(SSGX Origins) Let's make a Wipeout game! This time for real!

Xpand, hm no actually not. But don't get me wrong, your track is awesome!! I am really looking forward to play this one.

But my idea was to have some kind of open-world track where there is no street like the Wipeout ones where the ship are "trapped" on the road. See Oryx post to get the idea :)

Sort of a rally mode with no set level? Interesting, but would be a nightmare to code unless the AI is completely changed from node based to a navigation mesh style. Both Blender and Unity can do this, as they are built in functions.

I think it would be cool to have this. I don't know how difficult it would be to program the AI or how "big" the track would be in filesize and rendering time since you have to create much more ambience and collision objects but maybe it would really be a fun mode and makes Slipstream a little bit more "special" and not just a Wipeout copy.
I'd say "let's give it a try" but since I can't code or model anything the work would be on you, guys... So, it's up to you :)


EDIT:
I also have an idea for the menu... I really like how the menu works and the animations, just the wall behind the ship looks kinda empty. How about to place some kind of certifications, finish line screen shots or prize cups when completing a stage in the career mode? So everyone would have his individual wall depending on his progress in the game...

It would'nt be too hard to get obstacle avoidance, its just it would splinter the game into WO and a cross between a pachinko machine and the speeder bikes in Return of the Jedi.

It would work better if there were more than ten racers (more like F Zero X numbers- so fifty!) ...it would make up for the lack of intelligence of each racer.

I'm not against the idea at all, in fact I thought of something similar quite a while ago. I wrote an entire game design document with backstory etc. I deleted it though when I realised I could never make it. Its a shame I did not hang onto it, as it would be feasible now and was an interesting read. One idea was that during stages, you would find special items in hidden places, and destroyed opponents could be salvaged by your team for parts (think of the balloons collecting mana in Magic Carpet). Some of these items were parts for the ultimate racer, so you had a balance of racing to win, and racing to find objects that might benefit you later.

One thing though. Can't one make AI follow a spline curve with a certain margin of error? Like, having a spline curve in the center of the track following its shape and then have something that calculates nodes at each point (vertex) of the "curve". That way we could easily change the amount of nodes by changing the interpolation of the curve.

My node setup does this partially and I suspect Zero's does too (i.e. nodes within a set distance are seen as being reached: so for exact path following distance would be 0, whilst for lazy pathfinding a large number can be used (like 50)).

I did think about this, but Blender cannot use splines in the game engine unless you you record each vertex somehow. I cannot speak for Unity though.

Hmm I wouldn't make it too much like a "treasure hunt"... Just a track without an actual road with simple start and finish. Otherwise, people would start to randomly fly around to find things and give it a second try to win the race and it also somehow has that GTA feeling in my opinion^^ You know... Driving around & looking for easter eggs and other interesting stuff.

About the number of opponents... Well the good thing about lots of opponents is that there is much more going on on the road, maybe resulting in more action. But the other hand is that if you have just about 7 opponents and they will always keep the same name, same ship etc. it's much more "personal". Like in career you will always know "Opponent random name here is always a hard nut to crack..."

@Xpand: Sorry, I have no idea what you are talking about^^ I hope the programmers in this forum can help you out :)

@Docfo4r
As much as I enjoy the idea I think it kinda goes away from the initial goal of the game of being the Amateur league simulation and more into the sandbox game zone... But those are just my 2 cents...

@Rotational
For what I've seen zero's system makes the ships follow a straight path between two nodes. That is a problem when you have twisting turns and stuff like that. That's why I thought of this. One can always make the node objects in the 3D software and make the game software assign progressive values to them, so the ships will follow it in order. This way you wouldn't waste too much time setting up each and every node in the track. Maybe just adjust the ones that need adjustment.

Hehe it's alright. It's just an idea of mine. Since I can't really "work" on something, I'll just think about ideas... What do you think about my idea with the menu? The changes on the wall behind the ship?

Oh and another thing: While I played the demo with the AI on Cassandra, it happened twice already that two AI ships crash together and then keep staying on the exact place of the track, they won't move any more. Only after I hit them hard, they get rid of each other and keep on racing...
Don't know if that is a known problem, next time it happenes I will take a video.

Xpand, I made an awesome logo in Adobe Illustrator today. It started as an accident honestly, but it is still awesome. It doesn't have anything to do with Slipstream GX though (it is a FEISAR logo). I was just messing around in Illustrator, I was using the Eraser tool on a Circle. I noticed that I made an awesome looking pattern, so I turned it into this. Sorry that I couldn't make one for one of the teams in Slipstream.

The one on the left is the old one, and the one on the right has a Glow Filter added to it with Paint.NET
What do you think?

------

If you need to talk to me, you can find me on Steam.

Xpand: There is an old Python script that takes a spine and records all vertex positions, and each vertex can hold a property (would have to be set automatically though!) so theoretically it is possible.

I cannot speak for Zero though. His system is probably more advanced than mine and may be more suitable.

A crazy idea I had for node placement was for the nodes to be placed by the player as they racing: as they go round the track in a test mode, they could drop nodes like mines so the nodes are placed on a racing line. But this would require a system to record where they go, and would be a lot of work for little gain.

The node mesh version automatically computes the shortest route, apparently, and does what you describe pretty well. The downside to this on Blender is that the agent that follows the path is not set up for racing. People who have used it for racers have used the agent as a moving node, i.e. the AI tracks to the agent as the agent moves round the track. In fact I tried this approach (but with an agent that followed keyframes) when I first started this :blarg

docfo4r: that design document was from ten years ago- I was naive ;)

Heee, well I had this curve navigation idea at the beginning of A.I scripting but someone said something that it is hard and other stuff so I've chosen different path... easier one and more demanding as each collider is hell of a b**** if it comes to CPU usage... I know one thing, imported curves are not visible in Unity so I assume that they're just not supported :| (no words about it on google either)
I have one thing on mind though, I can return to my old AI navigation I was using earlier (node distance check navigation (no colliders)) but setting this up will be much more "painful" as not only I have to name each one incrementally (1, 2, 3... gosh) but I have to add much more of them and rotate each one in direction the track is facing as I'm using them for ship respawning xd... or I'll just look for another method for respawning (still there is naming...)

Oh, and there is another way too (just thought of it :P): Create a script which will follow track on race start and place a node every n-meters while naming each one incrementally, but if track will have a jump or something which will make this script to lose track of the... track, it might fail (my new track has some big holes in track)... well, maybe i can add some modifiers which will keep it on good way :)

You can always make a system that checks the ship distance to the track's wall and corrects the ship's direction using the nearest face normal of the track wall, that way the ship velocity will always be perpendicular to the wall normal. But I guess that might take a lot of memory... I'll try it here...

Still i need checkpoints to check the ship's distance from finish line... well, maybe i'll just check how it works and then think what next :g

Xpand: Wouldn't that mean that the ships would always race exactly in the middle of the track?

BTW: What happenes when a ship got turned around in some way. Does it know it's facing the wrong way or will it become a ghost rider then?

Not if you gave them a threshold. What I said would only align then properly when they got to a certain distance of the wall, doesn't mean that distance has to be right in the middle. Actually having them flying exactly in the middle of the track would be impossible because the system would try to align the ship following the nearest normal, but because the ship is in the exact middle the distance would be the same between the two sides of the track, ergo the system would become confuse and you would probably have a spinning ship in the middle of the track if not a system crash in the worst case scenario....

Xpand: The other assumption is that the track sides are uniform and are upright, something which can be faked.

Zero: Your old AI method sounds like my current AI, except that nodes do not require alignment!

Nobody is going to say anything about my pictures? I can try to make one for some of the Slipstream GX teams if you want.

Yeah, I make an invisible vertical wall for the tracks to detect collisions, the guidance system can use that as well.
@Amaroq: The picture is nice, but yeah, it's kinda off-placed in this thread. Try to make some for the game, like adverts and stuff like that. Or even try to come up with new teams.

I... might not be able to do that.

I often need a base to work with, and there are very few things I actually do from the ground up.

------

EDIT: New Team Idea;
Name: [UNDECIDED]
This corporation specializes in the development of androids, crash-test-dummies, and safety systems. They recently formed an AG team, in an attempt to prove that Androids were equally good if not superior AG pilots, and that the death tolls of the sport would be greatly reduced. The team didn't start competing until the FX200, but have sponsored the FX150 and F9000. Due to their high success rate, they have been qualified for the FX300, but wether or not they will actually participate is unknown (hint: they didn't).