2003-04-26 11:56:44 +00:00
|
|
|
/* image.c
|
2002-10-12 11:37:38 +00:00
|
|
|
*
|
|
|
|
* $Id$
|
|
|
|
*
|
|
|
|
* ***** BEGIN GPL/BL DUAL LICENSE BLOCK *****
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public License
|
|
|
|
* as published by the Free Software Foundation; either version 2
|
|
|
|
* of the License, or (at your option) any later version. The Blender
|
|
|
|
* Foundation also sells licenses for use in proprietary software under
|
|
|
|
* the Blender License. See http://www.blender.org/BL/ for information
|
|
|
|
* about this.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program; if not, write to the Free Software Foundation,
|
|
|
|
* Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
|
|
|
|
*
|
|
|
|
* The Original Code is Copyright (C) 2001-2002 by NaN Holding BV.
|
|
|
|
* All rights reserved.
|
|
|
|
*
|
|
|
|
* The Original Code is: all of this file.
|
|
|
|
*
|
|
|
|
* Contributor(s): none yet.
|
|
|
|
*
|
|
|
|
* ***** END GPL/BL DUAL LICENSE BLOCK *****
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <fcntl.h>
|
|
|
|
#include <math.h>
|
|
|
|
#ifndef WIN32
|
|
|
|
#include <unistd.h>
|
|
|
|
#else
|
|
|
|
#include <io.h>
|
|
|
|
#endif
|
|
|
|
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
#include <time.h>
|
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
#include "MEM_guardedalloc.h"
|
|
|
|
|
|
|
|
#include "IMB_imbuf_types.h"
|
|
|
|
#include "IMB_imbuf.h"
|
|
|
|
|
|
|
|
#include "DNA_image_types.h"
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
#include "DNA_texture_types.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
#include "DNA_packedFile_types.h"
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
#include "DNA_userdef_types.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
#include "BLI_blenlib.h"
|
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
#include "BKE_bmfont.h"
|
|
|
|
#include "BKE_packedFile.h"
|
|
|
|
#include "BKE_library.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
#include "BKE_global.h"
|
|
|
|
#include "BKE_main.h"
|
|
|
|
#include "BKE_image.h"
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
#include "BKE_scene.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
#include "BKE_texture.h"
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
#include "BKE_utildefines.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
|
2005-07-27 07:43:48 +00:00
|
|
|
#include "PIL_time.h"
|
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
/* bad level; call to free_realtime_image */
|
|
|
|
#include "BKE_bad_level_calls.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
void free_image_buffers(Image *ima)
|
|
|
|
{
|
|
|
|
int a;
|
|
|
|
|
|
|
|
if(ima->ibuf) {
|
|
|
|
if (ima->ibuf->userdata) {
|
|
|
|
MEM_freeN(ima->ibuf->userdata);
|
|
|
|
ima->ibuf->userdata = 0;
|
|
|
|
}
|
|
|
|
IMB_freeImBuf(ima->ibuf);
|
|
|
|
ima->ibuf= 0;
|
|
|
|
}
|
|
|
|
if(ima->anim) IMB_free_anim(ima->anim);
|
|
|
|
ima->anim= 0;
|
|
|
|
|
|
|
|
for(a=0; a<BLI_ARRAY_NELEMS(ima->mipmap); a++) {
|
|
|
|
if(ima->mipmap[a]) IMB_freeImBuf(ima->mipmap[a]);
|
|
|
|
ima->mipmap[a]= 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
free_realtime_image(ima);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void free_image(Image *ima)
|
|
|
|
{
|
|
|
|
|
|
|
|
free_image_buffers(ima);
|
|
|
|
if (ima->packedfile) {
|
|
|
|
freePackedFile(ima->packedfile);
|
|
|
|
ima->packedfile = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
Image *add_image(char *name)
|
|
|
|
{
|
|
|
|
Image *ima;
|
|
|
|
int file, len;
|
|
|
|
char *libname, str[256], strtest[256];
|
|
|
|
|
|
|
|
strcpy(str, name);
|
|
|
|
BLI_convertstringcode(str, G.sce, G.scene->r.cfra);
|
|
|
|
|
|
|
|
file= open(str, O_BINARY|O_RDONLY);
|
|
|
|
if(file== -1) return 0;
|
|
|
|
close(file);
|
|
|
|
|
2003-04-26 11:56:44 +00:00
|
|
|
/* first search an identical image */
|
2002-10-12 11:37:38 +00:00
|
|
|
ima= G.main->image.first;
|
|
|
|
while(ima) {
|
|
|
|
strcpy(strtest, ima->name);
|
|
|
|
BLI_convertstringcode(strtest, G.sce, G.scene->r.cfra);
|
|
|
|
if( strcmp(strtest, str)==0 ) {
|
|
|
|
if(ima->anim==0 || ima->id.us==0) {
|
|
|
|
strcpy(ima->name, name); /* for stringcode */
|
|
|
|
ima->id.us++;
|
|
|
|
ima->ok= 1;
|
|
|
|
return ima;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
ima= ima->id.next;
|
|
|
|
}
|
|
|
|
|
|
|
|
len= strlen(name);
|
|
|
|
|
|
|
|
while (len > 0 && name[len - 1] != '/' && name[len - 1] != '\\') len--;
|
|
|
|
libname= name+len;
|
|
|
|
|
|
|
|
ima= alloc_libblock(&G.main->image, ID_IM, libname);
|
|
|
|
strcpy(ima->name, name);
|
|
|
|
ima->ok= 1;
|
|
|
|
|
|
|
|
ima->xrep= ima->yrep= 1;
|
|
|
|
|
|
|
|
return ima;
|
|
|
|
}
|
|
|
|
|
2005-09-16 20:17:09 +00:00
|
|
|
Image *new_image(int width, int height, char *name)
|
|
|
|
{
|
|
|
|
Image *ima;
|
|
|
|
|
|
|
|
ima = alloc_libblock(&G.main->image, ID_IM, name);
|
|
|
|
|
|
|
|
if (ima)
|
|
|
|
{
|
|
|
|
ImBuf *ibuf;
|
|
|
|
unsigned char *rect;
|
|
|
|
int x, y;
|
|
|
|
|
|
|
|
strcpy(ima->name, "Untitled");
|
|
|
|
|
|
|
|
ibuf= IMB_allocImBuf(width, height, 24, IB_rect, 0);
|
|
|
|
strcpy(ibuf->name, "Untitled");
|
|
|
|
ima->ibuf= ibuf;
|
|
|
|
|
|
|
|
rect= (unsigned char*)ibuf->rect;
|
|
|
|
for(y=0; y<ibuf->y; y++) {
|
|
|
|
for(x=0; x<ibuf->x; x++, rect+=4) {
|
|
|
|
rect[0]= rect[1]= rect[2]= 0;
|
|
|
|
rect[3]= 255;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
ima->ok= 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return ima;
|
|
|
|
}
|
|
|
|
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
void tag_image_time(Image *ima)
|
|
|
|
{
|
|
|
|
if (ima)
|
2005-07-27 07:43:48 +00:00
|
|
|
ima->lastused = (int)PIL_check_seconds_timer();
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void tag_all_images_time() {
|
|
|
|
Image *ima;
|
2005-07-27 07:43:48 +00:00
|
|
|
int ctime = (int)PIL_check_seconds_timer();
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
|
|
|
|
ima= G.main->image.first;
|
|
|
|
while(ima) {
|
|
|
|
if(ima->bindcode || ima->repbind || ima->ibuf) {
|
|
|
|
ima->lastused = ctime;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void free_old_images()
|
|
|
|
{
|
|
|
|
Image *ima;
|
2005-07-27 07:43:48 +00:00
|
|
|
static int lasttime = 0;
|
|
|
|
int ctime = (int)PIL_check_seconds_timer();
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
Run garbage collector once for every collecting period of time
|
|
|
|
if textimeout is 0, that's the option to NOT run the collector
|
|
|
|
*/
|
|
|
|
if (U.textimeout == 0 || ctime % U.texcollectrate || ctime == lasttime)
|
|
|
|
return;
|
|
|
|
|
|
|
|
lasttime = ctime;
|
|
|
|
|
|
|
|
ima= G.main->image.first;
|
|
|
|
while(ima) {
|
2005-07-27 06:21:50 +00:00
|
|
|
if((ima->flag & IMA_NOCOLLECT)==0 && ctime - ima->lastused > U.textimeout) {
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
/*
|
|
|
|
If it's in GL memory, deallocate and set time tag to current time
|
|
|
|
This gives textures a "second chance" to be used before dying.
|
|
|
|
*/
|
|
|
|
if(ima->bindcode || ima->repbind) {
|
|
|
|
free_realtime_image(ima);
|
|
|
|
ima->lastused = ctime;
|
|
|
|
}
|
|
|
|
/* Otherwise, just kill the buffers */
|
|
|
|
else if (ima->ibuf) {
|
|
|
|
free_image_buffers(ima);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
ima = ima->id.next;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
void free_unused_animimages()
|
|
|
|
{
|
|
|
|
Image *ima, *nima;
|
|
|
|
|
|
|
|
ima= G.main->image.first;
|
|
|
|
while(ima) {
|
|
|
|
nima= ima->id.next;
|
|
|
|
if(ima->id.us==0) {
|
|
|
|
if(ima->flag & IMA_FROMANIM) free_libblock(&G.main->image, ima);
|
|
|
|
}
|
|
|
|
ima= nima;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-04-26 11:56:44 +00:00
|
|
|
/* *********** READ AND WRITE ************** */
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
void makepicstring(char *string, int frame)
|
|
|
|
{
|
|
|
|
short i,len;
|
|
|
|
char num[10], *extension;
|
|
|
|
|
|
|
|
if (string==0) return;
|
|
|
|
|
|
|
|
extension= "";
|
|
|
|
|
|
|
|
strcpy(string, G.scene->r.pic);
|
|
|
|
BLI_convertstringcode(string, G.sce, G.scene->r.cfra);
|
|
|
|
|
2005-05-20 12:18:11 +00:00
|
|
|
len= strlen(string);
|
2002-10-12 11:37:38 +00:00
|
|
|
|
2003-04-26 11:56:44 +00:00
|
|
|
/* can also: sprintf(num, "%04d", frame); */
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
i=4-sprintf(num,"%d",frame);
|
|
|
|
for(;i>0;i--){
|
|
|
|
string[len]='0';
|
|
|
|
len++;
|
|
|
|
}
|
|
|
|
string[len]=0;
|
|
|
|
strcat(string,num);
|
|
|
|
|
2005-11-20 14:32:07 +00:00
|
|
|
if(G.scene->r.scemode & R_EXTENSION)
|
|
|
|
addImageExtension(string);
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
}
|
|
|
|
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
void addImageExtension(char *string)
|
|
|
|
{
|
2005-07-17 08:58:42 +00:00
|
|
|
char *extension="";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
|
|
|
|
if(G.scene->r.imtype== R_IRIS) {
|
2005-11-20 14:32:07 +00:00
|
|
|
if(!BLI_testextensie(string, ".rgb"))
|
|
|
|
extension= ".rgb";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
|
|
|
else if(G.scene->r.imtype==R_IRIZ) {
|
2005-11-20 14:32:07 +00:00
|
|
|
if(!BLI_testextensie(string, ".rgb"))
|
|
|
|
extension= ".rgb";
|
|
|
|
}
|
|
|
|
else if(G.scene->r.imtype==R_RADHDR) {
|
|
|
|
if(!BLI_testextensie(string, ".hdr"))
|
|
|
|
extension= ".hdr";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
|
|
|
else if(G.scene->r.imtype==R_PNG) {
|
2005-11-20 14:32:07 +00:00
|
|
|
if(!BLI_testextensie(string, ".png"))
|
|
|
|
extension= ".png";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
|
|
|
else if(G.scene->r.imtype==R_TARGA) {
|
2005-11-20 14:32:07 +00:00
|
|
|
if(!BLI_testextensie(string, ".tga"))
|
|
|
|
extension= ".tga";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
|
|
|
else if(G.scene->r.imtype==R_RAWTGA) {
|
2005-11-20 14:32:07 +00:00
|
|
|
if(!BLI_testextensie(string, ".tga"))
|
|
|
|
extension= ".tga";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
|
|
|
else if(G.scene->r.imtype==R_JPEG90) {
|
2005-11-20 14:32:07 +00:00
|
|
|
if(!BLI_testextensie(string, ".jpg"))
|
|
|
|
extension= ".jpg";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
|
|
|
else if(G.scene->r.imtype==R_BMP) {
|
2005-11-20 14:32:07 +00:00
|
|
|
if(!BLI_testextensie(string, ".bmp"))
|
|
|
|
extension= ".bmp";
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
}
|
2005-11-20 14:32:07 +00:00
|
|
|
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
strcat(string, extension);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-04-26 11:56:44 +00:00
|
|
|
/* ******** IMAGE WRAPPING INIT ************* */
|
2002-10-12 11:37:38 +00:00
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
/* used by sequencer, texture */
|
2002-10-12 11:37:38 +00:00
|
|
|
void converttopremul(struct ImBuf *ibuf)
|
|
|
|
{
|
|
|
|
int x, y, val;
|
|
|
|
char *cp;
|
|
|
|
|
|
|
|
if(ibuf==0) return;
|
2003-04-26 11:56:44 +00:00
|
|
|
if(ibuf->depth==24) { /* put alpha at 255 */
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
cp= (char *)(ibuf->rect);
|
|
|
|
for(y=0; y<ibuf->y; y++) {
|
|
|
|
for(x=0; x<ibuf->x; x++, cp+=4) {
|
|
|
|
cp[3]= 255;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
cp= (char *)(ibuf->rect);
|
|
|
|
for(y=0; y<ibuf->y; y++) {
|
|
|
|
for(x=0; x<ibuf->x; x++, cp+=4) {
|
|
|
|
if(cp[3]==0) {
|
|
|
|
cp[0]= cp[1]= cp[2]= 0;
|
|
|
|
}
|
|
|
|
else if(cp[3]!=255) {
|
|
|
|
val= cp[3];
|
|
|
|
cp[0]= (cp[0]*val)>>8;
|
|
|
|
cp[1]= (cp[1]*val)>>8;
|
|
|
|
cp[2]= (cp[2]*val)>>8;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
/* used by sequencer, texture */
|
2002-10-12 11:37:38 +00:00
|
|
|
struct anim *openanim(char * name, int flags)
|
|
|
|
{
|
|
|
|
struct anim * anim;
|
|
|
|
struct ImBuf * ibuf;
|
|
|
|
|
|
|
|
anim = IMB_open_anim(name, flags);
|
|
|
|
if (anim == 0) return(0);
|
|
|
|
|
|
|
|
|
|
|
|
ibuf = IMB_anim_absolute(anim, 0);
|
|
|
|
if (ibuf == 0) {
|
|
|
|
printf("anim_absolute 0 failed\n");
|
|
|
|
IMB_free_anim(anim);
|
|
|
|
return(0);
|
|
|
|
}
|
|
|
|
IMB_freeImBuf(ibuf);
|
|
|
|
|
|
|
|
return(anim);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
/*
|
|
|
|
load_image handles reading the image from disk or from the packedfile.
|
|
|
|
*/
|
2004-01-26 12:37:34 +00:00
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
void load_image(Image * ima, int flags, char *relabase, int framenum)
|
|
|
|
{
|
|
|
|
char name[FILE_MAXDIR + FILE_MAXFILE];
|
2004-01-26 12:37:34 +00:00
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
if (ima->ibuf == NULL) {
|
2002-10-12 11:37:38 +00:00
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
// is there a PackedFile with this image ?;
|
|
|
|
if (ima->packedfile) {
|
|
|
|
ima->ibuf = IMB_ibImageFromMemory((int *) ima->packedfile->data, ima->packedfile->size, flags);
|
|
|
|
} else {
|
|
|
|
strcpy(name, ima->name);
|
|
|
|
BLI_convertstringcode(name, relabase, framenum);
|
2002-10-12 11:37:38 +00:00
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
ima->ibuf = IMB_loadiffname(name , flags);
|
|
|
|
}
|
|
|
|
// check if the image is a font image...
|
|
|
|
// printf("Checking for font\n");
|
2002-10-12 11:37:38 +00:00
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
if (ima->ibuf) {
|
|
|
|
detectBitmapFont(ima->ibuf);
|
|
|
|
}
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2005-03-09 19:45:59 +00:00
|
|
|
static void de_interlace_ng(struct ImBuf *ibuf) /* neogeo fields */
|
2002-10-12 11:37:38 +00:00
|
|
|
{
|
|
|
|
struct ImBuf * tbuf1, * tbuf2;
|
|
|
|
|
|
|
|
if (ibuf == 0) return;
|
|
|
|
if (ibuf->flags & IB_fields) return;
|
|
|
|
ibuf->flags |= IB_fields;
|
|
|
|
|
|
|
|
if (ibuf->rect) {
|
2003-04-26 11:56:44 +00:00
|
|
|
/* make copies */
|
2002-10-12 11:37:38 +00:00
|
|
|
tbuf1 = IMB_allocImBuf(ibuf->x, (short)(ibuf->y >> 1), (unsigned char)32, (int)IB_rect, (unsigned char)0);
|
|
|
|
tbuf2 = IMB_allocImBuf(ibuf->x, (short)(ibuf->y >> 1), (unsigned char)32, (int)IB_rect, (unsigned char)0);
|
|
|
|
|
|
|
|
ibuf->x *= 2;
|
|
|
|
/* These rectop calls are broken!!! I added a trailing 0 arg... */
|
|
|
|
IMB_rectop(tbuf1, ibuf, 0, 0, 0, 0, 32767, 32767, IMB_rectcpy, 0);
|
|
|
|
IMB_rectop(tbuf2, ibuf, 0, 0, tbuf2->x, 0, 32767, 32767, IMB_rectcpy, 0);
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
ibuf->x /= 2;
|
|
|
|
IMB_rectop(ibuf, tbuf1, 0, 0, 0, 0, 32767, 32767, IMB_rectcpy, 0);
|
|
|
|
IMB_rectop(ibuf, tbuf2, 0, tbuf2->y, 0, 0, 32767, 32767, IMB_rectcpy, 0);
|
|
|
|
|
|
|
|
IMB_freeImBuf(tbuf1);
|
|
|
|
IMB_freeImBuf(tbuf2);
|
|
|
|
}
|
|
|
|
ibuf->y /= 2;
|
|
|
|
}
|
|
|
|
|
2005-03-09 19:45:59 +00:00
|
|
|
static void de_interlace_st(struct ImBuf *ibuf) /* standard fields */
|
2002-10-12 11:37:38 +00:00
|
|
|
{
|
|
|
|
struct ImBuf * tbuf1, * tbuf2;
|
|
|
|
|
|
|
|
if (ibuf == 0) return;
|
|
|
|
if (ibuf->flags & IB_fields) return;
|
|
|
|
ibuf->flags |= IB_fields;
|
|
|
|
|
|
|
|
if (ibuf->rect) {
|
2003-04-26 11:56:44 +00:00
|
|
|
/* make copies */
|
2002-10-12 11:37:38 +00:00
|
|
|
tbuf1 = IMB_allocImBuf(ibuf->x, (short)(ibuf->y >> 1), (unsigned char)32, IB_rect, 0);
|
|
|
|
tbuf2 = IMB_allocImBuf(ibuf->x, (short)(ibuf->y >> 1), (unsigned char)32, IB_rect, 0);
|
|
|
|
|
|
|
|
ibuf->x *= 2;
|
|
|
|
/* These are brolenm as well... */
|
|
|
|
IMB_rectop(tbuf1, ibuf, 0, 0, 0, 0, 32767, 32767, IMB_rectcpy, 0);
|
|
|
|
IMB_rectop(tbuf2, ibuf, 0, 0, tbuf2->x, 0, 32767, 32767, IMB_rectcpy, 0);
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
ibuf->x /= 2;
|
|
|
|
IMB_rectop(ibuf, tbuf2, 0, 0, 0, 0, 32767, 32767, IMB_rectcpy, 0);
|
|
|
|
IMB_rectop(ibuf, tbuf1, 0, tbuf2->y, 0, 0, 32767, 32767, IMB_rectcpy, 0);
|
|
|
|
|
|
|
|
IMB_freeImBuf(tbuf1);
|
|
|
|
IMB_freeImBuf(tbuf2);
|
|
|
|
}
|
|
|
|
ibuf->y /= 2;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2005-04-28 09:57:21 +00:00
|
|
|
void ima_ibuf_is_nul(Tex *tex, Image *ima)
|
2002-10-12 11:37:38 +00:00
|
|
|
{
|
|
|
|
void (*de_interlacefunc)(struct ImBuf *ibuf);
|
2004-01-26 12:53:40 +00:00
|
|
|
int a, fra, dur;
|
2002-10-12 11:37:38 +00:00
|
|
|
char str[FILE_MAXDIR+FILE_MAXFILE], *cp;
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
if(ima==0) return;
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
strcpy(str, ima->name);
|
|
|
|
BLI_convertstringcode(str, G.sce, G.scene->r.cfra);
|
|
|
|
|
|
|
|
if(tex->imaflag & TEX_STD_FIELD) de_interlacefunc= de_interlace_st;
|
|
|
|
else de_interlacefunc= de_interlace_ng;
|
|
|
|
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
if(tex->imaflag & TEX_ANIM5) {
|
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
if(ima->anim==0) ima->anim = openanim(str, IB_cmap | IB_rect);
|
|
|
|
if (ima->anim) {
|
2004-01-26 12:53:40 +00:00
|
|
|
dur = IMB_anim_get_duration(ima->anim);
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
ima->lastquality= R.osa;
|
|
|
|
fra= ima->lastframe-1;
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2004-01-26 12:37:34 +00:00
|
|
|
if(fra<0) fra = 0;
|
2004-01-26 12:53:40 +00:00
|
|
|
if(fra>(dur-1)) fra= dur-1;
|
2002-10-12 11:37:38 +00:00
|
|
|
ima->ibuf = IMB_anim_absolute(ima->anim, fra);
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2003-04-26 11:56:44 +00:00
|
|
|
/* patch for textbutton with name ima (B_NAMEIMA) */
|
2002-10-12 11:37:38 +00:00
|
|
|
if(ima->ibuf) {
|
|
|
|
strcpy(ima->ibuf->name, ima->name);
|
|
|
|
if (tex->imaflag & TEX_FIELDS) de_interlacefunc(ima->ibuf);
|
|
|
|
}
|
|
|
|
}
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
else printf("Not an anim");
|
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
} else {
|
|
|
|
// create a packedfile for this image when autopack is on
|
|
|
|
// for performance (IMB_loadiffname uses mmap) we don't do this by default
|
|
|
|
if ((ima->packedfile == NULL) && (G.fileflags & G_AUTOPACK)) {
|
|
|
|
ima->packedfile = newPackedFile(str);
|
|
|
|
}
|
|
|
|
|
|
|
|
load_image(ima, IB_rect, G.sce, G.scene->r.cfra);
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
if (tex->imaflag & TEX_FIELDS) de_interlacefunc(ima->ibuf);
|
|
|
|
|
|
|
|
ima->lastquality= R.osa;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(ima->ibuf) {
|
|
|
|
|
2003-04-26 11:56:44 +00:00
|
|
|
/* stringcodes also in ibuf. ibuf->name is used as 'undo' (buttons.c) */
|
2002-10-12 11:37:38 +00:00
|
|
|
strcpy(ima->ibuf->name, ima->name);
|
|
|
|
|
|
|
|
if(ima->ibuf->cmap) {
|
|
|
|
|
|
|
|
if(tex->imaflag & TEX_ANIM5) {
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
if(tex->imaflag & TEX_MORKPATCH) {
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
/**** PATCH TO SET COLOR 2 RIGHT (neogeo..) */
|
2002-10-12 11:37:38 +00:00
|
|
|
if(ima->ibuf->maxcol > 4) {
|
|
|
|
cp= (char *)(ima->ibuf->cmap+2);
|
|
|
|
cp[0]= 0x80;
|
|
|
|
}
|
|
|
|
}
|
Biiig commit! Thanks to 2-3 weeks of cvs freeze...
Render:
- New; support for dual CPU render (SDL thread)
Currently only works with alternating scanlines, but gives excellent
performance. For both normal render as unified implemented.
Note the "mutex" locks on z-transp buffer render and imbuf loads.
- This has been made possible by major cleanups in render code, especially
getting rid of globals (example Tin Tr Tg Tb Ta for textures) or struct
OSA or using Materials or Texture data to write to.
- Made normal render fully 4x32 floats too, and removed all old optimizes
with chars or shorts.
- Made normal render and unified render use same code for sky and halo
render, giving equal (and better) results for halo render. Old render
now also uses PostProcess options (brightness, mul, gamma)
- Added option ("FBuf") in F10 Output Panel, this keeps a 4x32 bits buffer
after render. Using PostProcess menu you will note an immediate re-
display of image too (32 bits RGBA)
- Added "Hue" and "Saturation" sliders to PostProcess options
- Render module is still not having a "nice" API, but amount of dependencies
went down a lot. Next todo: remove abusive "previewrender" code.
The last main global in Render (struct Render) now can be re-used for fully
controlling a render, to allow multiple "instances" of render to open.
- Renderwindow now displays a smal bar on top with the stats, and keeps the
stats after render too. Including "spare" page support.
Not only easier visible that way, but also to remove the awkward code that
was drawing stats in the Info header (extreme slow on some ATIs too)
- Cleaned up blendef.h and BKE_utildefines.h, these two had overlapping
defines.
- I might have forgotten stuff... and will write a nice doc on the architecture!
2004-12-27 19:28:52 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
IMB_applycmap(ima->ibuf);
|
|
|
|
IMB_convert_rgba_to_abgr(ima->ibuf->x*ima->ibuf->y, ima->ibuf->rect);
|
|
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
converttopremul(ima->ibuf);
|
|
|
|
}
|
|
|
|
|
|
|
|
if(R.osa) {
|
|
|
|
|
|
|
|
if(tex->imaflag & TEX_ANTISCALE) {
|
|
|
|
IMB_clever_double(ima->ibuf);
|
|
|
|
IMB_antialias(ima->ibuf);
|
|
|
|
}
|
|
|
|
else if(tex->imaflag & TEX_ANTIALI) IMB_antialias(ima->ibuf);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
if(ima->ibuf)
|
|
|
|
ima->lastused = clock() / CLOCKS_PER_SEC;
|
|
|
|
else
|
|
|
|
ima->ok= 0;
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
for(a=0; a<BLI_ARRAY_NELEMS(ima->mipmap); a++) {
|
|
|
|
if(ima->mipmap[a]) IMB_freeImBuf(ima->mipmap[a]);
|
|
|
|
ima->mipmap[a]= 0;
|
|
|
|
}
|
3D view orbit option: Around Active
This fixes the active object in place when orbiting the view.
Choppy 15fps demo can be seen there: http://www.elysiun.com/~theeth/bf/around_active.html
Image Memory Grabage Collection
This adds memory handling to the image code. An image is tagged each time it is used.
During a collection cycle (frequency of cycles is user defined), if an image is older
than a user defined limit, its buffer gets deallocated. This also applies to gl memory buffers.
Images that are loading in GL memory needs to go through two time outs before being fully deallocated: the first time out deallocated the gl memorry, the second the buffer in ram.
Notes:
Image buffer loaded from python gets tagged as permanent upon load. That tag is removed when python stops using the image.
I might have missed some tagging spots, especially in the rendering pipeline. Someone with more knowledge about this code should check to be careful.
Tagging is done on every access, for rendering, this will probably be a performance hit. A scheme should be developped to only tag when the rendering is completed.
Collecting is called in draw_object, most likely not the best place to do it.
Safe from undo, since using undo deallocates memory anyway (like when loading a blend file with one currently opened)
Userpref DNA changes:
I've changed a couple of flagging variables from short to int. Some because they needed more space, others to keep SDNA happy.
Info window changes:
I've grouped a couple of buttons in aligned blocks and changed the color of mutually exclusive options to make them clearer.
Matt didn't do any changes on that in tuhopuu, so hopefully I'm not stepping on anyone's feet with this.
Also changed double constants into floats with f in a couple of places (mostly space.c) to make compiler happier.
2005-07-06 00:33:41 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|