Ah, now I understand. A front-end for LLVM that took a pseudo-Z80
assembler would be kind of cutting-edge cool, wouldn't it?

Davor


Chris Wild wrote:
> The main thing is too keep a code base that looks almost identical to
> the original z80. I like the fact that if you look at DOS LOM code it
> is practically the same as the original z80 version, but the option is
> there to add little tweaks as I have done with bug fixes and options.
> Porting to any other language changes the dynamic massively.
>
> With DOSBOX working nicely, then the urgency is less...
>
> Chris
>
>
> On 9 May 2009, at 06:04, Davor Cubranic wrote:
>
>
>> On Tuesday 05 May 2009 00:28:13 Chris Wild wrote:
>>
>>> So, I was thinking that a new approach is needed. I'm thinking of
>>> going back to the original code and porting it to a simple BYTE CODE
>>> implementation. In essence create a small language that LOM and DDR
>>> can be ported to that keep the faithful nature of them. Then we could
>>> write many virtual machines to play them...
>>>
>> Isn't that what JVM, or Python, or just about any decent cross-
>> platform
>> language will give you by default? Why would you want to create your
>> own?
>>
>> Hell, there are even Z80 and x86 emulators done in Java. Given LOMs
>> low
>> resource requirements, my guess is that with some minimal work to
>> accommodate graphics and keyboard input these could do the job just
>> fine.
>>
>> Davor
>>
>> P.S. Or, as you said in your last email, just use DOSBOX with the
>> original LOMDOS.
>>
>>
>> ------------------------------------
>>
>> --
>> The Midnight Group - For further midnight information www.icemark.com
>> Yahoo! Groups Links
>>
>>
>>
>>
>
>
>
> ------------------------------------
>
>