Anonymous ## Mod
04/25/17(Tue)16:22:37 No.562047

Anonymous ## Mod 04/25/17(Tue)16:22:37 No.562047

Click here to open this thread.

ALL /beg/ questions belong in /beg/

File: negativeshape.jpg (142 KB, 1955x1235)

142 KB JPG

>>982416
>What tips should I know to help me better understand and visualize the things I want to make

When you're observing a reference of something and making judgement about the spatial relationship between features,
pay careful attention not just to how the thing you are making line up to what's adjacent locally but look at the distance from that feature
to several landmarks around your subject. Finding out that something is off in your representation is easier if you compare it to the whole.

When you consider a curve or an outline of a feature don't just consider the shape of the feature itself but consider the shape of the outline
itself from the negative empty space side. Your brain has numerous preconceptions what a human is supposed to look like, but if you abstract the
shape from the negative side it is easier to see it for the shape as it actually exist in your reference.

That later is a quick trick to increase your accuracy making your eye measurements much more precise by simply doubling the data you're considering.

Sometimes your mind get tired and you need to walk away from a piece to see it with fresh eyes, but if you are under time constraints and need to judge
your shape more accuratly right now you can do things like flip the canvas upside down, invert the colors etc to stop seeing the surface as a familiar face

Comment too long. Click here to view the full text.

ayy lmao

>>980142
Do it with this model: https://3d.si.edu/object/3d/1903-wright-flyer%3Ad8c62e5e-4ebc-11ea-b77f-2e728ce88125

I know you can’t because the scanner messed it up and become impossible to use it.

File: pc88.png (549 KB, 640x400)

549 KB PNG

>>979843
same way they mastered pixel art:
by being a country of obsessed pedos

File: 1531464858239.jpg (1.21 MB, 1920x1929)

1.21 MB JPG

>>979843
It's all in the texture

>>979843
Why is it so hard to find lowpoly mannequin rigged?
Every time I find sometimes it's either:
>under price tag
>actually has 50,000 faces (but is still described as"low poly")

>>982399
>actually has 50,000 faces (but is still described as"low poly")
Lowpoly generally means anything that's not a sculpt

>>982073
>For some reasons some very, very tame Pixiv artists were b&n earlier this year (and later this previous years). Most have no idea why. We're talking as tame as Ganseki.

>>981485
>does anyone knows what happen to gansekiiwao on pixiv? They nuked his account!!! Where I can find his stuffs now?

I think I may know why. I read their terms yesterday and they do seem to disallow pictures that look "photorealistic" and it says it can lead to the suspension of the account.

>>982214
>>982218
It was an image set featuring a character from a porn VN. They wanted more of her. It ended up being about 30 different images. Told a little story. I don't want to boast too much, but I get told my stuff doesn't look like DAZ, which is probably the highest, albeit, back handed compliment one can get. I do allot of custom morphs using Marvelous designer and Zbrush so clothing and skin aren't so static.

There's allot of money in the VN market. I've seen 50 to 80 dollar commissions routinely go to people using Koikatsu and Honey Select ONE.

>>982396
How did you get approached, like where is your main platform for people to notice you?
Also what is VN?

>>982355
Fuck, maybe I should learn how to make custom Daz sculpts. Last time I spoke to Toto, his queue was 1 year long.

>>982400
VN is a Visual Novel.

I float around some f95 VN servers and post here and there. I'd rather just make a VN and post it to steam and collect residuals than have a patreon.

It's not unheard of for a 2 person project collecting 6 figures in a single year.

>>981937
There is nothing wro

File: B.png (3.11 MB, 1674x984)

3.11 MB PNG

taking a little break from my previous model, having a bit of fun sculpting, currently writing down a few things I've learned from my previous project to apply them on this one.

>>982391
What is this supposed to be?

>>982412
it's a 3D model of a Buddhist monk.

>>982417
He should see a doctor for that thing on his forehead.

File: 1695958764877116.png (700 KB, 755x425)

700 KB PNG

>>961913
Episode 2 is releasing, how will it fare compared to the pilot?

>>982220
I just watched the first e

the writing is kinda decent but some of the humor is trying too hard and the textures and lighting are atrocious, well animated though

Just watched ep2. It's cool I just cant fucking stand the lack of highlight compression in the rendering.

Episode 2 OUT NOW

https://m.youtube.com/watch?v=4ofJpOEXrZs

>>982367
No, Cris. The bestest are the jannies.

>>982368
No, you are the only person who is the best

>>955958
This is the best one

>>971886
This is the second best one

>>982371
:(
I will try harder next time

>>982374
here's a tutorial if you need it:
https://youtu.be/B0J27sf9N1Y

>>982370
Looks great. Good job, anon!

>>982370
insanely real, rispect, if you've reached that level of realism, go the extra mile and get inspired by this type of dough >>982326 to improve it even further

File: 10.png (1.47 MB, 1080x1080)

1.47 MB PNG

Guess we're all doing donuts now

>>982392
i actually give up on sculpting, too tedious compare to normal modeling

>>982010
>>982011
It's nice that you agree with yourself.

File: preview.png (832 KB, 1280x728)

832 KB PNG

>>980013
this was my first attempt, never tried again

File: topologySucks.png (955 KB, 1920x1033)

955 KB PNG

>>982405
...still suck at topology :3

>>982406
>>981725
Use front and side references from here until you develop understanding and u own style

>>982081
The way I do it is the follo

>>981282
That's not slop, are you cra

>>979039
why not render directly in houdini with arnold or karma?

File: file.png (17 KB, 250x208)

17 KB PNG

big coomer energy on this talk:
https://www.youtube.com/watch?v=rPdOjkEwxQM

>>982401
still far far far too complicated and they never discuss setting up the face or integration with mocap. Its like they don't actually want anyone to use it

bump

>>976621
Don't jump directly into the high poly/subsurface model. Do your low/medium poly that has EVERY detail you want in the final product, with decently clean topology. Only when you're COMPLETELY done that, duplicate all the parts and start working on your high poly (should be pretty quick since you've already planned out all the geometry) with any necessary supporting geometry and edge loops.

Lots of people jump right into trying to make things with subsurf from the get go and work themselves into a corner.
There are a handful of common 'problem' shapes that you should probably learn, things like circular holes in curved surfaces. They're not so bad once you've practiced them a little but if you don't know them you'll end up using way more geometry than is needed and your mesh will become difficult to work with.

File: lines.jpg (1.29 MB, 4000x2828)

1.29 MB JPG

>>976621
Line work is the base of any car model.
Start with line work. Lay down lines and curves in 3D space according to your blueprints. Start with lines that are present in a lot of views, these have the greatest "confidence" and they'll help you lay out other lines.
Be aware that most blueprints aren't 100% accurate, so one point might be off in one of the views.

I have placed colored spots in pic related, these lines (and all of the wheel well edges) are the ones I would start with.
Keep in mind that lines which aren't connected to other lines could be "reflection lines" - these are unimportant and should not be modeled, at least not right away.

After you're confident with all your lines start connecting them with faces and work out the surfaces. In this stage blueprints won't help you as much, go online and start looking for photos, full body and close up, where the paint shows the surface curvature.

Don't do panel work (cutting out doors, roof, fenders etc) until you are 100% satisfied with the proportions and topology of your mesh. Cutting out panels is the last step.

>>981472
That's useful knowledge. Anything else I should know? I want to learn low/medium poly modelling because from what I have seen, you can at least add a modifier to your mesh if you want to have high poly. One of my mates does low poly SubD modelling and they say you can edit it any time you wish to remove all the modifiers and turn it into something else

>>982346
>you can at least add a modifier to your mesh if you want to have high poly
that's correct, but then you're in topology territory and you'll have to pay extra attention to how your edge loops flow in order for the subdivision algorithm to give you a nice looking mesh. Cars are among the hardest subdivision models to get right, if you're starting out I'd stay away from that for now and focus on getting a nice low or mid poly shape out first.

>>982345
Poser and Bryce weren't nearly as popular as AI today.They were only known in the relatively small CGI community back then and not used by every normiefag like AI generators today. But it's true using Poser was looked down at by a few of those that could even model, rig and animate a character.

>>982345
For most of the zoomers participating in this “movement”, it started when this particular video showed up in their home page on YouTube [https://www.youtube.com/watch?v=JK8ilaPZbKE]. This isn’t a guess, this is THE answer to the question from someone who’s in the know.

It may not mean much to you since this was just the mundanity of everyday life back in the day, but I think it’s something special when you look at it with fresh eyes. There was something really magical about watching this video for the first time, with the ethereal music and the surreal visuals. For people who grew up playing low-poly games, watching 3dcg cartoons, and looking at the Windows pipes screensaver, “bad graphics” is something they can look past and actually appreciate what’s there. These images could be recreated using modern software to get something that looks cleaner, but then the dreamlike quality they have would be gone.

For a moment, imagine being a teenager during the 2010s. Most of what was at the forefront of “aesthetics” was corporate minimalism, mobile games AAA realism made by people with no artistic sensibilities, and pop stars and rappers encouraging people to be degenerates. After a decade of anti-aesthetics, I like to believe a lot of young people are copying things like old CGI, Y2K, Frutiger aero (dumb name), etc. because the things they experienced during their childhood were legitimately better in all regards than what they got later in life. I suppose the 2010s had vaporwave, but that got old quick when everyone started jumping on it without any creative ideas, just like what’s happening with this. Ultimately, that's the problem with OP’s pic. It could be due to a lack of ideas, but I could also just be because he doesn’t know how to use his program, and is limited to only the most basic tools.

>>982369
Have you tried getting a degree in psychology instead of posting?

>>982373
No, what for?

>>982369
I'm the guy you responded to, thanks for providing a legit sounding answer. I suppose the people nostalgic for this are the ones who are maybe half a decade younger than me and this is part of their connection to what then remained of a 'pre-internet' type world.

Obv the internet was already there for them but it was in it's infancy without the social networks, tubesites and smartphones etc.
I guess I understand what they say when talk about how legacy media and games had more 'soul' I'd tend to agree with that statement.
Where I'm lost is when they point to things like 'Frutiger aero' and 'Bryce 3D' type aesthetics as representative of that 'soul'
because to me that would represent the peak 'soul-less' of that era they're reaching back to.

Maybe it's more about having something easily reproducible that's just iconic for the time. Like how ugly juxtaposed neon colors & slalom patterns can invoke certain 80's nostalgia.

I may be saying something retarded cause I legit don't know 3d other than the most basic stuff
but shouldn't OP be exploring shaders?

>>982329
>I may be saying something retarded
yes you are, anon

>>980532
He needs you clothes, boots and your motorcycle.
Don’t burn you passion up. Doing this every day for 4 years sounds harrowing, even with such time that’s probably not enough man hours to get such a job done.

>>980473
>"3D Animation and Game Arts"
Which university? Why did you pick that university?

>>982365
It seems like op doesn't have the fundamental workflow down. His high poly and renders look good, but ignorance about correct workflow and best practices is going to fuck and burn him out before he ever reaches the animation stage.
Op, I recommend you check out courses for creating game and animation ready models, before moving onto basic animation in maya. Spend an intensive month or two on them. You'll acquire the core skills, and it'll allow you to be more confident and waste much less time pursuing your berserk project. Nevertheless, I'm rooting for you!

Can't wait to report your next github

>>981667
Can’t wait to see you disappointed when I unleash my next GitHub.

File: Axolotl Men.png (970 KB, 3000x1640)

970 KB PNG

Axolotl Men.

File: Fygoon’s Nemesis.png (104 KB, 3000x1640)

104 KB PNG

Here’s the enigmatic Warlord, the main antagonist of my game and Fygoon’s nemesis.

File: Invasion.png (1.84 MB, 3000x1640)

1.84 MB PNG

Finally made a new render, and it’s just to shit on the /vg/ crowd.

>>982377
>>982378
that sounds sensible, thanks for the help

File: dzung-phung-dinh-supporti(...).jpg (1.06 MB, 3840x1529)

1.06 MB JPG

>>982379
i found these images by some guy on artstation to be the best reference for what you need to do, it isnt necessary to copy this 1/1. some basic things i learned after doing my own style based on this
https://www.artstation.com/artwork/eNq43
more here
>loops around the major points of deformation
>clean planes with all quads to fill the areas between loops
>where you have 3 edges connect to one vertices (planes converging from different directions like a cube) you will end up with a 5 edge connection close by
>dont go over 5, creates too much stress, this is achievable
>VERY IMPORTANT: do not create loops with odd number of vertices,
now what I mean by this is that if a loop exists on a shoulder and the loop at the wrist, you are going from a large area to a small dense area, you will need to reduce the density of the faces and verts, this is achievable with clean topology but only if you dont go from even to odd numbers or vice versa. If you 32 verts in a loop around a thick thigh, but go down to say 16 verts around the knee, it will be possible to connect them with all quads using some 5 edged vertices, this also is relevant with shoulders and hands/wrists where you might end up with 1 vertices left that you cant connect cleanly if you dont have even to even loops. i hope that makes sense
>ALSO IMPORTANT: do not create vertices with more than perfect squares around the center of the body
make sure there is only one "horizontal" edge going to the center verts, if you have two converging edges going into a center vert it will create a "spider web" effect when you mirror it and end up with 6 verts connecting into the center verts
I hope all of that makes sense

File: uploads_1567094114010-Qua(...).png (60 KB, 1280x989)

60 KB PNG

>>982382
>>982379
heres what i mean by reducing topology from a large are to a small area, you can see that it where you have 8 to 6 verts, or 5 to 3, or 10 to 8. its just easier that way

>>982383
>>982382
>>982379
shouldve also said that its because when doing fingers you will probably do like 6 or 8 vertices loops, to get a nice equal amount of deformation, and texture size, etc. so the rest of the body will need even numbers too

>>982382
>>982383
saved and screencapped, thanks