Track Conversion To Assetto Corsa: General Discussion

[Ricardo Rey]

After several conversations here at ACM and elsewhere, it has been agreed that there is a need for a track conversion tutorial (at some point a car conversion tutorial). This thread is the start of what I hope will ultimately be used for a tutorial. For now, this thread is a general discussion to get feedback for said tutorial.

It must be said up front, and will probably be emphasized again; Get permission. Why? For one it is the courteous thing to do. I think the biggest concern is not that a person would mind giving permission, but if they have spent months, or even years, it’s their baby, and they would probably like to see it done properly. If you simply just ask; can I have permission to convert your track, you may get a yes, but IF it would be me/my content, I personally would want to know how you intend to go about it, and why should you be the one to do it. It has been my experience that if I present a game plan, detailing how I will go about it, I get a yes, and in fact will usually get help from them. More on this later.

Why a tutorial? When you look at the total number of racing sim enthusiast, the majority are not modders (I have no exact numbers, it’s probably 1%?). Their interest is content. I read how (on RD for example) people have an opinion of racing simulation games, in many cases, based on poorly converted content. People are going to convert tracks, no matter your or my personal opinions, pro or con. So why not provide a guide to do it right? I have found several tutorials that can help convert a track. However, they don’t discuss the details in doing it right.

The problem. Most of you here at ACM are involved in your own project(s). I have volunteered, as a rookie, to head up this endeavor. All I ask from the giants here is to add a comment here and there. I will sort out the details. Any contribution, in any order in the conversion process, would be appreciated.

My proposed outline:

· Get permission and how to go about it (although this is first, you should know the following)
· Research your project. What will it take to accomplish a quality conversion?
· Make sure that a fresh new scratch track is not on the horizon. Why waste your time if a scratch track is in the works?
· How to choose a track. How to evaluate the quality of the track using various tools.
· Creating a plan, a workflow including tools and resources.
· How to begin the conversion process. Folder structure, etc.
· How to extract the track mesh to smooth track. How to add back the track.
· How to update textures
· How to do “surgery” on broken parts.
· How to get your track conversion in AC.


I know I have missed a lot. Your comments are welcome. I am not easily offended, but please be constructive. I can’t do this alone.

 

[Johnr777]

This is a very grey area...

I think the library of possible "legal" conversions is small, and that library is full of lower grade content from older game engines. Getting that older content to look like something @Pixelchaser , @luchian and @LilSKi can create is a lot of work! I see this tutorial being used by people who'll rip codemasters, SMS, studio-397 stuff and charge $$ for it.

I would rather streamline a process where we can use lidar, GPS, Google Earth to get newbies building accurate scratch material without too much of a learning curve. The eye candy stuff to bring it up to AC/Kunos standard is already all available...

If someone wants to do a proper conversion, they'll find all the info they need, they already have your outline in mind for their project.

My 2 cents

 

[LilSKi]

The amount of work required to do a 'proper conversion' is almost more work than building from scratch. Plus most old tracks from other sims are far from accurate.

But in any case the largest problem with conversions is the material/texture/object list. Usually when coming from another sim the material list and the number of separate objects is huge. This causes a very large number of draw calls (DIP) which causes high CPU usage and when you add AI to the equation things go south very fast. The amount of work to consolidate those materials and objects is again almost more work than starting from scratch.

 

[Ricardo Rey]

I appreciate the input. I'm still new with AC and have only done R1 conversions many years ago from mainly ISI stuff. I don't want to spend a lot of time if it's not worth it, and I certainly don't want to promote piracy.

 

[luchian]

Guys, without wishing to be disrespectful, but I think you are not looking this from the eyes of a beginner also.

Short story (If TL;DR = we all start at point 0).
Some years back, I was looking for a little help with something in LFS. Wanted to display the status of our server in a different style and format. And a colleague forum user tells me: hey, that's really simple; you just need 2 lines of CSS in there and you're all set. And he gives me a link to some CSS tutorials. It's like I was asking to help me write "Hello" on a cardboard, and one would suggest me to go to school and after one year I would know to write it myself.

So what I'm trying to say here is the same: conversions might just be the first step one could take towards the awesome world of modding. When you see that something "you've" tinkered around with actually works, you might wish to see more of that. Maybe improve the quality of shaders, maybe even learn a bit of 3d for small modifications, maybe something on your own, who knows where it could go.

Again, no disrespect @LilSKi I love your tracks And works, and I could only wish to see only content like yours out there (..that would be bad business for Kunos tho as nobody would buy official tracks anymore :d) BUT not everybody has the skills you have. Not saying they cannot be built with patience and time, but you need to start somewhere. Sure, the performance will be bad, sure the track will look bad, but that's what I read into @Ricardo Rey 's post: structure the best practices and give solutions on how to do it properly. There's nothing scarier than the empty scene and the entire mountain to climb.. / @LilSKi

More than that, I don't think that everybody out there has bad intentions. I would even say, albeit naively, that some people might not even be (very) aware of the whole Intellectual Property thing. This would be an opportunity to hear about it.
When I first starting to #mod games, I had this image of a very "dirty" modding scene. I had the feeling that all the content out there is ripped from somewhere: 1/I was not even believing that someone could dedicate so much time for free; 2/I was wondering how somebody can achieve this at home, when studios are spending "years & millions" to do; 3/I was probably finding the wrong sites, that didn't really cared about what's ripped or not, but "to have quantity". 4/ok, I was stupid, but still..

But then I continued and I'm glad I have; I have met so many amazing people on the way !

Few scenarios:
- let's say I'm a "dark-without-even-knowing" modder. I wish to convert something from somewhere to AC. Like you said, the information is out there already. Wouldn't it be better I find this page tho', so that I can read about these aspects too ?
- there are good modders out there that are simply not interested in Assetto Corsa. But they did do very nice and built from scratch tracks for other sims. rFactor2 is an example. Maybe with a simple permission and some hard work, a very good track (that you might actually know and love in rFactor2) could find its way into Assetto Corsa..
- old tracks from old sims. There are lots of these. Some for historic versions of tracks which lack reference anyway regarding "accuracy". Maybe some of these are so little or remote, that no one will ever start them from scratch, because there is little interest for them. But when nostalgia kicks in.. , you would just love to drive'em. I am sure you MUST know the feeling when finding that little track where you once did a "track day". You would like it to be laser-scanned of course, but you are happy you have found it still .

Sure, some people might use this for the wrong reasons, but so is almost everything on this world (not being philosophical). And should somebody really want to do it, they will find the information.

All we can do here is to encourage fair use, respect for other modders and respect for others' work. Show that it is possible to do it "the fair-way". And I am sure also that all the inspiring projects around point exactly in this direction.

If you have reached this point and are still awake, thanks for reading. I'll go back to my #fairyland now

 

[luchian]

I would rather streamline a process where we can use lidar, GPS, Google Earth to get newbies building accurate scratch material without too much of a learning curve. The eye candy stuff to bring it up to AC/Kunos standard is already all available...

One does not exclude the other .
You might have seen them already but @DanTDBV did a lot in this direction, actually here, here and here.

 

[LilSKi]

I understand that viewpoint but that is just not how I started out. I started with that blank scene and started building Bridgehampton from nothing while using blender for the first time. It was not until later that I started messing around with conversions, for various reasons, and realized this is just too much work to do it right (for me). Now yes you can take a rf1-2 track and have it running in AC in the matter of a few hours but it won't be right. Now that may be all you ever want out of it but that is not how I work.

I'm not really saying don't do it. I am more saying don't expect it to be easy IF you want it to be right. It doesn't matter if you do it from scratch or a conversion. To do it right will require lots of work.

 

[luchian]

I started with that blank scene and started building Bridgehampton from nothing while using blender for the first time.

I don't actually have something to add, but wow !

 

[Ricardo Rey]

Let me say that whatever is decided, I don't plan on being a conversion guru. I know how to convert tracks to AC. They were not good at first, but found out a lot of what would be entailed in doing it right. I have been helping someone else with track conversations, also with the help of the original track builder.

I am actually spending most of my time building a scratch track for AC. But the conversion experience got me to look at Blender, and now learning to use it.

I felt that if conversions were going to happen, why not put together a way to do it right? It may actually discourage doing it. But I respect you all here, so good discussion...

EDIT: I think I need to be a little clearer. There is already a good tutorial by Rainmaker that tells how to use tools to convert tracks. I was mainly concerned with making sure a person got permission and more work was needed than just converting the tracks. The tutorial I am thinking of would be more of a work plan and more detail than the available tutorial; not to compete, or duplicate what I learned from his tutorial.

 

[cercata]

As @luchian says, I'd be interested in converting some tracks for nostalgia ...
For example the Montjuic MOD for Stock Car Extreme !!!!!

So I hope the tutorial advances ...

I see this tutorial being used by people who'll rip codemasters, SMS, studio-397 stuff and charge $$ for it.

You don't have legal problems if you do that ?
I'm gonna start a business then ...

To do it right will require lots of work.

Yeah, that's what many people don't get, quality/realism/polishment gets a lot of time.

Personally I don't aim those levels ...

 

[Ricardo Rey]

Don't know how many people would pirate tracks, and was not saying we should police it. I'm not about to ask everyone who builds a scratch track with 3ds max to see their license either. I'm sure there are some "illegal" scratch tracks as well

I'll put together more detail when I have more time. Most of the information is already available. I was simply proposing to organize it and add some detail to whats out there.
The top downloaded tracks at RD that are scratch come from you guys and I respect you opinion(s). Many of the top ten downloaded are conversions as well.

 

[GT VIRUS]

The info is out there for anyone who is interested, and there aren't any tracks left that need converting.

 

[Ricardo Rey]

Apparently I had left a draft here and can't delete it

 

[GT VIRUS]

So whats your point?

 

[Ricardo Rey]

So whats your point?

Didn't mean to quote you. So i deleted

 

[luchian]

The info is out there for anyone who is interested, and there aren't any tracks left that need converting.

Very well, so then the ripping part is solved (since it's done already). The tutorial will be used to polish things up. Win-win

PS: If you guys need anything deleted, let me know (you should be able to do it actually).

 

[Ricardo Rey]

The tutorial will be used to polish things up. Win-win

I have learned that the tracks I have been working on will be used by leagues/servers. (Not entirely). There is an apparent request for many more. Their big concern is track surface & light textures. I still want to decorate them up a bit. I do know that all have permission from original track maker. I honestly don't want to do many more and would like someone else to learn what I am doing.

 

[Ricardo Rey]

One of the issues that I mentioned in the Proposed Outline is evaluating a track. I have been working on a track for over a week and frankly, had I not committed to help, is not one I would convert.

@LilSKi - I have followed your advice and basically have reworked the entire track. I merged and re-faced some areas. I split the track. Went into node grid under utilities in the editor and changed everything to 0. I loaded the track and it looks great. The track is smooth except in about 3 track sections. I went back into Blender and there are no apparent problems. I even tried cutting those section out and then adding back new faces/polys. It is a flat surface track. Any ideas? Could there be something other than the track mesh causing the problem? I have spent way too much time on this and want to move on.

EDIT: Since it is flat, Is there a way to just smooth/flatten the entire mesh. What I found on Google re-oriented the axis

 

[LilSKi]

One of the issues that I mentioned in the Proposed Outline is evaluating a track. I have been working on a track for over a week and frankly, had I not committed to help, is not one I would convert.

@LilSKi - I have followed your advice and basically have reworked the entire track. I merged and re-faced some areas. I split the track. Went into node grid under utilities in the editor and changed everything to 0. I loaded the track and it looks great. The track is smooth except in about 3 track sections. I went back into Blender and there are no apparent problems. I even tried cutting those section out and then adding back new faces/polys. It is a flat surface track. Any ideas? Could there be something other than the track mesh causing the problem? I have spent way too much time on this and want to move on.

Is it a single large bump or an area that is crazy rough in general? It may be easier for me to see the track myself.

 

[Ricardo Rey]

Is it a single large bump or an area that is crazy rough in general? It may be easier for me to see the track myself.

When in AC driving, I can see a wave on the edge of the track where the bumps occur. Found a few more. In Blender I assumed it would be the areas I "patched up", but they are in the middle of sections that are straight; for the most part. I can get you a link in a minute.