[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: TransWarp IIgs GAL's - Upgrade now possible



Eric Smith wrote:
Henry wrote:

Cloning the TransWarp IIgs but in to a new FPGA would be the path we would take. I may be wrong, but as stated earlier the "configuration code" for the XC2064 on the TransWarp IIgs is stored in ROM?


Yes.


So, again I haven't even started any research on this, I would assume that if the code is readable then converting it in to something logical (i.e. - HDL or schematic)
should just be a matter of time.


The FPGA configuration bitstream format is not documented anywhere,
beyond that it has a header, a certain number of frames, and a trailer.
The content of a frame is not documented.

For an XC2064, it consists of 12,038 bits (just under 1505 bytes),
broken down as 160 frames of 71 bits (+ 4 bits framing overhead), and 38 bits of header/trailer overhead.

Bits in the frames configure the lookup tables, flip-flop, and
mulitplexer configuration in the 64 CLBs (Configurable Logic Blocks)
and the 58 IOBs (I/O Blocks), and all of the routing matrix
interconnections.

It's a significant challenge to convert it back to logic equations.  It
is probably easier to treat it as a black box, attach a logic analyzer
to the pins, capture traces of the the TWGS in operation, and deduce
what's going on inside the FPGA.

Having a schematic of the board together with the logic equations of
the GALs would help very much in figuring out what functions the
FPGA performs.

Very interesting.

So the engineering interface to FPGAs is solely their design tools?
That's quite remarkable--like not documenting the object code of a
processor but only providing a C compiler!

I would have expected the low-level interface to be documented so that
third parties could produce design tools.

-michael

Music synthesis for 8-bit Apple II's!
Home page:  http://members.aol.com/MJMahon/

"The wastebasket is our most important design
tool--and it is seriously underused."