One thing I’m always seeing Jonathan Williamson and Kent Trammell (and a bunch of others) do in their tutorials is turning the models around so they can see where changes need to be made. Another thing I see is often the work done in perspective mode, unless working from an orthographic view of a sketch.
I know when I’ve tried to match reference, I’ve always had a problem getting things “right”. A lot of that is simply practice, but it’s also learning how to see and gauge what you see.
You’re doing great, for the most part. Just needs a few tweaks here and there to match up.
(And, I’ll freely admit, when it comes to my actual experience, I don’t have a lot when it comes to character modeling. My comments are based on my own understanding of what I have learned so far and from the observations I’m making. So feel free to take what I said with a grain of salt, if necessary )
Yeah, rotating more freely around definitely helps keep it from feeling boxy. I was adopting a more polygon centric workflow for this project because I felt accuracy was the most important. Now I have some general form down and can begin sculpting. I feel much better with a warm tablet pen in my hand!
To this point, it seems you’re successfully capturing the “feel” of Scully’s face.
if you’re going for realistic, the more reference pics, the better. If you are strictly going for a pose, then use only a few reference pics. Less confusing.
going realistic also is great for learning real physiology. Sculptors really delve into anatomy and muscle and bone structure. Subtle underlying things that affects outer look. Exploring that is worth it in itself. Again, Doris has a very good tutorial on realistic head modelling. Ken Trammel’s stuff is great too. Doing that, you’ll understand things like “Why and how” the jaw muscle mass is located in relation to cheek bones.
However, be warned, one of the weird mental things which occurs whenever you make even a single change to your sculpt, is that in your mind, you’ll feel you’ve reached the completed end state. That with a single change to eyebrow or lip contour, you’ve arrived! Don’t know why but it happens to me every time. But then after a day or two of cooling, you realize, holy crap, not even close. Its an odd mental dynamic.
A good tip I’d forgotten was to work in thirds. Concentrate on thirds of the face, once the eyes are set. The eyes placement in relation to the nose are the reference point from which to build on. I would actually duplicate a third of the head I’m working on, like the eyes and brows, then work on them separately, in fine detail. Sculley’s eyes placement is good, by the way.
to help people critique a long running project, make a check list of facial parts. And next to those put a short description of changes done, but as well, where you’d like to go. Just For example:
Brows - Form appears good. But Striving for more angularity Eyes - Too lifeless at this point. Trying to create a more questioning anxious look
Nose - Good to go. No changes needed
Cheekbones - Need more definition for uncertain how to retain softness of contour
lips - Fullness Good to this point, Obtained good bow shape. Matches the pics and has the Look I’m going for
jaw - angle from chin to jaw heel is too shallow, Wanting a steeper rise of jaw plane from chin to jaw heel.
etc -
This also fixes more firmly in your mind where you’re at. And where you want to go.
If you make changes but then realize you’ve gone “too far” off the mark, don’t sweat it. You ACTUALLY LEARN MORE DOING THAT than making repeated subtle adjustments for fear of ruining the work. Moreover, Your recognition that you’re overshot the mark is a sign you’re grasping the real contours of the face. Now you can go back and fix it with more certainty.
The process of creating a realistic face is real discovery. You “learn” the face of the subject. And exploring anything thoroughly takes time. but the effort is way worth it to an artist. Moreover, everything you learn will apply to your next work.
Yep, I agree again, but this is not just important for sculpting mind you, for texturing the under skin layer and “meaty muscle” layer you need to know the anatomy, bone structure, muscle system and vein system well.
This is why I always advise other artists to have coffee breaks in between 1 to 2 hours after a 3 hours session, even if you think your on a role.
hehe, this go’s for all projects, never enter a blender session without a notepad :eyebrowlift:
“it’s not the destination, but the journey that brings satisfaction” - quote “some guy”
Thanks everyone. This will surely be a great learning project for me as it’s only my third attempt at a head. I have 22 refs so far, all from varying angles. I tried to select images that had the light falling over her head pretty softly, but still casting some shadows so that I can see the contours of her face. I have seen Kent Trammell’s tutorials on creating heads, as well as some of Jonathan Williamson’s tutorials. I can’t, however, see Doris’ tutorials because I don’t have a citizen account at cgcookie. Does she have any free tutorials available, or is there another good resource you could recommend?
I also asked for advice over on CGtalk as well, and a user named Kanga mentioned to me that my problem might be in my workflow, as it’s typically more common to sculpt first, then worry about the polygons later.
In any case, thanks for all your help so far, guys!
I like the coffee break rule. Also look at Doris’ sketchbook. She’s got some good sculpt works to peruse.
As far as workflow, IMO that’s a matter of experience. “Experienced” modellers seem to want to sculpt a work concept first and then retopologize after.
However, I think starting out, learning the tools of pulling and pushing verts and faces around is important. And once you get the knack, then shift to sculpting first.
Again, Doris and Michaelis would be good guides for solid work flow.
Last, you asked before about a look at the wires. Yeah, that would be good to see the topology and density of your mesh.
Definitely. Taking a break gives you a whole new perspective.
I can see the workflow going both ways. Not sure what would have been better for me, but at this point I think I’m past that decision (unless I decide to start over, and even then I can still use what I have here).
I’ll check both of those out, thanks!
Here’s a wire view of my mesh. One of the things that bugged me is that there are a lot of long rectangles, which won’t make for a very good base when sculpting the skin using alphas. So I’ll have to work on making things more square.
A perfect example of “artistic jamais vu”, and the importance of stepping away and coming back with a clear mental palate: at some point in time, I actually thought each of these previous versions looked good.
Who knows how bad this current iteration will look in comparison to the next version!
It’s like shooting free throws in basketball. It’s the misses that make you a better shot.
Ohh, and you know immediately which version feels right. Also if you redid scully from scratch you’d knock it out in one tenth the time it took for your original
Very nice look.
Slightly Broader face and particularly width of the jaw.
Nice feminine sweep from cheekbone to chin
chin is slightly more prominent like scullys
brow looks better
overall much improved
PS - once You add hair, delete the ears. Unnecessary geometry
Thanks FXR. Fixed the chin. Now I’m wondering about my workflow a little bit. Should I paint my skin texture before or after my sculpt? On the one hand, it makes sense to paint based on the wrinkles and pores added in sculpting. On the other hand, if I want to paint mirrored (to save time), I will need to paint on the symmetrical mesh to avoid misplaced strokes on the opposite side.
My two cents is do both. You haven’t worked enough yet to determine your “work flow”
I’d start working on the sculpt and then when you want a break go to the skin map and work that for awhile. That way you can peridically update your texture map to get a feel where you’re at.
I’d like to convert my high res sculpt to a displacement map so I can apply it to my low res mesh. This is mainly because I have always had lots of problems and artifacts when changing and adjusting the base mesh once I’ve sculpted, and I want the freedom to change the mesh later (especially when adding an expression). However, Blender’s baking tools aren’t working well at all either. I’m getting tons of artifacts (and just for clarification, I am baking to a 32bpc image). Does anyone know the best way to go about doing this and getting a clean result?
)
