If the answer isn't here, ask and it will be added...
See readme.txt.
From any SimTel mirror, in the directory gnu/djgpp/v2tk/, or from the Allegro homepage, http://www.talula.demon.co.uk/allegro/.
See allegro.txt.
An MSDOS port of the GNU C/C++ compiler, by DJ Delorie and friends. Great stuff...
From any SimTel mirror, in the directory gnu/djgpp/, or from DJ's site, http://www.delorie.com/djgpp/.
See the readme.1st file that comes with djgpp. Most of the problems that people have with Allegro are due to them not installing djgpp correctly, so please do be sure to read this!
Make is part of the standard djgpp distribution, in v2gnu/mak*b.zip (whatever the latest version number is). You should get this from wherever you got djgpp.
Are you sure that you have installed djgpp correctly? Check the instructions in the djgpp readme.1st file, and in particular confirm that you have set the DJGPP environment variable and added the djgpp\bin directory to your path.
Are you sure that you are using the GNU make program, and not some other version like the Borland one? If in doubt, run "make -v" and check that the GNU copyright message is displayed.
You shouldn't be running make from the Win95 explorer! Bring up a DOS box, change to your Allegro directory, and type "make" at the command prompt.
Are you sure that your system time and date are set correctly?
As the error message suggests, you need to provide more memory for the compiler to use. The go32-v2 program will tell you how much is currently available. If you are running under DOS, try to free up more disk space for use as a swapfile. When using win95, increase the DPMI memory limit in the properties for your DOS session to 65535 (you'll have to type this in by hand, because the pull down list doesn't go above 16384).
This seems to be a problem with the replacement linker from the RSXNTDJ package. Either go back to using the standard djgpp linker, or remove the "-s" flag from the Allegro makefile. If anyone knows why this happens, and how I can avoid it, I'd like to know about it!
You haven't read the docs, have you? :-) You need to link your program with the library file, liballeg.a. First, make sure you have installed everything properly (running make should do this for you). Second, if you are compiling from the command prompt or with a makefile, add -lalleg to the end of your gcc command line, or if you are using Rhide, go to the Options/Libraries menu, type 'alleg' into the first empty field, and make sure the box next to it is checked.
The grabber needs to be linked with the code in datedit.c. But you shouldn't have to worry about doing this by hand: just run "make" and that will build everything for you.
No, sorry. For starters, liballeg.a is about 450k, but you'd probably also want various utilities like the grabber, sound setup, etc. And what about all the example programs? If I included compiled versions of everything, a binary distribution would be over 7 megs: way too big to be practical! More seriously though, there just isn't any reason why you can't build it yourself. A compiler is a deterministic process: given a particular input file and command line, it will always produce the same output. If this isn't working, you either have the wrong input files (ie. your copy of Allegro is broken in some way), or the wrong command line (it is hard to see how that could happen, since all you have to do is type "make"...) or your compiler is broken, ie. you didn't install djgpp properly. You need to find and fix the problem, not just sweep it under the carpet by getting someone else to compile Allegro for you...
Nope.
Yes. It has been ported to run under win32 using DirectX, and to the Linux X-Windows system. See the Allegro homepage for links to these versions, and the latest information about any other porting projects.
Your mileage may vary. Some people have reported problems, while others say that it works fine. I don't have access to NT myself, so there is nothing I can do about it either way, and to be honest I don't care very much :-) If you want to run DOS programs, use DOS!
Yes, but with some caveats. If you are using the OpenDOS version of EMM386, you must disable their DPMI implementation (specify DPMI=OFF on the EMM386.EXE device line in your config.sys). You should also make sure the PIC=OFF flag is set, but this is the default so it won't be a problem unless you have specifically enabled it.
WIP stands for "work in progress", and refers to any changes that are more recent than the last official release. WIP versions of the library can be obtained as patches from the Allegro website, and are usually quite stable, although obviously not so well tested as a final release version.
Do you have a copy of patch.exe? If not, go and get it from the same place that you downloaded the rest of djgpp: this tool is a standard part of the compiler distribution. If you do have the patch program but it isn't working properly, make sure that you are installing the patch over the top of an unmodified copy of whatever Allegro version it is intended to update (this will usually be the most recent official release from before the WIP was made, but check the text file that comes with the WIP to be sure).
You can't. The limit is imposed by the VGA hardware, not by Allegro. To access more than 256k of video memory you need to use an SVGA mode, which means either switching to a higher resolution or getting a copy of the SciTech Display Doctor, which provides several low resolution SVGA modes.
This is just the way that the video hardware works: there can only be one palette in use at any given moment. Either convert your images so they all use the same palette, or switch to a truecolor graphics mode.
See the Allegro homepage for some utilities, for example FixPal and Smacker.
The VGA hardware only uses 6 bits for each color component, which means the red, green, and blue values in the palette structure range from 0 to 63, not all the way up to 255. That gives you a potential 2^18 = 262144 different colors, or 64 shades of grey. If you need more than this you could try using VESA function 0x4F08 to select an 8 bit wide DAC, but Allegro doesn't support this directly and I don't know how reliable it will be across different hardware.
When you are in a 256 color mode, the VGA card displays color #0 around the border of the display area (in truecolor modes it displays black). Your funny color will go away if you change the palette so that color #0 is black.
See the Allegro homepage for some utilities, for example DTA and Smacker.
In your favourite paint program, get hold of the RGB sliders and drag the red and blue ones up as far as they go (usually to 255, but this will depend on what units your software uses), and the green one right down to zero. The result is a special shade of Magic Pink, or as some people prefer to call it, magenta.
One of the GUI objects has changed the text_mode() setting while it was drawing itself. Call text_mode() again to select whatever background color you want to use.
Remember that the vertex positions are stored in fixed point format, so you must use the itofix() macro or shift your coordinates 16 bits to the left.
Remember that the angle of rotation is stored in fixed point format, so you must use the itofix() macro or shift your coordinates 16 bits to the left. For example, rotate_sprite(bmp, spr, x, y, itofix(32)) will rotate the graphic by 45 degrees.
It depends on exactly what you are doing. If your images are totally opaque, there is no advantage to using an RLE sprite, and it will probably be faster to use a regular bitmap with the blit() function. If your graphics contain masked areas, an RLE sprite will almost always be both smaller and faster than the draw_sprite() function. Compiled sprites are in general quite a bit faster than both the others for masked images, and slightly faster for opaque graphics, but this is far more variable. They are at their best with small sprites, on older machines and in mode-X, and may actually be slower than blit() when using SVGA modes on a pentium (the large size of a compiled sprite is very bad for the cache performance).
You need to make sure the game logic gets updated at a regular rate, but skip the screen refresh every now and then if the computer is too slow to keep up. This can be done by installing a timer handler that will increment a global variable at your game logic speed, eg:
volatile int speed_counter = 0;
void increment_speed_counter()
{
speed_counter++;
}
END_OF_FUNCTION(increment_speed_counter);
void play_the_game()
{
LOCK_VARIABLE(speed_counter);
LOCK_FUNCTION(increment_speed_counter);
install_int_ex(increment_speed_counter, BPS_TO_TIMER(60));
while (!game_over) {
while (speed_counter > 0) {
update_game_logic();
speed_counter--;
}
update_display();
}
}
Add a call to save_bitmap() somewhere in your code. See the save_bitmap() documentation for a discussion of one common pitfall when doing this, and some example code.
Call srandom(time(NULL)) at the beginning of your program, and then use random()%limit to obtain a pseudo-random number between 0 and limit-1.
The VBE/AF interface already provides this for the video drivers: see the FreeBE/AF project on http://www.talula.demon.co.uk/freebe/. For more general things like the sound, VESA, and mode-X routines, this would be very difficult to do because the drivers depend on a lot of helper functions from the core lib. The djgpp DXE format is nowhere near flexible enough to support this, and I don't want to make Allegro dependent on any other dynamic linking packages.
There is no need. The linker will only include the parts of the library that you actually use, so if you don't call any of, say, the texture mapping or FLIC playing functions, they will be left out of your executable. This doesn't work perfectly because a lot of the Allegro code uses tables of function pointers that cause some unnecessary routines to be linked in, so the majority of the graphics functions will be included in every executable, but I have tried to keep this to a minimum. See allegro.txt for information about more precise ways to remove some of the graphics and sound drivers.
No. I think this sort of hardware support would be most useful as part of a proper 3D API, which Allegro is not, and will never be. If you want to do some work on this, the MESA library (a free implementation of OpenGL) is IMHO the place to start.
Several very good ones already exist, for instance the JGMOD package. See the links on the Allegro website. You are not allowed to suggest that one of these libraries be merged into Allegro, because this topic has already been done to death on the mailing list and I am tired of it.
That would be a lot of work, and I don't have time to do it myself. There are several networking packages currently in development or floating around on the net, though, and in my opinion this sort of code is more useful as an external library than it would be as part of Allegro.
Unisys has a patent on the LZW compression algorithm that is used by the GIF format. I want everything in Allegro to be freely usable without any restrictions whatsoever, which means I can't include any code that is subject to licensing or the payment of royalties.
It does for some people, but not for others. The problem is that Creative Labs refuse to release any specs, so we don't know how to write a driver for it. Complain to them, or buy a different card from a more reasonable manufacturer.
This might be because you have the volume set very low: try changing this in the setup program. Also, Allegro is mixing several sounds into a single output buffer, unlike the Windows sound player that only plays one sample at a time, so each individual sound can only get a smaller percentage of the total output volume. This is just the price you pay for multiple output channels. If you don't like it, use the setup program to tweak the number of channels: this can be any power of two less than or equal to 64, and the smaller you make it, the louder the sound will be.
Try using a FreeBE/AF driver (http://www.talula.demon.co.uk/freebe/), or the commercial SciTech Display Doctor (http://www.scitechsoft.com). If it still doesn't work, post a description of your problem to the Allegro mailing list, along with a copy of the output from running the afinfo and vesainfo programs.
Perhaps. Try to isolate the smallest fragment of code that is able to reproduce the problem, and I'll have a look at it. If you can send me a 10 line program, I will fix it. 100 lines, and I can probably fix it. 1000 lines, and I don't have a chance :-)
Sure. See the giftware terms in readme.txt. I don't mind what you do with it, and there are no problems with commercial usage.
Whenever it is done! A little encouragement is always welcome, but I don't have any completion deadlines and I'm not going to make one up for you :-) As soon as it is finished, it will be released.
The grabber can import directly from GRX or BIOS format .fnt files, or you can draw them onto a .pcx image using any paint program. See the Allegro website for a utility (ttf2pcx) that will convert Windows TrueType fonts into this .pcx format.
See the Allegro homepage for some links. You can use Gravis patches (.pat format), or SoundFont 2.0 (.sf2) files, but the latter must be converted into a patches.dat file with the pat2dat utility.
You need to download the makertf conversion utility (ftp://ftp.coast.net/Coast/win3/winhelp/mkrtf104.zip), and the Windows Help compiler (ftp://ftp.microsoft.com/Softlib/MSLFILES/HC505.EXE). Make a temporary directory, copy the allegro.txi file from the allegro/docs dir, and run the commands "makertf --no-warn allegro.txi -o allegro.rtf -J allegro.hpj" followed by "hcp allegro.hpj". The second command will give a lot of warnings, but they can safely be ignored.
The allegro.rtf file can be read directly into Microsoft Word and printed from there, but you should right-click and update the table of contents and index fields to fill them with the correct data first. Alternatively you can install the TeX package and use the tex and dvips programs to convert allegro.txi into Postscript format.
Check the Allegro homepage, http://www.talula.demon.co.uk/allegro/. If you have anything to add to this, please send me the URL!
It simplifies the maintenance of your program, in case the value of PI ever needs to be changed. Also it will make your program more portable to other compilers that use different values of PI.
See http://www.talula.demon.co.uk/allegro/maillist.html.