Work on the Subdivided SkinMesh support engine has ground to a complete halt while I vainly try to understand why the code is failing during retrieval of optimized ibmesh.
In the meantime, the world keeps turning, and I've taken m$'s particle example and built on the structures they used for particles and particle systems...

Basically what I've done is add the LinkedObject header to the structures for particles and particlesystems, and added fields to the particle structure that make it a "particle template"..my particle template contain fields which determine all the attributes for a unique particle, including Gravit, Durability (to impacts), TTL, and different "termination effect" rules can be applied to determine the outcome of a particle terminating, say, due to impact or due to TTL expiry. Then it also includes fields determining the type of particle to spawn under those rules.

I intend to now design a number of predefined particle-types using this template. For example, if a "FireParticle" terminates due to impact with the ground, it has a 30% probability of spawning a maximum of 3 new particles of type "FireParticle". Of course it wouldn't spawn if it impacted with water !!
A magic weapon similar to a firework could spawn if it terminates in the air, but fizzle out if it hits the ground too early.

This allows me to create a quick library of smoke, fire, snow, sparks, grenades, blood spurts, projectiles and anything else particle-based that I can imagine.
I can quickly and easily then generate these effects in realtime.
Of course, I'm looking to create a system for Particle Emitters which will be my next step. I'll be looking to embed the Emitter Object in the structure I am using for Frames, allowing me to attach a particle emitter to any existing Frame. This in turn allows the transformations applied to the Frame to be applied to the Emitter itself. This constrains the emission orientation to the object to which it is attached. Thus if we stab a character in the neck with a headless barbiedoll, the blood spurting from his neck outwards will rotate as his lifeless body slumps to the floor, and spray upwards as he lays there, like a pretty little fountain. Heh.
OK so : particles move, particlesystems move, but emitters never ever move, they are stuck to something which MAY move.
Anyone interested in this yell out.
Posted on 2002-11-19 20:45:21 by Homer
Afternoon, EvilHomer2k.

Sounds Interesting:).

I still don't know when I'll have time available to help out though.

Since getting back from the UK, I now have a 'puter with which I can view the other half of the M$ DX examples I never was able too before :alright: .

I was also able to enroll in a few courses at the Game Institute, and purchase Programming Role Playing Games in Directx by Jim Adams.

The idea for me is to build a proggy in C++, and then convert chunks of it into masm32 libs.
Hopefully by the end of the courses and the book, I'll have enough knowledge to get the linkedlist/skinnedmesh up and running:).

Cheers,
Scronty
Posted on 2002-11-19 23:50:06 by Scronty
Work on implementing programmable Rules has led me to become interested in developing a limited scripting language for the purpose of defining particlesystems using a textfile. I plan on using the PrivateProfileString stuff meant for ini files to parse the data, not elegant, but easy to code. Basically the ini file structure will reflect the template object it was designed for.
The notion of using more scripting may become more important later but for now, this represents a simple way to implement new subsystems into the main code with the ability to be edited easily without recompiling the binary. I'm sure this will become more important as compile time increases, and especially if anyone decides to help out with modules of my project.
I'll post something on the Programmable Particle Engine soon.
Posted on 2002-11-20 06:46:57 by Homer
I have not produced the goods as far as the particle engine is concerned - it is undergoing a major overhaul for good reasons...
I did release something on the script engine though.
Seek and ye shall find.

What's so special about 42?
It's supposed to be the time it takes to fall through the earth to terminal velocity at its centre. and out the other side to (roughly) the same height that you began at, on the opposite side of the planet...42 seconds, give or take...:tongue:
Posted on 2002-12-02 01:28:10 by Homer
Working on the ParticleEngine...
I have all kinds of implementation ideas I'm trying and yet to try...
among them is a faked softbody physics and advanced particlesystem management including attractors and other behavior modifiers :cool:
I'll post something soon, I'm wrestling with some of the DX helper macros which seem to be awry - I'm using Caleb's dxmath files as reference...
Posted on 2002-12-18 07:52:22 by Homer
Almost there... glee !!

Soon I will publish the fruits of my labor in the spirit of open source.
Happy New Year to all.

Hint: the darker the drink, the heavier the hangover !!
Posted on 2002-12-31 03:14:30 by Homer