[Coco] The next coco3.....wish list

Gene Heskett gheskett at wdtv.com
Sun Oct 5 02:12:07 EDT 2014


On Saturday 04 October 2014 21:58:42 Mark McDougall did opine
And Gene did reply:
> On 5/10/2014 9:44 AM, Aaron Wolfe wrote:
> > Coco3fpga barely fits on the DE1, which I believe has 20,000.
> 
> Not quite true. The constraint on the DE1 with the Coco3FPGA project is
> the on-chip memory, not the logic. Last build I did it was only 66%
> logic.
> 
> Of course the 2C5 is way too small in terms of memory, and also too
> small logic-wise, which I mentioned a few weeks ago but that comment
> appears to have been ignored as people have continued to entertain the
> idea of the 2C5 dev board... oh well.
> 
> > Of course none of the speculation or "wish list" discussion actually
> > matters, because only Gary Becker with his coco3fpga has done the
> > work and created anything.
> 
> Again, not quite true. I finished a Coco1/2 FPGA implementation a few
> years ago, with hard drive (HDBDOS) emulation via CF. There's a DE1
> port of it, but since the SD card implementation on it is read-only
> atm, it's of little use to anyone but myself.
> 
> > Its easy to write about what "they" should do or how "somebody" could
> > make this or that, but it seems very hard for the list to understand
> > that there is no "they".  " someone" is you and me.  Either make it,
> > or help someone make it, or forget about ever seeing it happen.
> 
> IMHO what is really holding the 'Coco3FPGA' back is a lack of a
> cycle-accurate 6809/6309 core. If one ever surfaced, then it
> (Coco3FPGA/DE1) would tick a lot of boxes for a lot of people. But
> it's a tall ask; CPU cores - especially something like the 6809 -
> aren't exactly trivial.
> 
> Regards,

I am glad to see this pointed out. Not enough of us truly understand that 
the MC6809EP was quite an achievement by the moto engineers.  There are of 
course differences of opinions on that, as many as there are basic 
architectures back in the day.  And each has its proponents.  And I have 
programmed on several other architectures so I can well appreciate the 
advantages and disadvantages of each.  IMO the motorola stuff up to the 
68040 was hands down, the best of the complex instruction set arena.  The 
only thing I'll fault about the coco's use of it is the total lack of DMA 
capability in the coco's. There are dma enabling peripherals for the 6x09, 
and proper usage of those could both divorce the grfx from the cpu clock, 
but would also allow the enhancement of its display.

To me, that addition/modification to the reference design of the coco3, 
and tightening up the address decoding in the I/O page, would constitute 
the "coco4" in its entirety.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS


More information about the Coco mailing list