How to know is your mesh after retopo fine or not ?

focus_method

Hi,
I was wondering how do I know is my mesh after retopo fine (ready for production) or not ? how can i test it ?
Are there any kind of guides ?

and what is the ok polycount usually  for low poly , semi realistic and hyper realistic characters?

sacboi

Info is readily available, just by doing a bit of "own work* will pay dividends in the long run, instead of relying on others todo it for you...

So my advice is too 'mine' the Wiki for all it's worth -  ReTopologyModeling

DavidCruz

sacboi said: Wiki for all it's worth

^ Just going to cut off any discrepancies (letting the mind (of an outsider looking in) say this guy is a d!k. )

Usually what everyone new and old should do before posting is review the wiki otherwise we are always repeating ourselves, plus everyone (not me apparently) is too busy to repeat themselves to everyone.

So do not take it personal Focus, generally lot of people come on here to see the great artworks and maybe help someone that looks promising to lift up, otherwise everything has been pretty much asked before, unless the issue is some bug or general complaint, or plugin that might help the rest of us.

focus_method

thanx but i haven't found this useful 4 me yet. Even if there is an answer for my question i have to go through dosens of articles. I need something precise,  that's why i opened this topic.
Of course, if you think this topic is not ok , you delete it, im fine with your decisions.

Alex_J

You only know that it is fine by testing for it's specific application.

If you want to know in a general sense, "is my topo fine for animation?" then what you can do is download a few free animations from mixamo, use a quick rig solution like humanIK, and test it out that way.

For a lot of games if you can run, kneel, put arms overhead, and make basic facial expressions without ugly stretching or clipping, it's good.

pior

"I need something precise"

Then you do what everybody else does (or should be doing) : studying models from high caliber games and learning from that.

Blaizer

For a CGi character, a production model, subdivision modeling with a good topology for deformations and UV seams. You can see hundred of wires in Pinterest, the wireframes tend to be very clean and "quad only". For other assets, such as props, you can see standard meshes or subdiv models, all depends of the object shapes.

For video game "low poly" models, is practically the same except for subdivision modeling. There are hundreds of tricks in Pinterest or Polycount Wiki. Nowadays, polycounts are not a limit, so you can use a subdivided model and colapse rings/edges o delete loops of edges to clean unneeded detail. Polycounts surpass easily 100k tris in too many characters, and we use the needed polygons for a good silhouette/shape and good normal map shading (all depends of the object/character). The unique constraints are UV maps size and cuantity.

sacboi

Wiki's are a gold mine, typically first point of call for anyone researching a given topic which I should've in hindsight expressed succinctly minus the frustration, sorry about that focus_method

Panupat

Mostly experience. Yes you can look up guides and articles to give yourself some foundation. However once you work in a team you will get feedback most critically from riggers. They will tell you change flow, add loops, increase/reduce polygon in certain area and imo that's where you will learn the most. It's also quite specific to each rigging system.

poopipe

^^ That's the crux of the matter. 

You only know whether you've done something right by understanding what the next person in the chain needs to do with it. 
Character artists need to be able to  build a basic skeleton and skin their model to it just like riggers need to be able to animate at a basic level. 

This applies to every discipline and it works the other way along the chain as well