Hi,
if memory matters, we could also strip the sounds,
manuals of course, and some of the level packs?
And we might deactivate the thumbnail-creation,
but I don't know if this would be a good idea:
Either no thumbnails at all, or never save
thumbnails, always create them from scratch?
Greets,
Andreas
Pipeline schrieb:
Just to clarify, your not planning on releasing this before april
right? I don't think he's planning on releasing 1.10 until around
then. Or perhaps do a build of 1.01 source + performance patches.
The installer works nice, although maemo internal jffs filesystem is
already compressed so including the data internally is still only
around 6meg footprint (xerces + enigma + data) even though the actual
filesizes are around 16megs (uncompressed). Moving to memory card,
will however free up maybe 2-3 megs internal memory (at a cost of 9-12
megs on uncompressed mmc).
Have you been able to test a tilt device with the build? If so what
were your results? I would imagine the 'hollows' levels which are now
optimized would work well with tilt. I'd be interested in knowing how
sokobon levels do.
----- Original Message ----- From: "Till Harbaum / Lists"
<[EMAIL PROTECTED]>
To: <[email protected]>
Cc: "Ronald Lamprecht" <[EMAIL PROTECTED]>; "Pipeline"
<[EMAIL PROTECTED]>
Sent: Monday, February 18, 2008 2:55 PM
Subject: Re: [Enigma-devel] Maemo Port for Nokia Internet Tablets
Hi,
i have finished the first set of changes and i think i have a nice
maemo port
of the latest 1.10 r1040. It includes all the features expected from
a maemo
app incl. icons in the required sizes and installation routines. I
have only
done what i think is the minimal requirement to have a nice maemo
port. This
includes:
- removed/disabled all graphics besides 640x480
- removed/disabled all things related to change the resolution
- removed/disabled all things related to change fullscreen
- removed/disabled all mouse cursor related stuff
- added maemo specific icons/config files/setup routines/...
- added pipelines touchscreen code as well as the tiltstick interface
As the footprint is still quite big (~17MB) i have decided to add
some install
dialog that asks the user during installation whether he wants to
move the
data branch onto the internal memory card. This should be fully
transparent
to the user. This is of course cleaned up on de-installation.
Pipepline, you'll find a demo version in my maemo repository, so you
can test
it if you want to. All the dependencies should also be there, so the
entire
thing should be a one click install.
I can now either post some patches or start a new branch in svn
(however this
is done ...)
Regards,
Till
Am Sonntag 17 Februar 2008 schrieb Ronald Lamprecht:
Hi,
Ronald Lamprecht wrote:
> I thought I had eliminated all algorithms with quadratic behaviour on
> the number of actors :-(
>
> But I never checked the hollows thoroughly! Indeed I just dected by a
> quick code review a totally unecessary quadratic algorithm that
can be
> replaced by a linear one. Thanks for the report!
Gosh - it can even be done in O(1) with a constant that is less a
hundredth of the old algorithms constant!
I commited an unfinished patch with r1040 to the trunk. It still needs
some cleanup and thorough compatibility testing. But first I am
interested on the performance reports for the Maemo. The force
calculation should now be dominant, but this should cause no
problems if
other levels with the same number of actors do perform well.
Greets,
Ronald
_______________________________________________
Enigma-devel mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/enigma-devel
_______________________________________________
Enigma-devel mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/enigma-devel
_______________________________________________
Enigma-devel mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/enigma-devel
