Yeah, that’s was my next question, how to scale and orient it correctly... since you can’t control the uvw coordinates of a submap
that is the awkward problem with it. you cant scale it correctly in normal situations. I have a plan though. details for that when (if) I complete the plan.
Pixel great work on the textures so how did you go about getting the detail in the grass like the gravel areas, offtrack areas and the one at the side of the track to dirt is really cool I'm trying to re create that my self, all they all just re uv mapped with that texture or what? I'm currently testing now with my own green grass to dead grass 512x4096 So i'll see how it goes and for the grass you did your self is that also just uv mapped ontop of the sat image?
you referring to the 3dgrass texture test images ? if so ...its the variation texture used over the whole area of grass. but it was never intended to be used this way. impossible to match with any main diffuse below on main scenery, which was the objective when I was testing it.
I think he is talking about this... I think the masking system is not used often enough, can be a bit overwhelming at first but can produce very organic results. I used @LilSKi / @luchian guide for photo quality roads here: http://assettocorsamods.net/threads/photo-quality-road-surface.834/ Same principle applies to any texture/shader you want to mask. Seems to be something we talk about more often than not... Hopefully soon I can help out and write a simple guide just for grass.
ahh ok, I take my satellite image and turn it into something else with various post processing techniques to remove detail and maintain a standard over the whole canvas, I also do a lot of indexed colour reducing, down to low number colours. and eventually getting down to 4 or 5 for the mask channel layers. this uses a lot of the shape info retained from the google images for where you want the grass and where you want the dry ground etc and it removes the photographic colour in favour of what I apply which are generally muted colours that will work with the colouring from the detail textures. iam using a really high pixel per metre ratio for external scenery. so iam getting about 10 pixels per metre, so the smallest size a clump of grass could be about 10 cm wide (size of 1 pixel) etc which kinda is seen in the real world. and thus detail is built at that level regards what I post process in photoshop. and because my diffuse and mask is made with each other in mind I can get that level of detail to appear naturally with what I end up with and the detail of the grass Is just an average detail texture , might even have used a kunos one for it. the important points are pixel per metre ratio and what the most detail can be and applying suitable colouring to the preserved shapes already within the satellite image. the skill is in retaining what info you want from the satellite and playing with it to get the desired result.
Aha everytime i come back and read on here my mind gets blown there's soo damn much to learn xD, Yes I was on about what John posted that variation which is really cool. And the other one i was on about was the asphalt to dirt one the 512x4096.
Please do these masks are confusing me aha, i keep reading it but I still don't understand what is actually happening a guide would be a huge help.
the mask allows you to place detail textures, that blend with your diffuse. where its green it places a chosen texture, and same for red and blue. so in the image of my track quoted earlier, you see multiple detail textures blending together, a light sand texture, darker sand texture and a green grass texture.
Damn I'm still way out of the loop I think i need a video I've read all the things on here but still can't understand how it's done aha. Yes currently I need some offtrack areas for Pats acres the thing I don't understand is as you put above where the red and green is how does it know said texture goes there? That's what I'm finding incredibly difficult to grasp at the minute I can use alphas on trees but that's basic but when it comes this complex I'm getting confused. At the moment I'm currently modelling dirt patches in with a seperate texture however it doesn't look great and this is why i would like to learn this masking method. So for an example I can take my google earth sat image I have here and use this for grass, Offtrack, dirt areas etc? *Picture below* There sure is a ton to grasp aha, but I think it's one of those things when I know how to do it it'll be fine it's just really confusing to your head around and understand it. To get my grass at this current time I'm literally just selecting polys on the google map and applying a tiling grass texture straight on to it.
ok I'm not sure how to explain this. the mask is just an aspect of the shader that runs on the same mapping as your diffuse that you have there. like sometimes we add normal map files into the chosen shader, the multilayer shader however allows us to do crazy shit like add more texture and "BLEND" them with the original diffuse. in the shader we add a mask which is a copy of the diffuse, but its not a copy at all. its changed into a red green and blue file, in the game where its green it blends a details textures with the diffuse. where its blue, wheres its red too. other than doing this for you I not sure how to explain it anymore. so its like masking tape, imagine you are painting a room in your house. but the window sill is made of wood, so you wouldn't want to paint over that. you put masking tape on it to stop it receiving paint. well the game is doing similar but its blending the diffuse (not blocking it out completely) with the detail channels and the mask is what controls that operation. the mask isn't read by the game engine like a texture. its a positional thing that tells the graphics, paint this here(red) paint this over here (green) and this here (blue), but its all derived from the original diffuse. when the game engine sees red green or blue, its interpreted by the game engine as texture and that texture is in the detail channels, and it simply blends them with the diffuse, so the multilayer blends detail textures with your main diffuse texture in the game. this means from far away you see the google map and when you get close it turns into the detail channel texture. the mask being a copy of the diffuse but changed to red green, blue placements patches is what controls it because its a copy, because it has the same mapping. send me your diffuse and ill send you a mask for it. yesterday on discord we had a good discussion about this. where I made a diffuse from a mask for one of lilski tracks. there was no diffuse before, just a blank file (as is a technique to using it) so I replicated what his track looks like in game by just using the mask and his detail channels in photoshop to make a diffuse... in your case though you are starting with a diffuse so its a bit different.
Yeah, wonderful . "Great" feature for ths site.. in some aspects it was a good idea. In others (like this one) it sucks. This is of course "imho" . What he needs is a clear (and simple) example on how to actually do it. How does one create the mask. With less words and more photos, most probably. I will do one, hopefully this weekend.
I'm slowly starting to understand, When Luchian replied to me on my thread. I thought all of it was done in photoshop aha
it is all done in photoshop ontop of your typically uv mapped diffuse model whatever. well the topic yesterday was really, how to make a texture like a ripped iracing texture. I think you should calm down a bit Luchian. I was about to post those images and set up a system to show him this properly. the problem isn't the fact there isn't info available. its a lack of understanding on a users part. everyone else got it. but now I'm to upset to continue tbh. lochness ive sent you the track in pm, have a look see for yourself.
Not confirming nor denying my state of calmness, my point is different: I do not agree with throwing the cat into user's garden before asking myself these questions first: 1/ does one want to learn ? 2/ if yes, how can I explain it in the simplest form I think we are sometimes tempted to forget how stuff can become overwhelming in the beginning. Especially since Lockeness showed he is putting some effort into this. But you don't have to agree with me, that's only my view.
discord is discord, you shouldn't get hung up on info on ideas being delivered over there, its likely advanced stuff not suitable here until the time to discuss it pops up here. I relate this to tv science programs and how they don't tell of the full table of physics discovery`s in a way the leyman`s terms can not describe. but behind the scenes they are incredible discoveries being made that will eventually be explainable to novices via tv science programs when the time is right or is required. Advanced modders need their place too so that's what it is. and through it will become even more information available that will naturally appear in such places as this as and when the level of teaching required is apparent.
