ZBrushCentral

just painting a bump map ...

The effect I showed in the above pictures are the result of the render engine, in this case Lightwave, and cannot be shown in the zscript. But the bump maps themselves are quite simple to make.

It is best if you create special alphas for the various shapes but you can get similar results by using the default zsbrush alphas and the various settings in the alpha palette.

I recorded 3 scripts. The first shows how to set up your object for bump mapping and is annotated. The other two are just some bump mapping examples.

I would have used my own alphas for this as it would be quicker, but the filesizes would be too large to upload. The scripts were recorded rather quickly so please excuse any holes in the explanation.

Anyway, I hope the scripts are of some use.

thank you so much for this, lots of things started to make sense now.
Stupid questions tho: You add the bump using color instead of zadd or zsub, two diff. shades of gray. What is the difference between this and zadd/zsub? And wuth the gray, should i just set two colors and simply just “switchcolor” every time i need to reverse it?
Also in the 3rd script, you not using PM to add bump, you just switched tools and started paly around with the alpha, i should use PM if i want the sphere to stay as 3D tool when applying alpha bump, right?

Thank you again, this is helping alot.

All three scripts use Projection Master to add the greyscale colors. As soon as I switch tools I have dropped my 3D tool(object) with Projection Master. I did not write notes for the last 2 scripts, which is needed to indicate I dropped/picked up with Projection Master.

The colors I used range from RGB 1,1,1 to RGB 255,255,255. I simply picked light values to raise the bump and low values to lower the bump. You can start with a texture map of RGB 0,0,0 and only paint raised areas if you wish. How your render engine uses the texture map is down to its abilities and what settings you apply. Try Marcus_Civis’ Zswatch 3.1 script for changing colors.

Why use RGB only and not Zadd/Zsub? There are a couple of pros and cons.

  1. Drawing detail as a texture map, in this case a bump map, is nearly always going to be easier on your computer. You do not need to subdivide your object as you are simply painting the detail, not modeling/displacing the geometry to create detail.

  2. To get optimal detail from Zadd/Zsub your object needs to be sufficiently subdivided. Increased polygon density in detail heavy areas is a must. But sometimes you just cannot subdivide enough and then a bump map comes in handy for that extra detail. Back in Zbrush 1.55b we were limited to around 500.000 polygons so a bump map came in handy for the finer details. See SkyCastle’s Hand for an example.

  3. Painting a bump map in Zbrush will result in a 8 bit texture. Generating a displament map from modeled detail (Zadd/Zsub) results in a 16 bit texture. If you need the extra fidelity and your render engine supports it a displacement map used as a bump map is preferable.

  4. A bump map is ok but a normal map is better. A normal map is calculated from the same geometry as a displacement map (Zadd/Zsub).

Basically it all depends on your needs.

Also check Aurick’s post in this thread.

Perhaps another related thread to checkout is Grub’s tutorial >> HERE <<

Sven

Thank you TV, you have cleared out the mass(some of it) in my head. this is exactly what i need, cause my PC cant handle tools with more then 1mil polyes and even then its slow, freezing up and might crash.

Amount of polygons on the model still should be pretty high, right? Enough for the model to look good enough without the details, correct?

You need as much geometry as the job calls for:) Where is the model going to end up when finished? If you are going to stay within Zbrush there are more options available to you such as drawing with 2.5D pixols directly on the dropped object.

For a displacement map you still need a fair amount of polygons, at least 500.000 - 1 mil. all depending on the initial polycount.

There is also the possibility of breaking the model down into smaller pieces in order to subdivde each piece further for more detail. But that is a more involved process. Check the Making of Lord of Darkness by Tony Jung on CGtalk for more info on that.