tebe Posted January 7 Share Posted January 7 (edited) https://atarionline.pl/v01/index.php?ct=nowinki&ucat=1&subaction=showfull&id=1704572629 Death.of.the.Right.Side.Garbage.356x240.zip Edited January 7 by tebe 10 2 Quote Link to comment Share on other sites More sharing options...
Rybags Posted January 8 Share Posted January 8 So, I'm guessing that there's some timing code in there + the right instruction sequence to ensure we have the wanted pixel data on the bus when Antic is doing that last graphics fetch? How about the last scanline in hires? Is there a fix being implemented to stop the screen warping on a real machine? From memory you just set DMACTL to turn off the display at the appropriate time to get rid of it. 1 Quote Link to comment Share on other sites More sharing options...
tebe Posted January 8 Author Share Posted January 8 hires_356.zip 2 Quote Link to comment Share on other sites More sharing options...
phaeron Posted January 8 Share Posted January 8 Nicely done. There is a reference in the forum to needing either to preset ZP or BIT abs to put a value on the bus for two cycles. I'm guessing this is due to 1-cycle jitter from STA WSYNC -- this can probably be avoided by either using INC WSYNC or tracking DMA cycles to appropriate switch between STA WSYNC and STA WSYNC,X. 2 1 Quote Link to comment Share on other sites More sharing options...
mono Posted January 8 Share Posted January 8 1 hour ago, Rybags said: So, I'm guessing that there's some timing code in there + the right instruction sequence to ensure we have the wanted pixel data on the bus when Antic is doing that last graphics fetch? How about the last scanline in hires? Is there a fix being implemented to stop the screen warping on a real machine? From memory you just set DMACTL to turn off the display at the appropriate time to get rid of it. Exactly. I'm pushing pixel pattern at Virtual DMA cycle using BIT $x0 or BIT $x0x0 instruction as @phaeron already explained. Also I've implemented 240-hires scanlines bugfix using your algorithm. So, as you can see I didn't invent anything, but just sit and did the job @phaeron Thanks a million for your "Altirra Hardware Reference Manual" - I studied its ANTIC scanline diagrams for different modes. It's very helpful. 2 1 Quote Link to comment Share on other sites More sharing options...
Rybags Posted January 8 Share Posted January 8 For an encore / extra you could stuff the PM registers manually, change positions etc and have them outside the 240 scanline area. Though it is a for show type of thing with not a lot of practical value. 2 Quote Link to comment Share on other sites More sharing options...
mono Posted January 8 Share Posted January 8 Yes, there is an picture made by Irwin years ago named "Impossible is nothing" https://demozoo.org/graphics/62641/. 3 1 Quote Link to comment Share on other sites More sharing options...
Rybags Posted January 8 Share Posted January 8 Cool - I think I've seen that before somewhere but didn't realise it was using the PM in the border trick. Quote Link to comment Share on other sites More sharing options...
ivop Posted January 8 Share Posted January 8 2 hours ago, mono said: Yes, there is an picture made by Irwin years ago named "Impossible is nothing" https://demozoo.org/graphics/62641/. Interesting! Sophia 2: Top and bottom left are black. Altirra 4.20: More stuff is wrong, and the small line in the lower right corner changes pixels constantly. No matter what -vert-area option I pass to atari800, it won't display more than 240 lines. 1 Quote Link to comment Share on other sites More sharing options...
tebe Posted January 8 Author Share Posted January 8 G2F source file impossible is nothing by irwin.zip 3 Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.