![]() |
![]() |
#1021 | |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
Quote:
Now we'ree wasting a lot of time with blitter waiting since, so far, no interleaved bitmap used. We're doing well, a lot of spare dma cicles. There is a bug on winuae 3.3.0 that prevent me to copy my copperlist Last edited by sandruzzo; 15 June 2016 at 11:23. |
|
![]() |
![]() |
#1022 | |
Code Kitten
![]() Join Date: Aug 2015
Location: Montreal/Canadia
Age: 52
Posts: 1,178
|
Quote:
![]() |
|
![]() |
![]() |
#1023 | |
Going nowhere
Join Date: Oct 2001
Location: United Kingdom
Age: 50
Posts: 9,016
|
Quote:
|
|
![]() |
![]() |
#1024 |
Code Kitten
![]() Join Date: Aug 2015
Location: Montreal/Canadia
Age: 52
Posts: 1,178
|
UAE also has very handy copperlist dumping debug functions which can be even more practical than a native debugger.
Cf this post from Tony: http://eab.abime.net/showpost.php?p=963797&postcount=9 |
![]() |
![]() |
#1025 |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
|
![]() |
![]() |
#1026 |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
here My copperlist
Copper: dc.w $01fc,0 ; slow fetch mode AGA compatibility dc.w $0100,$5200 CopScreen: dc.w $8E,$2c98 ; DiwStrt dc.w $90,$f481 ; DiwStop dc.w $8E,$2c81 ; DiwStrt dc.w $90,$2cc1 ; DiwStop dc.w $92,$0039 ; DdfStart dc.w $94,$00aa ; DdfStop dc.w $0108,$0000+4 dc.w $010a,$0000+4 dc.w $0102 CopScroll: dc.w $0000 dc.w $0104,$0001 CopSpr: dc.w $0120,00000 dc.w $0122,00000 dc.w $0124,00000 dc.w $0126,00000 dc.w $0128,00000 dc.w $012a,00000 dc.w $012c,00000 dc.w $012e,00000 dc.w $0130,00000 dc.w $0132,00000 dc.w $0134,00000 dc.w $0136,00000 dc.w $0138,00000 dc.w $013a,00000 dc.w $013c,00000 dc.w $013e,00000 CopBpl: dc.w $00e0,00000 dc.w $00e2,00000 dc.w $00e4,00000 dc.w $00e6,00000 dc.w $00e8,00000 dc.w $00ea,00000 dc.w $00ec,00000 dc.w $00ee,00000 dc.w $00f0,00000 dc.w $00f2,00000 dc.w $0180 Color0: dc.w $0000 dc.w $0182,$0310,$0184,$0700,$0186,$0900 dc.w $0188,$0c00,$018a,$0c70,$018c,$0eb0,$018e,$0eed dc.w $0190,$0eca,$0192,$0ca8,$0194,$0a87,$0196,$0864 dc.w $0198,$0444,$019a,$0577,$019c,$0799,$019e,$09bb dc.w $01a0,$0455,$01a2,$0852,$01a4,$0c94,$01a6,$0efb dc.w $01a8,$0852,$01aa,$0064,$01ac,$0095,$01ae,$05c8 dc.w $01b0,$0c94,$01b2,$0420,$01b4,$0520,$01b6,$0630 dc.w $01b8,$0efb,$01ba,$0840,$01bc,$0950,$01be,$0d94 SprPos: ds.w 40*ParallaxHeight dc.w $ffff,$fffe Here my tab used to generate copper sprite SprPrxS: dc.w $0140,$0048 dc.w $0148,$0050 dc.w $0150,$0058 dc.w $0158,$0060 dc.w $0160,$0068 dc.w $0168,$0070 dc.w $0170,$0078 dc.w $0178,$0080 dc.w $0140,$0088 dc.w $0148,$0090 dc.w $0150,$0098 dc.w $0158,$00a0 dc.w $0160,$00a8 dc.w $0168,$00b0 dc.w $0170,$00b8 No voodoo here.. Last edited by sandruzzo; 19 June 2016 at 12:38. |
![]() |
![]() |
#1027 |
Registered User
Join Date: Dec 2015
Location: Poland
Posts: 189
|
Ok, as far as I can see there is the following cost of different techniques:
1. 32 colours 5 bitplane front page - no dual paralax (only front gameplay area). Ryggar sprite fetches 8 words per scanline 8x32=256cc 2. 26 colours 5 bitplane front page - 128x100px background sprite paralax Ryggar is 4 bitplane BOB and fetches 3x29x32=2784cc Sprites fetches 8x2x100=1600cc Cooper fetches 32x100=3200cc Total cost: 2784+1600+3200=7584cc Using this method paralax costs us 7584-256= 7328cc (every frame)more than no paralax mode 3. 26 colours 5 bitplane front page - 64x100px background sprite 4-7 paralax Ryggar sprite 0-3 fetches 8 words per scanline 8x32=256cc Sprites fetches 4x2x100=800cc Cooper fetches 64x100=6400cc Total cost: 256+800+6400=7456cc Using this method paralax costs us 7456-256= 7200cc (every frame)more than no paralax mode I believe version 3 is bad idea because it actually fetches only little less colour clocks, at the cost of much smaller paralax screen. I think our approach is the most optimal in case of 5 bitplane screen. We finally will use a 256x192 screenmode, plus 16 pixels for scrolling, resulting in bitplane DMA for 272x192. It uses 5 bitplanes. This costs (272/16)*192*5 colour clocks, or 16320 colour clocks total for bitplane DMA. The Amiga has 226x313 colour clocks available per PAL frame, or 70738 in total. Memory refresh costs 4 colour clocks per scanline: 313x4 = 1252. Audio costs up to 4 colour clocks per scanline: 313x4 = 1252. Background paralax cost = 7328 So, finally we should have 70738 - 1.252 - 1.252 - 7328 - 16320 = 44586 colour clocks left. Each enemy bob costs also 2784, so ten of them would requie 27840 colour clocks, still plenty to deal with the rest of the game. I believe we can fit into 50Hz!!! Last edited by Trachu; 23 June 2016 at 00:28. |
![]() |
![]() |
#1028 | |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
Quote:
Who know maybe we can even try 6 planes... Keep in mind that since I don't do the classic coockie cut way: save screen->draw bob->restore screen. we have a lot power to use. I do this: draw-bob->restore screen Last edited by sandruzzo; 23 June 2016 at 11:49. |
|
![]() |
![]() |
#1029 |
Registered User
Join Date: Dec 2015
Location: Poland
Posts: 189
|
hows it possible to restore screen in situation you dont know what it was before because you have not save it????
Considering your request i have a few questions: 1. Since it is the mask which selectes the pixels can i use all 8 coloursexcept transparent?? 2, I believe in this situation what we are looking is two sets of 8 colours bobs, but one of this sets must use half bright colours hmmmm. Some bobs would use one set others would use other. |
![]() |
![]() |
#1030 | |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
Quote:
![]() 1) I think so, we have to try. 2) Yes, since we use half bright, we have to do some test. |
|
![]() |
![]() |
#1031 |
Registered User
Join Date: Dec 2015
Location: Poland
Posts: 189
|
Considering Amiga architecture limitations this is what i allocated
Bitplanes - colour register 000000 - 00 copperised sprite colour 0 000001 - 01 Foreground 000010 - 02 Foreground 000011 - 03 Foreground 000100 - 04 Foreground 000101 - 05 Foreground 000110 - 06 Foreground 000111 - 07 Foreground 001000 - 08 Foreground 001001 - 09 Foreground 001010 - 10 Foreground, BOB colour set 1, colour 1 001011 - 11 Foreground, BOB colour set 1, colour 2 001100 - 12 Foreground 001101 - 13 Foreground 001110 - 14 Foreground, BOB colour set 1, colour 3 001111 - 15 Foreground, BOB colour set 1, colour 4 010000 - 16 Foreground 010001 - 17 background sprite colour 1 010010 - 18 background sprite colour 2 010011 - 19 background sprite colour 3 010100 - 20 background sprite colour 1 010101 - 21 Foreground, BOB colour set 2, colour 1 010110 - 22 Foreground 010111 - 23 Foreground, BOB colour set 2, colour 2 011000 - 24 background sprite colour 2 011001 - 25 Foreground 011010 - 26 Foreground, BOB colour set 1, colour 5 011011 - 27 Foreground, BOB colour set 1, colour 6 011100 - 28 background sprite colour 3 011101 - 29 Foreground, BOB colour set 2, colour 3 011110 - 30 Foreground, BOB colour set 1, colour 7 011111 - 31 Foreground, BOB colour set 2, colour 4, BOB colour set 1, colour 8 .... .... 110101 - 53 Foreground, BOB colour set 2, colour 5 (half bright of colour 1) 110111 - 55 Foreground, BOB colour set 2, colour 6 (half bright of colour 2) 111101 - 61 Foreground, BOB colour set 2, colour 7 (half bright of colour 3) 111111 - 63 Foreground, BOB colour set 2, colour 8 (half bright of colour 4) |
![]() |
![]() |
#1032 | |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
Quote:
You did a hell of job, and they look so good! Now i'm working on buddies system. 8 colors for enemies seems to few. Do you have any test pic? |
|
![]() |
![]() |
#1033 |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
this would be the layout only for bobs':
000001 rygar color 1 000100 rygar color 2 000101 rygar color 3 010000 rygar color 4 010001 rygar color 5 010100 rygar color 6 010101 rygar color 7 enemies: 000010 enemy color 1 001000 enemy color 2 001010 enemy color 3 100000 enemy color 4 100010 enemy color 5 101000 enemy color 6 101010 enemy color 7 |
![]() |
![]() |
#1034 | |
Registered User
Join Date: Dec 2015
Location: Poland
Posts: 189
|
Quote:
Your enemies will use bright version of colour 0 which is copperised, so the enemies could use only 6. Mine version is better, |
|
![]() |
![]() |
#1035 |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
|
![]() |
![]() |
#1036 |
2 contact me: email only!
Join Date: May 2001
Location: Auckland / New Zealand
Posts: 3,187
|
Why do you want to limit the colours so much just to save slightly on the drawing? It seems crazy to sacrifice so many colours just so you don't have to cookie cut 5 bitplanes (assuming you're doing 3 cookie cut planes and then punching the mask into the other 2).
|
![]() |
![]() |
#1037 | |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
Quote:
|
|
![]() |
![]() |
#1038 |
Registered User
Join Date: Dec 2015
Location: Poland
Posts: 189
|
Codetapper: Sandruzzo wants to make the Ryggar technical masterpiecie and that is why i like him
![]() You started this thread by hipthetical question we turn back in time. Now we know ,in 1990 Sandruzzo would never finish the game LOL. This thread is becoming more than just this game, we are exploring unresearched Amiga possibilities. Converting the sprites to 8+8 is doable however not easy as long as you want to have good quality. Last edited by Trachu; 24 June 2016 at 10:47. |
![]() |
![]() |
#1039 | |
Registered User
Join Date: Feb 2011
Location: Italy/Rome
Posts: 2,344
|
Quote:
|
|
![]() |
![]() |
#1040 |
Registered User
Join Date: Dec 2015
Location: Poland
Posts: 189
|
considering 8c enemies this is what i have so far :P
|
![]() |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
rygar? | Prosonic | support.Games | 7 | 09 June 2013 14:18 |
two more that borrow heavily from Rygar & Black Tiger | NfernalNfluence | Nostalgia & memories | 5 | 30 September 2012 18:57 |
SCIENCE 451-RYGAR: same disks | mrodfr | project.TOSEC (amiga only) | 0 | 26 December 2006 15:38 |
Wordworth conversion | vertigo | support.Apps | 5 | 09 December 2003 14:46 |
|
|