Dynamesh for high detailed HD sculpt?
polycounter lvl 6
Hi.
I am new to zBrush and 3d over all.
Started with 3D sculpt and used Dynamesh but then got info from friend of mine that i should not use Dynamesh too often as i get too many polygons and it makes impossible to rig it later or build effect on it as it is too "heavy".
What workflow is the best to follow and why?
What do they use in production?
Example:
World of Warcraft cinematics.
Starcraft Cinematics etc.
As i am interested to create animation with character i create in zbrush rig it >> build effect on it and put together a small clip.
Is this correct way?
https://www.youtube.com/watch?v=n8t-SV8rU8o
I am new to zBrush and 3d over all.
Started with 3D sculpt and used Dynamesh but then got info from friend of mine that i should not use Dynamesh too often as i get too many polygons and it makes impossible to rig it later or build effect on it as it is too "heavy".
What workflow is the best to follow and why?
What do they use in production?
Example:
World of Warcraft cinematics.
Starcraft Cinematics etc.
As i am interested to create animation with character i create in zbrush rig it >> build effect on it and put together a small clip.
Is this correct way?
https://www.youtube.com/watch?v=n8t-SV8rU8o
Replies
Here they show the density of the cinematic models they made for the Overwatch shorts @5:00 and 6:30
I would not recommend working with extremely dense dynamesh meshes simply because it becomes very difficult to make large changes late in the sculpt if you don't have access to the bottom subdivision levels.
Currently have it like this.
It is 5 sub div levels total.
And dynamesh i did not use.
As i understand should use dynamesh only at beginning.
Later should not use it as it would go too "heavy".
re-doing all model at the moment.
I am trying to get correct info as when i am noob it is hard to understand what is correct way and what is not as Interent is full of information and i must think that not everything there is correct and there is a lot of worong / out of date and incorrect in formation.
This is why i ask my question here.
Thank you for your feedback.
there is no "this is how you do it and this is the only right way, every other way is wrong"
on some models it doesn't matter if you start with a dynamesh or a basemesh. I usually prefer basemeshes but it comes down to the style and the production. You can create perfectly fine assets starting from a dynamesh, or zspheres, or Sculptris, or basemeshes or using booleans.
What suits you best in which situation is something you will have to learn the hard way. By getting experience and experience is nothing you can shortcut.
Personally i use dynamesh for lose sculpts, noodling around if there is no defined concept. Once i want to work clean i retopo my meshes to have a nice and clean base. Yes i do this by hand, or if it is some generic humanoid i reproject one of many premade basemesh using Wrap3 or other tools.
If it is for production i most likely use one of these basemeshes as a starting point and work with them. In many productions you have a set of basemeshes and this is what you use. There is no freedom in this, because it helps with the whole production past your part in it.
If this is all for yourself, play around, find what you like, look what is efficient for you. Study other peoples work, do some tutorials, stumble upon parts in those tutorials which are just plain wrong... great you learned something!
on your noodling around video, i would say the same base principle applies like what you said with "As i understand should use dynamesh only at beginning. " without understanding why this is so.
You always work from rough to fine, from blockout to detailled.
Right now you add details before you have the basic anatomy layed out, step back a bit and work on the anatomy first before you start adding wrinkles.
Animating it (rigging) in this case is not mandatory,
Main goal is to create Ogre face sculpt, add a bit skin hair and texture it and then export model to Houdini where i would create an effect on to it.
For example Blizzard cinematic:
It contains many aspects of 3D and it does not seem to use low poly model.
I am currently trying to learn part 1 that is to create a model that can be used in cinematic trailer (yes i know all kind of models can be used there but i am after for high level of detail model).
In my case if i rig it in future, then only a bit to move the jaws or eyes. Nothing big and massive.
And this is where it went confusing at one point as some people suggest use low poly model but as i understand low poly model cant hold as much detail as high poly model and what they do is they create high poly model and then project data from that to low poly model.
I am planing to use camera close ups in HoudiniFX so this means that model should look detailed and good also when i zoom in.
And it is hard (due creativity) to create low poly model for clips where i do not use camera close ups and create only part of model fr camera close ups as i am not sure in what part i will use the camera close up-s.
Here is another example where you can see camera close up shots and 3d elements.
Here is one clip what i did in Houdini FX:
Because i have no cash to spend i was not able to use Vimeo services to upload video at that point so i had to use Youtube, and yes i know it reduces video quality dramatically.
Here is screenshot from that clip in better quality.