pouët.net

Optimus information 13647 glöps

• general:
• level: user
• personal:
• first name: Michael
• last name: Byte
• portals:
• csdb: profile
• slengpung: pictures
• zxdemo: profile
• demozoo: profile
• cdcs:
• cdc #1: Contrast by Oxygene [web]
• cdc #2: Robotnik by Rage [web]
• cdc #3: Fruit Cake by Deus [web]
• cdc #4: Heart Core by Illi Recentes ImperatoreS
• cdc #5: Stargazer by Orb [web] & Andromeda [web]
• cdc #6: Extatique by adinpsz [web]
• cdc #7: Batman Forever by Batman Group
• cdc #8: Artphosis by Hitmen [web]

• 256b MS-Dos Function by Desire [web] & Darklite [web]
• That looks cool now on youtube :)
• rulezadded on the 2014-10-05 02:40:04
• 256b MS-Dos Vaders by optimus
• Oh shite, I realize why the link says youtube (live audio). I was having the stream running behind and fraps capture that too. Epic fail, I am going to sleep :P
• isokadded on the 2014-10-05 01:45:40
• 256b MS-Dos Abductee by Red Sector Inc. [web]
• Fantastic!
• rulezadded on the 2014-10-05 01:15:16
• 256b MS-Dos Vaders by optimus
• Youtube video that shows how this should run at the compo soon. I am not sure why it run like a slideshow at the stream. I wasn't expecting this. They used max cycles, but now the same settings is fine here. But I am not pissed, just curious what went wrong and I cannot recreate the same bug on dosbox by changing core, cpu and all. It only runs like this if I manually set the cycle to something like 256.
  
  Anyway, it was never finished, there were few bytes left and lot's of places to optimize.
• isokadded on the 2014-10-05 01:12:57
• 256b MS-Dos Squibs by Dila
• Nice.
• rulezadded on the 2014-10-04 18:31:30
• 512b MS-Dos Lightbox by Dila
• That's better!
• rulezadded on the 2014-10-04 12:32:43
• 512b MS-Dos Lightbox by Dila
• Kinda slow on youtube and not very visible even. Wondering if it improves on a P3/P4 booted in DOS, screenshot looks interesting though.
• isokadded on the 2014-10-03 15:46:10
• intro Amstrad CPC Boules et Bits by Vanity [web]
• Hehe,. baaalllzz
  
  Anyway, it's just very clever unroll codes per sprite. Maybe few more stuff? On AtariST, Leonard was doing something like clearing blocks of screen where many balls might be there in previous frame. So, with many balls overlapping, you clear much less area than if you individually cleared every ball alone. I don't know if the same technique is applied here since there are much less balls to make a gain, although they are more close together if you check.
  
  And the unroll could be like 16 or 32 different unroll codes. As optimized for CPC videoram specific structure, sprite starting from 0 to 7 line on Y, also from pixel 0 to 3 in starting byte. So, lot's of memory. At least that's similar to something I tried for this challenge (but never finished and I was at 23 sprites only at the time). Also, I was doing ADD HL,BC with BC = 2048 to go to next lines and then -14336 + 80. But today it can be optimized with SET/RES bits.
• isokadded on the 2014-10-03 12:58:38
• intro Amstrad CPC Boules et Bits by Vanity [web]
• I would like to see what the spectrum can do. I always thought it could do a little bit more because of smaller vram.
• isokadded on the 2014-10-03 11:18:59
• demo ZX Enhanced GirlSun by Hacker VBI
• hehe
• rulezadded on the 2014-09-28 13:05:55