I agree, while it is fine to model your game after another, you want something to make yours special. Otherwise why would someone play your game and not a AAA title like The Crew?

As for the whole using real cars and such, as long as you don't take a direct rip from another game it should be fine. I'm not big on the whole legality thing, but as long as you make a generic "look-a-like" car without the obvious car model/brand I don't think the other companies can do too much about it. A red car with a similar design but no model/brand can be argued that it's just a generic red car that happens to look similar. Maybe. I'm not a lawyer so that's just my guess.

I'm not all that worried. I doubt someone would care if I did use assets of various games.. GT2 released in 1999..by the time I finish the game it would be 2032. No one would sue me..specially not for a freeware game. And here in Slovenia in a random village. It would be very disgusting if they sued me.

Oh okay.. thats something I should keep in mind yes.
I realize that when I'm racing against people living in south america..its kinda laggy. I mean in GameRanger - Need for Speed Underground 2.. Abu.. :D for now I assume I wont have any problems with multiplayer stuff.

Not even in the slightest.. Test Drive Unlimited did ofcourse use sattelite imagery and transformed roads in 3D..the issue then was that the roads were way to low poly. They solved this with The Crew..but the Crew isn't exactly the US 1:1 scale.. My racing game would feature the entirity of Slovenian Istria.. To make things easier for myself the GFX would be 1995 era styled. I just load up the point cloud lidar data and transform it to mesh..decimate the mesh to a really low poly count..load the satellite image on top do the little allignment and use the knife tool to cut out the roads.. subdivide..smooth..remove noise..decimate at a decent polycount.. and repeat for all 200 squre km.. Then I would manually add the woods and other minor stuff on the side of the road..then I would make 10-20 buildings to copy paste on every village.. Istrian villages look all the same.

The only thing I would think would be unique is the fact that the roads and villages would be more or less exactly like in real life.

Overall its a good package.. 600 cars, real world locations, freeware, mmo, car tuning.. and there is a charming simplicity in it.

Overall, to answer your original question, as a hobby project, I think this is a great idea. I'd be really interested to see how it turns out. It's good that you're realistic about how much time it would take though :P My experience testing MMOs and overall planning on creating network based games is that as soon as you add a multiplayer aspect to things, it gets exponentially complex.

I'm not fully sure I would play the game myself as a gamer, as I quite like good, recent graphics, however, I'd be extremely interested in following the project as a case study point of view, and as an interesting experiment!

And if the point is for you to take on this challenge, in a sens, who cares if people are going to play it? managing to make the game would be an incredible achievement on its own, IMO, and worthy of praise! I'm sure just for that, some people (like me) will be interested in the achievement itself, and play the game to see in more details what you actually accomplished.

I think it could be interesting to do a video that shows, in parallel, the real life roads and your game, to document the fact that it's actually the same road.

Thank you :) That I shall doo.. This is gonna be a lifetime hobby.
Even tho I'm a slacker, I'll be always coming back to working on the game.

If I'll do something I'll post about it here. Pinky promise.
I'm having some issues for now. I want to decimate all 200 stl files all in one process..but I can't find on the internet anything that would let me do it for now. I loaded all the stl files in blender and I saved the blend file..the blend file is 800mb.. I might have to add the decimate collapse modifier and apply the modifier on each and every stl by hand.. And when I'm thinking of doing that.. I'm just chilling and playing mount and blade xD..

I'll try to do something in the next 2 months and I'm sure to post it here.

If you can market this game well, I can see it being pretty popular. Definitely not like WoW popular and whatnot, but I can see it having a solid player base at least in the 100s on release. Then it'll be up to you to make it last or grow. Either way, good luck mate! If you need help or just want to bounce more ideas around, you know where to find us. :p

Thank you for thinking that.. and oh name of the game would be Racing United..
I will ask..but it would be more of a technical nature..

Yes it is a good idea, though like some of the replies an MMO takes work, and you have to secure rights to use the cars, etc...even if its free. The only time you don't have to secure rights is if there is a stipulation about educational purposes or you and your friends are only going to be be playing it. That being said, you could probably work with someone and make some unbranded cars that are modelled after specs of real cars, and that way start with a few instead of 200. Cheers.

Oh so.. all I have to do is make everyone my friend. :)
I just make the game not open to the public.. A Facebook closed group where mods control who is allowed or not allowed to play the game. You bring proof that you know how to carve atleast a semi-perfect racing line and you get allowed to race in the server.

Everyone would know about the existence of the game..but it would still be kind of private like this.

What's the benefit you want to get out of it? If it's gaining esperience on creating a MO racing game (first M is discussable). In that case you could start with a smaller area and a few Cars and see how it goes. If it goes well, you could expand/ iterate on your plan (legal issues, etc).

Take it step by step.
I don't know how experienced you are and how big the team is, but it doesn't sound like something you can do "between things".

There are not many MMO racing games out there.. We had Need for Speed World but it closed down. The Crew isn't really all that nice. I've been playing racing games all my life and I was never truly satisfied with any of them. So I thought I start making my own racing game and aim at making it MMO..

I'm making it step by step.. I'm working on it in a matter that I won't ever really have the heart to abandon the project. Four months ago I've been ripping manually every GT2 car and try to assemble the textures of it from 16 palettes of different colours in a 4 bit CLUT texture.. It was hard and I said to myself that It will take me 2 years and a half to rip all the cars.. I asked Leo2236 to help me out and he programmed this tool that is making everyone in the GT2 community happy.. The GT2CV that views and exports the cars with UVs and textures and a wheel.. Okay that solved the cars.. then the next step for me was to ask my goverment to send me the LiDAR data of Istria. I got the data and I subsampled the point cloud of every square mile and transformed it into mesh.. Now I opened all 200 square miles in blender and I applied the decimate modifier collapse to 0.1000., it is linked to others..and I have to manually click to apply to every square km.. I'm doing 15 square km per day.. My netbook is slow and gets out of ram fast..So I have to save and reopen the blend file..
When I'll have every square kilometer decimated to the same quality.. I'll be taking screens of the satellite images from a zoomed in perspective (highest of quality) and assemble them in paint. Apply each and every satellite image to its 3D square KM..when I have them all applied I go onto the next step. Using the knife tool and precisely draw edges for the roads in every square miles.. After this is done.. I would rip textures and vegetation models from racing games.. downsaple the textures in a low resolution that matches the textures of the cars. When I have enough textures and models to feel confident to be able to work on the mesh.. I'll start applying some textures and mesh.. And I would help myself using the street view from google..When all this is done.. I would start cleaning the mesh and fixing up any errors I did along the way. So now I would have all the cars, sounds, textures, models.. the location of Istria done.. I would just need to code the game. So here I would just ask for help xD..

I have little knowledge 3D knowledge and no programming knowledge and I'm all alone.
My experience was doing UVs for Gothic Reloaded Mod.. I've been applying transitional textures like grass to mountain and stuff like that. I've been doing easy 3D stuff like using the knife tool and extruding planks.pillars of houses and huts.. little eye candy stuff. But I'm confident if I make this my life time hobby..that I'll be capable of making a racing game. :)