Gasman demo
in [Commodore]
|
Prev: MPS 1270A compatibility
Next: plus/4 GEOS questions
From: Simon Scott on 15 Jan 2006 08:21 Matthew Westcott wrote: > Simon Scott wrote: > >> Why do I only see 8 (under fuse)? >> >> I hit pause and the most sprites Ive seen is 8. > > That'll be the terminal case of blanking-at-the-wrong-time. In FUSE you > can briefly see all 23 if you reset, but that's about the only way I can > find of showing it. Thought so. > >> Is that using all the CPU? Why 23 and not 24? > > Yep, 23 is as many as I've been able to fit into one frame. I can get 24 > if I set my emulator to Pentagon (a Russian clone with a marginally > faster CPU and no slowdown from the ULA when writing to screen memory), > if that counts for anything... no, thought not :-) > Has the superiority of hardware sprites been proven then, or should I go ahead with my plan to write a multiplexor that shows the remaining CPU time? :D >> Lastly, my PC is kinda slow and the demo looks a bit glitchy and not >> 50Hz. > > If you get a steady red border and sprites disappearing at various > points, it's 50Hz - if the border and sprites flicker, it's run into the > end of the frame. yah, works fine on fuse
From: spike1 on 15 Jan 2006 11:06 Simon Scott <simon(a)chrome64.org> did eloquently scribble: > Oh damn, that means I have to write a multiplexor now :) > > Is that using all the CPU? Why 23 and not 24? > > Lastly, my PC is kinda slow and the demo looks a bit glitchy and not 50Hz. > What's the best spectrum emulator for linux? Fuse. -- ______________________________________________________________________________ | spike1(a)freenet.co.uk | "I'm alive!!! I can touch! I can taste! | |Andrew Halliwell BSc(hons)| I can SMELL!!! KRYTEN!!! Unpack Rachel and | | in | get out the puncture repair kit!" | | Computer Science | Arnold Judas Rimmer- Red Dwarf | ------------------------------------------------------------------------------
From: Matthew Westcott on 15 Jan 2006 14:01 Simon Scott wrote: > > Has the superiority of hardware sprites been proven then, or should I go > ahead with my plan to write a multiplexor that shows the remaining CPU > time? :D Hmm - I would rather like to see a direct comparison of the same effect, as much as it is a foregone conclusion that the C64 will fare significantly better... you're welcome to wait until I've refined my routine into something actually worth watching, though. Well done on the scroller though, TMR... will have to check it out on my Windows box later, as I don't think MacVICE (with no sound and decidedly jumpy timing) did it justice.
From: White Flame (aka David Holz) on 15 Jan 2006 17:26 "Matthew Westcott" <gasman(a)raww.org> wrote in message news:42uu46F1kh593U1(a)individual.net... > It is doing dirty rectangles - a full-screen blank would take over 33000 > cycles out of the ~70000 we have in a frame, not counting the delays the > ULA would throw in. Oh, okay. Due to the blanking effect and the need for sorting it sort of looked like fullscreen clears. I'm not used to managing single-buffered animation, so I've got to figure out what the side-effects are that I'm seeing. :) > Well, yes, if we were intent on putting the Spectrum at the greatest > possible disadvantage in the comparison then 24x21 would do the job. > Somehow I think the C64 would be put at an even greater disadvantage if > we'd chosen 25x22 instead :-) Nah, you don't have to do difficult-to-render block sizes, just something closer in pixel count to the C64's 504-pixel sprites would be a better comparison. I'm guessing that it really wouldn't be much of a code change or disadvantageous processing to keep the width the same and make the sprite 16x32, for 512 pixels. 16x24 = 384 which would still be small, but still closer to the ballpark. -- White Flame (aka David Holz) http://www.white-flame.com/ (spamblock in effect)
From: Simon Scott on 15 Jan 2006 18:17
Matthew Westcott wrote: > Simon Scott wrote: >> >> Has the superiority of hardware sprites been proven then, or should I go >> ahead with my plan to write a multiplexor that shows the remaining CPU >> time? :D > > Hmm - I would rather like to see a direct comparison of the same effect, > as much as it is a foregone conclusion that the C64 will fare > significantly better... you're welcome to wait until I've refined my > routine into something actually worth watching, though. Naw, Ill get started sometime this week. Ill see if I cant make some sort of display which shows remaining CPU time. |