[hatari-devel] Hatari v1.6 features?

[Laurent Sallafranque]

Hi,

Would the hd6301_cpu.c code I started to write 2 years ago and I still 
haven't finished would help here ?
Maybe I could take a few days and improve it.

But as there's already Nicolas's method to emulate this shift, I've 
never filled urged to finish this piece of code.

Just tell me.

regards

Laurent



Le 16/10/2011 14:45, Nicolas Pomarède a écrit :
> Le 15/10/2011 15:30, Eero Tamminen a écrit :
>
>> Nicolas:
>> - Move to new hosting service...
>> - ?
>
> Hello
>
> Maybe rework the acia/ikbd communication parts, there're still some 
> problems there, we regularly get reports of programs that gets broken 
> after we "fix" something in the ikbd. Timings and emulation of the 
> acia control registers are not right, I would need to write some 68000 
> test programs to precisely measure several cases on a real ST and 
> check we get the same behaviour under Hatari (definitly a boring task, 
> but this has to be done :) )
>
> Let's see also if we get better sound options for the analog voices 
> mixing.
>
> IKBD and FDC should give huge improvements for people playing games in 
> Hatari ; so far my focus was mainly on demos, as they require very 
> precise emulation and exhibit a lot of video/cpu cases.
>
> But IKBD and FDC are 2 of the oldest components that were not really 
> refactored recently and they still carry some limitations/errors from 
> the original Winston sources (that's done for FDC now, but IKBD remains)
>
>
>> [1] Nicolas, recording and replaying mouse events would be
>>      really nice...  Do you have any thoughts on best way to
>>       do that?
>
> In ikbdc.c, you could try to intercept when we write the byte in 
> $fffc02. This byte is the result of pressing a key or moving the mouse 
> on the host computer (through SDL).
>
> So, if each time a byte is written to be "sent" to the ST you store 
> the VBL number, the video cycle number in this number and the byte, 
> you should be able to record a precise flow of the bytes sents to the ST.
>
> But if you do this in IKBD_InterruptHandler_ACIA / 
> bByteInTransitToACIA==true, you might record also the bytes sent by 
> the ikbd, which is not what you want, so you better record bytes in 
> IKBD_SendAutoKeyboardCommands when IKBD_UpdateInternalMousePosition is 
> called (this would filter mouse only events, but requires quite some 
> plumbing anyway)
>
>
> Nicolas
> _______________________________________________
> hatari-devel mailing list
> hatari-devel at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/hatari-devel
>