Re: [hatari-devel] IKBD regression with EmuTOS

[Nicolas Pomarède <npomarede@xxxxxxxxxxxx>]

On 10/02/2013 21:16, Roger Burrows wrote:

> Also, EmuTOS sending the 0x1a and 0x12 will *disable* mouse and joystick
> reporting, not enable them.  So perhaps what is happening is that on real
> hardware, they are ignored (because they happen too fast), but Hatari processes
> them, leaving mouse & joysticks disabled.

Hello

as explained earlier, there's a special undocumented mode in the IKBD 
where sending any 0x1a, 0x12, 0x14 or 0x08 during the reset will put the 
IKBD in a mode where it reports both mouse and joystick at the same time 
(but joystick button is reported in a mouse packet).
That is, even sending a disable command during reset will in fact be 
seen as an enable command.

It's rarely used, but still need to be emulated (for the game Barbarian 
for example, see ikbd.c for more details).

Regarding EmuTOS, I don't think this is the intended behaviour, so one 
should wait for a $F1 byte after the IKBD reset command before sending 
other commands.

Nicolas