Discussion in 'Track Modding' started by TChapman500, Dec 10, 2013.
Thanks for the update. Glad you've been able to optimize some things.
It's good to take a break when things bog down.
I think it's safe to say that the start of the open beta period is only a month or two away. Here's a blog post with some details.
- if I only had time......
Well, the system I'm going to put in place on the CTC website will allow anyone not participating in the beta to at least track its progress.
good to see you are back on track
Good luck. I just hope you are able to listen to users when they provide feedback. Do not follow the steps of Mr Pywell which made a useless software just because he followed his intuition instead of listening to its old time users. RTB is the worst software I have come across in my life.
Enviado desde mi ONE A2001 mediante Tapatalk
You can be sure that I'll listen when my users give feedback. Also, I'm checking out the reviews of my competition to make sure that I don't make the same mistakes that he did.
Don't worry... he will turn on you if you do the slightest thing "wrong".
Good to hear you're back at it.
I too get 'bogged down' every few months. My aiw~thing architecture is laid out in ~35 data structures (functions & data members). So.. when bogged down, I usually go work on a different struct... something that handles a different process. Eventually, a week or so later the 'elegant' solution to the 'bog' presents itself (in some 2am delirium idea), and I can proceed.
Another nice thing about this architecture (compartmentalization in data structs)... if any one of them crashes in maxScript, everything else is still 'alive', so mining enough data to solve the crash isn't solely dependent on whatever error messages that the compiler spews out.
Anyhow... sorry about the novel.. and good luck working out this month's "undocumented features"..
Shakespearian Logic: 2B | /2B = ?
Thanks. And interesting insights Bink.
Interesting case-in-point on that "unpredictable progress" part of the last blog post: Looks like my new arc implementation is completely broken and I'm having an interesting time trying to figure out what is causing the break. I only have about 5,000 lines of code across about 10 files to figure out what's going on. But it also looks like my implementation is incomplete. The arc is forced straight by default and there is nothing in place that allows you to change it. I figure fixing that should be top priority considering how Sandbox users will be most familiar with that type of segment path and the fact that it's the easiest shape to implement that will allow me to give you some screenshots.
Side Note: Sandbox is designed specifically for NR2003, but by using 3D SimEd, I was able to get my tracks to work in rFactor 1 and I even made a Las Vegas re-bank for NASCAR SimRacing (that was actually banked 20 degrees and not just retextured). None of those tracks can be downloaded anymore and none of my rFactor 1 tracks or NR2003 tracks made it past beta (except for one track that was requested for NR2003). Since I am most familiar with how Sandbox works, I am basically copying the track creation concepts of Sandbox into the Chapman Track Creator, and extending those concepts while I'm at it.
Ah, Sandbox brings back some memories.
Very unstable application and every time when it crashed, it corrupts project file as well. If you didn't remember copy project file to another location before crash, all your work was gone.
I kinda liked its UI and spline method to create tracks and got pretty familiar with it during those years.
Nowdays I can't imagine to go back that kind of methology after 3ds max has grown inside my spine.
On the other hand, most new moders can't imagine paying ~4k bucks to buy a license for Max.
It's a lot easier to 'get your feet wet' (get addicted) with something like Chapman Track Creator, or RTB.
If either (or both) eventually produce a similar quality of track / facility, there's less incentive for Max ( and the heartbreak of Jka's spinal growths ).
TChapman has an advantage in the ability to buy a copy of RTB, build a track, and try out each new beta that comes along.
I'd imagine Ferrari would love to test the current Mercedes W07 on any track, whenever they want.
@Jka: Yup. Sandbox is very unstable. That's actually one of the reasons I'm building this track editor. The constant crashes are quite frustrating. Especially when "File -> Save" causes the crash.
@Bink: Don't you mean "to buy a subscription to use max?" That's what it is now! Note to self: Make sure the Chapman Track Creator is cheaper than 3DS Max. Anyways I doubt that I'll be purchasing RTB any time soon considering the fact that I'm still trying to figure out how to get my Steam account deleted and RTB is only on Steam (as far as I know).
Progress report (not enough to warrant a blog post): The new arc implementation is almost complete. Currently, you can directly set the radius, position, or length. Setting the radius or length will override the position and heading while setting the position or heading will override the radius and length. Still debugging the SetHeading function, which does not work at all.
In the off chance that your controls aren't already set up this way... Remember that the radius controls need to facilitate start radius and end radius for corners that 'tighten' or 'relax' during their traversal. [eg: check out (attached image) Shanghai turns 1-->2 (tighten), and 12-->13 (loosen)].
Having written a little track creation script (for testing ai driving line fitting in my script), I found that start/end radius corners are not as easily constructed as they initially seem... because the center point that the radius swings around (the red dots in attached image) needs to move throughout the span.
From an overhead view of the (above mentioned) Shanghai corners, you can see that the sweep center point (that the radius line swings around) moves from the center of the start radius setting to the center of the end radius setting.
View attachment 20105
(Edit) Note.. The reason for this radius center point movement:
At any point throughout the corner, the (yellow) radius line must be perpendicular to the instantaneous track path vector.
Hmm. Looks like an Euler Spiral. I already have a dummy Euler Spiral object in the code, but it's not implemented yet.
Interesting. I do my changing radius corners differently.
A straight has 0 angle/length 5m.
A right hand corner maybe 2 degrees angle/length 5m.
I do segments by doing as follows:
0.0 deg /length 5m (straight)
0.0 deg /length 5m
0.0 deg /length 5m
.25 deg right/length 5m (radius ever tightening)
.50 deg right
.75 deg right/length 5m
1.0 deg right/length 5m
1.25 deg right/length 5m
1.5 deg right/length 5m
1.75 deg right/length 5m
2.0 deg right/length 5m (radius now becomes constant)
2.0 deg right/length 5m
2.0 deg right/length 5m
Separate names with a comma.