Fido1100 MCU

[Plaz]

Interesting find. I'm still scanning the pages but can't find information on how code compatible it is. 95%, 99%, 100%? That's the trouble with coldfire. There are 68k instructions that are also vaild on coldfire, but do a differnt task. It would be easier for compatibility design if those instructions were not valid at all on coldfire.

The programmable I/O's of the fido could be very important. I/O is as big an issue as code compatibility when trying to interface with the Amiga buss.

I'll keep reading and I think I'll post this over to the coldfire mail list and see what they think.

Plaz

[Plaz]

That space is reserved for chip ram so it seems like a complete show stopper.

Not necessarily. Banking methods might be used to work around this or there might be other built-in ablities we don't know about yet.

I see some one has posted as a "rep" for the FIDO. I'm going to write and see what they can tell us.

Plaz

[Plaz]

So to me it appears that the Fido could function as a souped up 66Mhz 68000 but other things that still are a mystery are;

Missing instructions can be handled by trap exceptions, and handled in software, so this is not a killer problem. (yet) A good replacement would have to emulate and 020. Again this might be doable with traps and libs.

It's been the problem in coldfire that some important addressing instructions DO exist but perform a different operation that would crash the system on Amiga 68k code.

I don't want to seem overly optimistic, but FIDO's not dead yet in mind. I'll send a message to Mr. Wilson and see what they're up to.

Plaz

[Plaz]

@madvax

I'll be sending a PM shortly.


Plaz