[Fred]

It's easy to say: where is feature A, where is feature B, why was feature C done this way, but it's harder to create these things from scratch. In making a track editor, many things must be taken into account. I don't want to enter in technical elements here, as it's not the right place. Rather I'd like to talk about the various methods of producing an editor, and get feedback from NSR fans and creators. As a brief overview, I see two schools of though on track editors, each one with varying levels of complexity added. I'd like to list them, then talk about possible ways of improving the base system, and list their pro's and con's.

1. Block based editors.
2. Free form editors.

Block Based Editors
These are usually the simplest to implement and therefore are the most common. The user places blocks, and the program validates the layout.
Possible additions are numerous, I'll list a few off the top of my head: Height variations (bending blocks to vary the limited amount of blocks)
Decor varations (same physical block with different graphics). The advantage of this system is very easy editing (from a user perspective) and sharing of tracks. The disadvantage is never knowing where you are on a level, or on different level because everything feels strangly familiar.

Free Form Editors
These are harder to produce for several fundamental reasons (Editor complexity difficult to master for most players, generation of assets much more complex than block based systems). The advatage's are totally unique levels allowing amazing amounts of control by the creator. The disadvantages are mastering the editor (the more people making tracks, the more vibrant the scene).

Both system facilitate sharing, as they both represent small datasets.
Another area to consider which is incredibly important (which NSR shows brilliantly) is the SURROUNDINGS. Having landmarks is really important for remembering the track.

Soooo, what do you all think? Work with a complex editor and have unique experience, maybe having to spend many (4 or 5 minimum I would guess)
hours to get an exceptable result, or have a simple (easy to learn, and quicker to produce aceptable results) editor, but not giving you total control over the creation process?

Etienne, or other forums masters, could this be turned into a poll?

Sorry If not totally coherent, too much Yugoslavian plum brandy

[Fred]

To reply to myself

In the free form editor I imagine something like as follows :

Define the track using a spline from multiple views, then generating the terrain around that, then placing objects using a variety of tools (drop, place, spray etc)

[Etienne]

Thanks Fred for this wonderful summary of the two possible approaches for a track editor.

I personaly don't want to hear about the first option (block based).
The main reason is because knowing where you are on the track is very important for gameplay and for quick learning of the track.
All the track editors i've seen so far are based on this approach. The best one i've ever seen was in V-Rally 2.
I prefer to have fewer orignial tracks than thousands of same looking-and-playing tracks (i'm not aiming any particular game here )
I'd like to mention too that we spent a lot of time tweeking every single bump and turn in NSR so it's playable and interesting, and I'm sure we could never have reached that level of detail with any track editor out there.

For me the best approach would be to do it this way :
1) Build the terrain and drop trees and other elements
2) Build the road on this terrain

Of course this seems simple but it's very complex to do technically speaking if you want (nearly) anyone to use it.
This is also why (I beleive) I've never seen it in any game (or maybe I missed it).

[Fred]

Or, how about a totally different approach?

Take a scanned image (hand drawn, or from a map etc), and use image recognition to re-create a spline. Use this basic information for the route, then apply height variations etc!!!

Adding waypoints etc to help the software do its job....

Opens up real track (not really NSR orientated, not til F1 cars can fly ) creation with simplicity abound.

thoughts?

[Fred]

I personaly don't want to hear about the first option (block based).

I wouldn't discount this method, as it does contain advantages, as well as being much easier to accomplish on a technical front. The technical point of view is important too, as many people would love to see this feature, and it is feasible to implement in reasonable time scales. The other advantages I mentioned are also fair (easy editor, fast results). If the elements you dislike so much can be overcome, it could be a viable alternative. I'm also very keen to hear what others have to say too

[Etienne]

Take a scanned image (hand drawn, or from a map etc), and use image recognition to re-create a spline. Use this basic information for the route, then apply height variations etc!!!

I'm not sure it's a totaly different approach. It's a good idea, but it's going to save 5 % of the work I think. Drawing the track from top is the easiest part in the design process. Tweaking heights, width and banking is much more work.
We could imagine to get data from google map or google earth as well, so we would have a terrain and roads very quickly too.
But designing a NSR track is quite different from standard roads, as these are designed to avoid jumps, and with maximum banking value that are way too small.

I don't want to do a blocked based editor, rather do an importer to get tracks from other editors that use this technique.

By the way, i'd like to talk about Bob's Track Builder (http://www.bobstrackbuilder.net) which is an interesting approach. Having a 3D tool specialized in building tracks for various racing games is a good idea as well I think. It's not powerfull enough yet to do loops for example, but i'll keep an eye on it.

[kic]

y'a moyen d'avoir une version FR ....

je speak pas very très good l'english !!!

[Seb]

Tu peux faire des copiers-colleer des textes que tu comprends pas dans le traducteur de google
http://translate.google.com/translate_t?hl=fr
Bon c'est pas toujours nikel, mais ça traduit pas mal et devrat te permettre de mieux comprendre.

Bonne lecture

[kic]

mais que deviendrait-on sans google ... ?

[DromEd]

I dont mind about an easy Track Editor like Trackmania games...
I just want to know if that will possible to make a track with a 3D program modeling (like 3d studio max) and export it in the game
that will be nice

[Baboy]

I think there are more reasons why block based editors should be avoided. These are the following:

1) you don't want this to become a TM clone, do you?
2) block based editor is easy to pick up, but it has a lot less depth
3) give an editor that gives you the possibility of making tracks like the ones you made, even exactly if that's what users want

A free form editor should be the best idea, but for that, take inspiration by the shooter Pariah. Building maps in Unreal Tournament is a hard job for an everyday gamer, but it has almost no limits - Pariah though keeps the no limits, to a very user-friendly ambient. If you guys didn't try Pariah's map editor, the game costs almost nothing by now - I really think you should try it out to see how is it. I'm sure that would work well in NSR.

[montana]

Yes, this is great idea Editor in Pariah is only good thing in the game.
Please, try and think about it guys

[calexico]

A block-based editor is also very deep. A lot of Trackmania tracks are piece-masters and they are very different. I think it's the better solution because if it's well-made you can do also a lot of things in your map and everyone can easily do a track. And it could be a good way for the tracks database to have the quantity and also the quality.

So I prefer something more restrictive but more simple

[Etienne]

I'll have a look a Pariah editor.
The fact that an editor is very easy to use is not a good reason for me.
What is important is to have good quality tracks at the end. I prefer to have 10 very good quality tracks rather than 100 000 uninteresting tracks (ok I'm exagerating a bit here).
And the most important thing is that every turn and element must be different, so that in 2 to 3 laps you learn the track and you know where you are. We've done tracks in NSR keeping that in mind, and also we've done tracks that are not too long. I think it works quite well.
Using blocks has the consequence that you'll find the same turns again and again. And if to avoid that, you use the blocks only once, you loose the advantage of using blocks.

[Baboy]

A block-based editor is also very deep. A lot of Trackmania tracks are piece-masters and they are very different. I think it's the better solution because if it's well-made you can do also a lot of things in your map and everyone can easily do a track. And it could be a good way for the tracks database to have the quantity and also the quality.

So I prefer something more restrictive but more simple

I know that full well - I made over 100 tracks for the various TrackManias... but it's also true that after a while I started to feel the limits... it was annoying, especially in desert, that you have to keep using the same 1, 2, or 3 block curves, or maybe bankings. CrashDay did not make much of a success, as it all tried was doing what TM did TOO. NSR, way I see it, tries more than that. Which is correct, because Nadeo has made several TMs, we don't need another one - I mean, I'd buy it if Nadeo made one, but only if they finally make new environments You've got to admit that a Pariah style editor would give endless possibilities, while keeping an easy-to-use approach. I'd love that.