[QUOTE=Cathbadh;18522500]Was there an openGTA1? Or is this like a painstakingly reproduced clone of GTA2?[/QUOTE]
I have a feeling its a modified version of the real GTA2 like multi-theft auto etc
[QUOTE=jA_cOp;18523961]It's actually really easy, especially in real-mode. There's so much less to worry about, the only thing you really have to learn is how to interact with the BIOS using interrupts, and how to make your own interrupts. Once you're there, you're good to go. Testing is also easy with a virtual machine (like VirtualBox).
It doesn't take long to get started anyway, just grab VirtualBox and an assembler for the boot image (I recommend FASM... I think that's what windwakr uses too). Once you've got the hang of it, you can start coding in a different language for your OS.[/QUOTE]
What about stuff like memory allocation and the like?
[QUOTE=noctune9;18524462]What about stuff like memory allocation and the like?[/QUOTE]
That's all you. You have to do it manually.
[QUOTE=Burning Ninja;18524172]a2h, the code that's erroring has "AUTO_INCREMEN" rather than "AUTO_INCREMENT" at least looking at the error people have posted.
hope this helps[/QUOTE]
mysql_error() cuts off the query to around the problematic area.
And this has already been mentioned, so why post it again? You must've seen people mentioning it if you've seen the error on the previous page.
So no, that didn't help.
[QUOTE=r4nk_;18524443]I have a feeling its a modified version of the real GTA2 like multi-theft auto etc[/QUOTE]
OpenGTA1/2 are re-implementations of the original games I think, a lot of re-implementations are called "open" + original_games_name. (Open stands for 'open-source')
Such as openmorrowind, openttd and a few others.
[QUOTE=Mattz333;18526009]OpenGTA1/2 are re-implementations of the original games I think, a lot of re-implementations are called "open" + original_games_name. (Open stands for 'open-source')
Such as openmorrowind, openttd and a few others.[/QUOTE]
Ah nice, I didn't realise. So it's a completely new engine that (sometimes) loads the old games data?
[QUOTE=Loli;18506824]So I took my Basic Physics and added Mass. [b]Please don't rate me dumb, I'm only showing these because this is the first time I have really proggrammed any physics[/b].
[media]http://www.youtube.com/watch?v=fBa-603UAds[/media]
Next I hope to add Collisions and maybe rotations.[/QUOTE]
You're aware that this is your first time programming physics, you've got pretty much nothing there and yet you're spending your time making these pointless videos and asking to not rate you dumb?
Madness!
[QUOTE=r4nk_;18526147]Ah nice, I didn't realise. So it's a completely new engine that (sometimes) loads the old games data?[/QUOTE]
Yes, and most of the time they have additions and improvements. For example in OpenTTD they added a graphics replacement system, configurable gameplay elements, extended existing tools to make them more efficient, custom AI, multiplayer etc.
[QUOTE=Spoco;18527621]You're aware that this is your first time programming physics, you've got pretty much nothing there and yet you're spending your time making these pointless videos and asking to not rate you dumb?
Madness![/QUOTE]
I think its good that he shows it off, might give him some motivation.
Started working on this fake os thing to give me a break from graphics programming, only during school days though. Got help, credits, exit, wait, out, run commands. Run loads a file and parses all the commands in it, its really awesome and I made one of those "don't press x" games.
Currently working on variables.
[QUOTE=a2h;18524543]mysql_error() cuts off the query to around the problematic area.
And this has already been mentioned, so why post it again? You must've seen people mentioning it if you've seen the error on the previous page.
So no, that didn't help.[/QUOTE]
woaaaaaaah,
sorry for partying.
[media]http://www.youtube.com/watch?v=sIdXL9lBDAg[/media]
I added some features, made it look better and improved the performance (multithreading support and optimization) of my 2D fluid simulator.
Try it out [url=http://5.latest.testbotrobert.appspot.com/jFluid.html]here[/url]
[QUOTE=Robber;18530166][media]http://www.youtube.com/watch?v=sIdXL9lBDAg[/media]
I added some features, made it look better and improved the performance (multithreading support and optimization) of my 2D fluid simulator.
If you want to try it, I'll upload the applet and/or jar.[/QUOTE]
That's pretty sweet
To the person making the OS: I've been learning C++ quite a lot recently, and I would like to try making an operating system like that. How do you create an iso/operating system out of binary programs?
That's cool, I wouldn't mind learning Assembler. I'll start that soon, once I get some more OOP down.
What I mean with that is (this is a dumb question, as I simply don't quite know how isos work), that obviously the kernel cpp needs to be compiled to be able to run, but how would you run what is compiled? I don't know how to explain that, but the kernel somehow needs to be run. And be able to build it all into something that can be opened in virtualbox, or burned onto a cd.
An ISO is just a disk image. So I assume you just "install" it on the ISO like you would install it on a hard drive.
This thread is now about making operating systems. :eng101:
[QUOTE=cas97;18533757]This thread is now about making operating systems. :eng101:[/QUOTE]
A LOT better than fighting over whether to use IDE's or not, or whatever the last fight was about.
A bootable image is easy to make. When the BIOS looks through your devices for a bootable one, it looks for the last two bytes of the first sector (a sector is 512 bytes) of every device. If those bytes equals 0xAA55 (boot signature), then that's a bootable device, it will then load the entire sector into memory at 0x7C00 (so use that offset when writing your assembly) and jump to that location, that's when your bootloader gets control. You can then use BIOS routines (interrupts) to boot your OS in whatever way you want. (Of course, you can write a tiny test OS in the boot sector itself, you don't have to load more code).
[QUOTE=windwakr;18533902]Also, if you plan on coding an OS, you should join a forum that has an OS development section or the whole thing devoted to it. You won't get much help here.[/QUOTE]
Sure you will.
[QUOTE=windwakr;18533902]I'll leave these here.
[url]http://www.brokenthorn.com/Resources/OSDevIndex.html[/url] (This tutorial focuses on C for the kernel)
[url]http://www.osdever.net/[/url]
[url]http://wiki.osdev.org/Main_Page[/url]
[url]http://www.nondot.org/sabre/os/articles[/url]
[url]http://perso.wanadoo.fr/pierrelib/index.html[/url]
[url]http://loonix.technigga.net/osdev.php[/url]
Some C, some assembly, all helpful and full of knowledge.
Also, if you plan on coding an OS, you should join a forum that has an OS development section or the whole thing devoted to it. You won't get much help here.[/QUOTE]
Thanks :)
[QUOTE=windwakr;18534355]I meant that most people here won't know anything about OS coding. He would be better off going to a place where that is the goal of everyone there, and not a place with just a few people who know it.[/QUOTE]
But couldn't you say that for any topic in this subforum?
[QUOTE=jA_cOp;18534461]But couldn't you say that for any topic in this subforum?[/QUOTE]
Not really, a lot of people get their questions asked, for example SFML questions. Now look at high6's threads on who knows what, the crazy undersides of C and C++ or whatever, how many useful answers does he get?
for anyone looking into OS development, C++ is usually looked down upon, however you should take a look at the [url]http://www.l4ka.org/[/url] Pistachio kernel, to get an idea of how C++ can be used :)
Realize it's a microkernel. I also believe the L4 kernel spec names about 7 system calls, so it's very simple.
[QUOTE=nos217;18533680]That's cool, I wouldn't mind learning Assembler. I'll start that soon, once I get some more OOP down.
What I mean with that is (this is a dumb question, as I simply don't quite know how isos work), that obviously the kernel cpp needs to be compiled to be able to run, but how would you run what is compiled? I don't know how to explain that, but the kernel somehow needs to be run. And be able to build it all into something that can be opened in virtualbox, or burned onto a cd.[/QUOTE]
I think I know what you mean, for the compiler to output raw binaries, you usually need to setup the linking stage manually.
With gcc you can have the linker (LD) cook up the object files as PIC (position independent code) using the command-line switch "-pie" or "--pic-executable", you use this when writing code for a target that doesn't support relocation (i.e a bootstrapper that knows nothing about this stuff).
You then instruct the linker to output .BIN files using the command-line option "--oformat BINARY"; in which case you usually supply a special linker script (.x file) to define the layout of your final image (where data/code sections should go, where to base code at, etc).
I'm not sure how easy/hard this might turn out to be if you plan to do it with the Microsoft tool-chain but I believe it can be done, just a different process that I'm not very familiar with.
I started messing with Python, working on some stuff on my WM phone with it. I'll post pics later.
[QUOTE=Jimmylaw;18540824][media]http://www.youtube.com/watch?v=v4_xZ6fRiuY[/media][/QUOTE]
I laughed at the exception at the end. :D
[QUOTE=Diaklu;18541496]I laughed at the exception at the end. :D[/QUOTE]
Not sure if you mean that he goes back to the beginning or the fact i realised you could jump again and again and again to stop you hitting the floor.
The game idea is that you start at one end and have to reach something in the map by creating a path to it. You will only be given a certain number of blocks to create this path with so removing them as you go along is a must! Just don't fall off as you will have to start from the beginning again :)
Sounds like a good idea, but removing them will make it a little too easy.
Finally got normal maps working.
[img]http://img695.imageshack.us/img695/613/screenshotfs.jpg[/img]
Normals are specified in "tangent space," which allows me to reuse parts of the same normal map for different parts of the mesh, which you can see with the cube (or I guess not. the other faces are in shadow).
For the past week I've been struggling with it only to today find out that I didn't have the first vertex attribute array enabled, so my tangent vectors ended up being all in the same direction. Shit, stupid bugs like that are hard to debug with shaders because they give you no interface in which to debug with.
Also, I need a different texture compression algorithm. S3TC makes my normal maps look like poop.
Sorry, you need to Log In to post a reply to this thread.
