ZBrushCentral

Finished work in ZBrush, can't get a good render from anything :(

Hi, I recently got ZBrush, and over the last week I have been playing around in it. I made a little test to see how well it output when rendering. I plan to sculpt in ZBrush, get a very high quality render whatever it takes, and then use photoshop for major post production work. Since I just got ZBrush, I decided to make a quick model, and try to render it in high quality. Unfortunately that’s as far as I got :cry:.

Since I don’t like the default way that ZBrush renders (I’m probably doing something wrong) I decided that I would try in Maya or C4D, but I never got that to look very good either. I read most tutorials, but my renders always ended up looking crappy. At the moment, the default view in ZBrush looks the best haha.

These are exports of my model in Zbrush without eyes or teeth.

http://yfrog.com/3o001t
http://yfrog.com/0q002t

I don’t need all these options and settings and whistles and bells when rendering. All I need is a way I can sculpt creatures and details in ZBrush, and then get a nice looking render, where I can setup light sources and materials (something that I find hard to fake in photoshop). Going through Maya just hurt my head. I like C4D because it has a much easier UI and I like the material options much better.

I just can’t get detail from my Displacement maps in Maya or C4D, I haven’t tried 3ds Max yet, does anyone know if that is easier? For one thing, their are actual in-depth tutorials on 3ds Max which I couldn’t find on Maya or let alone C4D.

Is Vray any friendlier to use? Basically what I am trying to achieve is a look like this:

http://www.zbrushcentral.com/zbc/showthread.php?t=038589

You will probably need more info as to what the renders actually look like in Maya and C4D, but before I go into that deep stuff, I just wanted to know if 3ds Max is any easier (or if the tutorials are more user friendly). And anything else that might help me achieve a nice render.

Thanks heaps, Max

What you need to do is buy a book on lighting…im guessing. Without the images that you’re getting from Maya, Max, C4D I can’t really tell you what the problem is.

Mental Ray (MR) in my opinion is a better option than Vray (Vr).

Anyway, since you’re doing still images then you should just decimate the model (new Z3.1 plugin). Then take the model into whatever package you end up doing the renders in.

You didn’t say if you have a background in 3D, or photography or anything else to let us know if you actually know how to light a model, real world or otherwise. There is a science/art to lighting something…be it a character, or a tin can (think pepsi).

Anyway, what I am trying to say is, that there isn’t really a “make pretty” button in any application. While there is HDRI…damn near the make pretty button, you still need to have rim, key and fills where needed…or light boxes. I mean, Photographers take years of classes to learn good lighting. In full lenght productions there is a team of people that light scenes and whatnot.

You did mention that you want to do a ton of post pro. Does that mean that you want to add color? blend different lighting passes? Add extra specularity to the object? Or just simply auto level, balance, and color?

Anyway, what I am saying is, you’re asking waaayyy too broad of a question. You basically asked, “How does a car work?”

Not much help, but with more information someone can break down what you want to get from the renders, and then tell you how to achieve it.

As goast sayd there is no make pretty button.
But hdris as he sayd are very very close.
With a studio enviroment hdri you wont even have to set any lights.
The thing is tho that hdri use isnt that easy in most programms either.

If you really want a make pretty with hdri render you will have to go
unbiased. With Maxwell or fryrender it is really hard to get renders that arent
impressive. The downside is that you will have enourmous render times.
Without a renderfarm backing you up rendering an animation is close to
impossible.

The following image i rendered in fryrender using standard materials and an
studio enviroment 3 point lighning hdri.(never mind the screen its the placeholder version)
I had to render 5 hours to get to this little grain on the image on my
core i7 920 and it couldve easily needed a few more. The thing is that u cant use it for animations in a
decent timespan and its another programm u have to get which is not
of so much use if ur not having archviz jobs that need it.
Best is to get deep into your renderer of choice and learn how to make
it look good. Its not that much of a secret. Photometric lights area shadows
ambient occlusion and global ilumination mostly do the trick. :slight_smile: :+1:

[medizin colo2r.jpg](javascript:zb_insimg(‘144250’,‘medizin%20colo2r.jpg’,1,0))

Thanks for the replies guys :slight_smile:

Ill try and answer all the questions, and clarify my vague explanation. Firstly, when I said i needed good lighting, I didn’t mean insane :p. Before I worry about lighting, i need to get the displacement maps to work. Secondly, I am not doing animation, and render time is not an issue for me. My background is in 2D, digital painting in Photoshop etc, so I am quite new to the whole 3D department :o. I am hoping to use the 3D software, specifically ZBrush to create photo realistic creatures, that can then be brought into Photoshop, where I can work on the atmosphere, colour adjustments, and any cropping/photomanipulation. The problem is rendering. ZBrush is one bank, photoshop the other. With rendering being the river I am finding hard to cross :cry:.

I’ll start from the beginning and explain my steps and problems etc.

I started my model in ZBrush, not in Maya/3ds (making a lowpoly model) etc like most tutorials show. I did all my sculpting and detailing, and added eyes and teeth subtools (although I deleted them to make it easier to export etc). I then went through many tutorials and looked into how to make a displacement map and export the tool. I used this tutorial all the way up to the point where he finishes in ZBrush, as this was the most in depth tutorial i found on displacement mapping and exporting tools. So now I have my displacement map.tiff and my tool.obj. Now comes the problem. I have to choose which to render it in, Maya or C4D.

In both, I import the low Sub-D, and add a material that uses the disp map. For Maya i change the alpha offset and gain, and for C4D I play with the strength etc. I followed the tutorials for Maya, but a lot didn’t make sense because they all involved starting in Maya, and creating the UVs in Maya.

So I have set it all up in Maya/C4D and render it, but it just looks like the low Sub-D version, with a few dents/bumps here and there. One thing that I noticed that might be bad is that the disp map from Zbrush looks almost fully gray. I flipped it vertically etc, but still no luck on the render.

One last question, which doesn’t relate to rendering. Is there any way I can sculpt in ZBrush, but paint in Mudbox. I like painting in MudBox better, and was wondering if there was a way to get a detailed model from ZBrush to Mudbox, then create the disp map in Mudbox (easier?), and then render in Maya etc. (maybe some way using decimation plugin?)

I just did a render in C4D and it looks like it’s on the right track, the disp map is in the right area, but something is wrong. Is C4D’s render engine any good with this sort of stuff, if i ever get it to work?

Anyway I hope I have helped a little, and I will post the render from C4D (Maya isn’t being nice atm) as well as the tool and disp map.

A NOTE: The tool that’s included has been decimated at a high poly using the decimation plugin, as to reduce file size. So it is not the EXACT same as the tool used to export the obj. However the obj IS the one I have been using, and the disp map included is for it.

Rendered in C4D with these settings:

Imported obj, and edited default material.
Ticked the Displacement box, and selected my disp map, then ticked Sub-D and entered 4, then rendered.

Result: http://img30.imageshack.us/i/head001.tif/
Comparison:http://img26.yfrog.com/i/002.tif/
As you can see, its low quality and there are unusual bumps and crevices, yet the map looks to be in the right spot :confused:

Tool: http://www.mediafire.com/?2nxmajikmtm
Object: http://www.mediafire.com/?nrjzqtondyj
Disp Map: http://www.mediafire.com/?ne0xjzk1mjh

Thanks, Max

Thanks for the replies guys

Ill try and answer all the questions, and clarify my vague explanation. Firstly, when I said i needed good lighting, I didn’t mean insane . Before I worry about lighting, i need to get the displacement maps to work. Secondly, I am not doing animation, and render time is not an issue for me. My background is in 2D, digital painting in Photoshop etc, so I am quite new to the whole 3D department . I am hoping to use the 3D software, specifically ZBrush to create photo realistic creatures, that can then be brought into Photoshop, where I can work on the atmosphere, colour adjustments, and any cropping/photomanipulation. The problem is rendering. ZBrush is one bank, photoshop the other. With rendering being the river I am finding hard to cross .

I’ll start from the beginning and explain my steps and problems etc.

I started my model in ZBrush, not in Maya/3ds (making a lowpoly model) etc like most tutorials show. I did all my sculpting and detailing, and added eyes and teeth subtools (although I deleted them to make it easier to export etc). I then went through many tutorials and looked into how to make a displacement map and export the tool. I used this tutorial all the way up to the point where he finishes in ZBrush, as this was the most in depth tutorial i found on displacement mapping and exporting tools. So now I have my displacement map.tiff and my tool.obj. Now comes the problem. I have to choose which to render it in, Maya or C4D.

In both, I import the low Sub-D, and add a material that uses the disp map. For Maya i change the alpha offset and gain, and for C4D I play with the strength etc. I followed the tutorials for Maya, but a lot didn’t make sense because they all involved starting in Maya, and creating the UVs in Maya.

So I have set it all up in Maya/C4D and render it, but it just looks like the low Sub-D version, with a few dents/bumps here and there. One thing that I noticed that might be bad is that the disp map from Zbrush looks almost fully gray. I flipped it vertically etc, but still no luck on the render.

One last question, which doesn’t relate to rendering. Is there any way I can sculpt in ZBrush, but paint in Mudbox. I like painting in MudBox better, and was wondering if there was a way to get a detailed model from ZBrush to Mudbox, then create the disp map in Mudbox (easier?), and then render in Maya etc. (maybe some way using decimation plugin?)

I just did a render in C4D and it looks like it’s on the right track, the disp map is in the right area, but something is wrong. Is C4D’s render engine any good with this sort of stuff, if i ever get it to work?

Anyway I hope I have helped a little, and I will post the render from C4D (Maya isn’t being nice atm) as well as the tool and disp map.

A NOTE: The tool that’s included has been decimated at a high poly using the decimation plugin, as to reduce file size. So it is not the EXACT same as the tool used to export the obj. However the obj IS the one I have been using, and the disp map included is for it.

Rendered in C4D with these settings:

Imported obj, and edited default material.
Ticked the Displacement box, and selected my disp map, then ticked Sub-D and entered 4, then rendered.

Result: http://img30.imageshack.us/i/head001.tif/
Comparison:http://img26.yfrog.com/i/002.tif/
As you can see, its low quality and there are unusual bumps and crevices, yet the map looks to be in the right spot

Tool: http://www.mediafire.com/?2nxmajikmtm
Object: http://www.mediafire.com/?nrjzqtondyj
Disp Map: http://www.mediafire.com/?ne0xjzk1mjh

Thanks, Max

I downloaded the files but i cant really help you.
Had to rename the tool with an ztl extension and its only one level
and has no uvs.
A plain gray disp map isnt right are you sure your model had uvs?

This is just not how it works u need multiple subd levels and uvs.
Sculpting highres and then making the low poly by decimating it
and exporting a disp map that has no information because it has no uvs wont work.
I suggest u have some looks at tutorials covering the workflow of low to high
back to low again

Okay, I’ve been testing C4D with the test human head, and I got it perfect, aced the render. Then I imported my model, applied the displacement, and hit render. It worked great. I found out what my problem was :o; I kept forgetting to apply a HyperNURB.

So I finally got it pretty nice looking in C4D, but then I hit render to Picture Viewer and BAM! “Not enough memory”. I have never run into this error before, does it occur only in C4D (will Maya/3DS Max do the same)?

Also, how was Pixolator (1st post http://www.zbrushcentral.com/zbc/showthread.php?t=20310&page=1&pp=15) able to get this quality in ZBrush? This is better quality than I have seen in anything from MR/Vr or PRman. :frowning:

And sorry Disco Stu for my idiocy and little knowledge :cry:, but thanks heaps for your help :).

Thanks, Max

No problem man.
Whats the polycount in the scene that cant render in the picture viewer?
To find out the polycount rightclick in the object manager in c4d
and presss scene information. The number thats right is the second
one in brackets(its the polycount with hypernurb included).
Maybe maya or 3dsmax are gonna handle it better but it still will be slow.
The first post isnt all that exiting its not bad but have a look at this.
http://www.youtube.com/watch?v=O7-IYE4B6lE
Assasins Creed 2 trailer and u wont believe it :smiley: be sure to watch it in hd :slight_smile:

Oh yay! It worked, I found that ticking round geometry speeds up the render and is less mem instensive. Although I was still getting OOM errors, so I got rid of the HyperNURB (found it made no diff with round geo ticked). I was then able to render to picture viewer /cheer. It also helped that I went back into zbrush and exported an obj at sub-d lvl 2 rather than 3 :confused:.

It still has some probs, like the spikes/ridges look a little glitchy, but for the most part it’s something I can work with. Intense 4 days, but Zbrush is sweet.

http://img14.imageshack.us/i/finallywoot.tif/

And the poly count on that obj in C4D was 6144 without the HyperNURB. With HyperNURB at level 4 it was 1579008. :confused:

Checked out the trailer, and wow, fantastic rendering and CG. Any of it done in ZBrush? Can’t wait for AC2! :D:D

Thanks again, Max