Date   : Mon, 23 Jan 1995 11:12:58 -0600
From   : barnett@... (David Barnett)
Subject: Re: Scrolling Swapping

David Gibert wrote:
>
>> >That is what I do - the scanline timing is calculated directly from the
>> >horizontal timing registers.  The vertical timing is also calculated that
>> >way - without
>> >any bodging the timing in the high res modes comes out at exactly the
>>time for
>> >312 scanlines per field.
>
>
David Barnett pointed out:
>> >
>> As a point of information.  If interlace is on (Depending on TV setting)
>> then it is 312.5 scan lines per field.
>
David Gilbert replied:
>
>I think that depends - some versions of PAL have 312 lines on one field,
>and 313
>on the next - some have 312.5 on both.
>The 312 pops cleanly out of the screen parameters programmed by the beeb's OS.
>
>So - where does the extra scanline appear in the timing?
>
I haven't looked at the MOS in detail.  Check what *TV does.

The 1/2 line is important to achieve the interleaving of lines from
successive fields (actually, all sensible people have interlace off because
there is no resolution advantage on the BBC Micro and it gives rise to
annoying flicker.  Mode 7 always has interlace on).  As far as BBC
emulation is concerned, the effect is 32us.  Only games that do pallet
switching would care and they usually have a fudge band becuse the
interrupt delays introduce significant uncertainty anyway.  My guess is
that the Games either force *TV0,1 and use the VIA timer in auto-repeat
mode or they reload the VIA on every Vsync interrupt (the more robust thing
to do).

David Barnett