This forums is for OS project announcements including project openings, new releases, update notices, test requests, and job openings (both paying and volunteer).
I got an idea a while back... what if we would all join 'our forces' and create an OS specially designed for Gaming.
I'm thinking bootable CD with game designed for our OS, this way it could be run on any computer, we could create several kernels for different processor types then a neutral bootlader load the apropriate kernel, This would make it possible for everyone to play games (even Mac and Linux users )
nice idea but, like another post, very harde (if impossible to implement)
there is two approach and two problems
the worst approach : creating a totally new OS which is not compatible with the current one, then develop games for it ..... hmmm i think u can understand why thi is a bad idea
the second approach : creating an OS which can run current windwos games : yeah easy to say but this mean that u must reimplement all windows API etc... and dont forget that Micro$oft will never give u the desing and documentation that u need to do this, u can think about emulation, yeah good but allready exist (wine winex/cedega)
i already thought about this but seriously i think it's impossible.
-----------------------
There are 10 types of people in this world... those that understand binary, and those that don't.
well yes ofcourse making winds games run isnt easy, but thats not what I had in mind. I was thingking of a completely new OS (or perhaps linux based..) that we create games for, games designed to run in this system. Ofcourse it does take time to designe games to! but wouldent it be nice, being able to run games on any computer
again,
as i said, the idea is nice, but what game will run on this OS, do u think u can conveince ppl to developp games for ur OS ? not sure...
also, if u think u will developp games for ur own OS, what type of games u ll develop ?
developping an "acceptable" game takes lot of time, and lot of ressources (developpers, designers ...etc)
ur speaking about a very big project, IMHO : if u want to deveop an OS, think about more simple project, and if you want to develop games, think about learning OpenGL (or DirectX) first, but it's really a bad idea to try to do both things in the same time ....
I can understand what ur aiming to, but it s IMPOSSIBLE to do.
finally if u want to make a gaming linux distro, maybe i can help u, but, i m not sure that such diro will be useful, since u can do the same thing with knoppix or gnoppix (in CD distros).
-----------------------
There are 10 types of people in this world... those that understand binary, and those that don't.
I have a different approach. I'm coding my own software engine in ASM (to learn 3D, ASM, and optimizations) but I separated all OS dependant code in separate headers, so you can theorically port it to any OS just by rewritting the headers. I also plan to design a version without OS later in the future (running in pure x86 cpu) to see what gains that brings. I plan on releasing the code once I have something useful. I think you should start out with something similiar, and progressing slowly. That way you may get what you want Cheers,
coding 3D engine in asm is a good idea, but it is not new, all 3D engines have lot of critical code written in ASM to improve performances.
but u must know that just coding in asm does not mean that ur code is performant u have to take advantages from cpu specific optimisations (mmx, sse, sse2 3dnow ....etc) ,
also dont forget that compiling C code with some flags can optimize it and increaze lot of performances (some times better than ASM )
-----------------------
There are 10 types of people in this world... those that understand binary, and those that don't.
Yeah. I know. I'm taking care to study processor specific optimizations, like cache-isntructions and SIMD instructions. But anyway this is mostly a learning project and I'm having fun with it Cheers,
Why shuld it be impossible???
Try to play around with the idea instead!
If we want an crossplatform OS running on a live-cd for playing games that already exists Win/Mac/PS/XBOX etc, that's realy though, we need:
1. adaption to several bootstandards, Multiboot BootX and more...
2. A Kernel that runs on several processors (several compilations for it)
3. Implementation of all API:s from all systems.
4. Implementation of emulators for different processors (playing Win-games on Macs) etc.
This is a really huge and advanced task!
Then we could think like this:
We want our own gaming platform that is hardware-independent and that can be loaded from a live-CD/DVD.
If we do like this it is possible:
1. Implementation of various bootstandards (needs some engineering).
2. Development of a generic kernel that is as simple as possible but also flexible.
3. Drivers for the most common hardware on different platforms.
4. Different native multimedia-APIs, OpenGl, sound, controldevices (joysticks) etc.
5 We need to make the games running nativly on processors
a. a simple JIT-compiler for bytecode.
b. a flexible JAVA-API that is staticaly linked to the native multimedia-APIs, and that is powerful, flexible and simple, totaly: smart design!
With those steps developed I could imagine it is a quite attractive platform for develop games on, even for the commercial market.
When we got the most osdev parts running, I think the rest of OpenSource community would like the idea, this could be big.
Those are nice ideas, but I still think it should start slow. If you plan for it to be big, it will get so complex nobody will be able to develop it or with it. OS development is something very complex, so complex that there is no "perfect" OS, as each different person wants his specific OS tweaked for his needs (that's one reason why Linux did suceed). For me the perfect OS would be a NoOS. It doesn't do anything, you just plug in modules to do what you want and config it the way you use it. I still suggest start simple, do something YOU like and you want and then others will come and help you. Cheers,
here is the problem, and the thing making it impossible,
think about windows API, how will u implement it, there is no docs, no public standards, no source code ..... nutin
one solution is reverse ingeneering, and this is "ILLEGAL"
another solution is "emulation" but this allready exists as i said before (WINE, WINEX, CEDEGA, VMWARE ...)
-----------------------
There are 10 types of people in this world... those that understand binary, and those that don't.
that's what I mean width the second Idea, develop an API for multimedia if it should be used as a cross-hardware platform, not an emulator/simulator for different systems.
wrong!!!
ALL windows APIs are VERY WELL documented
all you have to do is look at the microsoft documentation that comes with visual studio (or any other windows compiler) everything is extremely well documented (better than it is for linux -- though that isnt too bad)
the API and calling conventions and addresses are all well documented if it wasnt it would be imposible to write software for windows (of course you dont need any of this if you are using MFC but that is a very new invention) and the MFC and stdlib source code is there for you even if you use it so you can still see the true calls
sorry if i sound upset im just tired of people lying (intentionally or not) about MS documentation just so they can have something against them
unless of course you ment source code instead of API:
of course that isnt available but you should rewrite that anyway
btw: i will be providing full win32/winNT API support eventually -- the only hard part is that it is the largest programming interface in the world (with thousands of call variations)
)
