Re: [hatari-devel] Bug in Hatari FDC emulation

["Roger Burrows" <rfburrows@xxxxxxxxx>]

On 15 Dec 2014 at 19:53, Nicolas Pomarède wrote:

> Le 15/12/2014 19:12, Roger Burrows a écrit :
> > The FDC emulation in Hatari 1.8.0 contains a small bug: changes to the WP
> > status bit during insertion/ejection of a write-protected diskette in Hatari
> do
> > not happen in the same way as on real hardware (e.g. a 1040ST).
> >
>
> yes, I know this as I measured it too on my STF some times ago, but I
> was a little bit lazy so far and left the code which does a 0/1
> transition, which is indeed not the case on real HW.
> On real HW, it's impossible to see the difference between a WP disk when
> it's in or out.
>
And you can't detect any transition, as I found out during EmuTOS development,
which is why I know about it.  I have been working on adding media change
detection via the floppy VBL handler, and my code worked fine on Hatari, but
not on real hardware.  This caused me a lot of confusion ... of course, now I
understand what's happening, I can fix things so EmuTOS works on real hardware
too.

> I will add a change for this later for better accuracy, but I don't
> think some programs are not working because of this.
>
I agree, it's unlikely.

BTW, the info in FDC.C about status only being returned for type I commands was
very useful - thanks!

Roger