MX Simulator

Just the thought of MXSim going UE4 makes my heart skip a beat, that would be wicked as hell. I know the New Monster Energy Supercross is running on UE but the physics are just unbearable, garbage. Someone also mentioned that nearly everyone would most likely repurchase mxsim had it been converted to UE and i'll tell you right now, it's a yes for me.

Jeremy150 wrote:So what options are we limited to with materials currently? Is it just Specular and Mirror/Gloss? Is it just the extra options like Metallic, Roughness, Anisotropic, Sheen, Clearcoat, IOR/Transmission, and accurate Fresnel, that have been giving me much better results? I suppose if you just added those properties, then we would be able to achieve the same result. However, why not just implement the Principled Shader node as our main material setup? From the node that I posted you can see how easy it is to get something that looks pretty good.

You currently have ambient+diffuse+specular+optional enviroment shine layer. It's basically the standard OpenGL lighting only per pixel instead of per-vertex. The weakness of that is ambient + a single directional light is an extremely simple light setup which looks bad by today's standards. That's why I want to make the diffuse and specular parts use the environment map which will give you infinitely complex lighting. That will be a huge difference. Fresnel is the sort of thing that's a nice touch, but you have to look for it to even notice. It's an easy calculation so I'll do it, but I don't get the obsession over it.

The reason not to use the Principled settings under the hood is because they're a limited set of parameters that Disney chose to be easy to understand for their artists. They aren't the actual settings a rendering engine will use internally. It's a little like the stability vs advanced stability settings in this game.

If I support exporting principled node materials it'll be part of the export script, not in the actual model file.

Jeremy150 wrote:Again, I'll agree that there isn't a huge difference, and that PBR isnt "some kind of rendering magic". But it is pretty clear that you can achieve the same, if not better results, with a much simpler node setup.

Selling it as a way to simplify your materials is fine. I agree with that. I'm just saying that the standard nodes are more powerful. You won't be able to come up with a superior render for principled vs the standard nodes because the standard nodes can do anything the principled node can do. (But not vice versa!) It's a good user interface but that's all it is!

jlv wrote:The reason not to use the Principled settings under the hood is because they're a limited set of parameters that Disney chose to be easy to understand for their artists. They aren't the actual settings a rendering engine will use internally. It's a little like the stability vs advanced stability settings in this game.

If I support exporting principled node materials it'll be part of the export script, not in the actual model file.

Jeremy150 wrote:Again, I'll agree that there isn't a huge difference, and that PBR isnt "some kind of rendering magic". But it is pretty clear that you can achieve the same, if not better results, with a much simpler node setup.

Selling it as a way to simplify your materials is fine. I agree with that. I'm just saying that the standard nodes are more powerful. You won't be able to come up with a superior render for principled vs the standard nodes because the standard nodes can do anything the principled node can do. (But not vice versa!) It's a good user interface but that's all it is!

Isn't that kind of the point? It's not about having a bunch of different options to play around with and tweak, and keep adding different nodes to get something that looks correct. It's having that one node that already has the calculations built in, and is correct.

I guess that if you are still against implementing it, just being able to do more complex node setups will still be a big improvement over what we have now. I am really happy that you are setting time aside to improve the visuals of your game, and I know you just want what's best for it. With all of the optimizations you are going to make and implementing the use of HDR environments, we are going to be seeing a big improvement this year.

I think graphics are the biggest turn off to new players now. You've got gameplay down besides a few small things. Once you get graphics down you could make this game see another big surge.

jlv wrote:- I'd kind of like to have displacement textures even though that would hurt the FPS.

By "displacement textures," do you mean something like this?


Or maybe more realistically, like this?


Either way, that would be pretty cool. I'd love to be able to get more detail/depth in the dirt—3d tire marks, etc. Though balancing that with erode visibility is a challenge.

Image based lighting seems great for improving the look of bikes/objects. I'm assuming the same lighting will be applied to the terrain? Will we have the ability to use global and local lighting (https://chetanjags.wordpress.com/2015/0 ... -lighting/)? I'm thinking of situations like a night race where the track runs through pools of light from various sources, or a sx track with a section that runs outside the stadium.

I'd also love to see some improvements in the visuals of the actual tracks. After all, the thing players are looking at most during a race is the surface of the track itself. Another huge component, especially outdoors, is vegetation and crowds. I've relied heavily on billboards in the past because they are easy on performance and give an acceptable impression, but if you look any closer it's pretty crappy. I'd love to see more 3d vegetation/grass/crowds. A lot more is possible in the game as it currently is, but I don't know if anybody has really cracked it (though Benjii_D had some really promising grass: http://forum.mxsimulator.com/viewtopic. ... 55#p798594). Some wind effects and animation would bring things to life.

Unfortunately I don't really know enough to know exactly what to request, but I'm very interested in what can be done to bring more life and realism to the environments.

Somewhere along the way can we get bike/rider shadows?

Jeremy150 wrote:Isn't that kind of the point? It's not about having a bunch of different options to play around with and tweak, and keep adding different nodes to get something that looks correct. It's having that one node that already has the calculations built in, and is correct.

Just because you used the principled node doesn't mean it's "correct". If you want correctness you'd actually photograph the material with the light and camera at various angles and create a material that reproduces those measurements. That'd probably be easier with the standard nodes because their functions are well known, whereas the Disney parameters aren't. (I'm sure they're documented somewhere, but I read the slides and course notes and still don't know the exact definitions.) If you're just eyeballing it you're so far off that the edge on parts at the tails of the functions that the Disney shader addresses are just noise compared to the errors that are facing you straight on.

Here's what we're arguing about. Left is a plain old diffuse+glossy mix, right is principled using the numbers you provided. Is it obvious why I care more about environment lighting instead of some marketing buzzword like PBR?

ColtonD719 wrote:

jlv wrote:- I'd kind of like to have displacement textures even though that would hurt the FPS.

By "displacement textures," do you mean something like this?


Or maybe more realistically, like this?

My plan would be to use this technique.

The game could actually be a super power if it had the graphics as Supercross the game does but with it's own physics.

Gad damn that keen to see sim graphically updated.

onefoureight wrote:The game could actually be a super power if it had the graphics as Supercross the game does but with it's own physics.

+1

Even with a new graphics engine, I don't see people dedicating themselves to figure out the difficulty of the game if they haven't already. If they're too hard headed to try it out because the graphics aren't up to par, then they'll find a way to not be interested if the physics finally exceed their expectations but then realize it's a difficult game to master. You know what I mean? I'm trying to explain at 1am while sitting on the toilet and it's hard to concentrate. Personally an upgrade to the graphics department would e awesome, but as long as the physics remain similar or identical.

Ddavis wrote:Even with a new graphics engine, I don't see people dedicating themselves to figure out the difficulty of the game if they haven't already. If they're too hard headed to try it out because the graphics aren't up to par, then they'll find a way to not be interested if the physics finally exceed their expectations but then realize it's a difficult game to master. You know what I mean? I'm trying to explain at 1am while sitting on the toilet and it's hard to concentrate. Personally an upgrade to the graphics department would e awesome, but as long as the physics remain similar or identical.

more to it than visuals

JLV we don't care about fps drops from displacement maps, crank this baby to the next level!

thread smells like beef

swoodmx wrote:JLV we don't care about fps drops from displacement maps, crank this baby to the next level!

lol I am with you brother

Unreal has some great stuff coming this year - I mainly use it for all my work and its super powerful.

Go have a look at what they doing for the pc dakar game 2018 .
https://www.youtube.com/watch?v=2Ka2op5mc0o

This alone should show its power - Im not sure how many unreal users we have on here but i hope to be able to showcase some great work on it soon for its realism, main drawback is a great model to be able to place it in unreal with some exteriors and lighting and some great pbr materials.

If anyone is keen to have their setup displayd please give me a shout im still on the lookout for a free model to be able to do a realism project with.

MUHAHAHAHAHAHAHA