If you look closer at the note at 33:27, you can notice that POST codes indicate the last phase of startup that was completed successfully, which means that the error actually seems to occur at phase 06, not 05, so the SM-RAM is probably okay in your M3.
Dang, I had totally forgot this could be the case and you may be right that the SM-RAM is not the culprit. Looking at the Libretto 100 manual for phase 06h it does have a mention for "SM-RAM base rewrite" but does go on to do a number of other checks. Some operations with SMI, HDD hardware reset, CMOS defaults, and initialization of the keyboard controller. The Libretto 50/70 manual indicates the next phase 06h is ROM/RAM copy, while that is in phase 05h for the Libretto 100. I guess where I am stuck is what manual to even reference. The Libretto 100 seems most logical in that they both share the same "In touch with tomorrow Toshiba" screen at first power-on, and having nearly identical configurations for the CPU, RAM, 32-bit Cardbus, and Video controller. I think a next step would be examining a BIOS dump from the working M3 to better understand what occurs at this step and possibly comparing to the Libretto 100 BIOS.
I had a libretto 50 that would do a similar thing. Ended up being broken solder joints on the ram. Check the board for a flex and see if anything got loose. Good luck with repairs!
Yeah thats a tough one, if I knew for sure that was the case I could try baking/reflowing the board. I've only done this a couple of times successfully and that was a long time ago. Do you have a guide for this? I don't have a fancy reflow or toaster oven. The thing I am most worried about is how densely packed both sides of the board are, so this may be one of the things I try as a last resort.
I have a stark feeling you have an M3 that might be very touchy and needs that backup battery installed. but given that laptop also just made its way to the US from Japan, I'd also check the solder joints on your surface mount chips there for anything out of place with a loop and tweezers.
This is a really great suggestion, I will test this and follow-up with the results. The Librettos are very touchy overall so I could certainly see the BIOS battery being the culprit. Will also take a closer look at the board as well. Thanks!
I haven't given up on mine yet, hoping to do a follow up to this soon. Signs point to a RAM issue, but Toshiba stacked the onboard memory using an interposer type board making this difficult to probe and check for broken solder points. Are you seeing any parallel POST activity on yours?
Yeah, this is really annoying about the M3 specifically since I believe the dock is only compatible with that one model. The others like the Libretto 50/60/70 and 100/110 had multiple versions of the dock available and compatible with several models. Since the dock is "mostly" pass-through I may be able to get the pinout for the parallel port at least, in theory you could solder to the internal pins to get the POST code signals.
If you look closer at the note at 33:27, you can notice that POST codes indicate the last phase of startup that was completed successfully, which means that the error actually seems to occur at phase 06, not 05, so the SM-RAM is probably okay in your M3.
Dang, I had totally forgot this could be the case and you may be right that the SM-RAM is not the culprit. Looking at the Libretto 100 manual for phase 06h it does have a mention for "SM-RAM base rewrite" but does go on to do a number of other checks. Some operations with SMI, HDD hardware reset, CMOS defaults, and initialization of the keyboard controller. The Libretto 50/70 manual indicates the next phase 06h is ROM/RAM copy, while that is in phase 05h for the Libretto 100.
I guess where I am stuck is what manual to even reference. The Libretto 100 seems most logical in that they both share the same "In touch with tomorrow Toshiba" screen at first power-on, and having nearly identical configurations for the CPU, RAM, 32-bit Cardbus, and Video controller. I think a next step would be examining a BIOS dump from the working M3 to better understand what occurs at this step and possibly comparing to the Libretto 100 BIOS.
I had a libretto 50 that would do a similar thing. Ended up being broken solder joints on the ram. Check the board for a flex and see if anything got loose. Good luck with repairs!
Maybe a board flexing issue, broken BGA joints ?
Yeah thats a tough one, if I knew for sure that was the case I could try baking/reflowing the board. I've only done this a couple of times successfully and that was a long time ago. Do you have a guide for this? I don't have a fancy reflow or toaster oven. The thing I am most worried about is how densely packed both sides of the board are, so this may be one of the things I try as a last resort.
I have a stark feeling you have an M3 that might be very touchy and needs that backup battery installed. but given that laptop also just made its way to the US from Japan, I'd also check the solder joints on your surface mount chips there for anything out of place with a loop and tweezers.
This is a really great suggestion, I will test this and follow-up with the results. The Librettos are very touchy overall so I could certainly see the BIOS battery being the culprit. Will also take a closer look at the board as well. Thanks!
My Libretto 100 CT does not need a backup battery to pass POST. Also, I’ve seen no evidence of any other Libretto requiring it as well.
I have the same problem with my M3 😢 it was working few days after arrival from Japan and then it stopped like that 😢
I haven't given up on mine yet, hoping to do a follow up to this soon. Signs point to a RAM issue, but Toshiba stacked the onboard memory using an interposer type board making this difficult to probe and check for broken solder points. Are you seeing any parallel POST activity on yours?
@@RobertsRetro Unfortunatelly I just started my adventure with retro electronics 😓 I have no docking station and equipment to check that.
Yeah, this is really annoying about the M3 specifically since I believe the dock is only compatible with that one model. The others like the Libretto 50/60/70 and 100/110 had multiple versions of the dock available and compatible with several models. Since the dock is "mostly" pass-through I may be able to get the pinout for the parallel port at least, in theory you could solder to the internal pins to get the POST code signals.
@@RobertsRetro I will wait for any update from You :) good luck!