![]() |
![]() |
#1 |
Registered User
Join Date: Apr 2017
Location: Norway
Posts: 4
|
AGA Fetch mode 2 on A4000 vs A1200
Hi
I'm using a A4000/030 developing a game in Blitz Basic 2.1, and everything is working fine on that machine. However when testing on two different A1200 there was graphical glitches on the BOBS. Using fetch mode 1 instead of 2 fixed the problem on A1200. So what's going on here? Is it somehow more bandwidth available on the A4000? Code:
#tilesize = 16 #bitmapwidth = 384 Buffer 0, 32768 Buffer 1, 32768 BitMap 0, #bitmapwidth, 256*2 + 5*#tilesize, 4 BitMap 1, #bitmapwidth, 256*2 + 5*#tilesize, 4 BitMap 2, 320, 768, 4 ; playfield 2 background BitMap 3, 320, 16, 5 ; score panel ; Magic Eraok statement to adjust display Statement ResizeWidth{indexnum.l, amount.l, xadjust.l} diwstrt.l = xadjust * 8 diwstop.l = diwstrt + amount * 8 dUfstrt.l = Int(diwstrt/2) dUfstop.l = Int(diwstop/2) DisplayAdjust indexnum,amount,dUfstrt,dUfstop,diwstrt,diwstop End Statement Statement initgamescreen{} flags.l = $8 +$10 +$20 +$10000 +$2000 ; fetch mode 2 flags2.l = $5 +$00 +$10000 SpriteMode 2 InitCopList 0, 44, 236, flags, 8, 64, -4 ResizeWidth{0,-4,2} InitCopList 1, 284, 16, flags2, 8, 32, 0 CreateDisplay 0, 1 DisplayPalette 0, 1 DisplayPalette 1, 5 DisplayControls 0, 0, $1C20, Xor($11, $22) DisplayRainbow 0, 0, 2 End Statement |
![]() |
![]() |
#2 |
This cat is no more
Join Date: Dec 2004
Location: FRANCE
Age: 52
Posts: 8,458
|
no, it's because your screen bitplanes aren't aligned on 4 bytes. FMODE=2 needs alignment on 4, if by luck it happens it works, else you get that crap.
FMODE=3 requires alignment on 8 bytes. |
![]() |
![]() |
#3 | |
Registered User
Join Date: Apr 2017
Location: Norway
Posts: 4
|
Quote:
Thanks jotd, setting up these screens sure is tricky. For some reason it consistently works on A4000, so yeah something makes it work by luck. |
|
![]() |
![]() |
#4 |
This cat is no more
Join Date: Dec 2004
Location: FRANCE
Age: 52
Posts: 8,458
|
you might want to try alignment clauses when declaring your chipmem buffers. Or declare a buffer and compute its aligned address manually.
|
![]() |
![]() |
#5 |
Registered User
Join Date: Dec 2013
Location: Auckland
Posts: 3,561
|
Apologies if this is a little off topic - is there a good reason to use fetchmode 2 instead of 1 or 3? I've only ever supported 1 and 3 in Scorpion because I assumed (perhaps wrongly) that fetchmode 2 doesn't offer a performance boost over 1?
|
![]() |
![]() |
#6 | |
WinUAE developer
Join Date: Aug 2001
Location: Hämeenlinna/Finland
Age: 49
Posts: 26,611
|
Quote:
![]() https://eab.abime.net/showthread.php?t=86883 |
|
![]() |
![]() |
#7 |
Registered User
Join Date: Dec 2013
Location: Auckland
Posts: 3,561
|
Ah, thanks.
I never supported Fetchmode 2 as an option in Scorpion so I guess I won't bother trying to lol. |
![]() |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Screen mode with AGA only | udihow | support.Hardware | 12 | 07 January 2022 18:51 |
Copper timings and AGA fetch modes | earok | Coders. Asm / Hardware | 9 | 12 March 2021 00:15 |
AGA Fetch mode testing | earok | Coders. Blitz Basic | 1 | 29 September 2017 01:25 |
Amiga Indivision AGA A1200 or A4000 wanted | Seblington | MarketPlace | 0 | 02 December 2013 12:20 |
Details of the AGA double CAS fetch mode? | Leffmann | Coders. General | 3 | 29 July 2008 11:26 |
|
|