Welcome Guest [Log In] [Register]
Welcome to New Beginnings. We hope you enjoy your visit.


You're currently viewing our website and forum as a guest. You can currently only read posts and are subject to ads. Join for free and you can have full access + have an ad free experience.


Join our community!


If you're already a member please log in to your account to access all of our features:

Username:   Password:
Add Reply
Update 6/15/13
Topic Started: Jun 15 2013, 12:42 PM (196 Views)
Lucas
Member Avatar
Chief Programmer

Afternoon peoples

This week was busy-work with the perfect collision detection system. The goal: use the models themselves as collision meshes for the most realistic collision ever. What it is right now: not working :( Over the past week, I have solved several large and minor bugs with the system, making improvements where I can. However, my lack of mathematics knowledge hinders me. That being said, if anyone here is good at math, check out my problem here: linky.

While struggling with the perfect collision system, I worked on adding more functionality to physics objects. Specifically, angular rotation, angular velocity, and angular acceleration. All angular properties follow correct physics formulas. Now, it is just a matter if getting the model of the physics object to actually represent the angular properties. Speaking of things I want to do this week:

  • Angular properties should be reflected by model correctly
  • Texture LOD
  • Better separation and transition between God Mode and Player Mode (i.e., player model will only render/have calculations done to it when in Player mode)
  • Sky Sphere efficiency upgrade (move from display list to VBO)
  • Do once-over on brush system and fix any and all bugs
  • Build vegetation system (based on brush system)

But all that stuff is relatively easy. There are a few large and difficult problems I want to deal with this week as well:

  • Normals for the edges of chunks are incorrect. Find a way to correct them that doesn't take 5 minutes to compute.
  • Brush update system updates the entire chunk array instead of just the chunks that are loaded/deleted
  • Moving diagonally causes the chunk-loading queue to back-up to the point of eventually crashing the game.
  • Entire chunk/terrain/brush/world/vegetation/physics systems should be re-oriented to be focused primarily on chunks (the current setup works, but will become much more complicated as the game develops unless some reorganizing is done now. Could also increase performance in the future)
  • Figure out how to do efficient and fast object serialization in C++ (2 lines of simple code in Java, a taunting task in C++ :/ ). When implemented, will allow for a super-fast and efficient save/load system to be implemented

After all of that is done, I will either have figured out why the perfect collision detection system math doesn't work, or someone will hopefully have solved that problem for me :P So after this week, I should be able to go back on working on the physics engine again :D

Till next week ;)
View the progress of the game here!
RockSplitStudios.com
My Blog
Offline Profile Quote Post Goto Top
 
Nuke

I'm surprised that Lucas said stuff for the future instead of stating progress this week.
Offline Profile Quote Post Goto Top
 
Lucas
Member Avatar
Chief Programmer

Nuke
Jun 16 2013, 04:41 AM
I'm surprised that Lucas said stuff for the future instead of stating progress this week.
Did you read the first two paragraphs? Lol.
View the progress of the game here!
RockSplitStudios.com
My Blog
Offline Profile Quote Post Goto Top
 
Nuke

Lucas
Jun 16 2013, 09:51 AM
Nuke
Jun 16 2013, 04:41 AM
I'm surprised that Lucas said stuff for the future instead of stating progress this week.
Did you read the first two paragraphs? Lol.
There seemed to still be more future than past, lol.

It's not like it's negative, though. I do it all the time :P
Offline Profile Quote Post Goto Top
 
1 user reading this topic (1 Guest and 0 Anonymous)
« Previous Topic · Announcements · Next Topic »
Add Reply