Development Screens (Each post contains a picture OR contributing post)

Experiment:


To make a road and island, that was around 15 minutes, to make terrain be smooth at edge of road, more than 5 hours probably, I have done almost 2 hours now and it is not much progression.

Greatest thing of all, I choose save as, save and NOTHING appears to folder, should of known that. So I guess that only way to keep progress would be to save over template map, except save option is grayed out.

From some reason json has appeared to /levels/ folder not levels\devils_rally\ which folder I chose. There is no logic in save function at all, maybe that is a bug?

 

I haven't made a map in forever but iirc it should be stored in MyDocuments/BeamNG.Drive/Levels, then you can copy the files from there to the folder you want for the mod.

 

It does, but when selecting save as, one would think that it would save to place one chooses to save. Really weird. Maybe it works after moving files, so at some point I can put some more wasted effort to my low effort map

Update: Mandatory screenshot, both terrains are generated with Bryze 3D, just few clicks, but surely it takes practice until all these tools become really useful.

 

@fufsgfen I thought about toning down the amount of polys on the intersection, but also, consider the fact that I can just snip the needlessly long ends off of it and that takes it down quite a number. I could do less polys, yes, but the collision differences will inherently get noticed, and, well, I want to go full detail.
Good luck making your own maps buddy, back up, and back up often, as with this editor, you might be backing up a lot with how it gets sometimes. I swear mine had PMS last month, or two months ago.

You don't want to know how many polys are on the single-piece mesh (that's almost always full-detail because it's almost always in view in some shape or form) that makes up the base for a majority of the city roads & highways in West Coast is. This is like 0.01% of that. Even 100 of these intersections wouldn't be 1%... So I've got a long ways to go before I impact the FPS. Plus I will end up eventually putting LOD's and null-detail levels on this piece, and that will help considerably.
I don't want to make so many pieces the CPU struggles to keep up, but I want to make enough that I can free-form build my city inside the editor, much faster than I would have been able to had I done it inside a 3d modeling program alone. Also, these pieces will be again completely modular, with tapered edges (as you'll see in the screenshot), so they don't catch car tires or edges while sliding or driving harshly. These pieces will then again have another lease on life beyond just this map. To make something so detailed, and so time consuming, for so little result (e.g. only one map) would be a complete waste. I want to give back to the scene that has given so much, but is so limited with so much more to give with more 'pieces to the puzzle' used to build the scene.
Sure, it might end up being an even bigger download than Roane County in the end of it all (by the time it's done), but it will be so much more city than we're used to getting in BeamNG. It will in the end be a true successor to the 'Generic City' map that I pledged to reboot. There will be working train tracks, and one or two MORE forms of mass transit that I cannot disclose at this time - one which we don't eve have a suitable vehicle for yet! What starts as a simple road, a few bridges, this intersection, will in the end be a beautiful, sprawling city that hopefully doesn't hit the FPS as bad as West Coast does (due to the fact that West Coast is rendering SO MUCH in full detail so far out, because of the single-piece models used in the map). My city will render in as you drive. Each 100 foot (or less or more) section of road will render in, not multiple miles of it packed together, some out of view, all at once, causing a nasty hit to FPS. I surely will be tearing up the 'donor map' quite a bit to do what I need to do - but it will have a lot of things that for 2~3 decades I have only dreamed of making. Once I get most all the road pieces made, the rest should be pretty clear sailing (filling in the blanks, or making pieces as-needed).
Again I will mention, that due to the fact there are no pieces out there already like it, it will be a complete kit for making your own:
*Surface streets, both suburban (lacking sidewalks) and within city jurisdiction (mandatory sidewalks and drainage, etc). All roads with curbs have the curb-faces tapered back a slight bit, as in real life.
*Collector roads and arterial roads that side/surface streets empty onto, complete with a matching modular bridge kit. Collector roads are usually striped (have lines) and have signaled vs signed intersections, sometimes are state numbered also.
*Highways, of course we must have these, because going fast is fun, especially in a car accident simulator! Of course this comes with it's own bridge kit and will come with lots of nice retaining walls, and in time, some nice modular on-ramps too. Likely will add curved pieces to the mix later in the development. My goal here is to be able to free-form almost any intersection you can imagine. (Choice of different) Pillars will always be separate pieces, as they're in different spots on a case-by-case or 'bridge by bridge' basis.
*Other forms of transit not limited to roads, such as train tracks above ground (at-grade or RAISED!), and below ground for tunnels - pieces will aim to match road length on many of them for the obvious benefits.

All pieces will have built-in collision meshes, and at-least a dummy billboard LOD generated by the game engine, if not multiple layers of detail (when and where needed, these increase model size both on-disk and VRAM use!).
Pieces will be public domain free-use (only for BEAMNG.DRIVE maps) released when the map comes out, with textures, so that more people can make Bob-Quality (or similar) maps or hopefully something even better!
I have to do some cpu tests to see just how modular I can go, but I have a pretty good idea of what to do and what not to do, as in, when it's just too much for the cpu to process (*COUGH* Downtown Harriman *COUGH COUGH HAAAAACK*).
If you can run Roane County smoothly without reflections being on, you shouldn't have too much issue here running this future map. At the going rate I should have in-map shots within a few weeks and a thread started.
--Cheers!
Oh, and I did finally finish my four-way intersection modeling. I might just be able to texture this tonight.
What you can see of the poly or triangle count (that didn't get cut off) is included the collision AND visible meshes, inside the modeling program. So that number would be half of what it is, to render. So don't cane me yet - the ends of the mesh here in the picture are tapered to avoid catching tires at joints, for those that missed it.

 

new project who dis


editt:

 

What happend with that Honda insight?

 

gave up on it due to lack of ref photos. might come back to it at some point. I dont have the skill

 

oh k

 

modding live!
--- Post updated ---
@fufsgfen

 

I introduce to you the biggest prop in beamng

 

Can i test it?
I5 3570k,gtx 650ti boost,32gb ddr3 1866 oced at 2133mhz,anything runs at 50-30fps
--- Post updated ---

My second pc (laptop)is
Amd a8 apu+r5 internal graphics 1gb gddr3 766mhz,8gb ddr3 1366mhz,1tb hdd

 

Sorry for the VERY late reply, but I would recommend making the rear indicators (turn signals) Amber as the majority of mid-80s Japanese cars (Toyota Camry, Honda Accord etc) has them.

 

Actually I already did that, but haven't posted any pics of it

 

cycles render seems to distort UVs quite a bit...

 

I added the front pickup bed wall. It was a slight challenge to join it to the body because of the odd way that the interior mesh is built. Now I just have to work out the textures, make the mesh look a bit cleaner, and add the rear glass. I also plan on making a completely topless version, and the two body types will come together in the same mod.

 

Would be cool if you could ask @losr for uv maps of Etk k and make it for Etk k