2009-07-31, 09:13 PM
^A little late but better than never...
I've actually been focusing my time on this in a round about way pretty much this whole time... at this point I'm less interested in making new graphics than getting the existing graphics out of SoM. Because basically I need them (or at least one) for reference to make sure everything looks right.
I was going to write a little custom tool to help decipher the files (or actually add a few routines to my usual homebrew graphics suite) ...but I decided I should be making good on that promise to remake SoM some, so I decided I will start working on that, and use it to visually decipher the graphics files...
So in a round about way I'm now working on finally getting this new virtual reality (development) environment to draw stuff (yes it technically does not even draw stuff yet)
The reason it doesn't draw, is because the system is supposed to be fully modular, and I'm taking my sweet time to make sure the thinking behind the module system is very sound. The trouble is, I want the system to have zero limitations. So in other words, if someone is willing to program a module (if necessary) there should be nothing the system can not be made to achieve. So the module interface is very wide open.
The environment is very complex supporting multiple "games" (or networks) and computing environments spread out across multiple machines and clusters of machines performing the same distributed tasks.
But basically the library that does all the drawing is also a module. I've already programmed the interface long ago for the first rendering module, called opengl-core, and defined how the low-level graphics server interacts with it (or technically does not interact - directly that is)
I tend to work in reverse from the more complex down to the more simple stuff, or in other words from the interface end down the implementation chain. But in short routing the modules through the library that handles them is pretty much the last step, and I'm making headway, and will hopefully have something to show after a little while. Patience is the key when it comes to designing/implementing very flexible (really futuristic) systems. It's really the opposite of what I did as a kid, just wanting to sit down and make some graphics popup. I did ground breaking stuff that way, but it's hardwired stuff more or less. Now I want to build (am building) wide open systems without limits.
So in short, I've given myself a very simple objective and have chosen the least direct way to go about achieving it
I'm also doing this though, because I felt like SoM was subtracting too much from more substantial work lately.
I've actually been focusing my time on this in a round about way pretty much this whole time... at this point I'm less interested in making new graphics than getting the existing graphics out of SoM. Because basically I need them (or at least one) for reference to make sure everything looks right.
I was going to write a little custom tool to help decipher the files (or actually add a few routines to my usual homebrew graphics suite) ...but I decided I should be making good on that promise to remake SoM some, so I decided I will start working on that, and use it to visually decipher the graphics files...
So in a round about way I'm now working on finally getting this new virtual reality (development) environment to draw stuff (yes it technically does not even draw stuff yet)
The reason it doesn't draw, is because the system is supposed to be fully modular, and I'm taking my sweet time to make sure the thinking behind the module system is very sound. The trouble is, I want the system to have zero limitations. So in other words, if someone is willing to program a module (if necessary) there should be nothing the system can not be made to achieve. So the module interface is very wide open.
The environment is very complex supporting multiple "games" (or networks) and computing environments spread out across multiple machines and clusters of machines performing the same distributed tasks.
But basically the library that does all the drawing is also a module. I've already programmed the interface long ago for the first rendering module, called opengl-core, and defined how the low-level graphics server interacts with it (or technically does not interact - directly that is)
I tend to work in reverse from the more complex down to the more simple stuff, or in other words from the interface end down the implementation chain. But in short routing the modules through the library that handles them is pretty much the last step, and I'm making headway, and will hopefully have something to show after a little while. Patience is the key when it comes to designing/implementing very flexible (really futuristic) systems. It's really the opposite of what I did as a kid, just wanting to sit down and make some graphics popup. I did ground breaking stuff that way, but it's hardwired stuff more or less. Now I want to build (am building) wide open systems without limits.
So in short, I've given myself a very simple objective and have chosen the least direct way to go about achieving it
I'm also doing this though, because I felt like SoM was subtracting too much from more substantial work lately.