[Soc-2008-dev] Week 9 report

[Davide Vercelli]

Week 9
======

This week
---------
Last week's verifiable target, "add support for all diffuse types",
was successfully met. This required finding out a suitable bounding
function for each diffuse shader, so it was more a "theoretical" work
than a programming task.

Also added support for all lamp falloff types, custom curves excluded.
Custom curve falloff is currently trivially bound to 1.0; it could be
handled slightly better in some circumstances but I would say that
it's not a priority right now. The "Sphere" option is still missing as
well.

Minor changes:
- check that shadowing and raytracing are enabled before initializing
lightcuts (or crashes can occur); there's no user feedback, though, if
lightcuts doesn't get used (basically it renders as always)
- added error rate to console stats

While studying the diffuse shaders in shadeoutput.c, I also took the
liberty to enhance code readability a bit, using macros (such as INPR,
MIN2, MAX2, CLAMPIS) where possible and removing some unnecessary
code. I reviewed those changes several times but further code reviews
are welcome.

I also started a document outlining my integration plans for Lightcuts:
http://wiki.blender.org/index.php/User:UncleZeiv/SummerOfCode2008/Integration
(It's currently mostly empty, only contains some insightful remarks
from Matt Ebb and some random thoughts of mine on the matter).

Next week
---------
Possible projects for next week are: faster tree creation or basic
instant radiosity. I have yet to decide and I'm currently discussing
the matter with Kent.

Issues
------
I had no input on the "shadeSkyView" problem (see last week's report).

Schedule
--------
Possibly slightly late, but I'm currently discussing detailed plans
for the last part of the project with Kent.

Davide