| RE: [hatari-devel] TT emulation |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
> -----Original Message-----
> From: Eero Tamminen [mailto:oak@xxxxxxxxxxxxxx]
> Sent: Wednesday, July 11, 2012 19:13 PM
> To: hatari-devel@xxxxxxxxxxxxxxxxxxx
> Subject: Re: [hatari-devel] TT emulation
>
[...]
> In general it looks OK to me. Here are couple of quick comments:
>
> * bTTSampleHold is used in Videl drawing code, but it doesn't seem to
> be zeroed if user switches to Falcon emulation (which also uses videl
> code). This the only real issue.
Ok, I'll correct that.
> * Only bTTSampleHold seems need to be global, i.e. it seems that
> bTThypermono should be static...
Ok, I'll move bool bTTHypermono from screen.h to video.c as static bool.
I thought about that and wasn't sure if its ok. Finally I put both TT special video modes together in one file.
> * rest of videl.c code uses different brace ('{}') positioning convention[1]
> than your patch.
>
> [1] Different files in Hatari sources use different conventions depending
> on their originating project and who has written / maintains them.
Ok, I'll correct that.
> > In additional, I attached test application for SampleHold and TT
> > Hypermono mode.
>
> That seems to work. Are there any other demos using these modes?
I don’t think so.
But as I mentioned before, I'm trying to prepare small TT demo with those features.
> There are actually already several other TT demos (from tscc) that don't work
> properly in Hatari (as documented in compatibility list), for example:
> * http://pouet.net/prod.php?which=52339
> * http://pouet.net/prod.php?which=16426
Both demos needs unsupported by Hatari feature - TT-RAM. And in case of Ray's TT-Highres TT cycle exact mode.
Currently, access to ST-RAM (and hardware I guess) is too fast in case of TT emulation. The TT has 500ns access to ST-RAM (like ST) but under Hatari, Nembench shows that access is about twice faster (Falcon's style?)
Regards
Cyprian