[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Gzz-commits] gzz TODO
From: |
Tuomas J. Lukka |
Subject: |
[Gzz-commits] gzz TODO |
Date: |
Fri, 27 Sep 2002 03:08:30 -0400 |
CVSROOT: /cvsroot/gzz
Module name: gzz
Changes by: Tuomas J. Lukka <address@hidden> 02/09/27 03:08:30
Modified files:
. : TODO
Log message:
twid
CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/gzz/TODO.diff?tr1=1.151&tr2=1.152&r1=text&r2=text
Patches:
Index: gzz/TODO
diff -c gzz/TODO:1.151 gzz/TODO:1.152
*** gzz/TODO:1.151 Fri Sep 27 03:01:18 2002
--- gzz/TODO Fri Sep 27 03:08:30 2002
***************
*** 17,31 ****
gzz.mediaserver.AddToMediaserver -c).
This way, we can refer to them so that people can get their
own instances themselves.
- - unit tests for libcoords:
- - a general test which tests that a CoordSys implementation
- and its inverse match up (enabled for most coordsys).
- - tests for the hierarchy working correctly
- - a general test that transform() and performGL() do the same thing
- - implement DistortCoords::createInverse(). This is not easy, and
- it IS allowable to use numeric approximation for root-finding.
- Make sure you abstract this into libdistort properly
- (inverse should be in the functions themselves)
vegai:
- make papermill (libpaper)
into real python modules so
--- 17,22 ----
***************
*** 46,77 ****
- new vertex() routine, which works without
vertex program, both with and without paper
position.
- make nonlinearity() really work right for
distorted coord sys and affine coordsys on top
of it.
Check by setting polygonmode to lines in xupdf
(by a key) and seeing that the non-distorted
polygons are diced less.
- - make mstmpimg generate images also in a resolution which
- better fits
- 1024x1024 textures - currently a LOT of space is being wasted
- - note however that we DO want square texels;
- ask Tjl for explanation
- - fix utf8 string non-null end bug found by Benja
jvk:
- - javadoc Paper.hxx: how to use Vertex Program interface.
- fix irregularquad to work with new coordsys stuff
tjl:
- - fix coordsys to take an iterator, which does the lerping,
- so we can store parameters more naturally and also create
- coordsys independently of the CoordSet stuff.
- - move coordsys into their own header file, or at least
- coords header file...
- - this is needed also by making the inversions
- distorted multi-page PDF view with xu links
- - abstract distortion out into libfisheye, for buoy edges etc.
- - use templates to make sure that we get performance
- - vertex arrays?
- buoy edges, distorted
- clean up the xupdf.py file
- texture performance, look at imagecache
--- 37,53 ----
- new vertex() routine, which works without
vertex program, both with and without paper
position.
+ - javadoc Paper.hxx: how to use Vertex Program interface.
- make nonlinearity() really work right for
distorted coord sys and affine coordsys on top
of it.
Check by setting polygonmode to lines in xupdf
(by a key) and seeing that the non-distorted
polygons are diced less.
jvk:
- fix irregularquad to work with new coordsys stuff
tjl:
- distorted multi-page PDF view with xu links
- buoy edges, distorted
- clean up the xupdf.py file
- texture performance, look at imagecache
***************
*** 93,99 ****
- when a gl window is closed, call Binder.windowClosed()
+ fillet implementation handling all connections from a node
simultaneously.
- + implementation of frame rate counter magic Vob
- fix things broken by new coordsys stuff
- PLEASE REPORT THEM HERE!!! If something is vital, move this up.
anybody:
--- 69,74 ----
***************
*** 101,109 ****
0.8alpha3: Crucial demos for continued funding [Deadline: 4.10]
( unlike the other releases, the things here are in priority order )
! deetsay:
! - UML and APIs of ImageCache
! - implementation of ImageCache
jvk, tjl:
- think about using render-to-texture to speed up papers.
We now control the repeating units; rendering one such
--- 76,99 ----
0.8alpha3: Crucial demos for continued funding [Deadline: 4.10]
( unlike the other releases, the things here are in priority order )
! tuukkah:
! - unit tests for libcoords:
! - a general test which tests that a CoordSys implementation
! and its inverse match up (enabled for most coordsys).
! - tests for the hierarchy working correctly
! - a general test that transform() and performGL() do the same thing
! - needs unit test framework for C++
! + implement DistortCoords::createInverse(). This is not easy, and
! it IS allowable to use numeric approximation for root-finding.
! Make sure you abstract this into libdistort properly
! (inverse should be in the functions themselves)
! humppake:
! - make mstmpimg generate images also in a resolution which
! better fits
! 1024x1024 textures - currently a LOT of space is being wasted
! - note however that we DO want square texels;
! ask Tjl for explanation
! - fix utf8 string non-null end bug found by Benja
jvk, tjl:
- think about using render-to-texture to speed up papers.
We now control the repeating units; rendering one such
***************
*** 113,118 ****
--- 103,109 ----
number of passes as much. However, this brings
caching problems.
tjl:
+ + implementation of frame rate counter magic Vob
- distorted multi-page PDF view with xu links
- mouse clicks on buoys to work
- limit mousing area to the real area of the paper.
***************
*** 141,146 ****
--- 132,140 ----
of text bulged, from a single Enfilade1D. - bulging through new cs?
0.8alpha4: Cleanup release after crucial demos for continued funding
+ deetsay:
+ - UML and APIs of ImageCache
+ - implementation of ImageCache
tuukkah:
- move Modules/ to basalt (what about Java/ and C++/) (Tjl: remove them)
anybody:
***************
*** 171,176 ****
--- 165,176 ----
and Vector.
- 4-dimensional homogeneous type?
- docxx new additions to gfx/
+ - fix coordsys to take an iterator, which does the lerping,
+ so we can store parameters more naturally and also create
+ coordsys independently of the CoordSet stuff.
+ - move coordsys into their own header file, or at least
+ coords header file...
+ - this is needed also by making the inversions
mudyc:
+ make libcallgl use GLX_ARB_get_proc_address for
the functions in
- Re: [Gzz-commits] gzz TODO, (continued)
[Gzz-commits] gzz TODO, Benja Fallenstein, 2002/09/25
[Gzz-commits] gzz TODO, Matti Katila, 2002/09/25
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/26
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/26
[Gzz-commits] gzz TODO, Benja Fallenstein, 2002/09/26
[Gzz-commits] gzz TODO, Asko Soukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO,
Tuomas J. Lukka <=
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27
[Gzz-commits] gzz TODO, Tuomas J. Lukka, 2002/09/27