|
|
|
This thread is locked; no one can reply to it.
|
1
2
|
| Allegro 4.9.14 |
|
Peter Wang
Member #23
April 2000
|
Allegro 4.9.14 is available: When we have nothing better to do, we rename things. This is the last time, I swear, or your money back. Quote:
Changes from 4.9.13 to 4.9.14 (September 2009) The main developers this time were: Trent Gamblin, Elias Pschernig, Paul Ports: - Elias Pschernig and Trent Gamblin started a iPhone port. Graphics: - Added al_get_opengl_texture_size and al_get_opengl_texture_position - Try to take into account GL_PACK_ALIGNMENT, GL_UNPACK_ALIGNMENT when - Fixed all (hopefully) conversion mistakes in the color conversion macros. - Sped up memory blitting, which was using conversion even with identical - Make al_set_current_display(NULL); unset the current display. - Added ALLEGRO_LOCK_READWRITE flag for al_lock_bitmap (in place of 0). - Fixed window titles which contain non-ASCII characters in X11. - Added OES_framebuffer_object extension. Input: - Added a lot more system mouse cursors. - Renamed al_get_cursor_position to al_get_mouse_cursor_position. - Prevent Windows from intercepting ALT for system menus. Filesystem: - Make the path returned by al_get_entry_name() owned by the filesystem entry - Renamed the filesystem entry functions, mainly to include "fs_entry" in - Reordered and renamed ALLEGRO_FS_INTERFACE members. - Make al_read_directory() not return . and .. directory entries. - Renamed al_create_path_for_dir to al_create_path_for_directory. - Added al_set_standard_file_interface, al_set_standard_fs_interface. Events: - Exported ALLEGRO_EVENT_TYPE_IS_USER. Threads: - Added a new function al_run_detached_thread. Other core: - Put prefixes on register_assert_handler, register_trace_handler. - Added functions to return the compiled Allegro version and addon versions. - Added al_ prefix to fixed point routines and document them. - Added al_get_system_config(). - Renamed al_system_driver() to al_get_system_driver(). - Added 64-bit intptr_t detection for Windows. - Added work-around to make OS X port compile in 64 bit mode. Addons: - Renamed addons from a5_* to allegro_*. Image addon: - Renamed the IIO addon to "allegro_image". - Renamed *_entry functions that take ALLEGRO_FILE * arguments to *_stream. - Fixed off-by-one error in greyscale JPEG loader. Audio addons: - Renamed the kcm_audio addon to "allegro_audio". - Renamed ALLEGRO_STREAM and stream functions to ALLEGRO_AUDIO_STREAM - Renamed al_stream_from_file to al_load_audio_stream - Added int16 mixing and configurable frequency and depth for default - Fixed FLAC decoding and added FLAC streaming support. - Changed the function signature of al_get_stream_fragment() to be more - Fixed bug in kcm audio that caused data to be deleted that was still used. - Made ALSA audio driver work when the driver does not support mmap (commonly - Removed al_is_channel_conf function. Font addons: - Optimized font loading by converting the mask color on a memory copy instead - Made the ttf addon read from file streams. Primitives addon: - Fixed the direction of the last segment in software line loop, and fixed - Fixed the thick ribbon code in the primitives addon, was broken for - Various fixes/hacks for the D3D driver of the primitives addon: hack to - Various enhancements to the transformations API: const correctness, - Added support for custom vertex formats. - Flipped the v axis of texture coordinates for primitives. Now (u=0, v=0) - Added a way to use texture coordinates measured in pixels. Changed PhysicsFS: - PhysFS readdir didn't prepend the parent directory name to the returned - Set execute bit on PhysFS directory entries. Examples: - Made examples report errors using the native dialogs addon if - Added new example ex_draw_bitmap, which simply measures FPS when drawing a Documentation: - Many documentation updates and clarifications. - Fixed up Info and PDF documentation. Bindings: - Added a script to generate a 1:1 Python wrapper (see `python` directory).
|
|
Thomas Fjellstrom
Member #476
June 2000
 |
In case anyone is interested, I have the docs for 4.9 auto generating twice a day at http://www.strangesoft.net/allegro/docs The pdf version is available here: http://www.strangesoft.net/allegro/docs/refman.pdf It still hasn't caught up to peters changes (as of the time I post this message), but it will within a few hours. Slightly off topic, theres also source archives of svn generated twice daily at http://www.strangesoft.net/allegro/ -- |
|
Todd Cope
Member #998
November 2000
 |
I don't see anything obvious in the changes that would cause this but the audio in 4.9.14 compared to 4.9.13 is severely lagged under Linux (Ubuntu 9.04 default setup). In 4.9.13 the audio in the demo plays with no noticeable delay but in 4.9.14 it takes about 6 seconds for the audio to start playing after a sound effect is triggered. The music is also delayed by 6 seconds. |
|
kenmasters1976
Member #8,794
July 2007
|
That's a lot of changes since 4.9.13. I'm downloading it now.
|
|
Neil Walker
Member #210
April 2000
 |
Works a treat here, I've got it down to svn download, clean build and ready for testing in about 2 minutes btw, I'm sure it's all so trivial in unix, but the fprintf to stderr for the console output examples is a right pain in arse for Windows. For the examples with errors you have to run the command and redirect it in a special way (e.g. ex_audio_simple > tmp.txt 2>&1) then open the file and read the contents. Otherwise there is no output at all, just a return back to command prompt. ex_clip is a bit slow as well. I'm only getting 125,000 fps Neil. wii:0356-1384-6687-2022, kart:3308-4806-6002. XBOX:chucklepie |
|
X-G
Member #856
December 2000
 |
That's weird, Windows should have no trouble writing to stderr. -- |
|
Neil Walker
Member #210
April 2000
|
Unless there's an obvious way to do it, that's the only way I could get it to show me the output. btw, I don't know if it helps anything, but ex_font (I mentioned this in another thread yesterday) takes about 10-15 seconds to respond. I traced the delay to the clone method. I split it up and it takes about 10 to 15 seconds for the screen_clone on my computer. ab=al_get_target_bitmap(); Neil. wii:0356-1384-6687-2022, kart:3308-4806-6002. XBOX:chucklepie |
|
Arthur Kalliokoski
Second in Command
February 2005
 |
Neil, are you compiling as a console program? (for the stderr output to appear). GUI programs don't have stdio handles, and if they did, where would the output appear? They all watch too much MSNBC... they get ideas. |
|
Neil Walker
Member #210
April 2000
|
I'm compiling against whatever cmake is giving me for visual studio. I can only assume the examples not using windows are compiled as console apps, otherwise an alert should be called rather than a stderr fprint? But yes, you're right, sending output to a stream not read by a window is a bit daft. btw, two other minor things I noticed. ex_winfull doesn't work (see bottom-right quarter of the window frame in the top-left quarter of the screen) and ex_vsync needs a epileptic fit warning before you start it. Neil. wii:0356-1384-6687-2022, kart:3308-4806-6002. XBOX:chucklepie |
|
Elias
Member #358
May 2000
|
Todd Cope said: I don't see anything obvious in the changes that would cause this but the audio in 4.9.14 compared to 4.9.13 is severely lagged under Linux (Ubuntu 9.04 default setup). In 4.9.13 the audio in the demo plays with no noticeable delay but in 4.9.14 it takes about 6 seconds for the audio to start playing after a sound effect is triggered. The music is also delayed by 6 seconds. Can you try in debug mode and attach allegro.log? (Ideally also from 4.9.13... but if you don't have it any longer 4.9.14 one is enough.) -- |
|
kenmasters1976
Member #8,794
July 2007
|
I feel bad for I'm always complaining for one thing or another. I have an integrated Intel card and my programs, as well as several of the examples are crashing with 4.9.14. I've not looked into it so much, but it seems that it has something to do with the primitives addon with DirectX. The following program runs. It doesn't draw the line, however. But if I uncomment the second al_draw_line() call, the program crashes. The program runs fine when I use OpenGL. 1#include <allegro5/allegro5.h>
2#include <allegro5/allegro_opengl.h>
3#include <allegro5/allegro_direct3d.h>
4#include <allegro5/allegro_primitives.h>
5
6
7
8static ALLEGRO_DISPLAY *display = NULL;
9
10
11
12int main() {
13 if (!al_init()) {
14 printf("Could not init Allegro.\n");
15 return 1;
16 }
17 if (!al_install_keyboard()) {
18 printf("Could not init keyboard.\n");
19 return 1;
20 }
21
22 al_set_new_display_flags(ALLEGRO_WINDOWED);
23 display = al_create_display(640, 480);
24
25 al_clear_to_color(al_map_rgb(255, 255, 255));
26 al_draw_line(0, 0, 640, 480, al_map_rgb(255, 0, 0), 2);
27 //al_draw_line(10, 10, 64, 48, al_map_rgb(0, 0, 255), 2);
28 al_flip_display();
29 al_rest(2.0);
30
31 return 0;
32}
33END_OF_MAIN()
This is a backtrace I just did with the normal build, not the debug one. I'll try to build the debug lib tomorrow. (gdb) bt
#0 0x00000000 in ?? ()
#1 0x4fda3795 in d3d9!D3DPERF_BeginEvent () from C:\WINDOWS\system32\d3d9.dll
#2 0x4fd98acd in d3d9!D3DPERF_BeginEvent () from C:\WINDOWS\system32\d3d9.dll
#3 0x67f07d90 in _al_draw_prim_directx (texture=0x0, vtxs=0x22fd50,
decl=0x0, start=0, end=4, type=5)
at C:/allegro/allegro-4.9.14/addons/primitives/prim_directx.c:193
#4 0x67f0a64c in al_draw_prim (vtxs=0x22fd50, decl=0x0, texture=0x0,
start=0, end=4, type=5)
at C:/allegro/allegro-4.9.14/addons/primitives/primitives.c:98
#5 0x67f013f5 in al_draw_line (x1=10, y1=10, x2=64.5754929, y2=48, color=
{r = 0, g = 0, b = 1, a = 1}, thickness=2)
at C:/allegro/allegro-4.9.14/addons/primitives/high_primitives.c:119
#6 0x00401480 in _mangled_main ()
#7 0x66d3253b in _WinMain (_main=0x4012e0, hInst=0x400000, hPrev=0x0,
Cmd=0x241f16 "", nShow=10)
at C:/allegro/allegro-4.9.14/src/win/wnewsys.c:137
#8 0x004014cd in WinMain@16 ()
#9 0x004017d1 in main ()
(gdb) quitAny ideas?. The problem might as well be on my end. Can someone else with an Intel card try it?. Thanks.
|
|
SiegeLord
Member #7,827
October 2006
 |
kenmasters1976 said: Any ideas?. The problem might as well be on my end. Can someone else with an Intel card try it?. Works fine for me on my ATI card. So it would seem that the problem is on your end. Does ex_prim work for you? I assume this worked fine in 4.9.13 though, right? "For in much wisdom is much grief: and he that increases knowledge increases sorrow."-Ecclesiastes 1:18 |
|
Trezker
Member #1,739
December 2001
 |
And now I've synced Allua with the new Allegro. |
|
GullRaDriel
Member #3,861
September 2003
 |
This is how misha looks like in ex_draw_bitmaps, I think something goes wrong: {"name":"599406","src":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/e\/2\/e2c2452cd9fcaeaf56404e957d1f2476.png","w":1280,"h":800,"tn":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/e\/2\/e2c2452cd9fcaeaf56404e957d1f2476"} Edit: looking at the picture, it's like the height is overridden by the width. Dunno if it's at loading or inside the example. Edit2: It's a new functionality of ex_draw_bitmaps, okay. But hey, the width should be independant of the height ! "Code is like shit - it only smells if it is not yours" |
|
Neil Walker
Member #210
April 2000
|
That's what I thought until you realise there is a set of resize keys available for making the bitmaps bigger/smaller which then adds or removes more detail. Perhaps what would have been better would be to make the image the correct size of image to start with. [edit]Just spotted your edit Neil. wii:0356-1384-6687-2022, kart:3308-4806-6002. XBOX:chucklepie |
|
MiquelFire
Member #3,110
January 2003
 |
It seems my lack of time to mess with Allegro had me miss submitting a patch (which I still have yet to create...) In the CMake configs, there's a libvorbis_include_dir (or something like it) to point where to find the include files for vorbis, but there's no setting to find the libogg include files. The way I went to manually compile the last time resulted in me not getting a DEMO because of that (Just a message saying the DEMO is turned of or whatever, but no reason why). --- |
|
GullRaDriel
Member #3,861
September 2003
|
I got no demo too. "Code is like shit - it only smells if it is not yours" |
|
MiquelFire
Member #3,110
January 2003
|
Quick fix until a patch is created and submitted Copy the ogg folder from the libogg include directory to the libvorbis include directory. You'll need to wipe the CMake cache after you do this however. --- |
|
Thomas Fjellstrom
Member #476
June 2000
|
What are you talking about? It works fine here as long as the libraries are installed prior to running cmake. -- |
|
MiquelFire
Member #3,110
January 2003
|
You're on Linux, not Windows. --- |
|
Thomas Fjellstrom
Member #476
June 2000
|
I recall it working similarly in Windows as well, though its been a while and something could have changed. -- |
|
kenmasters1976
Member #8,794
July 2007
|
SiegeLord said: Works fine for me on my ATI card. So it would seem that the problem is on your end. Does ex_prim work for you? I assume this worked fine in 4.9.13 though, right? Yes, it worked in 4.9.13. And yes, ex_prim does "run", but this is what it shows: {"name":"599407","src":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/e\/3\/e35eaa69bd686060aabdab73f9c324e7.png","w":806,"h":633,"tn":"\/\/djungxnpq2nug.cloudfront.net\/image\/cache\/e\/3\/e35eaa69bd686060aabdab73f9c324e7"} If I switch to software drawing, it works fine. I'll try to check it on my brother's computer to see if it's a problem with my machine.
|
|
MiquelFire
Member #3,110
January 2003
|
Most libraries on Windows don't install themselves like they do on Linux. --- |
|
Thomas Fjellstrom
Member #476
June 2000
|
Miquel Burns said: Most libraries on Windows don't install themselves like they do on Linux. ogg and vorbis do if you build with msys/cygwin. -- |
|
MiquelFire
Member #3,110
January 2003
|
Not using cygwin. That shall never touch my system. (And if it does, only because there's a lack of a native Windows port) --- |
|
|
1
2
|
