Sysfault Blog

dynrules 0.0.14 has been released. This is a minor compatibility fix release, which adds Python 3.2 support.

dynrules 0.0.13 has been released. This is a minor bugfix release, which fixes a bug in the RuleManagement constructor.

Enjoy!

Another release, another backwards compatibility break.

LJGF 0.0.7 adds some minor feature improvements, such as vertical scrolling, rendering overloads to enforce redrawing, and fixes some serious issues with ParallaxBackground clipping and (viewport) location offsets for the GameField class.

Rendering was heavily improved by adding dirty area lookups for effected areas over all SpriteGroup layers, getting rid of left over artifacts on repositioning sprites and optimizing edge clipping for the GameField viewport. The RectCollisionArea code was improved by removing superfluous intersection testing, making it notably faster for medium to large sprite amounts to test.

Have fun!

Monday evening the students participating in this year's GSoC were officially announced. Pygame was in the lucky situation to get a slot from MoinMoin, so we made it and now have one student to work on Pygame this summer.

Julian Habrock will work on improving 2D drawing in Pygame. His proposal already mentioned several very good ideas, though this did not let him stop to come up with even more that could be realized on time. I'll mentor him and can't wait to see what he'll come up with.

LJGF 0.0.6 has been released. It is a minor update, which improves the main game loop to provide a more exact timer resolution.

Enjoy!

So yesterday the student proposal submission ended and I am heavily impressed in two ways. First, there were less proposals for Pygame than the years before, but that does not mean much, because it seems that the overall interest in the GSoC is not as high as the last years.

However, the handful of proposals submitted specifically for Pygame are (quality-wise) top-notch. Even worse, the students introduced themselves quite some time before submitting on the mailing list (something which only was done by about 60% of the students in the last years), discussed their ideas, made adjustments based on the feedback and some even planned ahead regarding the extensibility of what they want to accomplish.

This will be probably the hardest selection to be done and I would love to see them all being accepted and working on their projects. As this is hardly possible, maybe we can convince those left behind to work on their projects anyways, which would be really great.

It's time again for another summer of code. I'll be mentoring for pygame related projects again this year and despite the fact that there seems to be less overall interest I can't wait to start with it. This is mainly due to the fact that less proposals mean higher quality (although that's usually not the case). The few students who expressed interest and submitted a proposal so far, did an excellent on thinking about it and writing it down.

All in all, it will be a tough job on choosing a student for mentoring, but maybe the one or other, who'll be rejected, will work on the project anyways. That would be really nice, especially since there are quite some interesting proposals this year, such as implementing OSC networking support, a well-written and well-designed AI module idea and a solid drawing module redesign approach.

We were lucky with the students of the last year, of which two actually worked on their project, although they were rejected for the GSoC itself and thus were not paid for doing that job. Hopefully we'll see the same engagement in 2010!

cflow 0.0.6 has been released. This is a minor bugfix release, which fixes the operator handling for '!' and '|' in C source code files.

Enjoy!

dynrules 0.0.12 has been released. This is just a minor bugfix release, which fixes the file recognition within the C implementation of the Learnsystem class.

Enjoy!

LJGF 0.0.5 was released to the wild, adding fun bag of new features, such as

• an improved collision detection system, including circle collisions
• abstract and concrete sound engine support for PCM-based samples (WAV, AU, AIFF, …) and MP3 data
• an improved main loop system using yielding and sleeping threads

0.0.5 breaks backwards compatibility, though, as lots of classes were moved into different namespaces or renamed to match the new additions. A small disadvantage is, that the MP3 sound engine relies upon JLayer, which is LGPL software (despite the licensing, it is an excellent package for working with MP3 encoded audio data). This however does not have any impact on the rest of the sofware, which is still Public Domain.

Enjoy!