I did see Dave Haynie mention that there were some fundamental differences in the Dragonball that caused binary-level compatibility problems, but I don't recall the specifics, either.

On Fri, 03 Dec 1999 15:33:49 +0100, "g'o'tz ohnesorge"

rtechnik.de> jammed all night, and by sunrise was overheard remarking: I don't think DragonBall or any other Motorola 68K variation is fully synthesized -- Motorola made a fairly big to-do about the fact that ColdFire is, whole chip. MIPS has a killer (for embedded stuff) synthesizable core you can licence. I don't know of any decent x86 core (there are probably low-end things, and you can find an 8052 core in the public domain I think). I don't know any PPC offering, either internally or as a licensable core. There are tons of ARM cores; like MIPS, it's been simple enough to offer since back when these things had to be pretty simple (due to weak VHDL compilers). Ed Hepler's company offers a whole 40MHz 68K compatible in VHDL. > Sure. But something like ColdFire in VHDL is a start at least; the

So far, Motorola's not actively licensing the core (don't know if that's a "won't" or a "would, but not what we like to sell" thing), but of course the could if necessary for competition. Primarily, it's a way for them to release new version with different features faster than in the past. The did this without synthesis throughout the MC683xx line, using standardized component modules (design and layout largely prefabbed, more or less like standard cell), but it's much more efficient with VHDL. And of course, this way, any I/O block you design for ColdFire could be reused for PowerPC or any other design that comes along. That's a 6502-like thing.