Normal Maps, I'm pretty sure I didn't Imagine it :) but I might be wrong.

John.

Yes, although the current version for download won't let you.  An update to enable this should be available "within the week" according to Bagginsbill.

Here:
http://www.renderosity.com/mod/forumpro/showthread.php?thread_id=2778338&page=8#message_3493957

We plan on having a small hotfix available perhaps as early as today that will enable Normal Mapping. I'll post a new thread once this goes live.

Steve Cooper

That will be wonderful, thank you for the update.

Cheers people :) Thats brilliant cant wait to have a go at creating some in 3D Coat and seeing how poser handles it.

John

Good. The majority of my newer figures have normal maps but mostly DS3 users have been making use of them since they outnumbered the PoserPro folks, or at least it seems like it from our end. -Les

I'm starting to create and use normal maps in PoserPro itself.

Okay I don't want to sound like a newbie but what are "normal maps"?

Quick, dirty explanation: a bump map basically increases or decreases an amount of brightness based on the greyscale value of the map. This if fine if something is only viewed from one perspective and the bump is generated with that in mind, but it can be problematic when something starts to move because the differences the bump should be representing in the surface don't actually change with the lighting. Part of the problem is that bump is inherently a 2 channel thing: greyscale. Even if a color map gets shoved in a bump channel, it's still calculated based on brightness. With normal maps, each channel of an RGB image defines how light interacts with the surface of the mesh based on either vertex or polygon normals, so where bump might give some intial appearnce of detailing a surface, it's going to appear sort of strange at times, especially if you move the model around. Normal mapping, however, because it actually interacts with the lighting and helps redefine the calculation of a surface normals at render time, doesn't have those constraints. Just look at any really killer looking video game. The Gears of War stuff comes to mind immediately for me. You'll se some freakish amounts of detail, but those models are really low res, especially compared to what Poser users are used to. One of the primary methods of achieving that detail is normal mapping. I jumped for joy when it was adopted into Poser Pro, jumped even higher when it was adopted into DS3 and as far it being in a "regular" version of Poser... well, it's about high time. -Les

Quote - Just look at any really killer looking video game.

To me a really killer video game is pong

Thanks for the info on this.

The hotfix enabling normal maps in Poser 8 is now live.

does this hotfix fix  the problems with alternate_diffuse and specular?

 What problem, ice-boy? :) I can check if I know what to look for :)

I already said it doesn't fix it in another thread.

sixus1, I appreciate that you wrote all that to help people, but I must respectfully disagree. Bump maps and normal maps encode the same information, but in a different way. The normal map is faster/easier for a real-time renderer to decode, but the results are the same as with an equivalent bump map. Bump maps do react to light. In a software-only renderer like Poser which is spending tons more time on other things, the normal maps have no particular speed advantage. Furthermore, a bump map intensity/depth can be dialed at will, whereas a normal map will always produce the same intensity/depth - it's baked in. That's why it's faster - the effective change in the normal is hard-coded in a normal map. The effective change in the normal must be calculated in a bump map. The fact that it is calculated offers the opportunity to modify/alter the results of that calculation by adjusting the bump depth.

Uhm.. BB.. are you saying that bump maps are better than normal maps then? I'm still trying to figure out where those normal maps would be an advantage.. and now I'm even more confuzzled  :unsure:

Yes bump maps are better than normal maps, unless you don't have a bump map such as when you are trying to render game content.

 awaiting similar pics made with normal maps

I don't have any normal maps. I could make one but it would be difficult as I don't have any tools that do that easily.

I have asked repeatedly (this makes the seventh time) in this forum for anybody to make a pair of matching bump and normal maps, and show me renders where the normal maps are better.

Every tool that I know of that makes normal maps, does so by first starting from a bump map, and a user-specified depth. In other words, the normal map is completely generated from a bump map and a specified desired depth, which means that the bump map is more general, while the normal map is baked permanently at that depth.

The renders would be the same, but with the bump map you could change the depth on the fly direclty in Poser. With only a normal map, you'd have to go to the other tool and specify a new depth to create a new baked normal map. Poser is doing what those tools do, generating internally a normal map from a bump map and a desired depth.

BM.

A bump map does not specify changes in brightness. It specifies changes in height from the original un-bumpy polygon.

Here I used the bump map not only to alter the normals, but also to alter the colors and the specularity.

Again, you cannot do this with a normal map.

Nvidia has one that works as a photoshop-compatible plugin and it's pretty easy

http://developer.nvidia.com/object/photoshop_dds_plugins.html

I've gone so far beyond what normal maps can do now, it's not even a fair comparison anymore.

No contest. Click for full size.

 I really love that last one, BB! Most impressive.

So.. just to be sure I've understood it right: Unless you're animating or doing low rez game stuff, there's no need for normal maps? And since none of these two things are the obvious aim of Poser.. then why have they implemented normal maps at all? 

I feel I'm missing something here. It's like there should be some sort of advantage and I can't see it :)

I think it's so you can render using game textures - like if you're doing cinematics for a game, you use the figure and props with the in-game assets, which are likely to be normal maps.

One big reason for normal maps' popularity is NVidia's .DDS image format, which includes a channel for normal map data.
http://developer.nvidia.com/object/photoshop_dds_plugins.html

Like Bagginsbill I don't really see a an advantage to using normal maps over bump maps, as neither can actually disturb the profile of the geometry the way a displacement map can.  My fundamental technical understand is a lot weaker (like, zero) though.

Well, actually Nvidia has a particular demonstration of normal map usage that I don't think a bump map could duplicate.
http://developer.nvidia.com/object/melody_home.html

By demonstration, do you mean the rabbit screen shot? Or something else reachable from there?

I don't see anything about that rabbit that is a unique ability of normal maps. What about it seems unique?

That may be all well and good for Poser, but I have to say I've never been able to achieve the depth and detail from a bump map as I can with a normal map. In "other" render engines, that is.

Quote - By demonstration, do you mean the rabbit screen shot? Or something else reachable from there?

I don't see anything about that rabbit that is a unique ability of normal maps. What about it seems unique?

On second glance, that screenshot is misleading; on the left is an actual high-poly model.  So I'm with you, yeah.

Mike, maybe you can show a demonstration?

Quote -
Every tool that I know of that makes normal maps, does so by first starting from a bump map, and a user-specified depth. In other words, the normal map is completely generated from a bump map and a specified desired depth, which means that the bump map is more general, while the normal map is baked permanently at that depth.

Zbrush and Mudbox, in my understanding of it, don't do it like that.
They calculate the difference of the sculpted detail on the high res mesh from the unsculpted detail of the low res mesh and create a normal map of it based on the mesh's UVs, not from a grayscale image.

I couldn't claim to know mathematically how it works, but I use ZBrush and it's a fascinating thing to watch, as it generates the normal map - it literally examines every polygon and compares it to its high res counterpart.

Now I suppose there are cases where grayscale bump maps are used to generate normal maps for .dds files, but that's really a hack - most normal maps these days are generated directly from the geometry itself.

Quote -
Mike, maybe you can show a demonstration?

I can, yeah, but not right now.
But I'm really interested in Poser's implementation of normal maps and have brought the subject up a few times recently.
I'll bookmark this thread and throw something in here some time tomorrow, but it will be with Lightwave.

Mike I don't get what you're saying.

You said they calculate the difference of the sculpted detail on the high res mesh from the unsculpted detail of the low res mesh. What do you mean by "detail"?

In the renders I did above, the low res mesh was just one polygon. It doesn't get any more low res than that. What looks un-detailed to you there?

When you say "not from a grayscale image", now the words are being twisted. Forget the term grayscale image. Instead, think "topology delta matrix". Given a topology delta matrix, i.e. the difference between two meshes as measured along the surface normal vector from one to the other, you end up with exactly the data found in a bump map.

See what I mean? From that, you can construct all the normals.

Mike,
There are some here that use NormalMaps and some that don't., I personally don't believe there is a right or wrong answer to this question...it is, what it is.
You do or you don't.

If you are going to use NormalMaps I would encourage you to create and use them with TangentSpace.

Hope this helps

Quote - When you say "not from a grayscale image", now the words are being twisted. Forget the term grayscale image. Instead, think "topology delta matrix". Given a topology delta matrix, i.e. the difference between two meshes as measured along the surface normal vector from one to the other, you end up with exactly the data found in a bump map.

See what I mean? From that, you can construct all the normals.

There is a technical difference in that normal maps make use of three color channels (RGB) to disturb the normal on X, Y and Z axes, whereas bump maps only use one to disturb the normal on Y.  So what he's saying about grayscale vs. not-grayscale does matter, at least on paper.  However you're still stuck with not being able to disturb the actual profile of the geometry; both techniques just give you a shading effect.  Raytraced shadows will ignore bump maps for sure, and any place where the horizon of a model is visible, it will still be perfectly smooth.

Here's a good example I found of what MIGHT be possible:
http://www.game-artist.net/forums/support-tech-discussion/7756-bump-maps-vs-normal-maps.html

I don't know if Poser's shadows (raytraced or depthmapped) will accurately reproduce the kind of self-shadowing demonstrated there, I tend to doubt it.  I don't have anything normal mapped handy to test it with.

Quote - I personally don't believe there is a right or wrong answer to this question...it is, what it is.
You do or you don't.

Nonsense, there's either a real technical advantage, or using normal maps is quite a bit more work for no purpose (at least if you're really doing a proper x/y/z normal map).

Bagginsbill you'd like this thread, there is math in it.
http://boards.polycount.net/showthread.php?t=36160