but it's looking more and more like it's necessary to re-install the OS! I did have some success when I did that before except that I had only 640x480 video, and after I installed the proper Neomagic video driver nothing worked any more!

I've compared the BOOTLOG.TXT file from a successful boot with the P-II/300, to that from an unsuccessful boot with the P-III/500. The files are identical up to the point where the P-III boot attempt hangs, and the next entry in the P-II bootlog file is "SYSCRITINIT = VMM". A Google search on that produced quite a few hits, but mostly they were just in listings of "BOOTLOG.TXT" files. However I did find a page titled "Description of the Windows 95 Startup Process" at http://support.microsoft.com/kb/174018 with the information:

"SYS_CRITICAL_INIT (SYSCRITINIT):
Interrupts are disabled during this phase. This gives VxDs time to prepare for device initialization without being interrupted by the system. No file I/O is allowed during SYSCRITINIT, so all SYSCRITINITs are not written to the Bootlog.txt file until after SYSCRITINIT is complete for all VxDs."

This seems to explain why the log of the failed boot process gives no information of exactly what failed - not very helpful! The file from the successful boot with the P-II shows what's supposed to be happening - a bunch of 'vxd' files are being loaded - but there's no way to know which one is causing the problem with the P-III. After more Google searching I learned that the 'vxd' files required by a particular system are set up in a file called "C:\Windows\System\vmm32.vxd" at the time of installing the OS, and that if additional 'vxd' files are later required by added hardware they are accessed at boot time from the folder "C:\Windows\System\Vmm32", where typically they have been put during the hardware installation. So I placed ALL of the 'vxd' files extracted from the Win 98SE CD into that folder, (and also replaced the existing "vmm32.vxd" with the 'virgin' one from the CD which was the recommended procedure I had found for fixing 'vxd' problems!), thinking that that should allow whatever file(s) that might have been missing to be loaded. BUT, THAT MADE NO DIFFERENCE WHATEVER!

At this point the only idea I have left is to once again try a clean install of the OS. If anyone reading this has other suggestions I'd be very interested to hear them.

I don't recall reading any posts in this thread about the 600E having problems when a P-II was swapped to a P-III, apart from the L2 cache not being recognized properly, and a fix was developed for that. The 390E doesn't seem to have any problem recognizing the P-III, L2 cache included, at the BIOS level, it's just when booting the OS - and so far I have tried only Windows 98SE. Can anyone contribute some insights on this, or has anyone else tried putting a P-III in a 390E???

[kocoman]

I got the grub patched for thinkpad to work, my 600e now run from 600mhz (before) to 750mhz now. you need to add the tpad to the menu.lst (under defaults = 0), but not in the kernel line.. it took me a while to figure that out. I never touch the settings that grub uses, only the kernel part I am used to editing. thanks

[beatoem]

I"m running the ol 600e @918mghz stable.
it it comparable to a p4 running approx 1.3ghz
* tp 390x ran the same processor (850mmc2) no problem,as did the 650+750 cpus from its original p11400spec,(contary to the so called EXPERTS ,from ibm,hp that imply 400mghz max capable speeds. on 600e,770,390e(no need on 600x,390x,hp4150b or non b laptops.
cpu msr seems better to enable l2 cache than pleap programme,less conflicts,boots faster +utilises no memory,in xp os anyway.
save the modified file then copy it to documents and settings/allusers/start menu/programs/start up folder in the c or d drive index,then upon reboot win will prompt you to direct to the program that enables the file to function,with the pointer checked in the square box to allow it to always use the particular exe file in the process.
my self rebuilt 600e battery is still showing 100% charge +over 2hrs+40minutes sufing the net on wifi ,it was rebuilt 25moths ago
goes to show if you don't use ibms deceptive battery software tools for the 600 sereis,and manually force win xp power managment never to suspend on low battery reserve, they go for ever ,what ever you do do not trust ibm's recommendations or sofware in relation to 600 batterys
they seem to instigate a counter that bluffs the battery into thinking its time to die!
certainly proves that the 600 laptops are not venerable as some may imply because of the known battery issues,set em up properly i strongly beleive they are the most durable post p11 laptops ever manufactured(period),no regrets whatsoever!yep,they don't build em like they used to.
*so do youself a favour,get yourself a trusty ol laptop,do the mods + the coin saved put it towards a new desktop for gaming instead,regardless no notebook in the planet could ever match a decent desktop,a 300mghz laptop on 8mps broadband will be always faster than a 2.4 core duo on 1.5 mps broadband all bx chipset mobiles are now fully supported in xp
with fully certified +proven stable drivers
where as recent spec mobiles require the time honoured method of manually installing the (sometimes buggydrivers before your ready to network.

[Sharedoc]

Congrats for 918MHz.

[beatoem]

Congrats for 918MHz.

credit to you sharedoc
i tell ya it only posted on the 3rd 600e motherboard
also lost yet another too
fortunately ,snapped up a couple of 600e with dud screens for less than the price of a good 650 processor (if you need another good 600e bios let us know!)
at this stage temps hover around 49-66 deg with 600x heatsink on the netin windows,and stable for many hours,however in a super intensive cpu load programm after 7 or so minutes
it peaks to 90deg and powers off
on another 600e @540mghz 600e heatsink,same test it peaks @88deg but never crashes,(only 2deg less)
the other weird thing is the 600x fan on the 600e seems to run in a pulsating manner sorts like 1/3 speed up and down and it never reaches
peak speed except at power on time.
i am pondering to boost fan voltage with a bypass booster switch to switch it on in time of need because it seems the lower speed is good for battery life,do you know on the 600e which voltage source is safe to use?
maybe i should try a 120 watt psu ! what do you think sharedoc?
cheers

[Sharedoc]

I tried once to use 5 volts from the USB socket to drive the cooler blower. However, it was not a big success. There was no noticeable improvement in stability but a big increase in noise.

I am now running my 700Mhz processor @ 756MHz using quite low Vcore to make the laptop almost silent.

Stays silent about 30 minutes from boot before the cooler starts to run (in room temperature)

BTW: how do you measure the CPU temp?

[beatoem]

I fitted dual hp4150 thermal mats with arctic silver 3 + 1.58 vcore
it passed a virus loading test (cpu killed the process after it got real hot)a
instead of blue screen,then immedietaly ran sis soft sandra high stress 98% cpu load,burn in test for approx 45 minutes ,temps averaged 85c +peaked @89c on the virus.theres no way any windows program +multiple tasks get anywhere near as hot,when i surf the net on 54 g wifi with all components intact +about 28c ambient it sits nicely around 48c and when i do other tasks etc peaks around 58c,i am content with the cpu temps after all i'm not that allows a virus to run in windows (unless i want it to),it seems to run very silent too ,how long i havn't measured but when the fan turns on it speeds up and down like a yo yo,quiet to mid speeds.oddly the 600e fan doesn't do this variable speed.
i wonder if theres a program that i can set the fan cut in period to later
or if theres a way to eliminate the variable speed controller so the fan can run constant like the 600e fan?
happy new year.

[Sharedoc]

Basically you can drive the fan from +5Volts and you can lower the voltage to a tolerable noise level using a power resistor or cascaded diodes (drop of 0.7 volts each).

[Rich600E]

Hi guys
I have been reading all the posts and all the links provided to set the FSB in my TP600E - with the SG577DYB clock chipset and 300MHZ PII

So what I did:

1. Disabled internal Ram by changing the value in the hex editor to 81
2. Installed 64MB Ram 66MHZ original IBM
3. Lifted pin25
4. soldered 1Kohm resistor to pin25 and connected to pin 48
5. boot

Nothing happens, the processor spins a little and 2 indicator lights light up. Nothing else. No bios nothing.

Am really going crazy why everyone got it working and I don't!!!

Please help me out! Where is the mistake?

THANKS
Richard

[Sharedoc]

Your 300Mhz processor propably does not overclock to 450Mhz. So you need to either to try another processor or undo the modifications.

[Rich600E]

Thanks for the quick reply Sharedoc!

hmm... not happy to hear that...
From what I have read in some of the postings, some people got the 300mhz procs to run overclocked. rottenmutt and trillium for example. Is it only a specific type of PII that works?
Or could it be because the ram is only running on 66mhz?

[pincellone]

All of a sudden the screen of my 600E upgraded to P3 700 mhz became bright white. I can barely see the actual content of the screen now.

I've tried to connect an external vga monitor and it works. What happened? How can I fix this problem? Help please! Thanks Alfred