nEditor - new omsi2 map editor

Hi,

I’d like to share that I’m working on a map editor for OMSI 2. My goal is to make map creation very user-friendly and avoid wasting time on things that can be simple and enjoyable (like creating splines or placing scenery). I hope the new editor will help extend the lifespan of the OMSI 2 community 🙂

At the moment, I’ve completed the Objects and Splines modules. Before release, I still need to finish at least the Terrain, vegetation/flora, and AI traffic tile editing tabs. Further development is planned after release 🙂

I invite you to check out the latest video presentation as well as the full playlist where I showcase development progress:

https://www.youtube.com/watch?…CcSd3HbwoX6R_V60yU9hsiylg

Vegetation editing will be a dedicated module, and I won’t go into too much detail yet, but it will include plenty of advanced features to make creating forested areas simple and fast.

As for undo—haven’t you gotten used to it being just an unnecessary gimmick?

Just kidding—of course undo will be there, along with multiple save slots for a single map and autosave.

The option for random rotation will definitely be there (as a checkbox). I try to include all useful features wherever possible.

As for vegetation, I can reveal that you’ll be able to predefine your own sets of trees (and bushes, if you like) and “paint” with them—but I won’t say more for now. I’ll show it in a video once it’s ready.

Regarding importing objects and splines—you’ll need to shift your mindset a bit compared to the current editor.

In nEditor, during the initial setup, the entire SceneryObjects and Splines directories are loaded into the editor’s internal database. This allows for fast searching and smooth operations—you can find things by name or path in seconds.

On subsequent launches, or by using the “import all” button, only the differences are imported progressively. There’s no need to manually browse directories anymore.

As for performance: I have around 43,000 objects in my SceneryObjects folder. The initial import (done once) takes several minutes, while subsequent launches that only check for missing items take just a few seconds.

However, there’s another time-consuming step—thumbnail generation. There’s no magic here: each object is placed into an empty scene, a screenshot is taken, and then saved to disk. For this number of objects, it takes about an hour—but again, it’s a one-time process. Thumbnail generation isn’t required for the editor to work, but having even a small preview of what an object looks like is incredibly useful 😉

Zitat von Etrusan

Will there be any optimization tools/information, eg drawcall+triangle count per scene and per individual object/spline (displayed near the object origin for example)? LOD switcher/debug, LOD calculation for the sco maybe? Things like that?

Yes, there will be information about geometry count (vertices and triangles) as well as a rating that evaluates the optimization quality of a given object.

As it happens, I once carried out quite a few tests on what performs poorly in the OMSI engine, so as a fun fact I can share some examples and conclusions with you.

1. Geometry and materials

In OMSI, geometry alone is not the only issue. Of course, the more geometry an object has, the heavier it becomes—but the number of materials on an object can be just as problematic.

Here’s an example that demonstrates this clearly:

A simple house (76 triangles, 129 vertices, 5 .jpg textures).

The major problem with this house: it has 17 materials. //don’t ask how I managed to do that...

I placed 5000 of these houses on a single map tile.

(My setup: AMD Ryzen 7 8700F / RTX 3080 / 32GB RAM 6400MT/s)

Result:

Average: 15 FPS.

I then optimized the object by combining textures that could fit into a UV atlas, while leaving repeating textures unchanged:

Result:

30 FPS (2x more)

I also tested a full atlas version (only one material).

Result:

~100 FPS

As another interesting fact: even unused materials heavily impact FPS.

So please pay attention to material count—it really matters 🙂

2. Trees vs objects

What is lighter? A [tree] type tree, or a regular .o3d object with a single material?

5000 [tree] trees:

1300 FPS — exactly the same as on an empty map!

5000 trees converted to .o3d (with only one material)

130 FPS (10x lower).