[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k] [s4s] [vip] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Home]
Settings Home
/3/ - 3DCG

4chan Pass users can bypass this verification. [Learn More] [Login]
  • Please read the Rules and FAQ before posting.
  • There are 72 posters in this thread.

10/04/16New board for 4chan Pass users: /vip/ - Very Important Posts
06/20/16New 4chan Banner Contest with a chance to win a 4chan Pass! See the contest page for details.
05/08/16Janitor acceptance emails will be sent out over the coming weeks. Make sure to check your spam box!
[Hide] [Show All]

File: 1405555530237.jpg (613 KB, 2560x1440)
613 KB
613 KB JPG
Old one is on auto sage so I will create a new one.

I have a nice rigged model with various animations in a fbx files. However, it has too many vertices.
now, how the hell do I split the mesh in 2 or 3 parts without breaking the skeleton in 3ds? I know how to split static meshes and they work just fine.

can you list the steps in a complete modeled architecture to ue4 pipeline?
just keep proper scale and use grid snapping.
id recommend to keep things like floor tiles modular if you can unless you want to set tiling of floor texture to one object and it fits well.
there is a tutorial on youtube, it will cover most of it
Is there anyone here who can instruct a retard in how to make rigged ragdoll objects in Cinema 4D?

This program is fucking horrible and I recommend none of you ever use it for character animation, because I'm almost positive if I just took the time to learn Maya, I could accomplish this in like three fucking button presses.
I haven't been able to animate for half a week, because I've been trying to figure this shit out, but at this point I'm nearly at the point of just giving up and animating it manually.
I am very angry and frustrated. Please excuse my faggotry.

The specifics:
I have a rigged model with polygon hair, that I want to make simple fucking IK chain dynamics for, with simple collision and gravity, so it doesn't clip through the model.
The polygon hair has a ridiculous amount of joints controlling it, so I'd REALLY want to avoid having to manually animate that if possible.

What I've already attempted:
(cont in the next post. I maybe should've made my own thread for this.)


- Cinema 4D's built in IK dynamics, using the built in auto-collision system with a proxy mesh for the body
This is comically unreliable and inconsistent, to the point of being near fucking useless.
Each rectangle is weighted to a single joint, for easier viewing.
These are all the same exact settings, and SHOULD result in the same simulation.

- Proxy meshes for the hair joints, making the proxy meshes rigid bodies, connecting them all together, (affected by gravity, and collisions), and constraining each hair joint to its proxy mesh counterpart.
Also general Cinema 4D fuckery when it's applied to a rigged character.
There are two tutorials in existence on this.
One is for a ragdolling a rigged model using a proxy mesh, but it's by some mouthbreathing autistic fuck who can't articulate himself worth shit, so I couldn't understand it.
The other is by some british dude, but it's for more traditional ragdolling, so the parts do not behave as I would like them to.
This is probably the best option, but I don't know how to do it and make it look not terrible (Parts would rotate indefinitely for no fucking reason, even when constrained to a 1 degree angle limit, the proxy mesh would sometimes just explode for no reason, the actual polygon mesh would shake like it's having a seizure, generally awful shit)
Also it takes goddamn forever to set up something that should be relatively simple.

then why don't you just learn Maya?
No. I will eventually, but that wasn't even remotely the point of my post.
I also refuse to believe that a software like Cinema 4D doesn't have the capability to accomplish this.
sounds like you're setting artificial limitations on yourself my dude. one of the things I've learned in my short time working with 3D applications is that there is no one tool that gets every job done well; why not just put your stubbornness aside and save yourself the massive headache that you're putting yourself through?
Z brush camera navigation has been a bitch.
Been doing right click, because being limited to having to click empty space to move around is dumb. But right click has its own problems, letting go of the modifier key before the right click makes you right click. And the camera itself is getting unusable, I'm noticing that it literally zooms in instead of moving the camera nearer. So I can't zoom into any detail work if there's anything between that spot and the far off camera. Really missing blender's middle mouse navigation, and centering the camera around a selection.

does anyone here use Nvidia Tesla cards in their computer?
I guess I should really work masking what I'm working on into my workflow anyway
Can someone give me some insight to what I should pursue in college? I liked CAD in high school, as in, designing 3D shit inside of the program, and I could see myself pursuing that sort of major, but the problem is, most schools near me have these really long, extensive programs that are Industrial Design or Architectural Design, and I don't know if those are right for me, when I have the option of attending another community college for another associates or certificate in CAD for much, much cheaper. I feel like this would give me a leg into the 3D world and maybe I could use my career to transition into some other parts of a more creative sense.

Any thoughts?
3d art is not even guaranteed work. the state don't even recognize it as a degree. you could learn art everywhere and it wouldn't matter as long as you have good works.
forget about 9-5 job with art, its chaos. get a safe degree and don't play with your future
A dude in another thread was literally just telling me to ditch the idea of 3D art in favor of CAD work or more practical designing such as floor plans or interiors, due to it being more lucrative and assuring people jobs. I'm saying I could still hang on to 3D art as a hobby but really focus on CAD work instead.

Are you saying CAD work isn't needed or is a bad decision?
cad is way better for job security.
i don't really know about the education part but i do know that electrical is a pain in the ass. the quality of your certificate shouldn't matter alot either i believe but you will absolutely need it
So you think I could break into the industry without actually needing a degree for CAD? There's only 1 uni near me with a 'real' design program, but like 20 community colleges near me offer certificate programs in CAD. I would imagine it's more of who you know though, right?
if you are unsure.
google "cad jobs in x area" then google "3d artist jobs in x area"
look up the requirements too
Is there a comprehensive wiki with all the 3D cg terms explained?

I'm an architect so I know my way around modeling, but textures and materials are pretty alien to me and I'd like an overview to just read up on the different kind of maps and shaders and so on.
The link in the sticky is broken and old af

I want to learn UE4 and it looks very fun to make my own materials with substance or quixel but I lack a lot of basic knowledge
Thanks m8

Also is anybody here using Rhino? Seems like it's all Maya, Max and Blender and I'm curious why
no but same principals apply to all programs.
people that ask advice in blender get tips from max guys and so on
Anyone with fusion 360 experience here? How the heck do I reset the orientation after rotating a component? My movement directions are fucked up. In Max I could just reset my pivots and it'd be okay, but I can't figure out Fusion 360 for the life of me
File: image.jpg (248 KB, 1500x2208)
248 KB
248 KB JPG
Hey, if you're asking what I think you're asking it's easy - in just about every move dialogue you have the opportunity to just move the pivot at any point. There's a checkbox in the move options dialogue box to allow you to work with the pivot - it changes to a green checkmark (pic related) while you're working with the pivot. The pivot will then snap to various points like centre points and corners etc around the scene and will also inherit the average normal orientation of the point you chose which is really useful, or you can rotate the pivot manually. When you're done getting the pivot how you want, remember to click the green check in the options box to go back to translating/rotating scene objects.

Not watched it - but I'm guessing this page explains in more detail.


In addition I'd like to add that I'm happy more and more people are taking up Fusion360 around here after a few weeks of me hyping it up. I dig it.

The polygon modelling tools tho, are about 10 years behind Max and Maya, which I really don't understand - I guess Autodesk don't want to undercut their big earners.

Maya and Max dominate the entertainment industries - moving image and games respectively.

Max, however, is a very very capable Archviz platform. Maya - not so much. In fact, not at all. Max can import and work with CAD data from Autocad and Rhino and other platforms in many useful ways that Maya simply doesn't.

Also, Max has a HUGE industry attached to it for 3rd party plugins and utilities that have basically become essential to archviz artists - RailClone and ForestPack being the two main examples. Definitely research those two plugins if you haven't already. These aren't available for anything but Max. Same with Batzal Roof Designer, RichDirt and a whole shitload of other things that make an archviz workflow so much easier.

Also - I may be wrong but I don't think Corona rendered is available for anything but Max (yet) and Corona is shaping up to be the 'one-click = amazing results' magic wand that visualisation artists have dreamt about.

The dominance of VRay is being called into question by Corona. And it's still cheap because it's new. Look into it.

Substance and Quixel are geared towards creating lightweight real-time texture assets primarily for games, or at least for things that are going to be moving.

They won't provide an Archviz artist with the eye-popping sharpness, detail and clarity that defines quality archviz work... Yet.

Your time would be better spent learning RailClone and VRay and all the millions of plugins and tools dedicated to archviz work - like the free multi-texture map plugin from CGsource, just right off the top of my head. Again - Max only.

If you've already become confident with Autocad - Max is not gonna be too much of a challenge at all.
If you are an Archviz artist and can't produce eye-popping sharpness, detail and clarity with Substance or Quixel you're just a lousy Texture artist.
You're blaming the tool, when you should blame the artist.

Substance, Quixel, Mari, 3Dcoat all are very useful for creating High fidelity textures, no matter what for.
hey, thanks for the reply

I've not tried Autocad since my office has been using Vectorworks for decades, but I've done a little bit in Maya and got an introduction to Max once, but I'm mostly using Rhino these days.

I'm not so much interested in classic archviz, as in producing photorealistic stills. I'm more interested in creating realtime, possibly even VR applications where clients can walk around and manipulate the project in realtime (move furniture, change colors of walls etc.)

I will definetely look into these plug ins tho, sounds interesting.
i just use the move scale and rotate menu items on the right when i zoom in too close to drag over empty space

I believe he's referring to their resolution constraints. I don't know about Quixel, but last I checked Substance Painter only went up to 4K texture resolution and didn't support UDIM. Mari is the better option atm for those needing tons of extremely high res textures (though it certainly has its own drawbacks).
File: 1.jpg (449 KB, 2154x736)
449 KB
449 KB JPG
is this acceptable if i'm trying to save on polycount for use in a unity game?
Thanks, it's a little fiddly but I got it to do what I wanted. I still don't think it's as robust as Max's options but oh well.

I mostly got into it because of 3D printing. It seems like it's popular in that scene and it'd be good to be able to share my work in a way the community can easily reuse. I don't see the polygonal modeling tools being extended too much, it just doesn't seem like it's meant for artists that would use polygonal modeling. Parametric solid modeling or w/e they call it is pretty powerful in its own right and seems more than capable of serving any and all CAD needs.


A texture for just the ground plane in a photorealistic wide shot that is going to be rendered at over 1000dpi and printed in high quality at large scale (we're talking enormous, poster-size prints) needs to be over 25k, preferably over 30k.

Where's the setting in Substance to allow me to create a 25k map? I guess the versions I've used that only go up to 4k must've been broken.

Btw congratulations on having a rig that's capable of interactively presenting a procedural 25k map, it must've cost a pretty penny!

Oh... wait - I think you said all that incorrect stuff up there because you forgot that some 3D artists have to produce images for print didn't you? And your understanding of the level of quality needed in print work is minimal at best, isn't it?
Maybe shut up and listen when it comes to things you know fuck-all about next time.

Can't tell if troll.
File: texture maps.jpg (17 KB, 400x194)
17 KB
What is the best explanation of the various texture maps (normal, occlusion, curvature, ect) you've seen /3/? Youtube video, article, book, whatever.
>enormous, poster-size prints

When does that ever happen that it requires such high quality?

The largest "archviz" I've ever seen are those on the signs posted in front of construction sites, and those are just upscaled regular renderings and look like ass most of the time anyway.

At pitches for large budget contracts, some architectural firms have been known to produce A0 size c-type prints. These tend to be from files that have around 1000dpi resolution:

So back to my original point - Substance is not going to cut it for this. So shut up.
I'm a different person m8

I would not classify A0 as enormous tho. But it pretty much depends on the individual shot anyway. If you have a wide angle shot with a lot of ground in it yeah, you might need a big ass texture - for the ground. But not for everything else.
anyone know where to get x-particles 3.5 for c4d for free? ive looked fucking everywhere
probably not seeing as this is a thread for questions. When people don't know what they are doing they usually ask questions about it.

Fair point, but a question thread could also be the perfect place for casual bit of trolling couldn't it?

Whatever, I don't know, I'm not a troll and I don't really get why kids on the internet find it so fun to say incorrect stuff or ask dumb questions as a setup to laugh at people who answer or try to correct them. But they do and knowing that causes me to second guess a lot of shit I see people say online.

A0 is pretty fucking enormous when you're talking about 1000dpi, yeah.
Well yeah

But then again, who will even be able to tell the difference? Is it appropriate to create huge textures for a client who doesn't even know what a texture is? Who will probably not even notice the ground unless it looks like utter shit?

The issue depends so much on client and individual shot that it seems silly to condemn a tool because it might not be useful for that one specific niche job.
how would I get started with animating 2d images like this?
File: Untitled.png (15 KB, 546x524)
15 KB
no one answered my question and i want to learn more so i can get better so im going to shamelessly repost this

>Just threw this together, never really did anything with 3d before, but im preparing myself for a class im about to take by getting some practice in.

>When I switch to object mode from edit mode, why do these lines stay there? I don't think I have any duplicate vertices or lines there. Did I do something wrong?

this is not a question about how bad this is. i know i still have a lot of room for improvement. thanks in advance.
nothing about that is 2d. all 3d with toon shader
oh really? welp
why do you need the vertex in the center? wouldn't one large polly be lowest possible polly count?
>all 3d with toon shader
And you're basing this assumption off of..?

That artist uses proprietary tools that they developed and refined over the course of like 10 years or some shit.
We don't know what it is, but it's definitely not just 3D with a toon shader.

Shut the fuck up.
you have nothing to worry about.
just keep making models and uwnrap, you will figure eventually what is happening
File: 1453532735420.jpg (37 KB, 528x480)
37 KB
What do you do when you have no inspiration to do anything?

I have tried looking at ArtStation, dissected Junji Ito's works, dissected the workings of a Game Boy, watched numerous tutorials of various techniques for gamedev and cinematic asset creation, scrolled furiously past high fashion designs, looked at the upcoming indie games and their progress, looks at the upcoming AAA games and their promotional works, listened to a plethora of musical genres begging for something to ring a bell in my head.

Yet I'm still completely BLANK as to what to do. I mean I work at a production studio which produces assets for cinematic and AAA games, so (I think) I technically have the capacity to do something, but I simply don't know what to do. Environmental? Character? Mech? Prop? What?

tl;dr How do I reacquire my soul and conjure up an idea on what to model
you just gotta grow some balls and do whatever you feel like.

for example right now im doing a big fucking project and im not remotely confident about it, so i check export to the engine and redo to the models to see how it looks in different renders,different shaders, different light setups.
and hell who know it might end up looking like shit, but at least the next time it gonna look better.
don't constantly look for reassurance, its a mistake that alot of people do. you need have the essence in your style and take it all the way.
this is how it is.
you need to accept that you will be just shit for the first year or two, even if your'e a great artist
2.methods i would recommend

1. just start creating without any intention.
Do it as the Surrealist do - automatic without interpretation and thought.
Just let it flow freely.


Go into nature and eat some shrooms.
Do not ask for a vision or creativity, rather make yourself empty (like in meditation) and observe what is going on (in your head)
It sounds like you're already making stuff. Or are you not involved in those assets and was irrelevant to bring up.
You're giving yourself too much freedom. You don't know what to model because you're giving yourself too many options.
Set your own limit; decide what kind of thing you're going to model first. Environment, character, etc... Just choose one, don't think too much. Simple.
Then try adding more limits: what setting? Going for a certain mood?
Eventually you'll get an idea of what to model because you're thinking in a way more confined space, instead of trying to think about everything at once.

As for the thinking itself; look up diffuse thinking. It's basically the state of thinking where your brain makes connections out of things that aren't connected at all.
This happens most easily when the mind is relaxed, free of stress.
It's the reason why I get my most creative ideas when I'm dozing off or just waking up.

Good luck bud, you'll get through it. Just don't stress too much.
It's something we all go through form time to time.
how many game devs take psychedelics?
it doesnt seem like an industry like music where drugs are really celebrated
just curious if anyone has any fun facts or stories

Certain kinds of dmt arent too overpowering and last for a long time. And it makes everything feel waaaaay fucking real. This really helps you make seriously effective game design choices that engage people, and makes you feel a thousand times more immersed in the game universe, so you can see what it would 'really actually be like' in the game world.

>take drugs to receive inspiration

you nutcases shouldn't give advice
if you dont take drugs you should take all your media and burn them because thats where they came from
i smoked pot for 5 years, you can't tell me man

a little reefer for a moment of inspiration is fine, but you don't do party drugs when you are at home working.
File: 1473550131786.jpg (48 KB, 500x375)
48 KB
I have a fbx file of a dragon. the mesh is rigged and has skeleton and everything. now I just want to create animations as simple as possible, bake them, export them and use them in ue4.

looks like motion builder is the best thing for my purpose. and I could do key framed animations, but I would like to use the solver (same as in maya).

I can put all the bones to the rig it wants. but how the fuck do I tell it that its a quadruped? it only accepts the rig if its in biped fucked up t-pose.

and how the fuck does attaching extensions really work? I want to add for example the tail. do I have to add every single bone and node in the tail or just the head of the hierarchy? or do I have to "re-rig" the model in maya again? does it lack information? Im kinda lost here.
Not sure if this belongs in /3/, but I'll ask anway.

Do you think it's possible for a bunch of middle aged architects to transition from 2D CAD to something like Revit in a reasonable amount of time?

I'm doing archviz and it absolutely sucks ass to build 3D models out of the 2D drawings I get. With 2D you can draw crude as fuck and it won't matter, especially not in print, but in 3D you immediately see every little error, so most of the time I end up redrawing most of the plans before actually modeling anything. This is obviously a huge waste of time, so I want the firm to move to Revit or other BIM solutions. That way I can take the 3D models, fine tune them and go from there.

But I can't really judge how hard it would be to learn a 3D software for someone who has used 2D CAD for their entire career..
Im working on a sculpt. I learned about ptex textures and wanted to work on it in mudbox, but the old blender sculpt pulled too many errors. Re-doing it in zbrush hoping it'll work better, im just curious if i got the right workflow planned out. Do I import the zbrush model, subdivision leves and all into mudbox, then ptex setup, then texture paint the highest subdivision level, then bake out a displacement map. I dont need a normal map if im using displacement right? Then i dont really need the higher subdiv levels, just render the low poly one with nodes for displacement and texture. Can hopefully rig it and everything at that point. Trying to learn a workflow for movie quality animation, or at least character creation with animation focus.
ptex died a quiet death
File: 1483520451149.gif (1.48 MB, 268x270)
1.48 MB
1.48 MB GIF
how is the hair made ? are those all singular strands or did they use alpha ?
did they use another material just for the lips or was it a specular map ?
a combination of sculpted hair and alpha plains

a pain in the ass
SSS for the lips i believe, handmade highlights either spec or other something
that would create a massive ngon which unity will split up into an ass ton of triangles. this would be to keep it from doing that
the op wasn't 2.5D lol. they look completely different.
As far as i know pixar and disney are still using it. It looks pretty great for non realtime animation.
literally no one else is using it
>no one
are you baiting?
it depends, are you?
perhaps you should quit posting if your bait is this weak
thanks, I think I will
Don't know why. I'm gonna try it at least. I started doing regular UV map with substance painter, but didn't like the resolution limitations. mudbox ptex setup can let me do absurd resolution, with variable resolution for any face or group of faces if I want say reasonable resolution for the body but crazy detail for the face.
yeah no.
feel free to stop posting any time you want. Stop spreading misinformation.
why not.
mudbox is like after effects but worse if your little babby mind can comprehend this analogy. Mari is like nuke.
ugh. I seen mari had ptex tools too, mudbox had more tutorials and results when I searched though so I didn't look into mari too much. too many programs for every step of the 3D process.
Why did the art get worse?
File: decapitate.png (1.31 MB, 1181x719)
1.31 MB
1.31 MB PNG
so dont mess with dynamesh or zremesher when you've masked in zbrush, it irreparably separates your geometry.

Also anyone know if or how zbrush remesher can retopo higher than 100k faces?
File: 1483747844963.jpg (1.42 MB, 2800x1600)
1.42 MB
1.42 MB JPG
Did this project ever get finished?
i don't think many artists even have the balls to approach this.

and if so. you would have to preaper the asset library like few months before you even start adding detail
It was a big collaboration project on here quite a few years ago, and last I saw (a couple years ago) the modelling was over half way done IIRC.

You can use close holes on that. DESU Zremesher got fucked in later editions of zbrush. I use 3d coat now, much more control. I hope they fix it in 5.
tried that, but it didn't close the gap, just capped off both ends. It's alright, I also found out dynamesh destroys subdiv levels so I'm starting over anyway.
It was never really finished, then some guy took all half-finished pieces and threw it together to make a cheap and ugly animation, pretending it was all his work.
I saw it and it wasn't pretty.
End of story.
File: wip.webm (2.35 MB, 600x600)
2.35 MB
2.35 MB WEBM
Any tutorials on how to git gud at facial animation?
1. Practice more
2. Use reference
3. Once you're good enough, break the reference
4. be better
File: 1352339719714.png (41 KB, 386x387)
41 KB
in motionbuilder how can I make that the bone length cant change, as in when translating a node? it seems that something so basic that no one cares to mention it. like the range constraint but with a min == max length (while range doesnt have a min length for whatever reason).
and even when using adding some ik constraint like the chain it still deforms the bones. do I really have to constrain every single bone with the relation constrain for that to work?
Google explains it well enough.
in UE4 is there a way to have light shine through curtains?

does ue4 have a shader for see through light?
i'm a unityfag but you could try subsurface scattering
File: helpobiwan.jpg (68 KB, 957x1300)
68 KB
the crotch part with the flat arching shape, what's the easiest way to model that? I can't believe I'm struggling so much with it..
Booleans nigga
Make the crotch part fully round, then subtract a tilted box from it.
Then just do a little geometry cleanup and voila.
It's been about a year since I used booleans, back when I didn't know what topology was. Apparently they're quite useful when you understand a bit more.. it's worked great, ty anon, much appreciated.
Okay Beginner tier Blender Question

I want to add mirror modifier to a cube so the cube gets well mirrored

"whats the prob"

Well it doesn't mirror the cube
I added the effect
and I added a line in the middle of the cube using ctrl + r
So I know it isn't a normal issue

I look online for various answers for 2 hours and watch countless videos I don't get what I am doing wrong

Can someone with blender link me a basic scene "empty" with half a cube inside and with mirror modifier?
I just want to build a basic character from the cube, is it too much asked? I restarted my client several times and I got it to work once, then I misplaced the spawn point and I didn't know that shift + c resets it so I redid the project without saving, I do not know what I'm doing wrong
>file name
If you can't name this drink your not an alcoholist

Wow it's actully spelled ALCOHOLIC
I'm going to model without it I can't seem to activate it

Is it recommended to use it?
change to mirror from X to Y ,most common troubleshoot
Thx It Fixed it to bad I made a unqiue high polygon model

anyway redoing but low poly this time thx (reason been at this for hours taking to long)
I need a wireless mouse and keyboard suitable for 3D. Any recommendations?
File: CaptureX.jpg (91 KB, 1170x880)
91 KB
Is anyone here using Fusion360? I don't understand completely those Joints commands.
For example: what joint should I use to make assembly like on the picture?
The picture is from solidworks and there is very simple. Use coincident mate between bracket and alu profile, coincident mate between bracket and the other profile and Width (advanced mate) to centre it to the middle of the profile.
How would you make this in Fusion360? Especially the centering of the bracket to the middle of the profile.

What I want is that bracket stays centered in the corner even though I might change dimensions to the profile or bracket.
File: 1479198725158.png (110 KB, 1065x481)
110 KB
110 KB PNG
Retarded newbie here.
Is there any good explanation when to keep edges and when to delete them? Model on the left intuitively feels wrong but I have no idea why.
floating verts on the left. The right doesnt have triangles which is also bad
One on the left has an 8 sided shape (n-gon) in it.
One on the right is all quads, which will generally always be preferable. (Tris are fine too in certain situations).
On a flat surface though, I'm sure you could hardly tell the difference at all between the two. Avoiding n-gons is just a good habit for when you're making shit that will deform, because then it WILL cause you problems.
Also I think game engines throw a fit when you give them objects with n-gons.
>The right doesnt have triangles which is also bad
Why is that bad?

Thanks for the explanation, anon!
Hello, I downloaded Maya 2017 yesterday and was trying to play around with Bifrost liquid emitters, rendering with mental ray. Mental ray works beautifully until the Bifrost liquid comes into the scene. When I try to use mental ray on a scene with a Bifrost liquid, pic related happens. I have searched for an answer online and found a person with my exact same problem in Maya 2017, but nobody has replied to him on the forum. I am not a wizard when it comes to animation and modeling yet- just a pleb trying to understand if this is a stupid mistake on my part or something that I have no control of. Any suggestions would be helpful.
why are you using mental ray when arnold is built in to 2017
probably for the same reason anon's OS interface looks the way it does
I was just following tutorials and the person making the video was using mental ray as the rendering utility. I understand that this can be done (probably in a less painful way) by using Arnold but I was just curious as to why this was happening.
its built into it, but it doesnt come installed into it comrade
File: what the fug is happening.png (411 KB, 1025x1212)
411 KB
411 KB PNG
how do i fix pic related ? it happens in marmoset, and the engine i'm porting the mesh to (gamebryo), but it doesn't show up in 3DS max (in the viewport or when rendering).

i have no normals on the model, so the problem doesn't come from there. the smoothing groups work fine on the whole model except for this part.
if it's only broken in one part, have you tried deleting that geo and remaking it?
yes, doesn't work.
I'm on Maya 2016. Is it worth the download to upgrade to 2017 just for arnold?
Also does the student edition have arnold on it?
tell me something autodesk fags

why don't maya simply auto update your product and why they have to use 2016-2017 meme everytime it gets a few fixes and additions? does the jump from 2015 to 2016 really justifies it?
What 3D package auto updates? This is the standard way of doing things.
After a bit of googling, it also seems that Maya 2017 can't do batch rendering without watermarks unless you pay for an arnold license, which is hilariously shitty, so I'm sticking to 2016 anyway.

In this case, I am glad it actually doesn't auto-update, because I would lose the ability to fucking batch render.
well not directly patching your stuff which can be destructive. i meant like downloading the installer straight from the program like unreal does
If im unwraping a model that is going to be rendered in Vray, how much space should i leave between uv islands if im going to be using 4k texture? Should it be less than if the model was to be rendered in a game engine?
You didn't google properly because if you did, you would know that there is a new feature called Render sequence, which does the same as the Batch Render just without watermarks.
Hating on Autodesk is totally Ok, but your reasons should be based on reality not your own lack of knowledge.

well , i fixed it. i had to chamfer the border of the hole.
In Maya, what is the easiest way to make it so only the part of the object's surface hit by bifrost fluids becomes a different color?
research wet maps, though i don't think bifrost natively supports them so you'll have to get creative
Do you guys ever team up with programmers for projects? I'm currently a programmer with a dream game idea but I'm artistically retarded and it's simply too late for me to start the monolithic task of learning art to create my own shit.
tried several times and hit a brick wall.
exaggerated demands like open world environments,fleet of characters,spaceships etc.
poor communication and not responding to messages,shitty attitude in general,bad scope management,lack of ideas and passion.
make shit tier blocky 3d assets as placeholders. If the game doesn't even play well or the mechanics are retarded then it was never meant to be. Think of it as a proof of concept.
Anyone know where I can get 3D landscapes?
I need a Mars landscape, preferably textured, but a heightmap or anything would do as well. I'd prefer a modeled one tho to import into rhino.

File: Environment.jpg (748 KB, 1928x1042)
748 KB
748 KB JPG
I learned that when you work on an environment, you have to work on it as a whole, and not piece by piece.
So for my current environment I layed out some steps.
1. Gather concept art, do a blockout with modular pieces
2. Make high and low poly versions of those pieces, bake normal and ambient occlusion maps
3. Make unique assets, bake normal and ambient occlusion maps
4. Texture complete environment
5. Do lighting, post processing

My question is, should the baking actually be included in the texturing part?

Currently I am baking all the maps for the modular pieces and I find myself already editing the occlusion maps in photoshop to make sure they are all the same kind of color so the whole environment looks properly shadowed.

This looks like a LOT of work though, so maybe my process is wrong, can anyone elaborate?

Pic related.
I haven't tried it but look up World Machine. Might be what you want.
Looks great, I will give it a try. Thanks!
Thanks. I'll see what I can find. Or maybe just use realflow.
So if I had fleshed out a game using placeholders would that encourage an artist more do you think?
Any way to easily use two or more cameras when rendering in maya?
Say I had one file that held all of my animation, and I used different cameras for each shot. Is there a way to do one render, but switch between cameras at a certain time?

Cinema4D has a stage object that allows you to animate which camera is currently active, which is incredibly convenient, but I don't know if maya has an equivalent.

I know of three ways to do this so far.
One includes using a single camera and a stepped keyframe to switch between shots, but that can cause severe motion blur if I enable it.
Another is to separately render each camera for the frames they're active, and then stitch the separate renders all together in a NLE program, which would be a huge painful waste of time.
Lastly I read it can be done through batch rendering and using MEL, but I haven't got too deep into that stuff yet, so I'd have to spend some time learning how to do that.

Any easier way of doing this? How do people normally deal with changing shots?
How the hell do you use Fusion 360? Really?

The lack of control you have compared to a polygonal/mesh editor is amazing.
you should import the baked map's to your texturing program and work on them in combination.

>my process is wrong
not wrong. but if you are using unreal you should copy your current UV map's (of the walls especially) to a lightmaps. when you build lightning it calculates the shadows and light path based on the UV's of the lightmap. there are youtube tutorials on the subject and it don't take very long assuming you are still early in the project
File: Environment.jpg (728 KB, 1928x1042)
728 KB
728 KB JPG
Thanks for the reply.
It turns out I was baking wrong, when I bake correct everything is about the same color/brightness.
Also I know what lightmaps are, but that will only be useful when I actually start lighting the scene. This scene is currently lighted by a skylight only.
But thanks to proper baking I managed to make the AO look good.
Would still love to hear any artists workflow for environments though.
If an artist asked you to program a game but his art was utter shit, would you do it?
What if his art was amazing?

People don't want to waste their time
File: concept.png (28 KB, 957x677)
28 KB
did my 2nd environment last week and i found out something important about planning.
you don't want to rush the process by just adding the props in the end. so i make a little blockout of how the level should start/end and the points+measurements and i have some scale data for the props.
i found out that if you plan the props ahead the scene looks more cohesive rather than add as you go.
your scene AO came out nicely but i think its too harsh, i would add subtract node and tone down the AO abit, its not a big deal.
It's right click because it's meant to be used with graphic tablets and it works great like that once you get used to it.

What I don't get is why does every 3d software uses a different camera control scheme. Sure, you can get used to any one of them, but switching between them is such a pain for the first few minutes.
The geometry hiding tool is your friend. I forgot its actual name, control shift select area.
What they teach you in college will probably be garbage. Nobody looks at your diplomas when we're talking 3d, everyone will want to see your work so far, so a work portfolio is mandatory.

The exception would be in technical fields, where a diploma is still required.

If you really want to invest yourself in 3d, learn the programs by yourself (daily work will get you far) and take courses in any art domain that has history, to learn the principles. These are more or less the same in all fields and together with knowledge of your tool you'll have a solid base to work on.
Rhino works a different field than max and maya, oriented towards hard surfaces and technical modeling.

Solid program though, i laugh my ass off every time i hear an architect designing a house in AutoCAD or LT then takes it to Max for modeling it in 3d and rendering it. You can do all that in just in rhino which coats as much as LT with a much faster workflow.

You're not gonna hear much about Rhino or other CAD on this site because the people working with them don't have time to waste here.

Oh yeah, and grasshopper is amazing.
Vray has been yaking huge leaps forward recently thought. How does corona stand out right now? I didn't get to use it yet.
I'm not even using regular CAD anymore, going back to Vectorworks from Rhino feels so backwards. You have to click so many things... Rhino is so convenient with that command bar.

Grasshopper is a godsend. I just wish I was better at it. Got the basics, but it's so deep and I should probably spend a day or two just googling what the hell half the terms even mean.
Yeah I ran into that, but from what I can tell you can't render from it, only playblast.
That still requires some scripting, but I think that may be the only way to accomplish this.

It just feels strange that you have to dig into scripting to accomplish something in maya that is a single button press in an "inferior" program like C4D.
once everything is set up in the camera sequencer you need to go to create> ubercam in the sequencer window, then in render settings set the renderable camera to the ubercam
File: IK stretch error.webm (194 KB, 1098x724)
194 KB
Anyone have any idea how to fix this?

I tried applying an old rig I'd made for a class back in 2013 to this model I made more recently, however since the original rig had much different proportions, I had to stretch it out to fit this new model. Problem is that seems to have resulted in a funny issue with the leg IK's, which squash for a bit when moved upwards before bending, instead of just bending from the start.

I've been modeling for years but know very little about actual animation. Any idea how to set it so the bones stay static and don't squash?
your weights are picking up too many verticies
usually stretch IK uses a condition node to determine when to apply scaling to the joints, where the condition is set by the distance between the first joint in the chain and the IK handle. seems like the value for the condition node hasn't been set correctly

Okay, read through the docs, watched the video, and after about three hours of trying to figure out which variable I could change to adjust the stretch point for the IK, I finally decided the closest thing to a solution would be to just delete the measuring objects outright. While this mean the leg no longer stretches, after some further adjustment it now bends properly.

File: fill me pls.jpg (46 KB, 1300x863)
46 KB
ok , i have problems with lags when i try to make pose for my makehuman character. just a while ago i tried to import less complex model with lowpoly topology, its way better now , but still not ideal. can someone tell me some ideas what i can do to make it even better ?

i already tried all options , minimalizing tabs , shading etc. i dont know maybe it's bcs my pc fault:

- geforce gtx 755m
i7 4700 mq
16 gb ram

if so pls answer so i wont bother myself and if anyone have some more ideas to improve working with blender reply too.

SRY FOR BAD ENGLISH / Im just simple paesant from europe
Which part(s) of the hardware effect rendering the most?

your computer simply isn't powerful enough.

you ideally want dual processors.
Disable subdivision for preview if that's on under modifiers, set the 3d view to solid instead of material, make sure ambient occlusion is off in the 3d view

Can't imagine what the issue would be beyond that, MakeHuman poses fine for me on a 2010 HP DV7 laptop so long as I don't do it with a complex material active..
Maya user here.
I added a IK handle to my characters leg, but the Pole Vector is slightly off, so when I add the knee vector the foot rotates and gets misaligned with the mesh.
Is there a proper way to fix this? The only solution I can up with is to rotate the foot joint back as best I can and then freeze it.
here you go:

File: Gouf_R35_010.jpg (112 KB, 1440x584)
112 KB
112 KB JPG
I'm still a loong way from rendering this, but this question is annoying me: How do I give stuff a sense of scale ? I don't want this to look like a toy. Would some small noise on the normal map work ?
Anyone know what this guy uses for his 3d?

Does anyone recognize this environment? I don't know the different 3D softwares but surely somebody will look at pic related and instantly know what it is
actually the comments might be right on this one, it looks like Moi3D
Does anyone know how to achieve this rendering? I assume he models in Moi but renders in something else?
A result like that should be easy to achieve in any of the more popular render engines.
I think you are looking at the post processing more than the render itself. I guarantee the raw render looks a lot more "achievable" in most renderers you can use.

color correction and chromatic abberation are the most obvious at a glance.
You could do that with Freestyle in Blender, even the dotted translucent lines.

He lists Cinema 4D, Keyshot, VRay, Photoshop, ZBrush, Modo on some of his projects.

http://simultan-cristiano.cgsociety.org/art/sci-fi-zbrush-device-keyshot-cyborg-h1o-1156912 This is a similar project, lists ZBrush and Keyshot.
Cool, thanks for the cgsociety link, I didn't know that website existed
dont make it the exact level of glossiness that suggests plastic
there's all sorts of tips to create scale, maybe googling it can give you some answers

but here's three I think are important ones:

-Add recognizable stuff. Anything we know the scale of. These can be bolts, handles, levers, as well as doors, ladders, furniture, cockpits, basically anything relateable to a human

-Model some high-frequency detail areas onto big stuff: basically "greebles", look it up, they used them a lot on spaceships in star wars to portray scale because they couldn't rely on atmospheric fog in space

-Damage: basically, different sizes of objects will get different types of weathering. A metal jeep and a metal pen might get worn down almost the same way, but the scratches, edge wear, dirt they will amass will look very different. (if your mecha is huge adding dirt streaks can be a great way to sell this because these only tend to form up on big things like buildings or ships)
Can anyone explain to a dummy how the HD remastering process works for visuals? Or point me in the right direction of a thread or book?
I understand audio (worked in a recording studio for years) but would love to to learn how they work on an existing games lighting, colors etc.
Depends which game. I assume it's sometimes possible to use the high-res developer textures (which were downscaled to fit prior hardware requirements, especially for last gen consoles). For some (like the Bioshock remaster), some of the textures are simply recreated from scratch, on top of pre-existing geo. See: https://80.lv/articles/the-remastering-of-textures-in-bioshock-the-collection/

Lighting would be engine related, though a good example of recent lighting improvements would be the arrival of PBR in games. Due to the way PBR shaders work, it's possible (with some work) to convert old textures into PBR. See: https://www.youtube.com/watch?v=cwBstFgTWAo

I rarely see actual geo being touched for remasters, likely because that involves a lot more work. You gotta unwrap again, skin, bake, etc.
Well, it's mostly programming and math, if you know those, the rest is up to your creativity.
File: DAZ-3D-LY-Trini.jpg (118 KB, 960x1248)
118 KB
118 KB JPG
does daz3d still needs you to buy developer / indie license to use their stuff in own games ? i was looking for that but found only old informations
which model is this? (for the face)
LY Trini, like the filename says.

Awesome thanks!
looks cartoony. I'm looking for more realistic to work form
I dont, but it's forcing me to learn so it's great anyway.
Look up groups, arrays, stacks, lists and some trigonometry formulas and it'll help you a lot.

If you understand the mathematical definitions of the shapes you're working with, you'll know how to break them down and GH becomes much easier.
I have an animation from Mixamo which I want to use in UE4
The animation has translations and rotations applied to the Root bone, which is at the feet of the character, and nothing on the Hip bone.

To use it I need Root to have no rotations applied to it, is there a straight forward way to transfer the animation from Root to Hip? So the general motion is the same, but it moves Hip instead of Root.
I'm having trouble because Root is offset from Hip, so it's rotation causes a translation in the whole character (and Hip) which doesn't seemingly have any curves I could transfer.

Bones go Root > Hip > other stuff
what would be the easiest way to model a shoe or boot around a pre-made model?
So I've made a few little things in Sketchup before and thought to myself "Hey, maybe I should try something more serious".

Get blender.

Open blender.

Fucking hell grating your dick with a blank expression on your face has to be easier than this. Mesh editing here is a nightmare.

I don't give a shit about textures, animation, rigging or anything else. I just want to make nice meshes. Is every other program as obtuse with this as Blender, or am I stuck with Sketchup if I want simpler?
Just about everything will be hard the first time, once you learn it then it's easier.

Bunch of tutorials out there from teenagers for editing Sims 4 hair with Blender, you can learn it too.

Maybe look for an intro to modelling in Blender video or article. Once you pick up on the logic of the program it'll gel.
File: untitled.png (34 KB, 500x300)
34 KB
In Blender I got a base for sculpting going by making a lowpoly shoe-like shape around the foot with an extra edge loop near the bottom, subdividing, adding a shrinkwrap modifier set to project pos and neg and using vertex weighting to paint in where I want to conform to the foot more, then applying it and adding an extra edgeloop on the bottom and subdividing again, and then apply a solidify modifier and delete the inside faces OR edit mode -> select all -> shrink / fatten to expand it a little.

Could also use it to snap other elements onto.
that's really helpful, thanks!
Yeah this probably doesn't deserve it's own thread but I am a student who is going into a 2 year robotics program at a local tech school. Recently 3D printing has interested me quite a bit. Is there a field for that? What type of jobs are there? Idk what I'm really asking. Ig I just want to do something with 3D printing. Probably don't have the mind to become an engineer
Using robots is the latest development in 3D printing, because it allows to actually 3D print without the constraints or limits of now common 3D printers, which can only move in the x, y and z axis. Basically you could have a bunch of robotic arms with different tools and extruders, and simultaneously print, paint and mill an object, because the robots can get to the object from any direction, while also twisting it around.

There is definetely a field for 3D printing, but you should learn 3D and programming.

There isnt much of a market for simply printing other peoples designs tho. At this stage, 3D printing takes too long compared to other manufacturing processes, so it's only good for prototyping or niche products.
Would Anyone Recommend Blenders Game Engine?
I would recommend UE4.
I'm playing around with world machine, which is an awesome tool. Now I exported a mesh of the world and I want to wrap it around a sphere, so that it looks like a planet.

But I'm stumped on how to do this. I'm using rhino but I can get edu licenses of Maya or Max or whatever.

What I ultimately want to achieve is a sort of comic strip or pictograph (printed out in 2D, so not animated or anything, but that would be cool too) zooming in on the planet until eventually showing the landscape, where I will then place a large megastructure that I designed.
I want to do this for multiple generated landscapes, to show that the structure I designed can be placed on any planet/landscape without regards to topology.

So how would I go about this?
It sounds so simple and I'm sure I'm missing something, but the only thing I came up with so far is just wrapping the heightmap from worldmachine around the sphere and just rendering that. But an actual mesh would be nicer.
uninstall world machine and rhino
because rhino will give you aids
it gave me aids and now my anus is so painful I can't even get fucked in the ass by black hunks anymore like I loved to
What if I already have aids?
Seriously tho, any comprehensive reasons?

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.