Getting Started

[DiaWun]

I am new to ZBrush, and new to 3d virtual modeling. My background is fine art with a focus on figurative sculpture. I did some modifications for Sims, so I have a very basic familiarity with meshes and UV mapping. I’m posting this because I think some others might benefit from it as well. Specific questions or issues for comment are highlighted in yellow to save the busy experts’ time.

I began with the “getting started” tutorials. I had a great time with these and felt fairly successful, so I wanted to start a project and continue through some more advanced tutorials. OMG! There are so many! A great many are completely beyond my level. I also discover that MOST of them are for previous versions to my 4R5, and although I can still use them I am often left to hunt for buttons and processes that no longer exist or have been moved. In no time, I am lost and overwhelmed. I wonder why these tutorials aren’t ranked by levels, even subjectively, such as beginner, intermediate, advanced. No tags for versions, so be sure to check the date on each and every post and tutorial.

I go to the forum and attempt to search by terms as specific as possible… “delete curve” <way too late for alt z>, “join head to body”, etc., in an attempt to filter some of the massive amounts of information. I still spend hours sifting through posts.

In spite of all this… I am still so excited I can barely sleep at night. Each day I try to log what I’ve learned, because the list is lengthy and I want to “recap” some of the things I don’t use so often.

Today, I am wishing for a basic process tutorial… something organic enough to be generally applied to most projects a beginner or intermediate user might do. As an artist, I have learned that every medium has it’s basic process, and failure to understand that process causes massive hours of frustration, failure, or both.

What are the guidelines for using dynamesh? When should I turn it on/off? What is a good resolution to start modeling a simple/complex form? Where can I find guidelines for industry standards as regards resolution and poly counts? If I need to restrain my vision because the target market is maxed out at a certain level, I want to know, now. What should my target(s) be for various media. Please include this information in tutorials!

As a beginner, I would like to take 2 or 3 projects through to completion, with modeling, painting and animation. I hope to have something worthy of posting on these forums when I have passed intermediate levels. I hope to be able to market the skills I learn, at some point in the not-too-distant future. I’m fairly certain most users of this software are not hobbyists, or do not intend to remain a hobbyist. I know for a fact I will not be upgrading until I have at least learned the version I own, now. LOL! From the looks of it… that may be a few… erm… days. HA!

I can’t speak for others, but I pay well to save myself hours of frustration and numerous failures. That’s why I wanted Zbrush!

Thanks so much for your time and consideration!

[beta_channel]

What are the guidelines for using dynamesh? When should I turn it on/off? What is a good resolution to start modeling a simple/complex form? Where can I find guidelines for industry standards as regards resolution and poly counts? If I need to restrain my vision because the target market is maxed out at a certain level, I want to know, now. What should my target(s) be for various media. Please include this information in tutorials!

While I can’t find, or am too lazy to find any tutorials on the info you are looking for I can still answer some of what you’re asking.

Dynamesh is used for blocking out shapes. It has other uses as well, but for starters, it is for large form changes. Think of it like this, if your mesh is stretching while trying to pull out ears, nose, etc then dynameshing is probably what you want to do. Dynamesh is really only needed at the resolution that supports your forms. If your fingers are getting fused together, you need to increase your resolution. After you have your forms in place, then it is time to switch off dynamesh and start using regular subD levels. You don’t need your dynamesh so dense that you’re doing pore level details.

There is no good resolution for a simple, or complex form. The resolution you should be using is the resolution that holds the forms. I typically start with 32, if that isn’t enough, then I up it by powers of 2 until it will hold my forms.

Industry standard for resolution and polycounts…Well, that all depends on the industry. Games, movies, still images, 3d prints, etc. Each of these have different standards, but nothing is ever set in stone. Each instance is different. Lets take Left for Dead 2, each enemy is around 4-7k triangles for the final mesh, while a character model in Mass Effect can be upwards of 10k triangles just for the face. It all comes down to what is important to the game. Left for Dead is face paced so having less strain on the gpu for mesh skinning was important. It also allows for more characters on screen at any given time, while Mass Effect is slower paced and uses a ton of dialog sequences that are right on the character’s faces. Now mind you, the head model is swapped out for a lower res version actual gameplay.

Texture resolution falls under the same idea. There isn’t a set tricount for a fire hydrant, nor is there a set texture resolution for a fire hydrant. What is important is only using triangles where they are needed and maximizing texture resolution through properly laid out UVs. Both of these will maximize the look of your model and your textures.

Movies and CG are a different bag. Most tri counts come down to what does the tech artist that has to rig that character want to deal with. Most movies and CG are using adaptive tessellation on their models so you can have your cage as low poly as will allow for proper deformations of your model. Other things like cars, etc don’t have tri-counts really, use what is needed, but don’t be wasteful.

3D prints, whatever your printer can handle, either decimated or not.

Stills, doesn’t matter, but higher counts = longer renders

Mobile games are very limited in both tri-counts and texture resolution but this is due to all the graphics being offloaded on the cpu or apu.

Handhelds (3Ds, Vita, etc) vary just like modern games do with their tri-counts. It all comes down to what your engine/game can handle, or needs to handle.

As a word of advice, find references for the industry you want to get into. If it is games, start modding, you can learn a ton about the industry/standard practices, etc from learning how to reverse engineer. Skyrim has a complete kit, every WoW character can be viewed, exported, ect, Left for Dead, War-hammer 40k, usually if it is on PC you can mod it or find the community that is and learn how if it isn’t directly supported by the developer. Torchlight 2 just released Guts. Cyrtek and UDK are free.

If you can’t tell, I work in games, but anyway, enough of a rant from me. I am off to play some games.
I hope this helps.

[DiaWun]

Thanks so very much for taking the time to put these things down! Really, a tremendously helpful and informative response which I’m sure MANY noobs such as myself will appreciate!!!

While I can’t find, or am too lazy to find any tutorials on the info you are looking for I can still answer some of what you’re asking.

Dynamesh is used for blocking out shapes. It has other uses as well, but for starters, it is for large form changes. Think of it like this, if your mesh is stretching while trying to pull out ears, nose, etc then dynameshing is probably what you want to do. Dynamesh is really only needed at the resolution that supports your forms. If your fingers are getting fused together, you need to increase your resolution. After you have your forms in place, then it is time to switch off dynamesh and start using regular subD levels. You don’t need your dynamesh so dense that you’re doing pore level details.

There is no good resolution for a simple, or complex form. The resolution you should be using is the resolution that holds the forms. I typically start with 32, if that isn’t enough, then I up it by powers of 2 until it will hold my forms.

Industry standard for resolution and polycounts…Well, that all depends on the industry. Games, movies, still images, 3d prints, etc. Each of these have different standards, but nothing is ever set in stone. Each instance is different. Lets take Left for Dead 2, each enemy is around 4-7k triangles for the final mesh, while a character model in Mass Effect can be upwards of 10k triangles just for the face. It all comes down to what is important to the game. Left for Dead is face paced so having less strain on the gpu for mesh skinning was important. It also allows for more characters on screen at any given time, while Mass Effect is slower paced and uses a ton of dialog sequences that are right on the character’s faces. Now mind you, the head model is swapped out for a lower res version actual gameplay.

Texture resolution falls under the same idea. There isn’t a set tricount for a fire hydrant, nor is there a set texture resolution for a fire hydrant. What is important is only using triangles where they are needed and maximizing texture resolution through properly laid out UVs. Both of these will maximize the look of your model and your textures.

Movies and CG are a different bag. Most tri counts come down to what does the tech artist that has to rig that character want to deal with. Most movies and CG are using adaptive tessellation on their models so you can have your cage as low poly as will allow for proper deformations of your model. Other things like cars, etc don’t have tri-counts really, use what is needed, but don’t be wasteful.

3D prints, whatever your printer can handle, either decimated or not.

Stills, doesn’t matter, but higher counts = longer renders

Mobile games are very limited in both tri-counts and texture resolution but this is due to all the graphics being offloaded on the cpu or apu.

Handhelds (3Ds, Vita, etc) vary just like modern games do with their tri-counts. It all comes down to what your engine/game can handle, or needs to handle.

As a word of advice, find references for the industry you want to get into. If it is games, start modding, you can learn a ton about the industry/standard practices, etc from learning how to reverse engineer. Skyrim has a complete kit, every WoW character can be viewed, exported, ect, Left for Dead, War-hammer 40k, usually if it is on PC you can mod it or find the community that is and learn how if it isn’t directly supported by the developer. Torchlight 2 just released Guts. Cyrtek and UDK are free.

If you can’t tell, I work in games, but anyway, enough of a rant from me. I am off to play some games.
I hope this helps.