{ "numMessagesInTopic": 11, "nextInTime": 1659, "senderId": "B_1cCAjpQcPxfhRGG7ngNuhro99lfG2_Jzryo6i1uCdYf5gAQJFEbvw8gk3tX8BjMZwGjUQVsaquoGeoqWCB", "systemMessage": false, "subject": "Re: TTYS1??", "from": ""Don" <don389@...>", "authorName": "Don", "msgSnippet": "A few observations... The message string in the BIOS says Dolphin 2 BIOS V2.0 07/17/2001 Looking through the BIOS, at offset 229557 is the string", "msgId": 1658, "profile": "don389", "topicId": 1312, "spamInfo": { "reason": "0", "isSpam": false }, "replyTo": "LIST", "userId": 124309979, "messageBody": "
--- In magnia_sg20@yahoogroups.com, "Eric Dube" <ericdube@c...> wrote:
\n>
\n> > something on one of the boards is providing hardware enough for
\nthe
\n> > second serial line --
\n> >
\n> > ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
\n> > ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
\n>
\n> I'm pretty sure that's because it's in the chipset. It thinks there
\n> are 2 serial ports even though they're not all fully functional.
\n>
\n> > although, as you say, connecting headers to js1 or js2 may not be
\n> enough?
\n>
\n> Definitely not enough. If you look at the etchings on the
\nmainboard,
\n> you'll note in the signal path the resistors are missing as well as
\n> the RS-232 tranceivers. Even if you put in the headers, it wouldn't
\n> work without those other components. My guess is the rest of the
\n> componets are on the daughter card.
\n>
\n> > but I think these days, all of that stuff is completely integrated
\n> > into the southbridge: superIO or whatever they call it. the
\nphilips
\n> > chip on the mezzanine card is simply to make communicating with
\nthe
\n> > lcd dead-easy, I believe -- taking care of all the serial comms so
\n> > that it's just a question of dumping text directly into the com
\n> port.
\n>
\n> I think there's components on the daughter board. I did forget to
\n> write them down, and now I'm mad at myself for not spending more
\n> time researching it while I had it open (although since it's my
\n> primary email and web server I just can't have it down for long
\n> periods of time so I was in a hurry. I really need to buy a second
\n> one to play with.)
\n>
\n> > and obviously I was wrong to assumed that the rear 20-pin block
\nwas
\n> > serial -- so now I am even more interested to know what the heck
\n> that
\n> > little thing is!! my dream is that it's a hidden vga or usb,
\n> but.... hehe.
\n>
\n> I actually have a *theory* on what this connector is. I think it's
\na
\n> unified plug for general case to mainboard stuff like HD/Power
\nLEDs,
\n> reset/power buttons, etc.. The number of pins would match that
\n> standard motherboard header. Again - it's only just a guess, but I
\n> can't think of anything else that would fit it. It might just be
\n> used for testing at the factory for instance (also, given it's
\n> label "panel" that might just be the case.)
\n>
\n> > I have made a complete inventory of the known and uknown
\nhardware,
\n> I
\n> > really need to get around to posting it on the t-m site -- it
\n> would be
\n> > great if one of the toshiba ppl who read this group could
\ncomment,
\n> but
\n> > I do understand if there are things preventing them from speaking
\n> > about the guts of the beast.
\n>
\n> I doubt anyone will speak up. Perhaps if you knew someone
\npersonally
\n> and either bribed them or took them out and got them drunk. But you
\n> never know.
\n>
\n> > there was one guy on this group already who said that he'd
\n> > saved/edited the bios -- he'd modded it to allow usb support to be
\n> > turned on, I think?
\n>
\n> Yeah, I found that message. He mentioned that the SG20 might use a
\n> PCCHIPS mainboard as he found that string in the BIOS. I've been
\n> meaning to load it up with modbin6 and poke around in there myself
\n> to see what I could find. Another thing he mentioned was the model
\n> number, although I think that was a red herring as that was
\nprobably
\n> left over from an older bios they used for the board. The model he
\n> mentioned was a 5-series which doesn't seem to fit the date of that
\n> chipset. The boards in that time frame all come with 7-series part
\n> numbers which would make more sense.
\n>
\n> > but since the bios chip is soldered on and not socketed, it'd
\n> either
\n> > have to be cut out and put into another machine, or else do
\n> something
\n> > like boot dos/windows and use available award-type tools, as you
\n> > mentioned, to get it.
\n>
\n> Actually, this is easy to get. The bios flash utility has an option
\n> to backup your bios. You can just save a copy from the bios chip
\n> that way, no need to remove anything. You're correct in assuming
\n> you'll need to boot up with DOS, but that is easy.
\n>
\n> > the only distinctive thing I remember from the bios screens was
\n> when
\n> > it booted it said something like "dolphin 7/1/2001", but that
\ncould
\n> > just as well be some name given by the toshiba engineers.
\n>
\n> Yeah, that's probably an internal code name like you said.
\n>
\n> > anyway, I'm glad that other people are interested in this
\n> particular
\n> > aspect of the magnia, I think one day we will finally know the
\n> truth
\n> > of things ^_^
\n>
\n> Really all I was looking to do was get a UPS hooked up to it
\nwithout
\n> buying a ton of extra hardware (i.e. network serial device,
\nparallel
\n> serial converter, pcmcia card, etc..) I figured I might be able to
\n> do it cleanly by adding a second serial port.
\n>
\n> I'm almost tempted to buy a similar chipset motherboard to steal
\n> parts from to try to get it working. It might be possible if I
\ncould
\n> find one with the same reference implementation for serial. That's
\na
\n> real long shot given how many different manufacturers of parts are
\n> out there. What I was think is I might be able to get the values
\n> right for the resistors.
\n>
\n> Regards,
\n> Eric Dube