Saturday, November 24, 2018

Project Rhiannon Update 2

John B. Marine | 7:31 PM | | | |
My Godot racing project, Project Rhiannon, lives on! I recently went from trying to make modular voxel land to making polygon land. This was inspired by Gran Turismo 5's Course Maker feature. Seven environments make up GT5's collection of environments. To practice designing levels, I did some box modeling in Wings3D. I tried to closely follow some of the design nature of custom tracks in Gran Turismo 5 in trying to design modular assets for a race track. What I practiced mostly was simple modeling of such a course, rather than build a proper course. Here is a look at what I've come up with:

Project Rhiannon game development
^ Instead of voxel land, I tried polygon tracks. I haven't yet found out how not to make things extremely shiny in Godot.

For some reason, I wanted to make some kind of non-tarmac course initially. I then decided to build an entire tarmac course in Wings3D and then scale it greatly in Godot to be used. I did not include a pit road, so this track isn't COMPLETELY practical. I wanted to make the road have subdivisions and loop cuts to allow for the possibility of making individual elevation changes to pieces of the track geometry. I am also learning that you do not have to model objects into the main model themselves. You build the track first followed by adding whatever objects you like. This can even help in the case where not rendering objects can help in the frame rate department as well as in reducing draw calls.

I still have a lot to learn, but I feel I can possibly make a proper track that I can eventually implement into racing games. It is a slow process that will eventually pay off into something amazing.


Driving Model.

While I am still using my voxel model cars for my test car, I am slowly wanting to develop my own proper 3D vehicles to use as my testing vehicles. I still have somewhat of a "to do" list to enhance my driving model. The driving model in Godot does not really allow for meticulous driving models. That is, unless you are incredibly adept at coding/logic for a proper model. The basics of the Godot Game Engine for vehicles are the commands of engine force (acceleration), brake (braking), and steering. I was looking to try to find some alternative methods to utilize these features. Everyone who has designed their own driving model through Godot GDScript has made his/her/their own spin on the vehicle model. These methods have been expressed in various games, and no one model is perfect. A driving model is only as perfect as the one(s) who deem it as perfect.

Engine Force
I allowed for more interesting acceleration by dividing the vehicle's Engine Force by a number less than 1. What would happen in this instance is that the car gets off from its starting position and go fast- almost like a drag racer. What I want to do is set a base speed that increases incrementally. So I want to properly simulate going from a standing start to properly accelerating away. I also want to develop a speedometer to know just how fast I am going as well as a tachometer to simulate the RPM range.

Braking
However, I have been struggling a bit to properly slow down the vehicle. The brakes work fine with my models. What I want to do is make the braking power more efficient. If I wanted to be more advanced, I would want to simulate brake balance. So a car with strong front brake balance could have some understeer dialed in. A car with strong rear brake balance can induce some oversteer. Too much either way can upset a vehicle's handling behavior.

Steering
Steering can be a bit slow. When I decided to meddle with the steering values, the vehicle may not even turn or only turn so slowly. Once I reached a numerical value that gives me some semblance of control, I basically left it alone and not chance ruining the steering angles.


I still need to get around these issues in trying to develop a driving model I can be proud of.


Extra Insight.

I mentioned Gran Turismo 5 and its Course Maker. Gran Turismo 5 allows you to construct race tracks purely on a set number of parameters and features. The environments differ. While you do not have meticulous control of each environment, your track is given a set number of variables you can control. There are seven locations you can choose from in Gran Turismo 5:

• Toscana (Tarmac)
• Eifel (Circuit)
• Mt. Aso (Tarmac)
• Liège (Gravel)
• Alaska (Snow)
• Eifel (Kart)
• Tokyo Bay (Kart)

The tarmac environments are primarily designed to be run by almost any kind of car. The dirt environment of Liège is meant for gravel rally racing. The snow of Alaska is meant for racing on snow and ice. The kart environments are short courses suited mostly for go-karts. Each environment includes a number of different features unique to that environment. For example, the Toscana environment features mostly a stroll along the Italian countryside in beautiful Toscana, complete with Italian Cypress trees. It also has fireworks going off at night.

If you have actually played with Gran Turismo 5, though, you know many of the environments appear very sparse. Almost like incomplete terrains. That is kind of like what my environment deal is like at present. Implementing some of my work will require some extra geometry edits as well as triangle tesselation. I do feel good about designing a practical course that could be implemented into games or certain other projects.

What I did was study how the roads for the environments were set up. I looked at differences in elevation, the placement of the road, grass, barriers, and things like that. Go back to the Toscana tarmac example for instance. You have a mostly wide road that has little edges on the side close to the rail fances. Down the way are usually either bushes or trees. These object placements do not have to be uniform all the way around the track, but you can put certain features in to make the environment a bit more entertaining and lively. I hope to learn from these in building my courses from scratch. Maybe if I learn certain algorithms to design procedural locations, I could create locations without having to do everything by hand.


Now you know a little more about my development of racing/driving games of late. I have yet to find a winning balance.





Now you have an update about my racing/driving game development. Thank you for reading! Take care and be well.

Subscribe to My Blog(s), and/or Follow on Bloglovin!
Subscribe to John's Blog Space (JBS) Subscribe to John's Creative Space (JCS) Subscribe to John's Race Space (JRS) Subscribe to StyleSpace (SS) Subscribe to John's Life Space (JLS)
Support My Creative Works!
JMDesigns Patreon Soundcloud Bandcamp Twitch OpenGameArt TurboSquid
Get Social With Me!
Facebook (Friends) Facebook (Fans) Twitter YouTube Instagram Google+ LinkedIn (professionals only)
Questions or Comments? Contact Me (serious inquiries only)...
Contact Me via E-Mail




Share this article

0 comments:

Post a Comment

JohnMarineDesigns on TurboSquid

Would you like to donate to support my creative work? If so, please show your support using this PayPal item:
I am now on Patreon! Support my creative work on Patreon today at: www.patreon.com/johnbmarine!
 
Copyright © 2015 John's Creative Space • All Rights Reserved.
Template Design by BTDesigner • Powered by Blogger