home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Aminet 28
/
Aminet 28 (1998)(GTI - Schatztruhe)[!][Dec 1998].iso
/
Aminet
/
util
/
misc
/
RunLame.lha
/
RunLame1.28
/
runlame.lha
/
RunLame
/
Docs
/
RLInfiltrator.history
< prev
next >
Wrap
Text File
|
1993-08-25
|
4KB
|
99 lines
History:
-------
V1.0ß: ·Not released.
V1.1ß: ·Not released.
·Due to improved system conformity the little PAL screen has a name
now.
V1.2ß: ·Not released.
·The PAL screen name is now identical for all forthcoming versions,
so it is not needed to update any databases where the screen name
has been registered (NewMode, Domino promotion system, ...).
·Like RunLame now clears caches after having moved the VBR, I
strongly recommend NOT TO USE earlier versions of
RunLameInfiltrator.
·Funnier version string, making all authors of lame programs blush.
V1.3: ·Released to the public.
·The internals of RunLameInfiltrator have been extensively improved,
so I strongly recommend NOT TO USE earlier versions of
RunLameInfiltrator, they may not work properly on some Amiga
models!!
·Bug fixed: The PAL screen tag list was not terminated with
TAG_DONE. I wonder why this worked:-)
·The PAL screen now is in PAL:Low Res (non interlaced).
·The gb_MemType is set to BANDWIDTH_1X now on AGA machines. This
has been suggested by ID of Sphinx. The tip came originally from
Randell Jesup.
·Changing and restoring gb_MemType is now safer (no task switching
in meanwhile). When another task changes gb_MemType after we did,
the old value is not going to be restored, so we don't kill the
environment of the other task (BTW, nobody had problems, but I
think it's just smarter this way).
·gb_MemType is only changed during the opening of the little PAL
screen. Both happens while only interrupts are allowed but no
taskswitching, thus reducing the chance another task can open a
screen with the wrong bandwidth. Furthermore all screens opened by
the program that has been infiltrated will have the former system
bandwidth, but these programs should be aware of that when they are
systemfriendly (I know only 2 demos which do open a screen using
OpenScreen(TagList)()).
·Heavy internal changes to the processor degrading/restoring code.
It is now much more safer and no inconsistencies in multitasking
with other tasks changing the same registers should occur (though
nobody reported any problems).
V1.4ß: ·Not released.
·Bug fixed: By accident the PAL screen has had no name in V1.3.
·The PAL screen name is now identical for all forthcoming versions,
so it is not needed to update any databases where the screen name
has been registered (NewMode, Domino promotion system, ...). This
should have been happened since V1.2ß, dunno why not.
·Bug fixed: Due to usage of the function CacheControl()
RunLameInfiltrator crashed under lower Exec versions than V36. I
didn't have a V34 Kickstart available , so I didn't see that this
function is not available under V34- (and the autodocs don't point
out that it's a V36+ function).
·Access to graphics.library base smarter now.
V1.5ß: ·Not released.
·The sprite resolution for the PAL screen is explicitely reset to
low resolution. This might not be needed, because the sprite
resolution for a that PAL screen should be lores anyway. In case
of false promotion of that PAL screen, at least the sprites will be
set to lores now.
V1.6: ·Second public release.
·The sprite DMA of the mouse pointer will point to a NULL graphic,
thus reducing the chance of sprite flicker if the lame program does
just disable the sprite DMA in the wrong moment. This does only
work if the user doesn't activate any window in the execution
phase, because all windows keep their pointer for greater
convenience.
·Checking if the PAL screen opened really in PAL lores. Results in
an error message if a promotor did change the screen mode (only if
you have OS2+ installed, but think twice, have you ever seen a
promotion utility for OS1.3-?)