Re: [hatari-devel] New video code with STF wakeup state :)

[Anders Eriksson <ae@xxxxxx>]

On Fri, 10 Jun 2016, Nicolas Pomarède wrote:

> step 3) could be coded without doing 1) and 2), but there would a risk of 
> dealing with wrong pixels data (timing-wise if DE start/end is not correct) 
> and to come with a bad rendering (mixing low/med parts on the same line 
> requires to change resolution precisely when first IRx will be reloaded, else 
> you end up with some kind of stabiliser and planes shifting, which is not the 
> expected result).
>
> step 3) will be the icing on the cake, as I think few programs used mixed 
> resolution per line.
>
> But step 2) will really improve a lot of never emulated demos with bad 
> stabiliser / non standard timings (omega fullscreen without stabiliser, death 
> of the left border has no stab but a hi/lo during 16 cycles, closure has also 
> no stab on the right, ...)
> This would also "automatically" emulates the "4 pixel hardscroll" used by 
> Alien.

Thanks for the detailed description. I'm looking forward to the day with 
low/mid in one line and maybe a debug screenmode (showing the entire 
512*313 field, would be great to be able to see raster timing across all 
cycles :)

In your original post, you mention that "Death of the left border" should 
now work in all ws. Perhaps I have different copy, but I'm always getting 
bitplane errors no matter which ws I select. You also mention black 
vertical bands every 32 pixels on real ST. At least on my STe, I can't see 
that.

--
Anders Eriksson
ae@xxxxxx     http://www.dhs.nu/
ae@xxxxxxxxx  http://www.atari.org/