Plan for Adrian Tysoe
by Adrian Tysoe · 11/01/2005 (4:50 pm) · 4 comments
This little racing game has been an ongoing pet project of mine for quite some time. Originaly started as a Leadfoot game, but things didn't work out as planned and I ended up starting over. By chance, Today I saw a job opportunity for a racing game. After preparing a new self playing demo I thought I'd take the opportunity to show it to anyone interested at GG too.
Heh Just read over my plan again and noticed the abundance of I's :). Infact Damien Sturdy did a wonderful job coding the core racing engine and hacking into ODE in order to get the most out of it and realise many of the concepts and ideas I had in mind during the development of the project.
It's supposed to be more of a casual fun party type racing game, rather than a serious sim or arcade racer.
Key features:
ODE Physics
Dynamic AI. Cars see walls and other vehicles and can dynamically adjust their racing line and speed in order to avoid a collision with a preference to hitting other cars over walls. They will also brake to avoid collisions. (good enough that 3 cars can navigate corners 3 abreast "most of the time").
Multiplayer up to 12 players on LAN (with a high end PC) 6-8 recommended for 1Ghz machines and a Geforce 3 generation video cards. LAN Multiplayer is auto negotiating coin op style setup where AI controls any additional NPC racers to fill in the empty positions.
You can play MP over the internet with mixed results. 2-4 player can be playable, the physics eating up the bandwidth somewhat.
Recently I've been seriously considering trying to port this game over to torque, with the new art pipeline the art assets will load straight into TGE which is kind of cool. Our biggest problem with the game so far has been internet multiplayer. Something torque excels at. In its currennt incarnation it can run quite well, but when it comes to playing over great distances, often 8000+ miles it can really bog down.
The one thing that is currently holding back trying to get the game up and running in torque is physics. So if anyone knows of any good networked physics solutions that are avaliable for torque users? I'm all ears. I once saw a Dracer video which left me pretty impressed, whether that technology will filter down to the community I don't know. What I saw looked very impressive. Other than that I noticed novodex used torque for their Ageia physics demo's which were very cool, but at this point in time I don't think thats an affordable solution.
If I could get it up and running to a similar standard as the current prototype, I believe something like this could do quite well on Xbox live. Which is one of my long term goals as far as indie games go.
So if anyone has any info regarding networked physics solutions let me know. Feel free to try out this self playing demo. 10mb
With the demo we pretty much maxed out the game as far as we felt it could go and still run at an acceptable speed on a 1ghz machine, the Older ODE physics implementation we used being quite a CPU hog.
s93153354.onlinehome.us/raceproto.zip
Quick info
I left the .ini file available so people could play with some of the settings we left open in there.
keys 1-9 change camera
ctrl+1 to 9 change car in view
you can tweak the .ini file to change a lot of settings
the game is moleboxed for security, as a consequence it may not run on some win98 systems.
if you want full screen:
screenmode=1
in the ini fixes that. And reducing the FPS on the mirror or number of cars in the ini will increase performance. (you can add more cars too, allthough I havent tried with more than 12).
Heh Just read over my plan again and noticed the abundance of I's :). Infact Damien Sturdy did a wonderful job coding the core racing engine and hacking into ODE in order to get the most out of it and realise many of the concepts and ideas I had in mind during the development of the project.
It's supposed to be more of a casual fun party type racing game, rather than a serious sim or arcade racer.
Key features:
ODE Physics
Dynamic AI. Cars see walls and other vehicles and can dynamically adjust their racing line and speed in order to avoid a collision with a preference to hitting other cars over walls. They will also brake to avoid collisions. (good enough that 3 cars can navigate corners 3 abreast "most of the time").
Multiplayer up to 12 players on LAN (with a high end PC) 6-8 recommended for 1Ghz machines and a Geforce 3 generation video cards. LAN Multiplayer is auto negotiating coin op style setup where AI controls any additional NPC racers to fill in the empty positions.
You can play MP over the internet with mixed results. 2-4 player can be playable, the physics eating up the bandwidth somewhat.
Recently I've been seriously considering trying to port this game over to torque, with the new art pipeline the art assets will load straight into TGE which is kind of cool. Our biggest problem with the game so far has been internet multiplayer. Something torque excels at. In its currennt incarnation it can run quite well, but when it comes to playing over great distances, often 8000+ miles it can really bog down.
The one thing that is currently holding back trying to get the game up and running in torque is physics. So if anyone knows of any good networked physics solutions that are avaliable for torque users? I'm all ears. I once saw a Dracer video which left me pretty impressed, whether that technology will filter down to the community I don't know. What I saw looked very impressive. Other than that I noticed novodex used torque for their Ageia physics demo's which were very cool, but at this point in time I don't think thats an affordable solution.
If I could get it up and running to a similar standard as the current prototype, I believe something like this could do quite well on Xbox live. Which is one of my long term goals as far as indie games go.
So if anyone has any info regarding networked physics solutions let me know. Feel free to try out this self playing demo. 10mb
With the demo we pretty much maxed out the game as far as we felt it could go and still run at an acceptable speed on a 1ghz machine, the Older ODE physics implementation we used being quite a CPU hog.
s93153354.onlinehome.us/raceproto.zip
Quick info
I left the .ini file available so people could play with some of the settings we left open in there.
keys 1-9 change camera
ctrl+1 to 9 change car in view
you can tweak the .ini file to change a lot of settings
the game is moleboxed for security, as a consequence it may not run on some win98 systems.
if you want full screen:
screenmode=1
in the ini fixes that. And reducing the FPS on the mirror or number of cars in the ini will increase performance. (you can add more cars too, allthough I havent tried with more than 12).
About the author
#3
11/02/2005 (5:45 am)
Midhir is right, there have been some posts on getting ODE to work in Torque. I think the main problems were getting ODE to play nicely over a network.
#4
I could be wrong though.
11/02/2005 (8:48 am)
I still wanted to keep the "wheels fall off" feature in there. Porting this to torque, I *think*, wouldn't be too difficult. As you said, We just have to figure out the physics. I have a feeling that adding ODE seperately to torque wont automatically network it- so we'd basically be ding it all over again.I could be wrong though.
Torque 3D Owner Jesse (Midhir) Liles
Good luck, and nice work =)