I just want to specify that I’m a newbie to Blender as well as the community (as a result any advice on either is greatly appreciated).
Recently I have been perusing Youtube videos of modelling to look at people’s techniques. I also like to check the comments in case anyone gives any helpful tips. One video (can’t link to it, the title is “Angelina Jolie - Blender Timelapse”), received some criticism from a user over the topology. He or she described one main flaw being the artist was “falling into the edgeloop/ring hype, which results in everything just having rings around it, which is suitable for cartoon characters but terrible for real faces.” When one of the commenters argued that loops around the eyes and mouth were good for animation, the other replied saying that his information was dated.
While many of the commenters seemed to agree with that, I also here that correct facial topology consists of loops around the eyes and mouth. Personally, I don’t have a preference or an opinion on which is correct, I just want to know which is better.
Snippets of information from a fractured conversation.
My take on it:
Define the result, chose the method that allows you to achieve that result.
Nothing wrong with creating the base mesh as per the video, or box modelling with or without modifiers, or kit bashing and using booleans, or starting with a sculpt and retopo workflow, your path will be determined by the result desired and personal preferences that you develop.
E-Winters reply is good. remembering that the retopo step is creating a new replacement low-poly mesh with adherence to correct edge flow for animation, I think others in the comments got to invested in arguing and lost focus.
Depending on your personal preference you may prefer texturing fine detail as apposed to sculpting,
here’s a nice example
If it’s not going to be animated (most busts aren’t) then topology doesn’t really matter beyond what helps you do the modeling. There’s no such thing as “correct” topology.
I’d be wary to get my information out of Youtube comment debates, but there is a point to the idea that people mistakenly believe that good-looking topology is good topology. Facial animation is really difficult to get right and I’m sure there’s many difficulties caused by a topology that’s too concerned with overall flow, i.e. sometimes you’d want to break flow in favor of control. You should get that sort of information from an actual facial animator, preferably the one you’re working with.
It always depends. Economically, the best option is the one that takes you to your end goal with the minimum amount of effort. Are you trying to win a wireframe beauty contest, or are you trying to produce a finished render?
Yeah, just don’t read the comments. Unless there are links to something useful, it’s a cesspool of crap like the above. Don’t trust the tutorials either, no matter how reliable the source is. Gather information from multiple sources, which adds to the information you already know and minimizes the effects of misinformation.
Just that the tutor is very popular and the youtube comments say he’s a god isn’t a guarantee that the information in the particular video is 100% correct. Of course he tries to make a good tutorial, which isn’t easy, but following a tutorial that gives 6/10 or 7/10 correct information can be worse than just a bad tutorial because he says and does things with a great confidence and a newbie doesn’t know any better.
Nothing wrong in using edge loops. Loop types, modeling specific kind of structure, is part of a larger topic that is topology. The timelapse video might not show the best example and have the best result, but irrelevant it isn’t. There are cases when topology knowledge is not needed, but it’s still a basis of readable structures, how tools work, incorporates functionality for the pipeline and end use, and because of all of those it’s part of modeling workflow you decide on.
Could sculpt the head and only focus on forms and proportions, but as mentioned, next step getting it through the pipeline is retopology to get a functional structure, which obviously requires topology knowledge and edge loops are very much part of it.
Edge loops/quads are in wide use for valid reasons. Their value is empirical. Especially when animating. And particularly in the areas you mentioned. That value has nothing to do with trends. It’s been tried, tested and refined over time. And it serves a purpose.
It will only be a “dated” method when it’s replaced, or rendered redundant. That has not happened yet to my knowledge.
If it’s static in nature, and it looks as you intended, then that’s the primary goal achieved. After that, you and others can talk about it’s subjective value.
You don’t have to change the topology unless it’s obstructing you in some way or causing shading artifacts. Certainly not to fit any perceived trend. It either works or it doesn’t. The object of the exercise is presumably to create something. Not obsess over neat or visually appealing patterns you won’t see. Certainly not in a static mesh.
Like JA12 said, it’s not a bad idea to get plenty of different views on the subject. Following a single individual could lock you into their particular method. Which isn’t a good thing if they have some bad habits.
