[Coco] CP/M OS-9 is back

[Luis Antoniosi (CoCoDemus)]

You should test on a regular coco. honestly I only tested on a 6309
coco3 and mess.

If the problem is the CoCoFPGA so you probably have a bad flag being
set for some arithmetic or logical instruction.

I would suspect the DAA instruction as well. I have a translation
table between 6809 -> 8080.


On Fri, Mar 28, 2014 at 11:09 PM, Joel Ewy <jcewy at swbell.net> wrote:
> On 03/26/2014 08:03 AM, Luis Antoniosi (CoCoDemus) wrote:
>>
>> Hi guys,
>>
>> I got some time improving mt CP/M emulator as you can see in the video:
>>
>> https://www.youtube.com/watch?v=FiwmUnLTcjY&feature=youtu.be
>>
>> It's emulating Microsoft Basic. Couldn't test it extensively but it
>> seems to be working fine.
>
>
> Hey Louis,
>
> This is cool stuff.  I thought I'd try it out on my CoCo3FPGA and see how
> fast the emulation is at a nominal 25MHz.  The speed looks great -- I was
> going to shoot a comparison video showing how well CoCo3FPGA runs CP/M, but
> I think I may have run into a bug.
>
> When I try to replicate your program in MS-BASIC, I type the first line just
> as you do in your video, but when I list the program, line 10 reads:
>
> 10 FOR N=0 TO 0
>
> Initially I thought I just mistyped the second value.  But in fact, no
> matter what values I use in the FOR loop, it always stores them as '0'.  I
> don't know whether the problem is with the emulation, the FPGA's CPU09, or
> just an unhappy interaction between the two. Performance-wise, it appears to
> me that most CP/M programs would be quite comfortably usable on the
> CoCo3FPGA, presuming that they worked correctly.  You should see the
> WordStar screen load up.
>
> Anybody have any ideas what might be wrong here?  Anybody else with
> CoCo3FPGA want to give this a go and confirm my results?
>
> JCE
>
>> I'll be releasing a beta version for you to play with. The emulator is
>> 100% assembly (lwasm) and is 7KB only. Only the i8080 is emulated
>> although I expect to extend it to z-80 extra instructions later.
>>
>> It's running on a higher speed than I previous thought. I will release
>> the assembly code as well and if you want to help to improve the speed
>> feel free to do it. The main bottleneck is the flag converting between
>> 6809 -> 8080.
>>
>> Although I created a lookup table to convert it, the 8080 has extra
>> flags or the flag is not changed on the 6809 equivalent instruction.
>>
>>
>
>
> --
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco



-- 
Long live the CoCo