Linux on a modified Tandy 100/102
Yef
I'm curious if any adventurous souls have every bothered
to produce an alternative motherboard for the Tandy 100/102
that had a 16- or 32-bit CPU. It would cost several hundred
dollars to produce the first run of boards, to be sure,
but these days that kind of thing should be quite feasible.
It's just a matter of learning how to interface with the
existing components -- LCD, keyboard, power.
It also occurs to me that a clever company could produce
a replica of this machine, make it as durable as the
original, but have a more modern processor in it and
flash memory for storage.
My own interest in this is that, although I am fairly
happy with my laptop, I know that it won't last more than
two years (it's a Toshiba), and anyway, when it dies
I'll be putting out another chunk of change that I don't
really want to.
When you get down to it, I don't need a lot of the new
software that's out there or the new features like video
and music. I just need the basics. </lament>
Anyway has anybody tried doing either of these two things?
Thanks.
Michael Black
the 300baud internal modem) is the small screen. No matter what
you do, you'd still have that small screen.
The thing I thought of years ago was that one could build up a small
computer, and use it with the Model 100, using the laptop, unmodified,
as the terminal. As CMOS ram density increased, I imagined how easy
it would be to cook up a 64K 6809-based computer, and run OS-9 on it
with the M100 as the terminal.
Nowadays, one could do the same thing with a much better CPU and a lot
more ram. I've seen 80X86 based "embedded" computers that could run
Linux which were about the size of an open hand.
BUt you have to look at the intended end product before coming up
with a design. The value of the M100, even to this day, is long
battery life, low weight, and a screen that doesn't need to be opened
up to see.
If one can live with what's in there, then an M100 is the best route,
But a whole slew of generations of laptops have come and gone since
the M100 twenty years ago. Getting a used one makes more sense
than trying to make the M100 into something it isn't. IN order
to run a better operating system, you need more resources, that
needs more battery power. If you want to run an operating system,
then the simplicity of the software (and the instand on and off) is
gone.
I got a Powerbook 1400C a year ago, for forty dollars. Needed
an AC adaptor (I made one out of a printer power supply), and
the batteries are dead. It's infinitely better than the M100,
but it also has low battery life, it's much heavier, and there
is nothing simple about using it. On the other hand, it's faster
(and has more RAM and a bigger hard drive) than every computer I've
had up to mid-2001, it runs a nice operating system, a great screen, and
may run Linux (depending on who's talking). It even has a CDROM
drive.
Michael
Yef
Michael Black wrote:
> The worst part of the model 100 (well except at this point for
> the 300baud internal modem) is the small screen. No matter what
> you do, you'd still have that small screen.
> ...snipped
I guess my point is that I'm tired of computers that have
built-in obsolescence. The 100 seems to have been made before
that was a consideration. Or its durability and battery life
maybe were a 'mistake'.
These days computer manufacturers never make such mistakes.
Even the ruggedized computers won't take much of a
pounding unless you pay for the $3000 unit.
Perhaps there's need for something new -- a new company,
a new computer and new software based on Linux, with no
hard drive, just flash, etc...
We need a Toyota or Honda to step in and make something that
will really last, even if it's underpowered. AND it should be
servicable with spare parts.
I guess it's kind of folly to pursue creating a motherboard
for the model 100, but certainly that design is a winner
and an inspiration.
My 2 cents.
NSM
"Michael Black" <et...@FreeNet.Carleton.CA> wrote in message
news:cvajb0$n31$1...@theodyn.ncf.ca...
> The worst part of the model 100 (well except at this point for
> the 300baud internal modem) is the small screen. No matter what
> you do, you'd still have that small screen.
A WorldPort modem makes a nice addition.
> But you have to look at the intended end product before coming up
> with a design. The value of the M100, even to this day, is long
> battery life, low weight, and a screen that doesn't need to be opened
> up to see.
>
> If one can live with what's in there, then an M100 is the best route,
It's like the Model T of laptops - rugged, reliable, but oddball. Some Model
T's run to this day. I haven't seen a running IBM 360 for some time.
> But a whole slew of generations of laptops have come and gone since
> the M100 twenty years ago. Getting a used one makes more sense
> than trying to make the M100 into something it isn't. IN order
> to run a better operating system, you need more resources, that
> needs more battery power. If you want to run an operating system,
> then the simplicity of the software (and the instant on and off) is
> gone.
I'm one of those who hope to crack and hack the Mivo 100 email appliance
with it's flash rom.
N
Michael Black
"Yef" (e9...@yahoo.com) writes:
> Michael Black wrote:
>
>> The worst part of the model 100 (well except at this point for
>> the 300baud internal modem) is the small screen. No matter what
>> you do, you'd still have that small screen.
>> ...snipped
>
> I guess my point is that I'm tired of computers that have
> built-in obsolescence. The 100 seems to have been made before
> that was a consideration. Or its durability and battery life
> maybe were a 'mistake'.
>
At the time, it was still a reasonably decent computer, laptop or
not. But it was also a time when there were a variety of computers,
and operating systems, so the fact that all the software was in ROM
didn't matter. People did not expect compatitiblity, because
there was no one major operating system.
You confuse "obsolescence" with other things. One is that people
want the latest and fastest, so they toss aside what they have already.
I haven't bought a new computer since 1989, and except for $150 in
2001, I've spent no more than fifty dollars on any computer I've
gotten in that period. I have no real interest in getting the latest,
so it doesn't matter that I don't. If it does what I need, then
it's not obsolete.
BUt there is also a need for faster and better computers. Not for
most people, but if you are doing critical things you may notice
a difference. But since advances in technology trickles down,
everyone gets those faster and better computers. But again, that
does not make the old obsolete.
The term "built in obsolescence" dates from a different time. Back
then, it was about things intended to fail, or run out of steam, in
a relatively short time. But with computers, they don't have built in
limitations or failure points, they become "obsolete" because the
filed is constantly moving. A toaster is a basic item, and there
is virtually no difference in one made today than one made decades ago.
It should keep on doing what it does for a long time. The fact
that they are now cheaper, and may be not as well made, may be
an exacmple of "built in obsolescence". But in a quarter century,
I've gone from a computer with 1K of memory, and a calculator style
keyboard and readout, and cassette interface, through various computers
that were increasingly better, to a 1GHz computer with 256Megs of
memory, and a 20gig hard drive. That KIM-1 from 1979 still works
(or if I actually fix the keypad it will), proving that it didn't
fail early. But it was limited back in 1981, when I got my second computer.
And so on. There is no way that back in 1977 when the KIM-1 first
came out that they could plan for the future that hadn't been seen
yet, or indeed provide the capability of the 1GHz computer.
I brought up that Mac 1400C because it's not "obsolete". It's a great
computer, and the only reason I'm not using it as a main computer is
that I have something better.
Michael
jhoger
get the attention you want on the m100 listserv.
Second, yes, it has been thought about and there are discussions in the
archives of the m100 listserv.
One promising possibility is to put a Cidco Mailstation board and
display into a T102 case. I haven't had much time for that, since I've
been focusing on DLPilot (a Tandy Portable Disk Drive emulator for the
Palm) and ReMem as my main Model T projects.
Steve Adolph will be going to PCB on ReMem, a RAM/Flash/MMU upgrade for
the Model 100, T102, T200 for our initial run of dev boards very soon.
I would imagine a new Model 100 compatible motherboard could be
designed around this. The question is what you would do about the CPU.
I would think you would want to use a "fast" Z80 instead. I wonder if
it would be possible to wall it off with a CPLD... the CPLD would
perform some necessary translations to make it appear to be an 8085.
ReMem is an "open" project in that all specs and firmware will be
release under open content licenses.
-- John.
jhoger
Alphasmart Dana. That said, I'm sticking to the Model T.
-- John.
Yef
Michael Black wrote:
> "Yef" (e9...@yahoo.com) writes:
> > I guess my point is that I'm tired of computers that have
> > built-in obsolescence. The 100 seems to have been made before
> > that was a consideration. Or its durability and battery life
> > maybe were a 'mistake'.
> You confuse "obsolescence" with other things.
Not really. I can think of 4 ways in which laptops
(my main subject) become obsolete where the fault
is the hardware manufacturer:
1. LCD backlights fade over time; I have a 5-year old laptop
that has seen heavy use and its backlight is almost kaput.
Now, this may not be the intention of the laptop makers,
but the artificially high price of replacement screens
makes stockpiling of screen(s) impossible at the time
of purchase.
2. Cheap circuitry. I've several laptops fail on me, all of them
ones that I had either bought used or got for free. Usually
it is the circuitry that fails, not the keyboard or case.
This is the hardware manufacturer's fault.
3. Non-exchangeable parts. While I can exchange a hard drive
when the old one fails, basically no other part can be swapped
out for a generic part. Not the motherboard, cpu, keyboard,
power circuit, power supply, or screen. RAM daughterboards can
but they don't usually fail. If a model-specific part fails,
getting a replacement can be a royal pain. I know, I've
pursued that.
4. Lack of ruggedness. Like I said, cases don't seem to fail
very easily in my experience, however some of the cases
that I've seen are pretty flimsy. Apple does a better job
than most by using polycarbonate, but then again, their
laptop keyboards are flimsy crap IMHO.
One that's not their fault:
*. Software becomes more and more bloated over time. This is
not strictly the fault of hardware manufacturers however.
But running older software is not always an option, for instance
when file formats change, or web standards change.
My 2 cents.
Yef
I owned one... for a few days. The PalmOS leaves A LOT to be
desired. Unfortunately the Dana uses the processor that doesn't
readily accept Linux -- it has no MMU.
NSM
"Yef" <e9...@yahoo.com> wrote in message
news:1108943307....@c13g2000cwb.googlegroups.com...
...
> 1. LCD backlights fade over time; I have a 5-year old laptop
> that has seen heavy use and its backlight is almost kaput.
> Now, this may not be the intention of the laptop makers,
> but the artificially high price of replacement screens
> makes stockpiling of screen(s) impossible at the time
> of purchase.
Allelectronics.com has some backlights - if they ever get 'your' size might
be worth grabbing some.
> 4. Lack of ruggedness. Like I said, cases don't seem to fail
> very easily in my experience, however some of the cases
> that I've seen are pretty flimsy. Apple does a better job
> than most by using polycarbonate, but then again, their
> laptop keyboards are flimsy crap IMHO.
Model 100's survive dropping and kicking.
N
jhoger
But I would say that Linux would not be my first choice as an OS for a
true portable. But then, neither would Windows or MacOS. I think
Alphasmart had the right Idea in choosing something targetted at
embedded systems. Linux is trying to go in that direction, and you can
run uCLinux on Palm and some power hungry handhelds... but Linux ain't
really designed for that.
Any replacement needs to be instant-on, the interface should be
data-centric rather than program or "OS navigation" centric, and your
data must always be easy to find and sync up. Ah, and it must be easily
programmable. I'd throw in some hobbyist niceties like A/D converters,
programmable I/O pins, and reconfigurable computing (on-board
reprogrammable logic circuits).
Yef
> > 4. Lack of ruggedness. ...
>
> Model 100's survive dropping and kicking.
Oops, that was a typo. I meant a good computer should be
rugged, not be lacking in ruggedness.
Yef
data-centric? I'm assuming that you don't want to pay
a licensing fee to use PalmOS or WinCE.
jhoger
BASIC interpreter.
Such a new unit would not be free to make. I don't think it would be a
big issue to add a software license to COGS if necessary, though I
think enough free stuff could be cobbled together.
Anyway, I wouldn't start with an OS, I'd start with a really small
kernel and add a Model 100 style interface, an interpreter, a loader, a
few apps and a flash and RAM file system. Steal some drivers for USB,
wireless from wherever.
Believe me this would be a significant project, software is just one
aspect of it, and it will cost $$$'s though existing free software
could be leveraged.
-- John.
Yef
jhoger wrote:
> Well as you know the original Model 100 didn't even have an OS. Just
a
> BASIC interpreter.
Yes, and people have made the language the center of systems
before besides that, like Forth and Smalltalk computers.
But their usability is questionable, I think. But feel free
to counter-argue that one.
But I understand your desire for a data-centric system.
Have you ever read a book called "The Design of Everyday Things"?
> Believe me this would be a significant project, software is just one
> aspect of it, and it will cost $$$'s though existing free software
> could be leveraged.
If you want to reinvent the wheel, then sure it will be a lot of
work. I don't think that Linux can't achieve instant on;
when booting from flash it would be very fast.
I see this issue as a matter of supply and demand. There's certainly
demand for cool Linux-based computers, and for ruggedized
computers. One can either leave the solutions up to some company
to produce, in which case consumers will be at the company's mercy
(which time and time again has proven disastrous), or,
users can solve the problem themselves.