Decision: T60p UXGA vs. X61s

[RealBlackStuff]

I have been involved with computers for over 50 years now, starting with the IBM 1401.
I can't possibly remember every how or what or why.
But I do know that my method works, so why try doing it differently?

[BigCatAndy]

Hello RealBlackStuff

I have been involved with computers for over 50 years now, starting with the IBM 1401.
I can't possibly remember every how or what or why.
But I do know that my method works, so why try doing it differently?

I think, nobody has doubts in your qualification and experience. We all also appreciate you help and hints. But we all know that every problem can be solved in different ways. Exactly in this case I could try "my" way with re-packaging ISO, and if I fail, I try yours. I have no problems to admit, that I was wrong - finally, others will also profit from it (knowing, that only your way is correct one). And I will just waste some of my time.

But in the case of re-packaging ISO I have one thing, that could lead to failure - checksum of the file. As far as I know, ISO-files have checksum attached. What happens if I change the content of ISO without recalculation the checksum?

My "create you own thinkpad"-project goes further. Today I bought T8100, RAM (4 GB, not is also on the way. Actually, now I must somehow install BIOS to enable the Penryn CPU, then replace the display, and finally I can start with moving the system.

Regards,
Andrey

[wujstefan]

Thanks for confirming about the 4:3 model requiring a different way of flashing, and the extra detail along with that. I never knew Lenovo would have the time to mess around with identical models.

R61 4:3 was... different.
I have middletoned / modded around 20 of those. W7 32-bit bios swap is actually the easiest and most reliable way.

You may find same problem FSB-modding those systems. Some will not accept 266, no matter what. But that's out of the topics' scope.

[unixed]

... I find that it could be done the same way with phlash16 under a DOS environment ...

Yes, you can do this in principle, just flash only the *.FL1 file which you can rename if you want. You may have to decompress it beforehand. Flashing from DOS/freeDOS is safest (wrt flashing from any other OS or CD) but phlash16 must be used with due care.

BigCatAndy - yes you can create a custom iso, but just swapping the *.FL1 file and hoping for the best is probably not the way to go. The boot CD actually uses DOS to flash the chip so you could wait to hear from Screamer what switches he used with phlash16 and whether/how he decompresses it. You can create an autoexec.bat that runs Screamer's flash command. You should proceed with caution and try and run an innocuous command first.

RBS is right though, his succinct summary of the primary references I quoted just works with minimal risk subject to requiring minimal competence, which is why I referenced them in the first place.

Winphlash and the usual BIOS update methods use wrappers around phlash16 or the 32 bit equivalent so that people don't flash the wrong BIOS etc so it's best to have some idea of what you're doing if you bypass these.
If you do decide to flash from windows make sure nothing else is running that can interfere with the BIOS update.

... I can only guess that the 4:3 R61 use the TI chip from R60 for 1394/CardBus, whereas the widescreen models use the Ricoh chip. Hmm... Does it mean that CardBus also stops working?

lspci output:
15.00.0 CardBus bridge: Ricoh Co Ltd RL5c476 II (rev b6)

interpretation:
Field 1 : 15:00.0 : bus number (15), device number (00) and function (0)

Firewire, if it did show up (which it doesn't) would be something like:
15.00.1 Firewire (IEEE 1394) Ricoh ....

So presumably cardbus works (appropriate kernel modules are loaded) but I'm unlikely to be able to confirm that soon by plugging in some devices. I'm running debian GNU/linux.

[dr_st]

OK, so my gut feeling was wrong - the R61 4:3 does use a Ricoh controller, so only Firewire is disabled. Is there not a BIOS option to re-enable it?

[unixed]

I can't check that soon either but I'm reasonably certain there is no BIOS option that would re-enable it.

The Ricoh chips are probably different (I don't have a 15.4" R61), for instance on my 15.4" T61p the Ricoh chip is identically identified except for (rev ba) and it handles Cardbus (0), Firewire (1), SD controller (2), Memory Stick adapter (4), and xD-Picture Card Controller (5) i.e. 15:00.x where function x has been parenthesized.
Even though the BIOS is supposed to be plug and play my guess is some offset is hardwired for the 15.4" R61 Ricoh chip. The benefit of running a penryn dwarfs this.

[Screamer]

My 15.4" R61i appears to have a different Ricoh controller as your 15" R61 does, like your 15.4" T61p (revision ba).

I checked out the latest BIOS meant for the 15" R61, and it seems like it is missing the microcode required to run a Penryn. Someone could probably add a single microcode meant for Penryn in there, then maybe losing FireWire would not be on someone's mind.

EDIT: My eyesight and attention span is great, I made a mistake.

[unixed]

There are different BIOSes for the 15.4" R61(i), your (since it is an R61i) machine type would have to be 8932 to use the 2.24-1.08 Middleton or official/stock BIOS which forms its basis.

I should point out that the 15.0" R61 is (rev b6) so if yours is in fact (rev ba) then you've confirmed it is the same Ricoh card as in the 15.4" T61p. Nonetheless knowing your machine type and Ricoh card identification would shed some light on the matter.

Adding the microcode (I've done it) makes no difference whatsoever, the early CPU init can't handle penryn.

[Screamer]

I see, my mistake. I initially thought you were referring to your 15" R61, so I accidentally mistook it as revision ba.

From what I can understand from your attempt: you manually added the required microcode into your 15" R61's BIOS image, and then flashed the modified BIOS. Then due to some strange reason, the processor initialization.... can not handle it?

[unixed]

Add the penryn microcode to the Middleton 1.24-1.00 BIOS (15.0" R61), flash it, put in a penryn CPU, it still won't run. Replace with the original CPU, it runs.
Adding microcode is not a sufficient condition for a CPU to run.

[BigCatAndy]

Hello,
I have good news: I could install 2.24 BIOS (using "classical" Windows7-method with 32bit OS) and I also exchanged the CPU. So, now I have T8100 in it.

The next step would be exchange the display, and then - after RAM has arrived - I could start to install Debian 64 bit on it, and finally move my whole system on it.

regards,
Andrey

[BigCatAndy]

Hello,

The next step would be exchange the display, and then - after RAM has arrived - I could start to install Debian 64 bit on it, and finally move my whole system on it.

hardware modifications is over - display exchanged, RAM installed, SSD from T60p moved. I also tried Debian Live CD - runs quite fast. So, next I will install Debian on SSD.

One question: what is S-Video-port for? I had it in my T43p (never used), T60p does not have it, and now R61i (2 generations after T43p) has it again. What can I do with it?

Cheers,
Andrey

[Screamer]

The S-Video port was generally used for older 4:3 televisions I believe, as an improvement over those composite cables. They don't have much use, other than being a better alternative to composite cables.

[BigCatAndy]

Hello,

The S-Video port was generally used for older 4:3 televisions I believe, as an improvement over those composite cables. They don't have much use, other than being a better alternative to composite cables.

OK... I see why I never used it Earlier I didn't have a real TV set, and now we all use HDMI or DVB-C.

Yesterday I installed Debian 64 bit on my R61i (it is actually a Frankenpad now? Or still R61i?). It works, also display. I'm slightly unsatisfied with a performance of SSD - I have a feeling that Ubuntu on T60p (with ordinary BIOS) was faster (in boot time) than Debian now (with Middleton's BIOS). I think, I will check later in Debian forums about it. But the shutdown time is unbelievable - something 2-3 seconds.

I want to thank all members here, who brought me to this crazy idea and helped me to implement it That was my first hardware experience of this scale, without you I would not start it.

Kind regards from hamburg,
Andrey