[Coco] 6309/6809 opcodes with mixed 8/16 bit registers

[jdaggett at gate.net]

On 22 Nov 2012 at 20:55, Mike Pepe wrote:

> I could have sworn the HCF on the 6809 was $FF. I recall that if you were to
> take a 6809, stick it on a breadboard, give it power and clock it'll run the
> address bus test routine- the data bus should float to $FF if it's not
> connected...
> 
> Hm, I should (re?) try that. I'm certain I've done this before.
> 
> -Mike

Mike 

Wrong

The only possible opcodes on page one that could be the HCF are

$14
$15
$18
$1B
$87
$8F
$C7
$CD
$CF

The rest of the unused opcodes on page one mirror and existing opcode.

example:

$00 and $01 are negate
$02 and $03 Negate/COM depending on the Carry bit state
$04 and $05 LSR
$0A and $0B DEC

and so forth for all the even odd pair from $00 to $7F. This is due to an incomplete decode of 
the opcode. That is the Least Significant Bit is left out of the equation in those instances. 

darn this discussion is rekindling the desire to reduplicate as much of the circuit in logic within 
an FPGA. Started that a few years ago and never completed it as an attempt to understand 
how the logic might have been done. 

james