[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y ] [Search | | Home]

Archived threads in /3/ - 3DCG - 69. page

This is a blue board which means that it's for everybody (Safe For Work content only). If you see any adult content, please report it.

File: shooter2.blend.png (105KB, 1350x689px) Image search: [iqdb] [SauceNao] [Google]
105KB, 1350x689px
Why cant i animate a figure made from normal geometric shapes e.g. rectangles, squares. icospheres etc.

I'm a N00b using Blender and I joined all the shapes together and put the armatures in but I didn't use the metaballs I made a stick figure

Is it possible to do this and if not, please explain why not so I can grasp the concept of "why I shouldn't don't do that again"
3 posts and 1 images submitted.
at least in 3DS Max, you can assign separated objects as "childs" and "parents" (ie fingers would be children of the hand, hand a child of the arm...) assigning joint types between them. I believe similar thing should be possible in Blender too. Doubt you need any skeletons nor skin modifiers for such mechanical animation.

YT is your friend.
it is possible

parent every separate object to the single bone, you want to move it with. You can even parent single points to a bone and move it with the bone. What you cannot this way is to "paint" bone-influence for different bones.
Another way, combine your seperate objects into one object and parent this to the armature and do the setting (painting) of the influence for every wanted bone for the movement.

I'm pretty good at environment art and modeling/sculpting in general, but when it comes to anything human related I actually have no idea what I'm doing...

What's a good resource to start learning anatomy/character art in general?
5 posts and 1 images submitted.
anatomy books
google body part and try to copy it
Study the skeletal system and know where the outer layer muscles begin and ends. Some books I recommend are,

Albinus On Anatomy by Robert Beverly Hale, Artistic Anatomy by Dr. Paul Richer
Figure Drawing by Michael Hampton

You'll have to learn how to break the muscles/body parts into basic shapes. You should can look up George Bridgeman as well.

File: 1488051368160.gif (2MB, 500x750px) Image search: [iqdb] [SauceNao] [Google]
2MB, 500x750px
I know this faggot is a /3/ user.

Tell me how much will it cost me for you to make me a model like this from concept art I will provide for you?

can you fags give me his tumblr?
11 posts and 5 images submitted.
How about no
no idea who made it, but the model looks fairly simple, the bets thing about the model is the texturing.
that model doesn't look good at that angle, And I guarantee you that model looks its best AT that angle. Waste of money.

File: 1477667090106.jpg (658KB, 2048x2048px) Image search: [iqdb] [SauceNao] [Google]
658KB, 2048x2048px
what should I make
6 posts and 3 images submitted.
File: 1474121931428.jpg (18KB, 176x134px) Image search: [iqdb] [SauceNao] [Google]
18KB, 176x134px
Furry porn

File: Rhino+close+up.jpg (687KB, 1000x667px) Image search: [iqdb] [SauceNao] [Google]
687KB, 1000x667px
What are the advantages of modeling with meshes over nurbs?
I'm somewhat experienced with Rhino and modeling architecture, but a total noob when it comes to modeling with meshes.
But with the exception of rhino, all 3D packages seem to be using meshes for everything and getting good results, so I would love for someone to shed some light on this.

To me, nurbs seem a lot more accessible and easy to control, and the mostly additive + booleans workflow is very intuitive and precise. Box modeling on the other hand seems to be the opposite, but idk.
4 posts and 1 images submitted.
You use meshes for animation purposes
Nurbs is really heavy to compute but really handy for mechanical pieces (for industry)

Zbrush or Modo give some good solution to create some meshes
most NURBS models will get converted to polygons anyway, at least at rendering, often even before.
I know nurbs are heavier to compute, but it's such an easy way to design.
Guess they're pretty useless for anything not to be manufactured tho.

And yeah, they will be turned into a mesh anyway, either manually to optimize a model, or simply into a rendermesh for rendering.

File: Ribosome FACES.png (574KB, 1088x814px) Image search: [iqdb] [SauceNao] [Google]
Ribosome FACES.png
574KB, 1088x814px
Anyone know how to dramatically reduce the number of triangles in an .stl?

My model has over 6 million triangles, and its 978 MB .stl file.

So big that makes me 3D printer crash.

Next pic is a close up of how many triangles and suggestions maybe
6 posts and 3 images submitted.
File: tRIANGLES.png (116KB, 1113x886px) Image search: [iqdb] [SauceNao] [Google]
116KB, 1113x886px
File: vdbMeshing.jpg (358KB, 1920x1017px) Image search: [iqdb] [SauceNao] [Google]
358KB, 1920x1017px
Right now it seems like you have a lot of polygons that aren't visible because they are intersecting, so that's a huge waste right there.

You could try a polygon reduction algorithm, I think all major 3d packages have them. Not sure how well that will work. The individual spheres don't look like they have a lot of detail, so you might be left with a jagged mess if you reduce the geometry too much. Might be enough though?

Personally I'd do this with Houdini. You can convert the whole mass of spheres into a volume representation and then convert it back to geometry with complete control over how many polys you end up with.
You could try detaching each geodesic sphere to its own mesh, then reduce the polycount on each individual mesh, or merge each individual mesh to a centre point and use those points to instance a less poly dense mesh.

File: Picture 3.jpg (154KB, 1280x960px) Image search: [iqdb] [SauceNao] [Google]
Picture 3.jpg
154KB, 1280x960px
Is this the right place?

I want someone to design me something for a 3d printer.

It's to hold this mini bike pump against my bike.

It needs to be open on all sides except for the two ends.

I've been using tinkercad but so far haven't been able to make much more than a box...

It's 20cm x 4cm
4 posts and 1 images submitted.
I'll do it for free, but I need more info.
Contact me at [email protected], I prefer skype to talk or discord or irc so send me info about one of this places we can talk there.
email sent!
You ppl are great. Way better than I could have done myself (after a week of study)

File: infinizoom0266.png (1MB, 1920x1080px) Image search: [iqdb] [SauceNao] [Google]
1MB, 1920x1080px
I made this simple animated thingy in c4d:


It's a recursive scene - think infinite zoom, like zooming into a picture that contains a smaller version of itself, and that smaller version also contains a yet smaller version, etc. - just in 3d.
Read the clip description on YT if you don't get it, or watch the clip.

It's not the final product or anything, just a test.

The ratio of the bigger scene to the smaller scene is 10:1. Meaning, each consecutive level of zoom is 10 times smaller, for simplicity sake. The camera zooms in. At frame 0, it has a keyframe that points its zoom level to, say, 1. And the last keyframe of each zooming makes the zoom level, say, 0.1.

Problem is perspective, though. If the F-curve connecting the two keyframes was set to linear, it would look like the zoom speeds up and then suddenly stops when it reaches the next zoom level, and repeating that pattern, instead of a relatively smooth-looking camera movement you see in the clip. I am not sure how I achieved that, but I set the F-curve to spline and fiddled about with the curve shape until I was like "Guess that has to do, I can't possibly do a better job", which took me a while.

But look at the blue car. It too has keyframes at 0 and the last frame of 1 zoom sequence: for position and size. Position changes in this fashion: 10 + 1 + 0.1 + 0.001 + ... etc, each level of zoom "adding" a 1/10th of its actual coordinates - since every consecutive iteration is 10 times smaller, right? Thus its size decreases from the factor of 1 to 0.1 to 0.01 to 0.001 ... etc. Since the camera's APPARENT movement is linear (while its ACTUAL movement, to zoom in, is this fucking spline that took me forever), the blue car's size should stay roughly the same throughout the whole animation. But with its F-curves set to linear (size & position), it doesn't look to smooth.

How do I compensate for the camera's movement?
7 posts and 3 images submitted.
waste of your/my time. AWFUL
Lol, the "art" is obviously just a placeholder. I didn't bother with textures or anything, I spent literally 2 hours modelling everything. Another 2 hours to render. That's not the focus.

I just want to figure out proper camera / movement splines when dealing with powers of 10 while zooming in - perhaps there's a function for that? Exponential, logarythmic etc? - so I can then use them to zoom into something way better.
File: cam f curve.png (225KB, 1021x919px) Image search: [iqdb] [SauceNao] [Google]
cam f curve.png
225KB, 1021x919px
Oh and here's how the spline currently looks. That has to be a crude approximation of a real math function I could use to figure out the shape of the spline. I just know it.

File: lulzbot.jpg (113KB, 1280x720px) Image search: [iqdb] [SauceNao] [Google]
113KB, 1280x720px
My Lulzbot Mini should be delivered to my house within the hour!

Does anyone have any experience using Cura? debating if i'll use it or a different slicing software

Also dubs decides my first print
16 posts and 3 images submitted.

Your first print shall be a dildo, you fag.
Why don't you 3d print an Ultimaker and then return that piece of shit OP?

File: 1485799278533.jpg (281KB, 1200x1520px) Image search: [iqdb] [SauceNao] [Google]
281KB, 1200x1520px

new blender guru video, this time about dynamic range, enjoy.
36 posts and 4 images submitted.
I didn't know that he had posted, thanks for sharing.
Stop shilling, Andrew.
>Set display to none and gamma to 2.2
>Save as OpenEXR
>Bring into program like LuminanceHDR or similar
>Tone map to your liking
>he's giving lectures at GDC now
>at the end of one he posts his work
>its shit

Is there a "VR compatible" application that one can use to organize files? Something simple, to cobble together pieces, and then you can import a document, make it look like a book, and put it on a virtual bookshelf.

I know there's some things which can put desktops or filesystems in VR, but I want a virtual plot of land I could customize and sort out for myself. Doesn't need to have decorative objects like figurines, even a basic system where a rectangular prism + color represents a book on a shelf, and I can walk up and left click on it to bring up the document internally or externally.

pic related, could look like shit
8 posts and 1 images submitted.
thats deep. I dont know of any yet but the basic vr windows thats pretty shitty and not fitting your request. but nice idea
what would be the point?
Microsoft Bob.

File: sunsetwater.png (792KB, 960x540px) Image search: [iqdb] [SauceNao] [Google]
792KB, 960x540px
It's pretty good for something made in Blender.
15 posts and 1 images submitted.
>Water doesn't look like water
>Alpha on leaves and grass is too high
>Shitty lens flare is the focus of the render
4/10 for effort
>It's pretty good
I don't think you get to decide that, hombre.
thats right, i do

I've been working on renders for a company and I'd like to get some feedback and thoughts on how I might improve.

This is made in Cinema 4d with Octane Render.

I realize it's under exposed and that the HDRI Lighting needs to be made a bit warmer.
6 posts and 2 images submitted.
This is a conceptual render for a technical part? I'm not trying to be rude when I say the people that look at this probably don't give a shit how real it looks, as long as they can see it in 3D they are happy.

This render is wasted potential for something better. As the render goes looks good, more orange couldn't hurt.
Exactly. My client want's it to look as real as possible so that rather than install the product, take the photos you need and hope you don't have to go back and reshoot, they have a rendered scene they can freely manipulate to give examples of installation.

You are right when you say most people won't care if it looks real or not but my client
Looks fine! I suggest you find a way to automate scene creation, modification, rendering, and post processing. - especially if you're visualizing multiple kits, identical camera angles and lighting will help. You can save a lot of time.

File: ted.jpg (79KB, 660x439px) Image search: [iqdb] [SauceNao] [Google]
79KB, 660x439px
Allow me to introduce you to Ted the Head. He is a sculpture modeled after my own head, and was one of my most challenging builds. I originally intended the head to be all yellow, but I ran out of yellow bricks. So I decided to make hair from black bricks! But then I ran out of those. Then I tried to make a hat with blue bricks, but ran out after the shade on the hat was made. After that I finally finished the hat with white bricks, and I also added a nice T decal on it. Hence the name, Ted the Head. His chin was a lot of trouble also. How do I make it the right size, but make it look like an actual human chin? It took me about two days to finish this model. A lot of people may think that is very short, but I did work long hours in order to get it done.
12 posts and 1 images submitted.
I'm afraid to tell you Ted this is a board for 3D computer modelling and related things, this is a real life model.
>o Ted the Head. He is a sculpture modeled after my own head, and was one of my most challenging builds. I originally intended the head to be all yellow, but I ran out of yellow
Is that a render? Nice!

File: Screenshot.png (646KB, 663x679px) Image search: [iqdb] [SauceNao] [Google]
646KB, 663x679px
Is there any software thats good for drawing textures on? Like drawing onto the object and then saving the texture as a uv map shape thing.
20 posts and 3 images submitted.
pretty sure any 3d program can texture in viewport and into the UV's.

this is why bodypaint was popular at the time, because there was no realtime painting
What you're looking for is Substance Painter
Thanks I'll have a look for it

Pages: [First page] [Previous page] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [Next page] [Last page]

[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y] [Search | Top | Home]
Please support this website by donating Bitcoins to 16mKtbZiwW52BLkibtCr8jUg2KVUMTxVQ5
If a post contains copyrighted or illegal content, please click on that post's [Report] button and fill out a post removal request
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.
This is a 4chan archive - all of the content originated from that site. This means that 4Archive shows an archive of their content. If you need information for a Poster - contact them.