🧵 A thread died for this
Anonymous at Fri, 5 Jan 2024 22:31:12 UTC No. 969472
Is anatomy good for you? Discuss character creation here while I post my "progress".
I believe in power of personal blogs since that's exactly how I got a coding job.
So here goes.
Am currently following the Anatomy for sculptors book.
Today's goal is to make something resembling a human. I've already spent an hour on this blank helmet.
Final goal for this year is animate two characters interacting with each other and environment in a silly manner.
Anonymous at Fri, 5 Jan 2024 23:01:28 UTC No. 969477
No complaints here. Here's hoping the rest of the face goes well for you.
Anonymous at Fri, 5 Jan 2024 23:20:47 UTC No. 969478
>>969472
where did you get the cool refs on the right?
Anonymous at Fri, 5 Jan 2024 23:30:17 UTC No. 969479
>>969477
No this time for sure, but it will go well sooner or later.
>>969478
Screenshots from the book
Uldis Zarins - Form of the Head and Neck 2021.pdf
I don't remember where exactly I've pirated it
Anonymous at Fri, 5 Jan 2024 23:41:30 UTC No. 969480
>>969479
>Uldis Zarins - Form of the Head and Neck
thanks! good luck in your scultps
Anonymous at Sat, 6 Jan 2024 01:20:46 UTC No. 969485
>>969480
Thanks =)
4 hours of work have yielded this.
Can't fix this one, it looked better an hour ago before I started trying to copy detailed features from the book and everything fell apart.
Spent half an hour moving around zygomatic bone alone. It just doesn't fit anywhere.
Gonna watch some more timelapses, take some notes, see you tomorow.
Anonymous at Sat, 6 Jan 2024 02:35:21 UTC No. 969490
You set your own limits for what is acceptable quality. If you wanted to *only* animate 2 characters interacting with each other and environment, you could have it done within the week of learning animation. But your actual goal, is to animate something that looks "good". Something that is of acceptable quality to your eyes.
That *could* take all year. But what if at the end of the year, you're not pleased with the quality of your work? Does the git gud journey continue? Or do you settle for what you have? It's good you have a plan and a regimen to keep you on the path to improvement. But I'm just trying to get you to think about a realistic outcome. Remember not to get too mired by 1 thing like anatomy. When the larger goal is the animation. Divide your time between studying each component you need to make the animation happen. One component being anatomy, and the others...? Animation for sure. So rigging. Probably a little hard surface modelling. Composition. Shading. All that stuff.
Of course, if you stated your goal was just to get good at anatomy, I would tell you to ignore all that other stuff and 100% focus on anatomy practice. It really depends on the goal you're shooting for.
Anonymous at Sat, 6 Jan 2024 02:45:17 UTC No. 969492
>>969485
how many tris your used for that?
Anonymous at Sat, 6 Jan 2024 03:25:36 UTC No. 969494
>>969490
Your psychological breakdown couldn't be more true, I do want my stuff to look good.
But this final goal is whatever, I have to set estimates to my work tasks so why not use it for a hobby in a vague manner.
Yeah tried getting models from smutbase, and they've got great rigs and textures, but it's not easy to just animate them right away, they've got bugs, I can't just slap a new face on them, proportions are not to my liking, what if premade rigs lack something I need, etc.
So, yeah, not in a hurry, might as well grind anatomy, make good looking faces and bodies for my characters. But occasionally checking out other stuff - why not.
Thanks for your input.
>>969492
It says 200k.
It's a default 1m sphere with remesh size set to 0.025m.
What's the point in counting tris in a sculpt anyway?
Anonymous at Sat, 6 Jan 2024 06:41:58 UTC No. 969507
>>969494
Not the guy you're replying to but the number or tris is a good gauge of what your limit for detail is. Also it's a factor in performance. I usually don't wanna go over 5million tris for a single object. You'll want to retopo and multires for finer details once you have a good base down. Here's the first video of a two-parter I like that shows how to create a character from sculpt to final mesh. You'll probably want to reference this later down the pipeline once you've sorted out anatomy.
https://www.youtube.com/watch?v=5mW
Anonymous at Sat, 6 Jan 2024 07:29:03 UTC No. 969512
>>969507
Ah yes that's fair.
Thanks for the video, added to my notes.
By the way your guy makes interesting decisions with his retopo. I'll probably find out later why is he so chill with placing polygons on top of intense geometry like in the pic.
Anonymous at Sat, 6 Jan 2024 09:53:47 UTC No. 969513
>>969472
sauce for pic on the right?
are those, like, a human bust modeled with fuckin NURBS??
Anonymous at Sat, 6 Jan 2024 21:18:52 UTC No. 969587
New day - new sculpt.
I'm gonna try a method from an /ic/ book
Betty Edwards - The New Drawing on the Right Side of the Brain_ A Course in Enhancing Creativity and Artistic Confidence-Tarcher (1999)
When your reference is a collection of alien shapes and you gotta simply copy them.
Anonymous at Sat, 6 Jan 2024 21:29:05 UTC No. 969588
>>969513
Here
>>969479
>Screenshots from the book
>Uldis Zarins - Form of the Head and Neck 2021.pdf
Anonymous at Sat, 6 Jan 2024 23:39:26 UTC No. 969596
Anonymous at Sun, 7 Jan 2024 01:30:06 UTC No. 969608
>>969595
I think you've fallen into a trap alot of new artists do. You're spending way too much time focusing on details way too early in the work. You gotta shit out a well-proportioned rough then start smoothing it out and adding detail. It's okay if it looks like blurry cottage cheese. It can be smoothed out and detailed later. That being said, things are looking alright. You just gotta keep practicing. What I did was sculpt a head as much as I could for an hour everyday as fast as I could. It took about a week for the insight to hit me. Then I started to consistently churn out a good base every day. Dunno how long it'll take for you. Also: Try not to get attached to what you're trying to sculpt at this stage. Once you accept that you're just churning out trash you're gonna delete anyway, you start to focus more on the process than the end product. Ironically this produces better results. Don't try to copy your reference. Use it as a guideline. That's how it works for me, anyways.
Anonymous at Sun, 7 Jan 2024 01:39:09 UTC No. 969610
Well, today's work is the closest to a human I've ever done.
The problem is I have deliberately went against reference features, poking at and fine-tuning proportions so they resemble a human somewhat.
But practice is practice so these 5 hours were definitely not wasted.
Anonymous at Sun, 7 Jan 2024 01:44:34 UTC No. 969611
>>969608
Thanks, it's all true, but I can't see if proportions are right until all facial features are present and something starts to feel off.
But it'll come in time, yeah.
You're 100% right about starting speed sculpting, and that's exactly what I'm going to do next.
Anonymous at Sun, 7 Jan 2024 02:10:16 UTC No. 969616
>>969610
>>969611
Yeah, Things are rough before you've lined your brain with eyes. Sorry I had Let out that Bloodborne reference. But, yeah! Things get alot faster and more intuitive once you brain can project your vision into your work. I was super frustrated when I started out. Everything looked like aliens or weird Juniji Ito abominations. Then one day shit just clicked into place. Like sculpting, it seems being able to "see" what you're doing is a skill in itself.
Your sculpting has improved, by the way. Though she looks more like a skinny Yakuza than a Caucasian woman, she does look like a human now. The chin is also closer to a natural position. Doing good.
Here's a vid of a guy doing a sculpt in 30min real time.
https://www.youtube.com/watch?v=W3G
He doesn't explain a whole lot in it but you can study his technique and strats.
Anonymous at Sun, 7 Jan 2024 03:35:07 UTC No. 969619
>>969616
Thanks! I watch him all the time
Anonymous at Mon, 8 Jan 2024 06:48:27 UTC No. 969748
Tried doing eyes manually, couldn't pretty them up before the hour timer rang.
This week I'm gonna do similar hour speed sculpts after work every day and then on weekend decide what to do next.
Anonymous at Mon, 8 Jan 2024 07:36:55 UTC No. 969749
>>969472
You have a good eye for shapes Op but try to be more aware how adjacent areas flow together, how shapes curve and merge into one another as if their lines continued to extend into those areas. Reason our anatomy looks the way it does is because how it gives mechanical advantage in manipulating the structure that is on the opposite side of where the muscles insert.
If you bring this awareness into your work you'll lose that feeling of having a 'neck inserting into a torso' and gain that feel of it being two interconnected surfaces that where shaped to go together as a whole in a combined interconnected system.
Trace the curves of your reference there with your finger or in your mind on the right and flow from one area to the next.
Notice how it all goes together and you can smoothly flow from one area to another as a continuous curve, where as in your variant you try to do this you kinda bump from one surface onto the next while they're quite shapely in isolation.
Anonymous at Mon, 8 Jan 2024 07:47:55 UTC No. 969751
>>969472
>>969749
And yeah anatomy is important but too early overemphasis can serve to lead you astray. You will need ever more intricate anatomical understanding as you progress but knowing
anatomy in great detail before you have a trained eye for shape, learnt to see what is actually there, can cause you to follow a formula you have learnt for what that area is supposed to look like as opposed to what it looks before you in the real world reference of a particular specimens physique you wanna replicate.
IMO Anatomy done right is informing you as to what you're looking at, what you're looking for, as opposed to being used to retrofit constructs that sit in your mind of what
such and such anatomy is supposed to look like.
You wanna be like "oh this curvature here is due to the zygomatic process and it's shaped like this in this guy visible to this degree" as opposed to
"and in this area the zygomatic process is located so I will proceed to draw in my idealized concept of the zygomatic process I saw in anatomy reference X in this area"
Anonymous at Mon, 8 Jan 2024 08:38:52 UTC No. 969754
>>969749
>>969751
Thank you.
>"and in this area the zygomatic process is located so I will proceed to draw in my idealized concept of the zygomatic process I saw in anatomy reference X in this area"
Funny you mention this, because I'm more than aware of the pitfall. Read about it long ago in a drawing book. Since then, I'm always on the lookout for lines, angles, lengths, intersections, etc.
And about the real reason our anatomy looks how it looks too. I often find myself at a crossroads:
- this fat pad on the reference goes like this
- but if I do it like this on my object it looks unrealistic
- in real life it would be drooping but in my case it's like silicon
- all because those landmarks A, B, C are off by a wide margin in comparison with the reference, but they are the part of another facial feature X, and that facial feature looks okay, so if I fix landmarks for Z, it would ruin X
- BUT if I make it flatter at the top it looks okay and is physically possible
- BUT I'm moving away from the reference again
So, believe it or not, I'm already trying to do those things you mentioned, I'm just not good at it at the moment. Keeping full picture in your mind is a challenge. But again, thanks for analysis, it truly helps.
Anonymous at Mon, 8 Jan 2024 08:59:31 UTC No. 969755
if you want to do portraiture you should watch scott eaton's tut:
https://www.bilibili.com/video/BV12
Anonymous at Mon, 8 Jan 2024 09:29:07 UTC No. 969758
>>969755
Thanks, added to my notes
This course is on rutracker btw, but with a terrible voiceover. I mean it's my language but the quality is bad.
https://rutracker.org/forum/viewtop
Anonymous at Mon, 8 Jan 2024 09:44:15 UTC No. 969761
>>969758
it's funny, originally it was uploaded in english with terrible quality, then it was uploaded in decent quality with russian voiceover, then someone put the english audio back on the russian one, then chinks subbed it and uploaded it lel
just a heads up, he does zero sculpting in this course. it's all observation and theory + he'll draw over images to show stuff. but despite that it's very good. imo you need something like this to actually use a book like anatomy for sculptors properly.
Anonymous at Mon, 8 Jan 2024 09:53:26 UTC No. 969762
>>969761
oh also, this is just about the face. he doesn't even cover the neck region iirc. for that there's a separate course about general anatomy.
there's also a third course where he is sculpting if you want to watch him work. that one assumes you're already familiar with anatomy though. there's a very brief recap before each section, but you won't get the msot out of it unless you've already seen the other two.
Anonymous at Mon, 8 Jan 2024 11:02:35 UTC No. 969767
>>969762
Well, now I know how I'll spend next couple hundred of meals in front of a computer. Thanks
>he'll draw over images to show stuff
This is my favoriute part
Anonymous at Fri, 12 Jan 2024 20:02:41 UTC No. 970287
Work week has been rough, finally, friday night, can do my hour sculpt.
Played around with forms, couldn't get to the detail level in an hour.
Tomorrow's gonna be a longer, deeper session after some anatomy lectures.
Anonymous at Sat, 13 Jan 2024 10:29:36 UTC No. 970331
Not even 10 minutes in, the lecturer asked viewers to skecth a skull. I can't draw so tried to sculpt. This is my result after 2 hours.
Started patching holes in edit mode and noticed bad geometry all over the place, couldn't fix it fast. I'm sure I'm not supposed to take even more than 15 minutes so I'll drop it.
I'll continue with the video until it makes me do another excercise.
Anonymous at Sat, 13 Jan 2024 13:48:15 UTC No. 970352
>>970331
>>970332
why are you bothering with topology for an exercise like that. just crank up the voxel remesh if you need more resolution.
Anonymous at Sat, 13 Jan 2024 14:57:30 UTC No. 970367
>>970352
It was distorting and making holes when I touched the place
Anonymous at Fri, 19 Jan 2024 19:03:42 UTC No. 971127
Aight, work week is over.
Been watching Scott Eaton's lectures while doing dishes and cooking. Bluetooth earphones are the best btw.
Learned so much about construction, let's check if it will amount to anything. Spent about an hour on this blank.
Anonymous at Fri, 19 Jan 2024 22:10:31 UTC No. 971139
>>971127
4 hours have yielded this. Still struggling with shapes and the big picture. She looked like monki until at the very end when I remembered about supraorbital rims.
But what you gonna do.
Plan for tomorrow is to take some screenshots from lectures so as not to forget basic construction stuff, and try again.
Anonymous at Fri, 19 Jan 2024 22:14:34 UTC No. 971140
>>971139
Looks good
>she
Oof. Well the anatomy is going well anyway.
Anonymous at Sat, 20 Jan 2024 09:02:47 UTC No. 971192
>>971140
Shit, now I see it. This is what happens when you add mass again and again while trying to fix proportions
Anonymous at Sat, 20 Jan 2024 22:27:08 UTC No. 971262
>>971192
It's ok. Shooting for accuracy is more important for learning anatomy. If you accurately capture a woman's face, then it will appear feminine by happenstance. You won't have to force it.
And besides, a lot of women don't look very feminine without their long hair and lashes anyway.
Anonymous at Sun, 21 Jan 2024 21:03:30 UTC No. 971386
Today, I've made a huge step closer to the reference, because the sculpt actually looks feminine.
Once more, it took me two hours of jerking around supraorbital rims and ROOF (this fat above top eyelids) to establish a somewhat human liking.
That's what I get for not doing course excercises. Goddamn I hate drawing, but I'll probably have to do them sooner or later anyway.
The plan for the week is as always: watch lectures, don't have time for speed sculpts, wait for friday.
Anonymous at Sun, 21 Jan 2024 21:17:47 UTC No. 971388
>>971262
>And besides, a lot of women don't look very feminine without their long hair and lashes anyway.
True, but it's no excuse, I'm not a western game developer forced to churn out average people.
Anonymous at Sun, 21 Jan 2024 22:03:53 UTC No. 971390
>>971386
work on your eye area lids particularly. it's easy to see you're not confident in that area at all and it's all just vague gestures.
>>971388
you're about to learn some hard lessons. hopefully you learn the right ones.
Anonymous at Mon, 22 Jan 2024 07:25:18 UTC No. 971425
>>971390
To be fair, I'm not confident in any area at all.
With eyelids, on top of anatomy ambiguity, it's a resolution question. This is 0.02m remesh resolution. I'm afraid to lower it and do indepth eye and ear areas, because clay strips brush already lags on 0.02 on large strokes, and the scupt isn't even a full body yet. All this despite having i5 11400F, 32GB 3200hz RAM, RTX 4070.
But again, it's no excuse.
I should probably try dyntopo or zbrush.
Anonymous at Mon, 22 Jan 2024 11:59:58 UTC No. 971436
>>971425
if you're running into perf issues at this mesh resolution it's probably better to just switch to zbrush now, especially if you want to pursue very realistic results.
i wouldn't want to use sculptris/dyntopo at this stage. it's too lo-res.
Anonymous at Mon, 22 Jan 2024 20:58:11 UTC No. 971489
>>971425
holy crap you run into performance issues at that resolution?
I'm sculpting this:
>>971377
with an amd fx6300 from like 11 years ago, seriously just switch to zbrush, it's good to see you're making progress so once you adjust to zbrush you'll do even better cause you won't be limited by mesh resolution.
Anonymous at Mon, 22 Jan 2024 20:59:13 UTC No. 971490
and also cause you absolutely need more resolution to start working on the eyes.
Anonymous at Tue, 23 Jan 2024 02:49:57 UTC No. 971516
>>971425
>>971436
>>971489
The Zbrush shills bothered me. So I wanted to see if I can diagnose your problem in blender. You have better specs than me, so you should be getting better performance than me.
So I just went into blender and started with the default sculpting option at start. The first thing I did was reach for the remesh resolution option. But immediately, I thought "I don't want to remesh after every stroke". So then I googled "blender sculpting making more polygons", and was reminded that dynotopo existed.
So enabling dynotopo, you can lay down strokes that increase polygons only around the areas you're touching. By default, dynotopo is set to "subdivide collapse" that options attempts to maintain a certain level of detail. But personally, I found that setting it to just "subdivide", ensures that you get enough geometry necessary as you dig into crevices, Or add more mass.
Once you think you have your crevice deep enough, or your mass bulgy enough, then click dynotopo off, and just shape the mesh. Turning it back on if you think you need more geometry.
Dynotopo has "relative detail" set as the default. What that means, is that when you zoom in, you will get finer sub division. Zooming out, you get broader subdivision. It's relative that way. So I set dynotopo to 4.0pixels. And it seemed to get just the right amount of geometry without going overboard. See here for my test sculpt: >>971508
I didn't have any performance issues. There was an initial problem when I first enabled dynotopo, because I set it to a ridiculously low number, and then brushed over a huge surface. But after restarting blender, I set to to a normal 4 pixels, subdivide, relative, and it all worked out. Nice and smooth. If you're still having performance issues, then I would look at what other wacky and random option you have enabled, and disable it. Or create a new file so everything is set back to default.
Anonymous at Tue, 23 Jan 2024 03:17:17 UTC No. 971521
>>971516
Ok but you've got to think long term. If you sculpt, you're going to want to switch gender at some point in your life. To switch to ZBrush is a first step in that direction.
Anonymous at Tue, 23 Jan 2024 17:35:33 UTC No. 971618
Op here, shit, I don't want to get into ZBrush, it's too hard already.
I don't know, is video related an acceptable level of freeze on 0.02m? It also lags much more with graphic tablet as opposed to the mouse.
>>971516
I've heard dyntopo is only for pros, but I'll give it a try before zbrush for sure.
Anonymous at Tue, 23 Jan 2024 19:40:35 UTC No. 971629
>>971516
damn you discovered something that was already mentioned and isn't a good solution to this problem. dynotopo/sculptris are appropriate for localised detail work; they aren't a replacement for voxel remesh / dynamesh.
>>971618
no it's not good enough and there's no reason to deal with this.
pirate zbrush, and follow along with this:
https://www.youtube.com/watch?v=_yK
it'll get you to the level you're at in an hour.
you can look up other stuff as you go.
Anonymous at Tue, 23 Jan 2024 19:56:54 UTC No. 971630
OP you should more heavily rely on references. humans are kind of complex where individual features being even slightly off can ruin the whole thing. consider grabbing PureRef and literally overlaying references over your model and making adjustments from there, I think it would be helpful
Anonymous at Tue, 23 Jan 2024 20:04:39 UTC No. 971631
>>971629
Thanks, guess I have no choice. I'll do the next sculpt in zbrush this friday.
Anonymous at Tue, 23 Jan 2024 20:52:03 UTC No. 971637
>>971636
i think you can already see a few fixes in there, the ear feels a little high, the neck a little short, face slightly long. maybe the brow could match silhouette a bit better. that's the darnest thing, the devil is in the details
Anonymous at Wed, 24 Jan 2024 02:02:31 UTC No. 971678
>>971629
>dynotopo/sculptris are appropriate for localised detail work; they aren't a replacement for voxel remesh / dynamesh.
Explain further. I would like to know the "why" of this, rather than just "NUH UH, DYNOTOPO BAD, CHANGE TOOLS"
What is voxel remesh / dynamesh. How are they better than dynotopo?
>dynotopo/sculptris are appropriate for localised detail work
That's literally what's described in the video you linked. I skipped ahead to the remesh part, and he's literally telling you not to go overkill on the whole mesh, and how to use it only for localized portions you need more detail for. So it sounds like pretty much the same use as dynotopo.
Also, the whole reason this matter came up, was to get more detail inside of the eye. Or in other words, to get detail into a localized region. He can turn on dynotopo for just the eye area, and then turn it off again when working on other bits. Only enabling it when needed.
>>971618
Something is definitely wrong, because performance shouldn't be *that* slow. Couldn't tell you what's wrong. But the webm will show you how it's supposed to look. My mesh has much denser geometry in the face, and it's not lagging there at all. Your mesh might be a larger scale than mine, so I tested it at 10x the scale, and the strips still brushed over the face as smooth as butter.
>I'll give it a try before zbrush for sure.
Yeah, at least try it first. Remesh can give you a baseline with which to begin your sculpt. But the problem is that for small areas like eyes, you won't get good detail unless you remesh the entire thing. Which creates way too many polygons than necessary. So remesh isn't something to be relied upon throughout the whole process. You need more geometry in isolated areas. And that's what dynotopo does. I'm not sure where the idea that it's for pros came from. It's just a means of getting more geometry where necessary.
Anonymous at Wed, 24 Jan 2024 02:25:36 UTC No. 971686
>>971678
>>971618
Oh, I think I just recreated your level of lag. How many polygons does your model have? Turn on statistics so you and I can see.
I noticed that your clay strips were getting finer detail than mine. That's why I tried upscaling before. But I neglected a step: I didn't remesh after upscaling. That's what I just did. now my mesh is 10x the size(which was already 10x life size when I began) So a total of 100x life size. Then, I remeshed by 0.02. It created 13.5 million faces. Or 27 million tris. NOW, when I use clay strips, it lags just like your webm. Could it be that you just have way too many polygons? You may need to scale down. Or alternatively, remesh to 0.2, rather than 0.02. idk.
Jesus, it's so slow, that even changing from object mode to sculpt mode takes a full minute.
Anyway, let me know if polygon count is the issue. If not, I have another idea. It might be window's ink interfering with your tablet.
Anonymous at Wed, 24 Jan 2024 03:15:20 UTC No. 971694
>>971686
blender was def not meant to handle that many polygons, you also don't really need to go that high imo
Anonymous at Wed, 24 Jan 2024 07:53:15 UTC No. 971718
>>971678
Yeah you definitely have more details. I thought, maybe you're not supposed to brush like crazy across the whole sculpt at this detail, but your strokes are still very smooth.
I'll look into it then. This is not okay at 200k tris, that's for sure. It can be anything, second monitor on, other programs opened, old drivers, particular tablet being bad, sculpt was cut off with box cut method and has wonky geometry at the bottom, anything really.
Thanks for taking the time to open my eyes, something's wrong here.
Anonymous at Wed, 24 Jan 2024 12:26:51 UTC No. 971728
Very nice work anon, theres been a lot of improvement since you've started this thread, don't forget about the ears though!
In a true side profile, the top of the ears will align with the brow ridge, and the bottom of the ears will align with the bottom of the nostrils, +/- 1/10th of an inch or so
The ears will also start at the middle of the skull, and there will be a decent chunk of space from the back of the ear to the back of the skull (>>969610 the distance from the back of the ear to the back of the skull in this sculpt is the best)
The overall shape of the ear can also be simplified fairly easily, the outside of the ear will have a ( ? ) question marked shape, while the inside of the ear will have a somewhat sharp ( y ) shape carved into the inside, while the rest is mostly hollow - The peak on the body (Where the two lines fork together) of the "Y" and the "?" are roughly the same height, sometimes from the front view you can see the "Y" and sometimes you can see the "?", but they are generally very closely in line with each other
Also from the front view you will see almost none of the forward part of the ear, this is because the area where the ear attaches to the skull is caved in slightly, making the cheekbone slightly wider than where the front of the ear attaches at
Its also important to know when looking at reference that photographers and models almost never take true "Side profile" photos, the models head is usually turned in some way, some good rules of thumb to keep in mind are - If you can see the opposite side of the faces' eyebrow, the models head is turned towards the camera, and the ear will appear to be further back on the head instead of towards the middle
The lower eyelid should be slightly further back than the upper eyelid - If the upper eyelid is behind the lower eyelid, the model is looking up towards the sky, and the ear will be lower as a result
Anonymous at Wed, 24 Jan 2024 13:02:11 UTC No. 971730
>>971728
Thanks! For kind words and the information.
I haven't listened to ear lectures yet that's why I've been ignoring them mostly. All I know is that the earhole is in the middle of the head, slightly lower than the zygomatic arch and just behind the ramus.
I'll combine knowledge you provided with what I learn next in the course, and ears on the next sculpt will be much better.
Anonymous at Wed, 24 Jan 2024 13:09:03 UTC No. 971731
>>971718
I don't know what you got going on over there. But 19gb of ram usages suggests a lot is happening. I only have 16gb myself. With firefox open(only 4 tabs), steam open, and blender open, I'm only using 37% of that. I opened up my sculpt, which is now at 994k faces.(It's full body now) And with that, only 5.9gb used.
What's gobbling your ram? I doubt I can help you. Computer stuff is far from my expertise. But thought you should know the ram usage is weird.
Anonymous at Wed, 24 Jan 2024 13:20:49 UTC No. 971732
>>971731
Nah it's normal, I have a lot of work stuff opened. Plus the more RAM available, the more windows just takes for no reason.
I'll test with everything closed though, for good measure.
Anonymous at Thu, 25 Jan 2024 11:20:28 UTC No. 971811
>>971678
>I would like to know the "why" of this,
because brushes react differently to different mesh density and you don't want to be babysitting varying mesh density over large parts of the sculpt. you definitely don't want to get bogged down at these early stages of the sculpt.
op's mesh is too low-res period, no part of it has enough resolution. dynotopoing at this sort of stage would just result in huge patches of inconsistent density.
again, no one's saying buy zbrush (espically not with maxon owning them now) - steal that shit and this just stops being a problem. sculptris is a relatively recent addition - you usually don't need it all in zbrush. scott eaton, in his decade old lectures, for example does something relatively unorthodox: he dynameshes once at the beginning of the sculpt in an absurd resolution and basically takes it to a finished sculpt (most people should probably not do this). zbrush has allowed for this sort of flexibility for a long time.
>>971730
there are no ear lectures in scott eaton course. he talks bit about placement generally when talking about other stuff, but there's no dedicated ear section. not sure if he just doesn't cover it, or if it was missed in the upload.
Anonymous at Fri, 26 Jan 2024 19:06:19 UTC No. 971959
Turns out, I was retarded all along.
Box trim creates a few 9999999999-gons where it cuts. When you have those you have lag, simple as.
They weren't affected by remesh because they belonged to a different face set. Then I clicked something in the Face Sets submenu and now remesh affects all face sets, even after I restarted blender to capture a video.
I'm so relieved I don't have to get into zbrush.
🗑️ Anonymous at Sat, 27 Jan 2024 04:49:41 UTC No. 972000
>>971959
Nice! Glad you were able to figure it out.
Remesh strikes again. I wish there was a way to just turn them off.
I've gotten into the reflexive habit of initializing face sets every time I change anything. Because I just assume they're going to interfere in some way.(they often do)
Anonymous at Sat, 27 Jan 2024 04:50:44 UTC No. 972001
>>971959
Nice! Glad you were able to figure it out.
Face sets strike again. I wish there was a way to just turn them off.
I've gotten into the reflexive habit of initializing face sets every time I change anything. Because I just assume they're going to interfere in some way.(they often do)
Anonymous at Sat, 27 Jan 2024 18:04:29 UTC No. 972063
>>969472
It could be, though most of the talented artists spend their lives, self-worth and credibility designing awfully grotesque models getting excited and penetrated in varying unnatural ways.
Anonymous at Sun, 28 Jan 2024 23:36:34 UTC No. 972177
Slapped some teeth in a hurry.
Decided against going into smaller details for now, the goal is maximum likeness to a reference proportionally.
Idk why am I not able to set up camera path animation like I always did so here's a screenshot.
Next sculpt is going to have correct mouth cavity, teeth and gums will be more well proportioned, and they won't just hang in the air.
Major goal for the next month - likeness to a reference. Then wrap it up and start body even if there's no progress with faces.
Anonymous at Sat, 10 Feb 2024 17:31:49 UTC No. 973546
Idk, 3 hours in, have been playing around with proportions, couldn't grasp them.
Too tired, gonna fix it tomorrow.
Glad I have coding experience under the belt which tells me it's okay to be slow and even hopelessly stuck, it's gonna work out in the end one way or another.
Anonymous at Sat, 17 Feb 2024 22:35:35 UTC No. 974461
Nah it's pointless, not one confident stroke was made during these four hours.
Okay, plan B, rewatch every lecture and this time do screenshots and take notes.
Also, do you see a temporal line on the reference? Because I don't.
Anonymous at Sat, 17 Feb 2024 23:43:41 UTC No. 974472
>>973546
I think you should watch some speedchar head videos, his heads look nicer than that after ten minutes.
He works low-res at the start, so you can probably copy his steps without too much trouble.
The basic proportions are something you can practice with very low resolution in quite short amounts of time.
Anonymous at Fri, 23 Feb 2024 02:52:35 UTC No. 975239
>>973546
use freakin refs