Messages

thanks

after investigation, I found why some configuration of XP sp2+ crash with some packers/crypters : DEP, Data Execution Prevention, the Microsoft protection built in to XP SP2+.
If activated for all programs and services => crash. It's possible to set exceptions
If only activated for windows programs and services => ok

I'm certainly not the only one XP user having DEP turned on !

Control Panel->System->Advanced->Performance->Settings->Data Execution Prevention

Btw I don't see a point to re-pack 20+ prods just because they don't work on one machine. XP is quite obsolete anyway. Let's rather find out what causes the problem! 

I'm certainly not alone (already one more XP : Celtic)
and Rules from Retro-Remakes :

Remake submissions:
The remake should be to run in Windows, please make it compatible with Windows XP, Windows 98 and Vista

so, don't use specials packers (AV issues, etc...)

yep, I tried with a debugger, dumped, but failed to fix the IAT
where am i wrong ?

and it's a little bit too long for 20+ prods packed like that

have you the UPX version used for ? I could unpack them and repack with normal upx

thanks

@padman : I just tested your prods from retroremake crunched with hacked upx : same crash now !

about Ruff'n Tumble, seems not packed. I have the message "Memory access violation".
Too bad, no more news from the author Shezzor :
http://www.youtube.com/user/shezzor

crash on "old" XP sp3 dx9 runtime vc2005-2008
same bug with all your prods (sorry) since I changed gfxcard and updated dx9 : some old prods crash with the same message "memory access violation", "memory cannot be written"

also with some of my prods crunched with fsg, mew or kkrunch (Pha CS1/2 remakes, paradox cracktros) : I recrunched them with upx, it's ok now.

But I have the same bug now with an old demo : "Natural Beat" http://www.pouet.net/prod.php?which=11058 : don't work anymore since the upgrade, but work fine under Vista DX11 and Seven 64 DX10 !!

Same thing with 2 others prods : "Ruff'n Tumble remake preview" dl here and "Ruff'n Tumble Mashup" exe version.

explanation, solution ??

XMPlay works fine
http://support.xmplay.com/
use Delix plugin (native now)

great xmastro : very funny and cool remix music 
thanx

waou!
wonderful! many thanks Rbz !!! 
you're the god of asm 

hello

I need the function getroworder from the last versions of uFmod, but I don't understand how works the new volume scale 
the values are no longer 0 to 64 (from original amiga volume min and max) but 0 to 25.
I tried to convert old volume to new one (scale /2.56) but that doesn't work.
how to use setvolume in the last version of uFmod with older values ?

thanks

 
awesome

great

I hope a good tuto about "timing" 60 fps in windowed/fullscreen.
I tried your intros with WINE under Linux Ubuntu and the speed is good. I tested my intros...err...Too Fast !
in windowed mode, I have always 85 fps like my screen properties. The good speed is only in fullscreen.

and also Tutos about Vectors

thanx 

I was thing about something like that (pre and post values)

have you drawn your own font, or is from an existing one ? (seems to be "Jellyka Estrya's Handwriting" http://www.dafont.com/fr/search.php?psize=m&q=jellyka+estrya+handwriting)

Hmm, I might even throw it together as a standalone proggy with a proper tutorial if people want...

that will be cool 

wonderful scrolltext ! love the font and proportional+overlap  great effect (how  )

oops... a little bit too late 

Code Select Expand

lea  4(a0),a0
move.w  d0,4(a0)
swap  d0
move.w  d0,(a0)
swap  d0

start of the routine missing :

Code Select Expand


move.l #BpVector,d0
add.l d7,d0
lea CLvect-2,a0
moveq #0,d4
moveq #7-1,d7 ; 7 lines


now it's more simple to understand what do the prog :
only install Bitplan Vector address (d0) into the Copperlist Bitplan registers pointed in A0 (CLvect=CopperList)

CopperList:

Code Select Expand

CLvect
dc.w $e0,0,$e2,0 ; bitplan 0
dc.w $e4,0,$e6,0 ; bitplan 1
dc.w $e8,0,$ea,0 ; bitplan 2
dc.w $ec,0,$ee,0

$ex = $dff0ex = bitplans pointers (screen)
bitplan0 : $dff0e0 long word ($e0 and $e2)
d0=address high.w low.w .
example : for BpVector at d0=$51000
move.w d0 = $1000
swap d0
move.w d0 = $0005
$dff0e0 .w = $0005, $dff0e2 .w =$1000 => $dff0e0 .L = $51000

One thing I'm confused about, is that there is this funny chinese character in one of the demo groups names (Moses # ex-setrox), but the fontmap that hellsangel ripped didnt have it in ( added it of course). Does that mean he made that font bitmap manually and just missed the character?  I thought it was taken directly from memory.

no, this character is not in the original font (not made manually! the font is like that in memory), no more in the Characters Table : "abcdefghijklmnopqrstuvwxyz1234567890-.:+^,!?><() "
the text is "Moses / ex-setrox". The "/" is missing.

I'll take a look at the source to see what the program do with the "/"

Edit : so, it's not a "funny chinese character" but a funny bug ^^
no test is done if a char is not in the table => the char address is the second bitplan of the font bitmap (char "A").
see the screenshot of the font in memory

work fine (XP and Vista) !
great (I missed to rip the pattern for the random pixels text fading, but not needed !)

LMB to exit missing

woaw! 

! courageous to retype all the text 

for all the questions, I disassembled the program after unpacked : I use megamon, a desassembler/debugger. When I have all addresses of gfx, text,... I can rip them with megamon and my fav gfx ripper 3rdday, and for the text, I "coded" directly in memory the "decrypter".
the text is just at the start of the program:
"bra StartIntro
dc.b '...' ; text
...
StartIntro:
... ; code
"
for the whole program, I disassemble it with IDA (Interactive Disassembler) under Windows. (shame on me, I dont use ReSource  on Amiga )