WIP Speedbuild

As i'm trying to refine the speedbuild method, i'll be posting a lot of tracks. Some of them will be progressed to the next stage, some not. But i like to get feedback on these. So expect 2-3 new tracks per week.. I'm posting these as BTB projects, you need this XPack on all of these:

http://cream.galleria.fi/BTB/XPacks/Default.zip (It's default XPack with CK-Tools embedded, highly recommended. To use as default XPack in your own projects: replace the BTB\XPacks\Default.zip with this one). Rev2 AIW needs that 140000 to fuel consumption
BlackKnight Lee said:
By the way, 143000.0 I found was a good working fuelusage figure & I re-named Rev2 as Isle of Speed ;-D
I also suggest that unless you change GDB, set tire/fuel to 0 wear/usage, otherwise you tires will cook off when testing.

Here's results from first two sessions (name, build time, length):
Rev2 (or Isle of Speed...): 8h, 14km
http://cream.galleria.fi/BTB/Speedbuild/Rev2.zip
SRSPeedbuild: 1h45min, 1,8km
http://cream.galleria.fi/BTB/Speedbuild/SRSSpeedbuild.zip

Honest feedback please. It means that if it sucks, it sucks and should be said so.. I'll also will be making tutorial to this method as we go. I can't fortunately make videos, so someone can do them if there's will. I can make pics and screengrabs to try to explain this method. I'm pretty happy so far with the results of streamlining the creative process. Speedbuilding is quite a rush with rewarding test drive session at the end... In that first project (rev2) i had a faulty texture that slipped by inspections, that bughunt is not included in buildtime but the first working version was done in just over two hours.. Note to self:use XPacker to add new textures, it has a red warning..or inspect every texture before adding.. With the second one, i got the idea just before the race that evening, crashed and started doing the track at 22:00. By 23:45 i was on the track and enjoying my ass off. The track works, i know it beforehand but the main fun of learning to drive on track is there. Rev2 is tested on BMW GT2, the SRSSpeedbuild is for Matech GT.

There's also a custom weathers for each track, i'll post those later.
 
Rev2 test procedure:
Load the latest version:
Replace My Documents\SimBin\Race07\User\LOG\weather.txt with http://Cream.Galleria.fi/BTB/Speedbuild/weather.txt and go to offline practice, set conditions to changeable!!!.
Car is GT Matech from Power Expansion. Load car setup from online.
Drive on to track. Start slowly, track has some runoff and protection. If you happen to find yourself off course, behind the wall, just drive towards the track. There is invisible ramp surrounding the walls that will help you get back on track. Watch as the weather progresses and try to stay on track with slicks as long as possible.. And imagine what it would look like if there would be working dynamic shadows..
 
The method basically is using fairly simple meshes as "layers" with slightly different parameters (mainly random positions in Evo, rFactor has more surfaces to play with) to create hugely complex shapes. The whole track can then be reskinned and possibly even with different surface parameters to create infinitely amount of variations in the least possible effort and filesize. Reskins would consist of textures and new .trk, .gdb. .aiw . cam (or whatever they are in whatever platform..) files. Under 10mb easily. And i would recommend the Autosim Weather Generator to create custom weathers/events for each skin, going as far as limiting the car class/mod. The tracks gain a new layer when it is desgned for certain conditions and in the back of my mind, i have something like driving 2011 F1 Australia weather in a specially designed track/car combo..
 
I like the idea, but can't get my head around it exactly ;) I think some of us would need some screenies to visualise what you're saying here Kennett.

(Btw, tried to reply re. Goodwood Hillclimb but your mailbox was full. Might want to empty it a bit ;) )
 
I like the idea, but can't get my head around it exactly ;) I think some of us would need some screenies to visualise what you're saying here Kennett.

(Btw, tried to reply re. Goodwood Hillclimb but your mailbox was full. Might want to empty it a bit ;) )

Yep, noticed.. fortunately it sends an alarm thru email. It's cleaned now..

I'll start making the tutorials, i'll first have to think a bit how to do it, in which order.. Easiest way to do it is if you guys load the project, it's pretty self explanatory.. I haven't thought of correct terms but closest i can think of is layers of meshes that reside on the same plane but have little variations in them. Try to think the track as a crossection, slice it and look what is in there.

With the project i suggest that you select track crossection tool and change it to Material. Click on track to see what is what and proceed to next track with "T" key. I make for ex walls for the whole length of track with one tool. Same goes with all the space between the walls and tarmac. It's all made from tracks with usually two textures per side. By overlapping their position in crossection and using different random values, they start to appear in some places thru the flatter surfaces. Adjusting track height in crossection edit and adjusting sligthly different cambers, different parts of those "interacting" meshes can be adjusted. And this all can happen in very long sections of the road. Also you can use track width to change only the walls width etc.

The protruding of more bumby mesh thru flatter one that is 1mm higher effectively creates much more complex shapes, it would've otherwise increase the polygoncount on single mesh so high that it would not practical. That's the theory. It also makes the use of different panel lengths to further make more complexity.. Try making the two tarmac/tar surface to 5 and 3 meter panel lenghts. The edges are in totally different place all over track. One could figure what is the difference between even or odd ratios in panel lengths. Even creates simpler but more strong bumbs, odd has more variations in all directions.

Change the active track to bumbs (i don't remember if i renamed them... if i did, it's B) Try then to change the random values in track crossection Shape tool. Again, long sections of the track can be controlled and position of the bumbs can be adjusted further. I used different numbder of points in crossections in purpose, to mix them more up. With camber one can make the bumbs appear in in/outside of the track. All this is applicable to berms, ditches, runoffs.. If you use the same texture in bumbs and normal road, the user doesn't get any enhancments locating the bumb "zones". By using more or less varied textures on both, the effect can be adjusted visually.

By increasing the details on track, both visually and at the FFB too, the overall detail needed around is much simpler.. The method creates more immersion and detail where it is needed, on track. Specially in Race07, where there's only one tarmac.. rFactor guys should be dancing around naked at this point, ASBO worth hulabaloo at least... I've tried with rumble and tarmac but then every driver needs to drop rumbles from FFB and sound needs to be modded to have a true second drivable surface to Race series. Grass and gravel won't work, the tire needs to be at certain depth (~2cm] before it is on gravel. If there's another drivable surface just under it [~1cm}, good luck.. Basically car needs to be on gravel/grass with full tire contact patch and with 20cm gravel surrounding.. Patches won't cause reaction.

Load the project, this is smaller: http://cream.galleria.fi/BTB/Speedbuild/SRSSpeedbuild.zip
Add this default XPack to the project folder: http://cream.galleria.fi/BTB/XPacks/Default.zip
 
TL;DR

Oh.. by turning track visibility in track select/copy menu the different elements can be identified.. That is the quickest way.. So, for those that think previous epic post is TL;DR, here's ADHD version.. Load the project, open track select/copy menu and cycle tracks visibility on/off..
 
And one can make a Ice track too! By changing the protruding track surface material to grvl, there could be "snow" patches in the track! And replace grass and gravel textures to something suitable, dirty/clean snow for ex.. In GDB decrease grip, tire temp multi and wear, change location and date/time.. Create a custom weather that has 0 celsius.. Bang.. Winter... if we could only change the rain to more snowy consistency, slush maybe if not snow..
 


A quick reskin, 20-30 minutes with lousy shops no exports.. Have to wait til tomorrow for testing the first wintertrack in Race07..

EDIT: Did a shader test and boy, if i change the snow road patch textures to follow the tarmac road add/multiply maps instead of custom spec/bumb.. Snow with grass's Add/multi maps blended with grass patches (shopwork).. Beautiful, i have to HQ them..
 
You really want to know?
OK. One of the things in the BTB community i hate the most is that 8 out of 10 builders do quantity instead of quality.
And your examples look like that. They look like a 8 year old has made it (don´t know if its only the two examples or if this is your imagination of quality).
If you wanna make it look really good then it takes the same time like "normal building" because one has to do more in max and simed.
One has to delete a lot of useless polys, connect others, etc.

Or did i get something wrong?
 
Nope, Pretty much covered what i know. They are speedbuilds, not final in any way. More of a method of quickly getting things done. If you wait for 10 minutes for something, the inspiration might be gone.There's is the issue of unsused polys which i hope i find some solution for it... If we all had max and knew how to use it, the sky is limit.

I do rather quantity, i haven't got time/patience to do one track/year. That level of detail and optimization i expect from game companies and more-than-one-people modding teams. It is absolutely great that there are builders that concentrate on doing great classic real life tracks no matter how long it takes. If the track is less used and the layout not-so-great, the level of detail doesn't need to be amazing. With fantasy tracks, you need to be absolutely certain that the track it self is worth the extra attention. I can't think of anything else but to produce a lot and wait for feedback to which you can respond quickly, right in the creative process.

What you're after unfortunately, there's no shortcuts. It takes a looong looong time to finish a high detail track. That will reduce the number of trackbuilders. A. lot. More streamlined, simple process that'll produce adequate results increases the userbase. From the base we get few real talents and we get more detailed high quality tracks. Simple. So for this method, speed is of the essence. There's more driving and less fiddling, which i do prefer as learning a new track is highly rewarding, even if it's one of you're own creation. The learnign curve of producing tracks is reduced. And in now way these fantasy tracks are here to compete with RL tracks.. That's why i don't automatically upload them to RD:s storage, i use my own.

I would like to see what can be done about cleaning the project. And how much you can delete/merge stuff but that's max-territory and i have no idea how that beast works.. But i think that the track surface layering does make much needed complexity to road shape. We usually have single panels on the road with bumbs on equal intervals, this with BTB. And if performance is not an issue then there's no need for 100% optimization. Memory is plenty, i haven't found GPU or CPU issues apart when two ripple to road material changes are layered together..( Strange FPS drops, CPU most likely. Physics going in to overdrive or something. Two roada's don't affect the game in anyway, nor does roada mixed with grass..). I know it's bad "code" to do "bad code", not completing certain established procedures. Depends on the level of releases, me thinks.
 

Latest News

How long have you been simracing

  • < 1 year

    Votes: 299 15.3%
  • < 2 years

    Votes: 210 10.7%
  • < 3 years

    Votes: 201 10.3%
  • < 4 years

    Votes: 150 7.7%
  • < 5 years

    Votes: 266 13.6%
  • < 10 years

    Votes: 230 11.8%
  • < 15 years

    Votes: 146 7.5%
  • < 20 years

    Votes: 117 6.0%
  • < 25 years

    Votes: 88 4.5%
  • Ok, I am a dinosaur

    Votes: 249 12.7%
Back
Top