Another one

https://www.allegorithmic.com/products/substance-painter

Dave

This seems to be a rather good programm, but i understand that i really should learn how to make a uv-map first if i want to create such textures.

Yes you will allways need a uv map to work from for any sort of texturing .

Dave

Even if you do your texturing in just Gimp or Photoshop, you need to have good UV maps.

Unfamiliar made some really easy to follow modelling tutorials, I would highly recommend them.

In those tuts he shows the basics of making a model, unwrapping (UV), and texturing. It is much easier than it looks, it is only difficult because we are unfamiliar with the process at first - pun intended

Reliquia....

Thanks reliquia

Using 3dsmax, i have no scaling problem when i import models made in Sketchup and export them into Game Guru.
With Blender i do and i don't know how to set up a template so that the scaling matches with Sketchup and hence Game Guru.
To be able to make use of only one texture and hence have a working -N and -S i resorted to using a texture atlas.

When i create characters in Blender i use an .obj copy of one of the default characters (masked soldier) to measure it against, i just import it, before i add a rig to the character, stand it next to my character and scale it up. When i scale i always press shift+c to take the cursor to the origin and set it to scale from the 3d cursor, then scale in edit mode, the reason i don't scale in object mode is because if you forget to set your loc/rot/scale game engines will scale it to 1:1 instead of the scale in the model, so it will come out the wrong size.

Will there be an illumination, or emissive map too?

An illumination map is something I'd also like to see. Or at least an equivalent. Only issue I can see here, is we will soon run out of memory with all these added texture types if we want PBR based games Were up to 7, 9 with the extra two mentioned above. Nine textures would near triple the memory cost of normal objects (if they are all of the same size).

Regardless of memory concerns I would say the PBR is looking good so far. It will be nice to see it in action.

It would only be one or the other, they're both the same thing, just different naming conventions. To be honest we could get by without the _ao map as we have SAO in the engine, and could probably get by without the height map.

Yeah, will be interesting to see the damage to FPS that's for sure.

As for emissive maps - they are a requirement imo - PBR will loose it's appeal if we can't light models up

And if we are adding this awesomeness then let's not do a half-baked job. I am sure Lee wants to get this right the first time round.


Reliquia....

I think it would be good if Lee includes them all,you only have to use the ones you want on your model.

Dave

Lee can you maybe show us how will the rocks of death valley look with the new dx11 upgrade if possible.

Here you go

https://forum.game-guru.com/thread/218545

Bottom few pics .

Dave

I'm really looking forward to being able to use PBR assets, i've had substance painter for quite some time, but never really learnt how to use it because it doesn't export simple DNS texture sets and it was simply easier to texture those sets in Blender than it is to work with painter and convert it's exported textures, but now we are getting PBR I've spent every moment watching substance painter tutorials and playing around with PBR textures, if we can get eventually even only half of the quality of the results i'm seeing in painter then it's going to be a real game changer for gameguru.

The thing i'm most looking forward to being able to do is to take a high poly model and project its geometry via textures onto a low poly model. Yes, i know we can already do this to an extent, but without proper height map information and normals working together it doesn't quite project through well enough, but with PBR we should be able to get some pretty sharp projections.

@Belidos - you can export DNS texture sets from Substance Painter, but the process is a lot more complicated than it should be.

Dave are u sure the rocks are from the death valley part.?

Sorry no,they are not i think.they would have to be re textured to look like the ones Lee put up (my bad)

Dave

I made a rusty old barrel to test in GG ,Looks ok but not poping out yet. (Might be better when more maps are added to the import )

Textured in Quixil

In Quixil




In GG




Dave

Looks good

. So what you say is all the original stock entities will not look better with dx11, they will have to be retextured. So if that is true then only the clever users that can texture would then benifid with the dx11. ? True or not?
Sory for bad spelling.

Exactly that, DX11 itself is not a graphics improvement, changing to DX11is more a change to the way graphics is handled, and it opens up access to being able to use other things that will improve graphics like PBR, and better shaders etc.

PBR is the graphics improvement, however you will only see a difference if the models you use are PBR models.

As many people have said many times all along, DX11 won't magically make the graphics better, it's not the engine that's bad it's the models (they're not bad, but they're not top quality), 50% of good graphics is placement, 40% is quality of models, and the last 10% is the engine itself.

That's why it's been voted down so often in the past, because unless other stuff is added alongside it DX11 really doesn't change all that much, fortunately we now have Lee working away at PBR alongside DX11 so it is worthwhile, but without better models there won't be much difference at all.

Even with triple A games, the stunning graphics isn't always because of the directX version being used, in fact i've seen some OpenGL games that look better than some AAA drirectx games, a huge chunk of it is the models and textures you use, and how you place them. Just look at Funcoms The Secret World, you can switch between DirectX9 and DirectX11, and there's very little difference, most of the difference is in the performance and the shaders used, but because they have great placement of models, and great models and textures, it's not so noticeable.

this is the first proper way to convert traditional textures to pbr
so all old textured models are not lost,

https://www.youtube.com/watch?v=cwBstFgTWAo&t=2875s

https://www.youtube.com/watch?v=kskdxhrM1FY

INHO, If DX11 will load the level faster and process the models faster - that's what we need. And beauty is not so important, if you focus on the gameplay.

Besides faster loading times and higher frame rate we'll also have PBR terrain and sky, EBE with PBR, so there is a visual improvement to be expected.
Isn't it also the case that we'll have unlimited dynamic lighting with DX11?

It's possible, within limits of course, but not sure if that will be part of this update. It's not finished yet, but not seen anything concrete about the lighting system so far. Lee did mention the light count could go up dramatically when he first played with DX11; but as far as I'm aware, we will still be limited at least temporarily, with 4 max lights. Let's hope I'm wrong

Yeah, more lights, mooooor lights

Reliquia....

In that case, as we say in dutch, 'wat in het vat zit verzuurt niet'.
Meaning, maybe not with this update, but the possibilities which DX11 offers, will be extracted.
I wonder if this means the elimination of using the lightmapper if we stay within the limit.

---------------------

I've read that the output of certain PBR maps in Substance is greyscale, would that be a problem for GG or is it possible to change a setting in order to export as RGB?

That should read: "the output of certain PBR maps in Substance is greyscale."

That's how PBR works. Most information maps such as height, metalic, glossiness, specular, emmisive etc. are generally greyscale in all PBR.

So, no it won't be a problem for GameGuru because that s exactly what GameGuru will be looking for.

Ahh, thanks Belidos

@ lee great screen shots
I would run that tree through the new simplygon and use the re mesh option
or just use the 3 stage lod basic,

https://www.simplygon.com/

It's not my tree Of course, if you were making a game and bought the tree from CubeBrush like I did, you would be free to use it in your game as normal, but any asset I provide inside GameGuru would need to have a sub-license term, and most artists don't like selling their model once and seeing it used thousands of times For this reason, I am using my own TGC artist to create a PBR tree for the final update, and I am pretty sure he will ensure we use the lowest number of polygons for the maximum visual impact (complete with the essential LOD levels). Glad you like the screenshot, first of many now the shader is mostly complete!

@lee when can we expect the pbr shaders to
come into gg beta, and a bit of documentation,
a lot of forum members and testers talking guesses
on figuring it out, would love to see some shiny metal
surfaces in the screen shots, is there going to be any
specific software for the workflow, is the shader going
to compatible with most pbr workflows, looking
forward to trying these out.

Lee what about displacements maps?
this example, you get the same result in UE 4, what about GG DX11, pbr etc
i mean REAL DISPLACE MAPS (parralax mapping) like this example:

https://www.rd-textures.com/gallery/#10274

others examples of displace maps

https://www.rd-textures.com/gallery/

and here directx 11 parallax mapping
https://www.youtube.com/watch?v=FWwUp7Oz1Lw

It must be just me and the fact I saw the Computer age from the start to present day but the point comes when it no longer feels like a game ... More like FMV and your just pushing your way through with a joystick ... Perhaps I'm getting to old for this kind of thing.

I may be wrong but isn't that just height maps

Explanation

http://polycount.com/discussion/105750/what-are-displacement-maps
(Map this map that,where will it all end )
Dave

from my reading and learning exercises this is how I perceive Height/Dissplacement I may be wrong but basically Displacement Mapping: Is a technique that aims to render bumps as true geometry, in a very fine mesh. Unlike bump mapping, parallax, and relief mapping which tries to "fake" bumps using normal maps, Displacement mapping actually displaces the surface, creating triangles between the texels. .

Height Mapping: is the same thing, but it's usually used in the context where a displacement map (also called height map) is applied on a terrain where the value are only used to modify the vertex height.

It can be implemented on the CPU or the GPU.

One common CPU approach is to read height or displacement values from a height/displacement map(texture) where each texel directly maps to one vertex. Where each texel encodes a height/displacement value. This is then applied directly to the geometry by displacing each vertex using the looked up value in a unique direction.

Choosing the direction can be in the Up direction (usually in case of terrains), which results in modifying the vertex Y value, or could be in the direction of the face normal usually used on objects other than terrains.

A GPU alternative is to use the vertex texture fetch feature (introduced in Shader Model 3.0) to have a the terrain mesh modified by accessing a displacement/height map The height retrieved from the texture is used by the vertex shading program to modify the vertex's location.

Other uses for Using a texture allows for faster manipulation of data for wave simulations and other animations to apply to the mesh.

Regarding adaptive tessellation:

One draw back of displacement mapping is that for large terrains you need a lot of polygons and vertices to model a detailed terrain which makes displacement maping somehow inefficient for large terrains.

This is where adaptive tessellation and level of detail techniques come to play to make displacement mapping more feasible, especially with the advancement of the GPUs and introducing geometry shaders, performing tessellation on the fly with this advancement has become the dominant technique. It is simple to program and on newer GPUs and has few drawbacks.

Other techniques like relief and bump mapping offer additional realism at a generally reasonable cost, but the fact that the base surface is unperturbed makes collision detection, and therefore object interaction, more challenging.

As a conclusion Displacement mapping and adaptive tessellation brings superior detail and quality with less draw backs at a feasible performance cost,
but only with dedicated DX11 cards and above.

Displacement maps are a type of height map, basically a normal map uses textures to fake details on low poly models, whereas a displacement map uses height details to actually push vertices in or out along the normal on higher poly models. Brilliant for rendering 3d model scenes for video etc. but can be a little resource hungry for games because you need higher poly models, games that use them tend to have a few prominent pieces strategically located to supplement normal low poly pbr models (usually terrain).

We have _height in our pbr already.

Yeah, I tend to agree with that (perhaps me be too old as well) I don't mind realism to a point, but I just like a game to take me somewhere unreal!

Reliquia....

well a world can be very realistic in therm of 3d/texturing/lighting, but unrealistic in an artistic way, imagine if GG can have the same results, it can be more profitable for both sides.

@imothep85, yeah, I agree there,

Not against having the ability to create realism or anything, just stating a preference is all.

I think the more scope we have to create what we want as individual developers, will make GG an even better engine. PBR is definitely going to add much to Game Guru's appeal.

Reliquia....

I was not planning to do true displacement (i.e. vertex/geometry shader texture fetching) as that definitely has a performance hit unless you GPU is pretty idle. My first attempt would be a form of parallax mapping, and only for models at close range, and only those chose by the level designer in places that would benefit from it. Having it for nearby terrain is a good option as the perspective lends itself very well to the effect, but adding it to every model presents its own problems, not just performance but as mentioned, collision and visual artifacts which destroys the illusion when viewed at an extreme angle. I am ever conscious that as much as the visuals are important, and need to have a 2015+ feel, performance was also a critical concern for many new GameGuru users (and existing ones) and the first update should attempt to address both. If then, displacement mapping becomes a big thing, we can look at it separately and weigh the costs of adding it in comparison to the version we aim to ship before Christmas (i.e. better visuals, faster game and quicker loading).

is the parallax mapping going into the alpha channel of the normal map.
as a grayscale heightmap. is this same as virtual displacement mapping.
or do you plan something else

are we looking at the effect in link bellow.


https://www.youtube.com/watch?v=c4RAG_Kn9ow

Not the most helpful demonstration of Parallax mapping You need more than a dual-core CPU to mitigate the performance cost of parallax which turns every single pixel into a sequence of texture lookups as you cast a ray through your height data stored in an extra texture channel. The height data will be provided as a separate texture (_height) by default but I am looking at techniques which support collapsing this greyscale data into existing textures to reduce the impact of swapping in lots of textures during the shading process.

@ imothep85. Dream on. If you want visuals like that you will need to use Unreal. You may achieve similar results in GG but not with displacement maps. GG is now a $20 game making tool, released in 2013, that is designed to make things easier for people. Unreal is a professional product that has been around since the mid 90's. It's free now but wasn't always so and guess what? It was a tad more than $20. It still is if you want all the bells and whistles. It's been the industry leader for some time now, despite other engines being released since.

Unreal, however is not particularly easy to use; sure you can get some stuff going to a degree fairly fast but for any kind of real game it will take time and effort or money and copypaste coding. There are license issues as well, although if you earned the cash they talk about it wouldn't be much of a hardship imo. I've tried it a few times and can safely say the learning curve is high and in honesty I just don't have the patience to attempt to delve into the actual coding side of it.

That's the main reason you see so many talented people using GG. Any other engine I have used such as Unreal or Unity are just a lot more difficult to use. For people doing this for fun GG is just easier, the coding is far simpler than trying to get into Unreal Blueprints or C++. In fact many people use GG and expect NOT to have to code at all, which is a big expectation in itself. Especially when everyone wants to make the next Battlefield clone.

I'm not expecting miracles with this update and really you should not get your hopes too high by watching Unreal demo's and expecting anything like Unreal can do. The DX11 update should be a big upgrade, but it won't change GG into Unreal I'm afraid.

@DVader, I agree with you. If someone is developing a game (and he is alone), the most reasonable idea a focus on the gameplay. Never reach AAA level graphics if you do not have a team and and considerable money.

It means means that only a good idea and successful gameplay can help. In such conditions, the main thing is the speed of the engine, but not beauty, imho. (I just want to express my thoughts, sorry if i said the trivial things ).

I like that the Game-Guru is able to do this. This is the best 3D-engine for this. .
I tried to try the Unreal and the Unity, but they are much more complicated in many aspects, and I rejected them.

@ Tauren. I agree. We need a fast engine to keep game play acceptable. Cosmetics come second to this for me, in Game Guru and with games I play. If a game is too slow, pretty graphics will not help. My above point was really just pointing out that although we will be seeing an improvement in graphics and speed (hopefully), we can't expect Unreal performance or features. The image above is unrealistic as to what GG will be able to achieve. It just shows how powerful Unreal is to be honest. We have to expect a little less for the privilege of using an engine that takes away most of the coding aspects needed when making games.

I meant no offence, but you have to be realistic here. Unreal has been around for over 20 years. It was a professional tool for most of those and cost a fortune. It has had plenty of money and time to develop into what it is today and the staff needed to update and work on new features will be vast. Game Guru has been around for a few years and is the work of mostly one man. Considering the drain on FPS with basic terrains already, having detail like this in them would cripple GG immediately. If you want top notch visuals like this without resorting to models, Unreal is the only option I know of.

@ DVader, I again agree with you.
Personally, I would be satisfied with the graphics for 2008-2010
Moreover, even then there were such wonderful games as Stalker and Witcher 1. Although, they even came out earlier, in 2007

And about Unreal - it's beautiful and powerful engine, but harder than GameGuru - and I'll just wait a year or a slightly more.
I think, the Game-Guru is is already close to become strong and truly flexible for single developers. A little more control over the game mechanics in LUA scripts and everything will be fine, imho.

Honestly, if you have the drive, determination and ability and are new to making games. You would probably be as well to try Unreal out. It's a lot harder to use, but I wouldn't say it's impossible for a lone dev to use, just not to exploit to it's full potential. Most people however will return to GG quick smart once they realise Unreal is nowhere close as easy to use. If you can wrap your head around blueprints or C++ coding, I'm sure Unreal is less daunting. For most of us that is not the case.

If you can make a half decent GG game then, if your willing to spend the time and have no issues with Unreal's programming, you will probably be able to make a decent Unreal game. From my limited use of it, it's almost as easy to create a map, just the logic side of things does not suit me at all I wish it did, I made a great little driving game that worked great, it even had a spline based road, which would be a great addition to GG I add. It was fine apart from if you got rolled over, it would sit there forever. I could not see for the life of me how to rectify this. If I had the patience I would probably have found out, but again, I do this for fun really. It would be fantastic to release a game, but I would not be happy with what I can achieve in GG to do this. I wouldn't have a clue about getting an Unreal level into a compiled game, I've never tried and until I knew about getting the game to work as I want, wouldn't want to.

Horses for courses. If I put as much time into Unreal as I have GG I'd probably be halfway to making a great game. Truth be told, I haven't and I probably won't. Blame GG I am getting too old to learn new stuff and GG works well for me It's been fun playing with it, whereas Unreal is impressive but not fun, at least not till you get something working ;p

Game Guru is fun. Mostly. Although I'm never going to make an awesome driving game with it, which is sad but true. Even thinking about this makes me consider getting Unreal again. Although I know it will lead to the same result