Tracks Any dedicated people interested in collaborating on a track together?

[BluePotato00]

In my research on building a track, it appears to be a huge undertaking. It appears most tracks here been the efforts of a single developer, but maybe I am wrong. I see many incredible completed mod tracks (like Sveg - great interview about the build process here) and I'd enjoy the process and feeling of accomplishment of completing a mod track of my own.

There is a lot to learn and I think this could be an enjoyable project for a small team of 2-4.

Proposal:

• Build a team of 2-4 dedicated developers. Criteria:
  • Collaboration in the English language
  • Ample free time to dedicate ~5 to 10 hours per week on the project
  • Does not need to come in with a wealth of knowledge. We can learn as we go.
  • Needs to have the ability and dedication to educate oneself and research. Given the tools we have and the quantity of resources available online (RD forum, assettocorsamods.net forum, YouTube, Lynda, etc.) we can learn anything.
  • Needed software: Blender (that seems to be the preferred 3D modeling software for AC track modders), Photoshop with DDS plugin, KSEditor, others???
• Decide on a track. I honestly have no preferences or agenda here besides the following criteria:
  • Proper road course (not an oval) with a pit lane
  • Prefer a circuit over point-to-point
  • Hasn't been done before in AC
  • Is not currently in development (that we are aware of) by another modder
  • Lots of reference material available (terrain mapping [LIDAR?], reference photos, high detail in satellite imagery)
  • Ideally geographically nearby to a member of our team
• Share files through a well-organized Google drive folder with subfolders for reference material, models, textures, etc.
• Agree on a free means of team communication. Slack? Discord? A forum thread here? It might make the process more enjoyable if we had a means of keeping the communication public for outside comments/suggestions/etc.
• Gather reference materials
• Plan, model, texture, build, test, debug, learn, repeat, repeat, repeat...
• Share it with the community. Share everything. Not just the completed track, but all of our track objects, textures, and other source materials.

About me:
I'm a 39-year old American living in America, but I've lived overseas in India, Thailand, and Cambodia. Married to a great woman and blissfully child-free! I hold an MBA and a PhD, so learning/research is embedded in my core! I am on university faculty in information systems. I've been a computer geek power-user since middle school when we had an 8086 desktop computer with a 4-color CGA CRT monitor at home . My hobbies include playing music, technology, home theater, basketball, tennis, ultimate frisbee, running, travelling, riding my Vespa, reading, and of course sim racing. Sim racing is a newer hobby for me of only 3 years. I race on iRacing primarily where I peaked at a 2,150 iRating, but love AC for its wealth of content and this active community. I have an aluminum profile rig with a DD1 wheel base, Heusinkveld pedals, and a 4-actuator SFX-100 system that I built at the start of this year (I learned a lot through that process as well and enjoyed the help of the community).

My technical skills (not sure if all apply to building a track mod):

• 3D modelling: Beginner/intermediate (hope to learn a lot!)
• Photoshop: Beginner/intermediate (same!)
• C#: Intermediate
• Python: Intermediate
• SQL: Intermediate
• HTML/CSS/JS: Beginner/intermediate
• Excel and other data analytics: Advanced
• Video editing: Beginner/intermediate
• Photography: Intermediate
• Photo editing: Intermediate

My soft skills:
I'm an introvert, but not shy. I aim to communicate clearly whether in front of a classroom of students, in a video tutorial, or in an email. I prefer short meetings. I'm organized and I'm a planner. I enjoy leisure time and aim for a lot of it, but when I'm working, I'm productive.

So, is anybody interested? I have no idea if the responses I get will be:

• [no responses]
• [100 responses - yikes!]
• "This will never work"
• "It's better to do something like this solo"
• "Screw you, loser. Crawl back under that rock you came from"

 

[Emery]

If the track is a North American classic from the '60s, I have an excellent research library. I've also got a good library for a few European '60s tracks, such as Reims, Rouen, Snetterton, & Brands Hatch. My modeling skills are lame, primarily sticking to simple objects, but I'm decent with texturing manmade objects. I intermittently finish small works because life interrupts large works.

Back in the rFactor 1 days, I released a couple tracks with Bob's TrackBuilder. They were better than average, but not good enough for this decade. Can use SketchUp, still struggling to learn Blender. Used CloudCompare & BTB to process data for rally tracks commercially, but dropped out of the team when my mother-in-law needed higher care.

 

[BluePotato00]

If the track is a North American classic from the '60s, I have an excellent research library. I've also got a good library for a few European '60s tracks, such as Reims, Rouen, Snetterton, & Brands Hatch. My modeling skills are lame, primarily sticking to simple objects, but I'm decent with texturing manmade objects. I intermittently finish small works because life interrupts large works.

Back in the rFactor 1 days, I released a couple tracks with Bob's TrackBuilder. They were better than average, but not good enough for this decade. Can use SketchUp, still struggling to learn Blender. Used CloudCompare & BTB to process data for rally tracks commercially, but dropped out of the team when my mother-in-law needed higher care.

Hi Emery. Thanks for introducing yourself and expressing interest. I have no preference for current vs. historic. I assumed that we would have better access to reference material for current tracks, but perhaps you have a nice repository for some historic tracks.

 

[elloLeo Kinnunen]

Something like this was proposed earlier this year for Bathurst: https://www.racedepartment.com/threads/bathurst-lidar-reboot.165005/

It doesn't look like things are moving very quickly now, but might be an idea to join up together?

Personally, I'd be keen to be a part of what you're suggesting. I've run through lots of the tutorials and made starts on a few different tracks but young family limits the time I can commit to less than 5-10 hours per week.

One question I have: if a number of people are working on numerous objects around the track how messy are the texture maps likely become and how will this affect performance?

 

[BluePotato00]

Something like this was proposed earlier this year for Bathurst: https://www.racedepartment.com/threads/bathurst-lidar-reboot.165005/

It doesn't look like things are moving very quickly now, but might be an idea to join up together?

Personally, I'd be keen to be a part of what you're suggesting. I've run through lots of the tutorials and made starts on a few different tracks but young family limits the time I can commit to less than 5-10 hours per week.

One question I have: if a number of people are working on numerous objects around the track how messy are the texture maps likely become and how will this affect performance?

Thanks for sharing about the Bathurst project. I'm willing to be a contributor on another existing project for sure. It sounds like that approach to Bathurst could result in an extremely realistic mod. The one downside for me is that Bathurst has already been done before in AC.

With respect to the question about multiple contributors, I don't have prior experience working on a 3D production team, so I don't know. My thoughts were to model individual elements (trees, barriers, pit objects, cones, etc.) in separate files and import and place them into the main mesh once we were happy with the track and terrain. That would keep our main model as manageable as possible until the end. But maybe the ideal workflow is completely different from what I'm thinking.

 

[BluePotato00]

@LilSki just posted this video. Looks like a great approach if we can find LIDAR data for a track.

 

[Emery]

One question I have: if a number of people are working on numerous objects around the track how messy are the texture maps likely become and how will this affect performance?

Depends on how you group the parts. Main thing is that items used repeatedly should be grouped on the same texture atlas... I believe this is one reason why it seldom is useful to reuse objects between projects because then you have to go back and remap the objects to a new texture atlas.

 

[Emery]

LilSki just posted this video. Looks like a great approach if we can find LIDAR data for a track.

<ahem> I can find lidar data (https://www.racedepartment.com/threads/emerys-public-lidar-database.97718/). Doesn't exist for all tracks, but there's a lot out there.

 

[elloLeo Kinnunen]

I think the reasoning for doing Bathurst again was precisely that it hasn't been done with lidar. What tracks interest you though?

Emery has compiled USA and UK tracks with lidar https://docs.google.com/spreadsheets/d/1TC3ZEGzZtE7Ywmzf2UOlrZRaMbj_BOoIKkJHDtIPSpM/edit#gid=0

Spanish tracks: http://centrodedescargas.cnig.es/CentroDescargas/buscadorCatalogo.do?codFamilia=LIDAR

I think Brno, Czech republic is available but I'm struggling to navigate the cz language website: https://ags.cuzk.cz/dmr/

 

[Emery]

Holland has lidar available. I don't speak Dutch, but I halfway blundered through their site. Zandvoort's been done pretty thoroughly, so I'm not too eager to dig into it at this stage.

France seems to have lidar available, but I really don't speak French, so haven't blundered into. I did find their historic aerial photo database and its even better than the USA.

 

[elloLeo Kinnunen]

Can you point me to France lidar Emery?

 

[garyjpaterson]

Spanish tracks: http://centrodedescargas.cnig.es/CentroDescargas/buscadorCatalogo.do?codFamilia=LIDAR

Um... Yeah, not super dense data but plenty good enough to capture the elevation (not much, ~20m) and crucially the corner cambers. Quite pleased with how its drives.

Thinking about Aragon too which is found on that site for future (yeah I know, I'll never get round to finishing any mod now )

 

[BluePotato00]

The road tracks listed in @Emery 's tracks with LIDAR data available that don't yet have an AC mod (or in development) are the following:

• Blackhawk Farms
• Brainerd International Raceway
• Heartland
• Nelson Ledges
• Old Bridge Township Raceway Park
• Roebling
• Second Creek Raceway

Any thoughts on these? I'm unfamiliar with all of them. I'll do a bit of research on these tracks to see what reference materials are available.

 

[Emery]

Can you point me to France lidar Emery?

I can only point you to Pixsim, who have downloaded Charade lidar https://forum.studio-397.com/index.php?threads/pixsim-tweets.50701/page-15#post-935394. I've not asked them where they got it, but considering the coverage area, it seems like it is a national source.

 

[Emery]

The road tracks listed in @Emery 's tracks with LIDAR data available that don't yet have an AC mod (or in development) are the following:

• Blackhawk Farms
• Brainerd International Raceway
• Heartland
• Nelson Ledges
• Old Bridge Township Raceway Park
• Roebling
• Second Creek Raceway

Any thoughts on these? I'm unfamiliar with all of them. I'll do a bit of research on these tracks to see what reference materials are available.

Brainerd was used for both TransAm and CanAm in the '60s. Current TransAm runs there these days. Out of that list, this would be my first choice because it would be easy to do both historic and modern versions. There was one version made for rF1 (or converted to rF1) in the dark ages and no one has used the lidar data.

Blackhawk Farms and Nelson Ledges are both old club tracks with a rich history. Nelson Ledges has almost always had an annual 24 hour race, lately run by 24 Hours of Lemons. Versions of both existed in rF1 (and earlier!), but definitely not worth converting. Blackhawk Farms also was done for a motorcycle racing game. Nobody has used the lidar data for either.

Heartland Park Topeka was used for SCCA Runoffs in the '90s to the 2000s. A decent version exists for rF1 that was ported to rF2, but it is showing its age.

Old Bridge Township Raceway Park is virtually unknown and, to my knowledge, never been modeled.

Roebling Road is mostly known for being Motor Week's TV test track. A nice little circuit, though kind of flat and lifeless. Was modeled once for rF1 without the lidar data.

Second Creek Raceway closed in 2005. Used to be one of Colorado's desired tracks. Can't remember if it was debt or noise complaints. Never modeled.

 

[elloLeo Kinnunen]

Brainerd was used for both TransAm and CanAm in the '60s. Current TransAm runs there these days. Out of that list, this would be my first choice because it would be easy to do both historic and modern versions.

Looks good to me. Nice long sweepers to get your attention in the old cars!

 

[LilSki]

Building a track with someone can be tricky. First for me personally would not even think of teaming up with someone who is new to the process. Sounds harsh I know but I just could not see myself teaming up with someone I would have to teach everything to. I might as well just do it myself.

I've been working on a project on and off with Johnr777 which is Road America. We split the project up in two basically. I've been doing the track layout, land, walls, trees, fences, etc. and he has been modeling the various track side objects. This process works fairly well as one can work independently of the other. After I sent him a base land model he can start placing his objects and even keep them in their own separate kn5 file which all I have to do is copy over to the track folder and it loads up with all his work in place. We are even working in different programs with him in max and me in blender. So it's doable but each party really has to have some experience for it to work out in my opinion.

 

[Stereo]

Ir Sindaco tried to farm out help on Luccaring and I think ended up getting about 10 buildings contributed by other people and built the other 150+ himself, it sounds good to collaborate on a track but it's actually a lot of work and most people who volunteer only mean to give maybe 10 hours of work max to the project, not the probably 100+ hours it takes to make a decent proportion of the work.

 

[BluePotato00]

Brainerd was used for both TransAm and CanAm in the '60s. Current TransAm runs there these days. Out of that list, this would be my first choice because it would be easy to do both historic and modern versions. There was one version made for rF1 (or converted to rF1) in the dark ages and no one has used the lidar data.

You're a wealth of knowledge on tracks! Brainerd looks awesome! Two layouts (2.5mi and 3.1mi) with interesting features in the infield (camping, walkways, a tunnel, and a lake). I was a resident of Minnesota for 4 years of my childhood in the early 90s, so the environment is very green and nostalgic to me. My Google image search however lacked many "detail" photos that would aid in modeling and texturing.

 

[BluePotato00]

I downloaded 2 tiles of LIDAR data to see what sort of quality/detail was available. Here it is in CloudCompare (Following @LilSki 's latest video tutorial)!:

I needed tiles 2726-19-55 and 2726-19-56 since the track spanned two LAZ tiles

I acquired the tiles from here. The two LAZ tiles together were 90MB. Hopefully that's acceptable for a level of detail for the region I'm looking at.