V-Ray normal map format? Why is this question being ignored?

[aaron_ross]

Hello,

Very simple question that needs a very simple answer. I have spent the last half hour trying to find the answer. Multiple users have asked the same question and there has been no response. For YEARS. Some of the posts are five years old with no response.

Simply put, what is the normal map format for VRay for Maya? OpenGL or DirectX?

My question is specific to the Substance Plugin for Maya, I need to know which normal format to choose.

Please just answer the question. It's not in the documentation.

Also, please stop putting circular definitions in the docs. This is so incredibly not helpful. Why bother putting in such useless non-information? It really gets under my skin.

https://docs.chaos.com/display/VMAYA...Normal+Mapping

Map Type – Determines how the Map parameter is interpreted.

Bump Map – Uses a bump map to determine the bump effect on the material.
Normal map in tangent space – Uses a tangent normal map to determine the bump effect on the material.
Normal map in object space – Uses an object space normal map to determine the bump effect on the material.
Normal map in screen space – Uses a screen space normal map to determine the bump effect on the material.
Normal map in world space – Uses a world space normal map to determine the bump effect on the material.

Really? That was the best you could do?

[intercard]

Hello !

The problem is not of VRay…it is of Maya or 3dsMax … for example the 3DsMax use DirectX and Maya use OpenGL but if you have bad result only you will need invert the green color of your normal map for change to DirectX or OpenGL … this is not a problem for me !

tips if you have the shadows invert in your render just need invert the green color +Y to -Y of your normal map…

but wait for the answer of some Chaos technician

the documentation about bump layer are okay for me too hehe

[Morbid Angel]

You need open_gl for maya. I agree, that the doc isn't helpful. I would like to see that each of the spaces described in more detail, why its needed and what it is used for.

[SonyBoy]

Yeah, it kinda irritates me as well. This occurs in a number of places throughout Vray's documentation, where it's almost as if the person tasked with writing the docs didn't know anything but had to write something. Simply describing the attribute's name again in a slightly reworded fashion is NOT documentation, and it's obvious that in many cases some practical/visual example is needed as well. I would say this was definitely true in this case.

[SonyBoy]

This is another example of this idiotic practice. Like we need to be told that this setting chooses the sky method, which is laughable to begin with, but then gives absolutely zero information about what the differences are between these options or why you should choose one over the other.

Hosek et al. – The VRaySky procedural texture is generated based on the Hosek et al. method.
Preetham et al. – The VRaySky procedural texture is generated based on the Preetham et al. method.
CIE Clear – The VRaySky procedural texture is generated based on the CIE method for a clear sky.
CIE Overcast – The VRaySky procedural texture is generated based on the CIE method for a cloudy sky.
PRG Clear Sky – The VRaySky procedural texture is generated based on the Improved method which has enhanced sunrise and

[Twist]

Hello,

Very simple question that needs a very simple answer. I have spent the last half hour trying to find the answer. Multiple users have asked the same question and there has been no response. For YEARS. Some of the posts are five years old with no response.

Simply put, what is the normal map format for VRay for Maya? OpenGL or DirectX?

My question is specific to the Substance Plugin for Maya, I need to know which normal format to choose.

Please just answer the question. It's not in the documentation.

Also, please stop putting circular definitions in the docs. This is so incredibly not helpful. Why bother putting in such useless non-information? It really gets under my skin.

https://docs.chaos.com/display/VMAYA...Normal+Mapping



Really? That was the best you could do?

Very good point. And I’m sure there are many more examples.

more tutorials would also be welcomed greatly

[yolov]

Hello,
Also, please stop putting circular definitions in the docs. This is so incredibly not helpful. Why bother putting in such useless non-information? It really gets under my skin.

https://docs.chaos.com/display/VMAYA...Normal+Mapping

We've been going through and improving many places in the docs. It's just too much information that takes lots (and lots) of time to update.
It helps to know what pages/topics need more information, so we'll try to work on that. Thanks for the pointer.

[yolov]

This is another example of this idiotic practice. Like we need to be told that this setting chooses the sky method, which is laughable to begin with, but then gives absolutely zero information about what the differences are between these options or why you should choose one over the other.

Hosek et al. – The VRaySky procedural texture is generated based on the Hosek et al. method.
Preetham et al. – The VRaySky procedural texture is generated based on the Preetham et al. method.
CIE Clear – The VRaySky procedural texture is generated based on the CIE method for a clear sky.
CIE Overcast – The VRaySky procedural texture is generated based on the CIE method for a cloudy sky.
PRG Clear Sky – The VRaySky procedural texture is generated based on the Improved method which has enhanced sunrise and

I'm not sure how well we can describe what each of the models looks like. We do have example renders a bit further down the page to illustrate the appearance of each model. Would it help to put in a longer description for each?

[yolov]

Hello,
Simply put, what is the normal map format for VRay for Maya? OpenGL or DirectX?

My question is specific to the Substance Plugin for Maya, I need to know which normal format to choose.

I'll try and fetch you some more info on that and add it to the docs.
In the meantime, see if that helps: https://substance3d.adobe.com/docume...196215180.html

Also a mention that we have a VRayNormalMap texture in Maya that can be used to swap channels for normal maps.

[SonyBoy]

I'm not sure how well we can describe what each of the models looks like. We do have example renders a bit further down the page to illustrate the appearance of each model. Would it help to put in a longer description for each?

Either that or remove this pointless section of the docs completely. I'm sure you'll agree it does nothing except waste the user's time.

[aaron_ross]

We've been going through and improving many places in the docs. It's just too much information that takes lots (and lots) of time to update.
It helps to know what pages/topics need more information, so we'll try to work on that. Thanks for the pointer.

Thanks for your reply. This is a problem across the entire creative software industry. In the early 2000s all of the experienced technical writers got laid off. In some cases (Autodesk), new people were hired who would work for less. In other cases (Adobe, apparently) the tactic was to solely rely on user forums to do the work for free.

So the statement "documentation is big and difficult" doesn't really address the issue. The underlying problem is the lack of respect for qualified tech writers. The underlying assumption is that teachers and writers are worthless, that developers can easily do those jobs in their spare time. But I am here to tell you that developers are the very dead last people who should be trusted with user-facing documentation. It takes a special talent, developed over years and decades, to translate technical language into something that is useful to users.

I'm not sure how well we can describe what each of the models looks like.

This is a prime example of what I am talking about. Developers, project managers, anyone directly involved with the creation of complex software, has enough on their mind. It is not their job to explain things. You need people who have devoted their careers to explaining things.

There is absolutely zero benefit in "describing what each of the models looks like". That is a complete waste of time and is utterly misunderstanding the entire purpose of documentation. If you have posted renderings of what the various models look like, there is no point in attempting to describe those images. We can see the images for ourselves. If you want to comment or call out specific visual phenomena that are not blindingly obvious, then, great. But describing an image is only useful for the visually impaired, which is not your audience.

What WOULD be helpful is defining what those various models ARE. You don't need to go into excessive detail, but if you're going to bother using up screen space, tell us something we don't know.

Bump maps and normal maps simulate fine surface detail without altering mesh geometry. Rendered pixels on surfaces, known as shading points, are pushed in different directions relative to the surface. This is a lighting effect that doesn't change the mesh shape, and so the silhouettes of objects are not affected.

The word "normal" has multiple meanings in 3D.

A surface normal is a line or vector extending perpendicular to each polygon face.

A vertex normal is a line or vector extending from each point on the mesh. At each polygon vertex, there's a vertex normal for each connected face. So at a point where four polygons connect, there are four vertex normals. Vertex normals can be oriented in any direction to determine the hardness or softness of edges, and how the surface responds to light.

A bump map or normal map is an image or pattern that simulates the lighting effect of a mesh vertex normal at every shading point, giving the illusion of surface relief.

Bump map: monochromatic, single-channel height map that deviates shading points in the direction of the polygon surface normals. The brightness or intensity of the map determines the amount of relief, with brighter colors corresponding to greater deviation from the surface.

Normal map: RGB map in which shading points are deviated in three directions, instead of only one direction as with a bump map. The intensity of each RGB color primary encodes the amount of surface deviation in a single direction in XYZ space.

Normal map in tangent space: The frame of reference for each pixel in a tangent-space normal map is the polygon surface normal. The XYZ axis tripod of each pixel in the normal map is positioned and oriented at a tangent to the corresponding polygon face.

Normal map in object space: All pixels in an object-space normal map share the same point of reference for XYZ shading deviation: the pivot point of the object.

Normal map in screen space: The frame of reference for each pixel in a screen-space normal map is the position in the picture plane of the rendered camera. The XYZ axis tripod of each pixel in the normal map is positioned at the corresponding X and Y coordinates of the picture plane, and oriented with the camera.

Normal map in world space: All pixels in a world-space normal map share the same point of reference for XYZ shading deviation: the world origin.

[yolov]

Either that or remove this pointless section of the docs completely. I'm sure you'll agree it does nothing except waste the user's time.

I agree. Thanks for chiming in.

[yolov]

What WOULD be helpful is defining what those various models ARE. You don't need to go into excessive detail, but if you're going to bother using up screen space, tell us something we don't know.

Best pointer so far, thanks!
Thanks for taking the time and effort, I truly appreciate it.
We had an internal discussion last week and we'll put some things in motion.
Let us work on that, share some progress and see if things are going in the right direction. We can very well use the same forum thread.

If you have other topics in mind, do share so we can adjust priorities as needed.

Thanks again!