That happens sometimes if you're on a route that has clipping for the route you're actually on. I've found it happen a few times when jumping off a bridge onto the main track.
Again, AC implements the bare minimum for the track to be simulated "accurately", but it still doesn't feel like a real place. There's so many invisible walls at any point where taking any sort of "interesting" detour gets kiboshed immediately. iRacing I think sends you back to pits and instantly charges your CC $5 if you try to do anything that's "interesting" or "adventurous".
This is more. This is fun and whimsical and makes Westhill feel like it could be a real place. Which is a huge thing to say considering nearly all Kunos' tracks ARE real places, yet they feel more fictional than Westhill.
The thing that I figured out, which is a great complement to LFS, is that the new track "feels" like a feasable race track in a real space. So many tracks in racing games are "the track", but they feel so dead becuase they are just the bare minimum to simulate the track.
LFS tracks have always felt a bit above this, especially with open configs, and this does take it to another level, making LFS feel much more like a real place in the world.
Plus this is a demonstration of what LFS tracks "should" look like if upgraded to a new standard of higher texture resolution and higher poly count. It shows that complex shaders aren't needed to make LFS look better (just better shadows, the baked lightmaps still look awful ).
Stall 9/10 in Carpark X, I encountered this: (First SS)
As well, I think the grandstands should have an invisible wall, preventing this from happening
I did run into a few places on the national (I think) Kart track where I swapped (lost) the route when I was off the track. Even on my (powerful) GTX 970, causes FPS to go from 250 to 100. The routes for those tracks might need to be loosened up a bit as well.
I also encountered a weird GFX glitch (that I didn't SS) when I was on international track (along the backstraight), then turned left up the access road and crossed the bridge. The land was seethrough until it decided I was in a "off track" route.
Currently, I've resorted to creating a 2nd HTTP server (built in PHP, not a webserver) that's handling WebSocket handshaking + connection management. It's a bloody nightmare to integrate at the core of an application.
I've built websockets into the insim relay - it's the ideal program to do this for and actually quite easy. I only had to detect websocket requests on the usual port and from there on wrap existing packets in websocket frames and vice versa. And done!
But yea, don't you always need some separate service to provide websocket functionality? Nginx or Apache, etc would only serve as proxy, no? I never even understood how that would work, technically ... in my eyes it's easier to just create my own connection handler :P
hmm going a bit off topic aren't we
We don't even have a web server in our stack anymore .
I'll figure it out, it was a bit of a rant because I dislike the upgrade process. I would prefer a separate connection rather than devouring the initial HTTP upgrade request. It's not been fun, especially trying to implement a websocket server that is "compliant" to the RFC.
[...] and I'm aiming to introduce the first version very soon.
(I'm very sorry Vic, but I just couldn't resist )
I realised it when i wrote it, but as i've taken some time off of other work to work on this stuff, I dared saying it
I'll keep the pitch forks handy, UNLESS you want to take a stab at adding WebSockets support to PRISM. I understand that you have a solution that's awesome that you use here on this site.
As someone who's currently implementing WebSockets inside of PHP... It's **** terrible ;p.
ReactPHP is great, which is what most of the "cool" PHP stiff is based off of, but WebSockets as a standard is so retarded in its implementation. Most people implement it in a insane way, which doesn't help, where they wrap their WebApp and WebSocket routes inside this massive "App" object, which I hate, but unfortunately is the way WebSockets is implemented. (As a true upgraded connection from HTTP, and uses the same initial connection as well without option to route to different socket)
Currently, I've resorted to creating a 2nd HTTP server (built in PHP, not a webserver) that's handling WebSocket handshaking + connection management. It's a bloody nightmare to integrate at the core of an application.
Telling Hyperactive to buy a better wheel is like telling the Queen she's a peasant.
Hyperactive already has a very expensive sim setup IIRC.
So is better then AC Dustin? I always 'found myself' in yours and Tristan's posts when it comes to physics and force feedback.. we seem to have similiar taste in that stuff, if you can call it like that.
I don't really have a great base of knowledge about rFactor. I did buy rFactor 2, tried it out for a week, thought the FFB felt absolutely terrible (no concept of having grip or not) so I took up their money back offer. It wasn't inspiring at all, and none of the mods I installed felt any better.
I never did buy rF1 though. For some unknown reason, It never interested me to buy it.
That said, I thought Hyperactive had a different setup than he did.. Although I could be getting him confused with Matrixi. All finns look the same after all.