IBM Thinkpad 600E bios mod for processor update

[Sharedoc]

Beatoem,

Hehe, you got a dead motherboard? Well, **** happens.

You need to recall what modifications you did before the motherboard died. Then undo the mods. If it does not help, you propably have a dead motherboard.

In TP600E Bios-EEPROM is soldered to the motherboard PCB, so it cannot be swapped as in desktop machines.

[Sharedoc]

Katch,

Your PIII being a disguised Celeron would explain L2 being 128k.

What does WCPUID program show for your processor CPUID?

My 700MHz-PIII gives
Family 6
Model 8
SteppingID 3

[Sharedoc]

Borg#1,

I am not sure if I understood correctly your message.

Are you telling that you have tools that can compress a modified bios image (=binary code file) back to IBM TP600 bios format (with correct checksum)? So that a modified bios can be flashed to EEPROM in the same way as a bios update from IBM website?

If this is true, then maybe we could try to make a modified bios to TP600E to correctly enable PIII's L2.

Any other features we need for the modified bios?

[Borg Number One]

Why converting a PhoenixBIOS image into IBM-compressed format????
It is not necessary to do this.


You just need to do following to get a decent, uncompressed, "modify-able" PhoenixBIOS:

Update your IBM-Phoenix-BIOS with the latest available BIOS-update & Phoennix-WinPhlash.
(use the mentioned method in the previous threads to get the "$*.fl1"-files)
->we reach to get the latest BIOS for modification.

If the BIOS-file ("$*.fl1) cannot be load with Winphlash,
then it uses a totally different compression, which Phoenix's WinPhlash does not handle.

In this case use the flash util(s)/instruction(s) which come(s) with the BIOS update itselfs.


After updating your BIOS take WinPhlash.
To create an uncompressed BIOS backup image (with some bytes of additional uninteresting data at the end) you have to use any* BIOS-backup file in the line:
"Specify new BIOS file"

*=I mean any Phoenix-BIOS file...
-it should be accepted by Phoenix WinPflash
-it should have the same final size like your current BIOS's content.
(256K/2Mbit, 512K/4Mbit, 1MB/8Mbit)
I already mentioned some examples in previous threads...(acer, medion and some IBM-BIOSfiles)



Then select: "Backup BIOS only" and choose a location for the BIOS backup image.


After this, press the button Advanced Settings.
If you get an error message, then the BIOS-file will not be accepted by Phoenix WinPhlash.
Then deselect all clickboxes and press Ok.

Now, press the button: "Backup BIOS

Do not use the foreign/external BIOS-Image to flash it...that would be really stupid.




After Backup you can load the BIOS backup image to Phoenix-BIOS-Editor.


How to build a new BIOS after modification is secret.
If the secret(s) would be publish here, Wims(the author of this forum) would get in trouble with Phoenix Technologies....


If you want to have a modified Phoenix BIOS editor, get in contact with me or take google to find the secrets.

[Sharedoc]

Borg#1,

Still not sure if I understand correctly. Which alternative are mean:

(1) start from IBM TP600E bios, open it, modify it and pack it to Phoenix bios format, flash it using Phoenix flasher

(2) start with some generic Phoenix bios, open it, modify it, and pack it to Phoenix bios format, flash it using Phoenix flasher

If (2) is true, how can I be sure resulting bios runs ok in TP600E?

Apologises for being a slow brain now.

[Sharedoc]

Beatoem,
Can you post/email a picture of the top side of 750MHz-MMC2-PIII?
I need to compare it to my 700MHz-MMC2-PIII.

[Borg Number One]

Sharedoc, I do not know how to explain it in another way which would be understandable for you.
I described the method step by step!!!

Why do you mention the method 2???
I did not described this and it would be really stupid to do this.
I will be angry!!!


Is this /are these your BIOS(es) ?

http://www-307.ibm.com/pc/support/site. ... SHY-46HLKQ
->ftp://ftp.software.ibm.com/pc/pccbbs/mobiles/spsdin36.exe

http://www-307.ibm.com/pc/support/site. ... SHY-43DNQ5

Which one?

[Sharedoc]

Borg#1,
Thanks for confirming (1). Just had to make sure I am in the right ballpark. Have to take a closer look to the tools&steps...

[Sharedoc]

My TP600E is 2645-4A0 series running the INET36WW-bios

This is the bios link

http://www-307.ibm.com/pc/support/site. ... SHY-46HLKQ

[Borg Number One]

1.)
Take the Winpflash Utility from here:
ftp://ftp.support.acer-euro.com/utiliti ... tility.zip

2.)
Take any PhoenixBIOS update, for example by/from acer:
ftp://ftp.support.acer-euro.com/noteboo ... _intel.zip
...
or other BIOS updates.

3.)
Add the wph file from the 3a06_intel.zip to the Specify new BIOS file -field.

4.)
Choose: "Backup BIOS only"

5.)
Choose: "Advanced settings"
Then deselect all clickboxes and press Ok.

6.)
Now press: "Backup BIOS" and send me the backup file.
(compressed!!! (rar3 -> progressiv; zip ->both in/with highest compression)


PS:
ftp://ftp.software.ibm.com/pc/pccbbs/mo ... sdin36.exe
is a self extracting imagefile-to-disk executable.
I could start it in Bochs to get the Disk-Image instead building a real Disk.
After this I could extract the "$0044000.fl1" -file from the new built disk image.

Well, I am afraid the $0044000.fl1 does not seem to be neither an Award nor a Phoenix BIOS....or it is protected/encrypted anyhow.

So, your BIOS backup is necessary.

[beatoem]

[quote="Sharedoc"]Beatoem,

Hehe, you got a dead motherboard? Well, **** happens.

You need to recall what modifications you did before the motherboard died. Then undo the mods. If it does not help, you propably have a dead motherboard.

In TP600E Bios-EEPROM is soldered to the motherboard PCB


DUNNO,i was trying all these different cpus then this 333cel
felt real hot,after that no post,lights+fan goes on but no post,i'm aware i need to solder the chip,admittingly i reverted the fsb to default,so no way was the fsb the problem, willing to have a go,but needs to be sure
chip location +which chip it is ,any idea?conversely you got the same ver36 bios as me, i got a full ballooned tp 600x bios + a unballooned ver36 bios i can mail them if it helps,theres also a 47 bios for the 600e,i wonder will that work maybe by chance its got p111 support,hp4150 is weird,aside that it doesnt detect coppermine 450-500 non speedstep cpu ,it fully detected+worked a treat with the 650+750 cpus (lower speeds)it scored some very impressive memory bench scores out doing a duall 933 dell rig despite 600mghz how do i upload a pic in the forum?,conversely can i mail the info (including the flash tools for windows
+the tp600x+600e files?conversely i got few tp600 (mmc1)tp lowers
floating about it would be wise if maybe the bios hacks can be tested
on these first (seens they have a lcd/video issue)it would be nogreat loss if it went amiss,thus if all went well (despite he fact theres no p3 cpus)for them,a bios is a bios +testing is testing.

[beatoem]

i can send the full 600x bios for evaluation,whats the addy send that file?i will try later to xtract the full 600e (36)bios ,unless sharedocs already done so, the 600x has the microcodes we need its about 1.4mb big.

[Sharedoc]

Beatoem,

Just a wild guess but if the motherboard died when you were overclocking with increased Vcore, then you may have toasted first the processor core and then either Vcore (1.60V) or Vio (3.3V) power supply regulator(s). Vcore is regulated by the MMC2 module, so trying with a different mmc2-processor should cure that. Vio is regulated by the motherboard, so if it is gone, it is gone.

Look at the bright side, now you have a good clock generator available for a turbo-pll, hehe.

----

I have unpacked inet36ww bios using some other bios tool earlier, so I have the binary code. I am not sure if the upper half of the code block is valid. So please send me the code and the 600x bios code as well, so I can compare. I pm my direct email address.

----

If you need to buy a new motherboard for your 600E, you might consider buying a 600X motherboard. It does not fit the housing completely, so you need to apply hacksaw and other power tools to fit. Also the modem board is different. But you will get full PIII support.

I have one 600X motherboard which I don't know if it is good, since it doesn't boot with my modded 700MHz-mmc2-PIII. I need to give it a try with another PIII some day. I bought it for 20? from eBay.

[beatoem]

[quote="Sharedoc"

Look at the bright side, now you have a good clock generator available for a turbo-pll, hehe.


carefull or else i'll hack in to the ol board + go turbo+you know who's
gonna help me?you ,conversely if you need to see how far the fsb can go i can send over a nice copprmine celeron 450/100 to see what fsbs are possible,you can try on 600x too,first file is on its way ,it is nearly 2mb bear that in mind.

[brokencase]

Greetings all, I am glad to have found this forum. I was about to give up on the old 600e. I am now in proccess of upgrading to a PIII as well. Much thanks to Sharedoc for his investigations.

One question I have is with regards to Linux and enabling the 256k cache. There is a kernel boot parameter that allows the user to override the cache detection. I beleive in this case it would be "cachemem=256K". Just wondering if anyone has had success with this.

Other thoughts...
Some have raised some concern about the higher heat output being trouble. I looked up some specs and most of the PIII MMC-2 have power dissipation close to the earlier PII MMC-2. At least when the are operated in the lower speed step mode. I believe it is 14.4 W for the 750/600mhz PIII. The 366 PII MMC-2 was 13.8W

Regarding SpeedStep. Sharedoc, am I correct in assuming that your mod is simply pulling #GHI low and hoping that with battery boot up that one is lucky to cross through the "Deep Sleep" state in the proccessor?
Maybe a simple add on circuit could be made to monitor bus clock and toggle the speed step line at the opertune time? I wonder if during a suspend one could toggle the line manually and then come out of suspend in the faster speed.

Seems like a lot of work, but some of us still like our 600e's