[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 / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / 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] [Search] [Home]
Board
Settings Home
/3/ - 3DCG


Name
Options
Comment
Verification
4chan Pass users can bypass this verification. [Learn More] [Login]
File
  • Please read the Rules and FAQ before posting.
  • There are 68 posters in this thread.

05/04/17New trial board added: /bant/ - International/Random
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.
[Hide] [Show All]




File: Houdini-Logo.png (16 KB, 520x520)
16 KB
16 KB PNG
Are there Houdini users on /3/? What are you learning or working on?
>>
I've been learning it for the past two months, so I'm pretty much still a novice. Currently, I'm learning to light in it, using Mantra, following Birn's Digital Lighting and Rendering.
>>
Houdini Procedural Lake Houses from Anastasia Opara
This gal has absolutely BLOWN my mind about what this package can do in hands of competent mathematician and artist.

Now I lament I skipped maths when I had a chance.
>>
Just finished up on a large scale crowds project, posted it here in WIP: >>633196

Other than that, just enjoying my time off between studios. I'll probably do some pyro rnd in the next couple weeks, and experiment a bit more with rebuilding solvers. Houdini is truly my favourite 3d app, it has opened so many doors.

>>633086
She's incredible. One of the extremely rare cases where someone is both creatively and technically gifted to an insane degree. Massive respect for her.
>>
>>633041
I actually just installed an apprentice version today. Getting used to the interface and all of the nodes
>>
File: OceanWIP1.png (3.96 MB, 3198x1748)
3.96 MB
3.96 MB PNG
>>633223
my wip on an ocean I decided to be my first project
>>
>>633041
quick maths
>>
>>633236
What do you mean?
>>
File: 1564578.jpg (8 KB, 229x250)
8 KB
8 KB JPG
Anyone had any luck exporting rigs and animations from Houdini to something like UE4? It seems that Houdini's way of handling skeletons (bone based rather than joint based) fucks up when I try to export it to a joint based system (UE4, Maya).

Is there a way to convert it on export or something? If this worked, I could finally ditch Maya for good.
>>
Do you people polymodel in Houdini, or elsewhere?
>>
>>633477
If you have 10 bux to spare, this third party tutorial might be helpful:
https://www.sidefx.com/tutorials/game-rigging-in-houdini-creating-export-friendly-character-rigs/

>>633569
I've personally always modeled in Houdini.
>>
I'd like to ask this to people who use Houdini. What made you choose Houdini over a more traditional modeling software? How hard was it to learn and how do you enjoy it now?
>>
>>633781
Initially I mainly did VFX, so Houdini was a no-brainer. When I started learning it properly, I realized I'd never go back to non node-based modelling, it's just so insanely convenient. If Maya decides to wreck your mesh for no apparent reason there isn't really much you can do without it being a huge headache, but if that happens in Houdini it's very much like debugging a piece of code, usually fast and simple. I think the best comparison is Maya being Microsoft Word, while Houdini is LaTeX.

The real benefit comes when you learn to model procedurally. You still get the above benefits if you model normally, but when you start actually using the procedural elements it gets even better. Also, it does fucking everything without needing plugins, best example is how I'd have to buy a plugin or use After Effects to compose a flipbook from Maya, but I can just set up Houdini to directly spit out a flipbook instead, saving me a lot of work.
>>
Is Houdini truly the endgame of 3d software? What can't it do? Everything I've read and seen about it so far just presents it as the perfect 3dcg production suite. Outside of price, is there even a reason why alternatives are considered over it?
>>
>>634111
It can do basically anything another package can do.

There are still some quirks (keyframes automatically bezier, why the fuck), but overall yeah if it continues on this trajectory it will be a staple in every VFX department.

Reasons for using other software at the moment are that Houdini is still mostly used for FX work, while other packages are used for animation, etc (though it is becoming popular in assets modeling). Big companies with a lot of time and effort spent on their pipeline won't switch to Houdini for a lot of this work for a fair while, and I'll bet animation will be the last to transition, even though RIGHT NOW Houdini is objectively better for rigging and animation.
>>
This might be a really dumb question but does VFX software differ between games and film/media industry?
>>
>>634262
No, but how you use it does. For movies and such you usually use the simulation itself directly, while for games you'll need to take several snapshots of it to make a flipbook sequence that you then bring into the engine and try to make it look like it's not just a sequence of flat sprites.
>>
>>634264
Here we have a true pro...
>>
Well I'm trying to make a monster emerge from the ocean but for some reason I'm unable to find any good resource on what a setup like that would look like, as in a flip tank driven by an infinite ocean around it with realistic foam. Does anyone know of something like that?
>>
>>634111

any of the big packages can do anything houdini can do, except usually 50x faster and you dont require a maths degree to do it

only exception is stuff that is fundamentally procedural....like fractals, generative modeling and such. houdini excels at that.

its like the difference between writing a term paper in microsoft word, or writing it in html and formatting it in css. sure, css is going to ultimately give you more customization options, but why waste time with that shit (unless totally necessary), when you could just use a cookie cutter solution that will get the job done way faster.
>>
>>634517
>when you could just use a cookie cutter solution that will get the job done way faster.
That's the thing. You use a cookie cutter when that's enough for the job. You use something like Houdini when you want to fit the process to your ends, and not the other way around.
>>
>>634517
you don't need math to do shit in houdini though
It helps, sure, but you can just work with wrangles and nodes and shit
>>
>>634517
>any of the big packages can do anything houdini can do, except usually 50x faster and you dont require a maths degree to do it
>only exception is stuff that is fundamentally procedural....like fractals, generative modeling and such. houdini excels at that.
How about VFX, you massive retard?

There's a reason Houdini pros are paid a lot more than others: Houdini is LaTeX where Maya/Modo/3DS/etc are Word.
>>
>>634685

You really don't know what your're talking about, correct? You just read term "wrangle" somewhere and now want to sound knowledgeable.

"Wrangles" are Houdini nodes in which you input Vex code, so you are basically writing snippets of programming code in them. Now you don't "need" math for programming, but when it comes to manipulating points in 3D space you kind of have to understand some of it. Vectors, arrays, matrices, dot product, trigonometry - you NEED that in Houdini if you want to use Vex and "wrangles" and shit.

You can get away in Houdini with just using shelf tools and basic nodes, but if you want to do custom advnced stuff you need wrangles and math and shit
>>
>>634698
nah, I do know what I'm talking about, I'm just being retarded and called an attribute vop a wrangle.
Also you don't really need to be good at math to use vex on beginner and even intermediate level, and vectors, trigonometry and matrices are part of the school program (at least in my country), which I assume everyone has studied.
I understand why someone would say something like that, but in al seriousness you don't even need Vex to be good at Houdini, that's for sure.
>>
>>634698
>Vectors, arrays, matrices, dot product, trigonometry - you NEED that in Houdini if you want to use Vex and "wrangles" and shit.
That's high school math. Please, tell me that when people are afraid of Houdini & math they are thinking of something more complex. Or tell me otherwise, and make my day, because I'll plunge right away into the program.
>>
>>634715
Trig, vectors and matrices all you need for basic procedural modeling and effects. Nothing more complex than you would need for e.g. building a high quality rig in Maya.

It gets more complicated when you want to do things like write your own fluid simulation. Which is something that can be necessary if you want to get to the highest levels of simulation. This is comparatively "easy" to do in Houdini, as the solvers are open box if you want to plug in your own algorithms, but you'll need to do physics math to take advantage of those features.
>>
Voxelization + Decimation seems pretty good.

I'm having trouble with the autouv thing though, it likes to generate a lot of tiny single poly islands
>>
This made me laugh: https://vimeo.com/282088899

What's the story behind the rubber toy?
>>
File: testgeo.jpg (169 KB, 1298x847)
169 KB
169 KB JPG
>>635471
I'm not sure there's any exciting story behind it. Houdini has a handful of built in oddball primitives. The pig head and squid show up in plenty of tech demos.
>>
>>634694

how about vfx? huh? are you seriously implying that other packages can't handle "vfx"

were you dropped on your head as a baby you stupid fuck?
>>
>>635485
The pig head is the weirdest one, it's not even a normal pig head, it's some sort of humanoid pig head you'd see in a horror movie.

>>635504
>how about vfx? huh? are you seriously implying that other packages can't handle "vfx"
Not even half as well as Houdini, you shit-smeared mongoloid.
>>
hey my dudus,
looking to get into houdini but a little bit confused as to where to start. i'm most interested in the non-dynamics based procedural stuff it can do, but i know i still need to have a handle on the core dynamics stuff before even approaching the other stuff.

anyone got recommendations for a specific course, or even a curriculum of sorts? i've been looking at:
-adam swaab's learn squared courses
-gnomon workshop's houdini kickstart
-rohan dalvi's stuff (but not sure which one to start with)

ideally i want something that touches most of what houdini has to offer, before diving into a specific area.

also, how good is the arnold implementation in houdini compared to something like the native mantra renderer? i know none of the courses/tutos use arnold so it's not a consideration there, just wondering how viable it is down the line.
>>
>>635524
>-gnomon workshop's houdini kickstart
Skip this one. Bad practices.

>-rohan dalvi's stuff (but not sure which one to start with)
Good enough. https://www.sidefx.com/learn/collections/introduction-to-houdini/

For procedural modeling, the gold-standard course may very well be Anastasia Opara's "Houdini Procedural Lake Houses" series. But you need to have some experience already with Houdini and especially VEX.

>where to start
I'd suggest this: https://www.sidefx.com/learn/getting_started/
Start with the book, then the Basics, then Dalvi's introduction, then go over http://www.tokeru.com/cgwiki/index.php?title=HoudiniGettingStarted in the presented order. If you do this while constantly referring to Houdini's docs, you'll get a good basic grasp of all areas of the program.

Also, the docs are a good way to start, however it can be a bit dry at times. Read at least the introductory sections -- "Getting Started", all of it; "Using Houdini", preferably all of it; "Nodes", go over them so you know is there available to you. https://www.sidefx.com/docs/houdini/
>>
>>635527
>so you know is there
*so you know what is there
>>
>>635527
>>635528
ta v.much. very helpful.
>>
>>635524
I recommend the Pluralsight intro. It touches most areas of the software, teaches best practices and the instructor is adorable.
>>
>>635541
Cheers. I might quickly run through the dalvi one and then do the pluralsight thing. The book and the dalvi tutorial don't appear to get into the sim stuff much at all.
>>
>>635541
>>635567
Just a quick question tho, I know Houdini has had some major changes in the last few years. Is it okay to look at tutorials from v15 without worrying too much about that there are better ways to do things in 16.5?
>>
>>635527
>how hard could it be?
>Hard surfaces sounds cool
>brutal Rohan Dalvi course from hell
>jesus christ
>maybe he has an easier one
>the hot fudge brownie tutorial
>fuck
Fudge brownie took me 3 weeks to finish and its one of the best.
>>635567
check out abstract effects in houdini by Niels PRAYER for a good intro to sims
>>
File: screenshot.png (48 KB, 913x280)
48 KB
48 KB PNG
No mention of H17 at SIGGRAPH so far, but a few TBDs in the schedule. It's about time we get a sneak peek, isn't it?

Also, pic related. Sounds very interesting.
>>
>>635656
Yeah would be about time now. I imagine there will be a lot of modelling and animation stuff, maybe I can finally ditch Maya for good.
>>
>>635657
I am re-familiarizing myself with Maya since we are switching our pipeline at work from Lightwave (lol).I don't want judge too quickly, but this application seems a bit old and crusty, which is funny because Houdini has been around just as long. On the surface it looks and feels like something from the 90s, I'm hoping it's not that bad once we are deeper into it.

>>635656
I have a feeling 17 will be another monster release, like 15.5 -> 16. I've read that they will be introducing a couple of new contexts as well (LOPs & TOPs), should be interesting.
>>
>>635663
>new contexts as well (LOPs & TOPs),
I didn't know about that, just read now on odforce some speculation about what they might be. I'm hoping LOPs have to do with lighting, which would make sense considering RenderMan 22 IPR is expected to come Q4 this year.
>>
>>635657
Please God let them push the animation and rigging forward. I am so sick of the Autodesk monopoly on animation tools.
>>
>>635663
Maya suffered a curse of popularity. They can't change a damn thing about it because there are so many piles of proprietary tools that are built around its strangeness. The strange APIs, bullshit nodes, MEL, it's calcified and untouchable. I mean the tool ecosystem is incredible, but the software is stuck in the 90s.
>>
>>635663
>On the surface it looks and feels like something from the 90s
That's because it is, and it only gets worse. Partly because of what >>635670 said, partly because Autodesk is incompetent (see: recent attempts at adding modern features into Maya and how horribly they were handled).
>>
>>633041
R20 seems to make houdini less interesting now
>>
>>634113
>RIGHT NOW Houdini is objectively better for rigging and animation.
>bone-based
>better than joint-based
Hell fucking no.
No.
I'll take stuck-in-the-90s Maya any day of the year.
>>
>>635715
I don't understand. "Bones" and "joints" are interchangeable terms AFAIK. Both refer to the thing that a game engine like Unreal does. What does Houdini do differently?
>>
>>635713
I suppose you're talking about C4D. Had to look up what new features it's got. They are good, but nothing new for a Houdini user. Anyway, the great thing about Houdini is the nodal-based process, and its programmability; nothing in the new C4D release (as far as I could see) gets it closer to Houdini in that regard.
>>
>>635748
>>
>>635656
>>635657
I missed it somehow. In the schedule, it reads "H17 Sneak Peek - PolyDraw 1.0".
>>
File: DdifiNXU0AAIWBK.jpg (428 KB, 1593x2048)
428 KB
428 KB JPG
>>635595
>mfw i go to his twitter and see he's making daz titty girls
>>
>>635780
kek
Yeah, was surprised to see what content he makes in his free time when he puts out Houdini tuts of that quality.
We even had a thread bashing him a while back: https://archived.moe/3/thread/621100/
>>
>>635784
had a good laugh while scrolling through his tweets.
heard about the rohan in an interview with serjan burlak and thought the dude would be doing some next-level autistic shit, so it's hilarious that he's so devoted to the dazgirls.
good for him tho.
>>
>>635748
Maya handles rigs the way it should be done, Houdini is fucking stupid instead.
AFAIK Blender is also bone-based.
Bone-based rigs are literally unusable.
Not being able to move joints freely means you can't adapt to model/proportion changes (just think about mocap) and rework hierarchies on the fly. Just so people can deal with a "more intuitive" concept like a bone instead of an oriented point in 3d space.
Fuck that shit.
Fuck, just thinking about it makes me angry.
>>
>>635811
Unironically better than 99% of DAZ output.
>>
>>635818
>creates most if not all of the assets himself
>uses a very powerful render engine
As opposed to normal DAZ content that's all purchased assets arranged in a scene rendered in Iray. Shame that he tries to force the anime look onto his mannequins though, the faces really look terrifying. He made one portrait of Widowmaker that actually looked pretty good, but I don't know why the rest of his DAZ Dolls are all bug-eyed and shit.
>>
File: options.png (70 KB, 835x935)
70 KB
70 KB PNG
>>634113
>(keyframes automatically bezier, why the fuck)
Does this solve it? (I'm no animator.)
>>
>>634113
>RIGHT NOW Houdini is objectively better for rigging and animation
really? any reels/making-of/tuts you can recommend? can it do animation layering?
>>
>>636518
https://vimeo.com/269110716

https://vimeo.com/273798604
>>
Houdini Customer Reel 2018. Enjoy!

https://vimeo.com/283047555
>>
>>633041
I'm also a studunt of the houdinis
I'm primarily focussed on building digital assets, and using vex. VEX is just amazing, its quick, its pretty easy to use if you are a bit versed in programming. The only downside is that documentation on it can be pretty obscure, notably in regard to intrinsic attributes.

The last thing I made was a vine gen with AI and optional gravity, so it could crawl along the floor and climb up things.

I'm getting into Dop Pop's next, so I can build some particle stuff for a unity project someone at the studio is working on.
>>
>>634746

Knowing a bit about quaternions is pretty handy to. I'm struggling with them but I find some functions VERY handy, such as the dihedral to get a correct qrotate going
>>
>>636475
Answering myself, since I'm learning to work with the animation tools. Yep, that setting sets the default tweening function -- I chose constant() and it works as expected.
>>
https://www.youtube.com/watch?v=nqlL-_o8ioQ
Blender BTFO
>>
>>638183
That course looks dope af. But why mention Blender, though?
>>
>>638189
because some mental midgets here seriously claim that Blender is a contender for Houdini
>>
>>633214
Any tutorials you can share? I do composting myself and really wanna open my skills up beyond nuke. Thanks!
>>
>>638279
Not who you're replying to but I got my kickstart with this

https://vimeo.com/194219623

Don't know if it's still up for sale but you can probably get it on cgpeers.

If you comp with nuke your transition to houdini will be much smoother due to node workflow. I even got to learn some things to use in nuke, such as creating null nodes to use as master switches for motion blur / defocus on my entire comp
>>
>>635568
ahhh It's hard to say, I would go through the docs first then look at tutorials but I've been reading the docs veraciously for the past week and I'm only about halfway through rofl
>>
>>635568
>>639023
The Illume webinar videos might be a good thing to reference when you are checking out older tutorials. Sometimes they cover topics where workflows or best practices have changed.

https://www.sidefx.com/learn/talks/
Change the drop-down from All Types to Illume Webinar.
>>
>>638729
I think Fusion 9 might be a second place contender for houdini pass off
>>
https://vimeo.com/286242378
>>
File: disintegration.webm (506 KB, 1280x720)
506 KB
506 KB WEBM
>>633041
Don't work on anything at home, usually doing small tests and example setups for work if needed. I work as FX TD for film however current company where I'm at is shit and now looking for new one. Almost decade of work experience.
>>
how good are houdini's hair tools, compare to say xgen or gmh plugin?
>>
>>639169
Traditionally hair and cloth have both been lackluster in Houdini but I think hair is now in quite good position. In next version of Houdini the cloth will get new solvers as well so maybe it will be good as well.
>>
>>633041
Okay, this is epic
>>
Peeps, do you participate in any of the Houdini forums? If so, and you see a question that can obviously be answered by reading the docs, do you bother to answer?

I'm of two minds here. I mean, on the one hand, I'd be helping someone learn, giving back, in a way, to a place from where I learn lots; but, on the other hand, I know the answer because at some time I read the docs, and this person clearly didn't, so I can't but think, "why should I spoon feed them?"
>>
>>639164
It's really bothering me that the air flows along the disintegration.... What's supposed to be happening? If the object is being disintegrated by a laser or something, then wouldn't the force move the particles backwards? Or if it's like a meteor, wouldn't the particles start disintegrating from the front and then to the back?
>>
>>635713
LOL just because C4D is better than Blunder doesn't mean its anywhere close to Houdini
>>
Should I use smoke solver or pyro for regular smoke stuff ? I don't really get the difference other than pyro solver already have some built in micro solvers like turbulence and dissipation
>>
>>640496
dunno lol I Blender only
>>
>>640496
Smoke solver. I use smoke solver for explosions too, the pyro combustion model is pretty slow so often it's better to make your own.
>>
File: hidden-services-h.jpg (206 KB, 1600x750)
206 KB
206 KB JPG
>>640199
Just a comment
>>
>>640554
What is that supposed to mean?
>>
File: debri_smoke_v02.0068.jpg (353 KB, 2048x1152)
353 KB
353 KB JPG
>>640533
thanks m8, any tips you'd like to share? I'm just messing around with each microsolver parameter to get a better grasp at what they do. I'm at the level where I can make some cool stuff, but if my supervisor asks "make it more X" I'd be lost
>>
>>640558
Not who you're replying to but I have a big tip for you anyway. Study nature and the science of the pheonomen you wish to recreate. Understand what is happening and why. Software knowledge is good but not essential. You can always search the documentation once you've figured out what you want to do.

Solver settings are great and all, but the key to a good simulation is a good source. Good shit in, good shit out.
>>
>>640179
Magic is happening! Bullshit magic that doesn't make any sense. But people eat it up because it "looks cool". There is an effect similar to this in A Wrinkle in Time and the new Jumanji
>>
>>635505

>Not even half as well as Houdini, you shit-smeared mongoloid.

oh gosh. he actually believes this. will someone tell him??
>>
>>640573
AAAAAlright, yeah I get, so it's being sucked by something, Jumanji style, it sort of makes sense then
>>
File: test_03.gif (858 KB, 480x480)
858 KB
858 KB GIF
building procedural ovens than crushing them
>>
>>640678
That’s super cool. Why are you crushing ovens tho?
>>
>>640613
>RRRREEEEEE ALL THE PROFESSIONALS AND THE ENTIRE MOVIE INDUSTRY IS WRONG USE MY SOFTWARE INSTEAD
Let me guess, you use Blender?
>>
>>640613
Blundertards and their crappy amateur package
>>
>>640613

Will someone tell him what? That blender sucks?
>>
>>640832
>>641065
>>641091
Why do you think he must've been thinking of Blender? He could be honestly thinking of Maya or who-knows-what-else.
>>
>>641092
it's /3 on Fagchan here
not a forum where any professionals waste their time
>>
>>640678
>>how you do that?
>>
>>641096
>not a forum where any professionals waste their time
But where else would professionals go to shitpost about anything and everything 3D related?
>>
File: 5435345636.jpg (36 KB, 400x300)
36 KB
36 KB JPG
>make mesh in Houdini
>export to Substance Painter
>horrid seams
>google it
>"you gotta pixel snap your UVs"
>apparently not doing so will lead to unsightly seams
>google how to do that in Houdini
>ZERO results
>no options in any of the UV-related nodes in Houdini
Someone explain to me how to pixel-snap UVs in Houdini, I've been losing my sanity over this.
>>
>>641215
Can you share a slice of the problematic geo, with UV mapping of course? No promise that I will be able to solve the problem, but I'd like to give it a try.
>>
File: seam.jpg (60 KB, 1399x593)
60 KB
60 KB JPG
>>641220
Not really much to share, last hour or so I've been trying to find the root of this issue I have just been using a simple sphere made in Houdini, quick-UV'd with the new game auto UV tool (but any other method has the same result). Just make a sphere, UV it however you want (leave an exposed seam so it's easier to see), export to Substance, and apply any material (or even worse, a smart material). As you can see on the image, the seam is there, in this case I simply applied pure copper to it.

I've tried tri-planar projection, doesn't work (it does help pattern-heavy materials ignore the seam so you can't see it very well as they kinda hide it, but it's still there). I've tried with baked maps and with no maps, same result (so it's not a normal or curvature map bake issue). Isn't the whole point of applications like Painter that since you are painting on a 3D mesh, it doesn't give a fuck about UVs and its seams? Quite sure Mari for example cares nothing for UV seams.
>>
File: Untitled-2.jpg (301 KB, 1058x1644)
301 KB
301 KB JPG
>>641223
Is the seam visible on a render? Substance's viewport can't really be trusted for this, it may show seams because it's not using texture maps as they will be when exported, that is, with dilation and diffusion. Pic related, it's how a similar setup renders in Mantra.
>>
File: Untitled-1.jpg (254 KB, 2308x897)
254 KB
254 KB JPG
>>641223
>>641230
Regarding smart materials, as far as I know, Substance Painter (or rather, the material in question) doesn't consider the shape of the mesh when projecting the material, it just fills the UV tile. The way around this is to carefully align the UV islands.

I assume it may be possible to do it more easily / automatically with Substance Designer, but I have no experience in that regard.
>>
>>641230
>>641231
>Is the seam visible on a render?
Yes, both in the Substance render and inside UE4. It seems to be some sort of normalmap issue, the seam goes away for opaque materials (e.g. concrete) like concrete if I use tri-planar projection in Substance, but persists regardless if I use a shiny material (e.g. pure copper). The same happens with exports from Maya, so it seems that it is something with Substance rather than Houdini.

I don't mind not being able to use smart materials as I make my own stuff, but base materials (which are a good starting point) and smart masks fucking up is very annoying, though I guess my options are to either move to Mari (that can use PTEX) or simply hide my seams better.
>>
File: types-of-love.jpg (112 KB, 487x390)
112 KB
112 KB JPG
>>638196
Blender + Houdini is consummate love.
>>
>>640179
>>640573
Goddamn you are stupid. Middle-schooler brainlets. Do you fucking understand what is "test" or "setup"? Do you fucking understand the purpose of just toying around? Kill yourself.
>>
File: IMG_1743.jpg (1.99 MB, 2448x2448)
1.99 MB
1.99 MB JPG
>>639152
Thanks for posting this.

>>641103
I want to know.

>>641749
Yo, fuck that. Look up dispersion rates of wood burning's gas production or shit. Just looks like a timelapse to me.

Personally, I'm learning Houdini so I can procedurally generate stupid, weird 3D printed art- but a lot of the basic tools may replace other bits I've used in the past.
>>
Yasss
>>
>>641867
LOL I wonder if X-Force manages to crack the license server in time or some P2p realease a hex.edited crack again and X-Force will be sulky like little girls again :"No Houdini 17 from X-Force beacause you've been bad!"
>>
>>641867
Oh shit son! Judging by the background art I'm guessing cloth is one of the things getting updated?

>>641870
Yeah that was really silly.
>>
>>641876
>Judging by the background art I'm guessing cloth is one of the things getting updated?
I guess so. It was like that with hair for H16. Plus in the Polydraw sneak peek there was a new tab visible, "Vellum", which some have hypothesized may have to do with cloth sim. Soon we'll know!
>>
why is importing a camera to houdini such a retarded process? I have to pull it in through a alembic archive node and after 5 subgroups it's there but I can't scale it to houdini measurements, nothing happens. In nuke you create a camera, link the alembic fine, fucking done
>>
>>641867
>>641876
I imagine there will be a lot of game development pipeline, modelling, and animation improvements, since Houdini is expanding rapidly in that direction.
>>
>>641876
>>641878
Yup I've used it, it's fucken fantastic
>>
>>641948
Great to hear. How did you get to test it, are you at a facility that gets the new toys early to test-drive?
>>
I know there are a few jokes in the docs, so I thought, well, maybe this has to do with the Marquis de Sade, and I did a search to make sure. As it turns out, the first results weren't about the Marquis, but a porn actress who goes by the name "Sasha de Sade". Coincidence?
>>
>>643193
>Sasha de Sade
>actress
Googled Sasha deSade : Shemale Pornstar Model
are you trying to get people off Houdini?
>>
>>641948
how it compares with marvelous designer?
>>
>>633041
>>641867
fuck, does this mean my sicrit cool software is gonna be overrun by mongs now? I mean it has been happening for 2 years now, but is this the straw that breaks the camel's back?
>>
>>643448
What makes you say that?
>>
>>643451
I'm gonna be nice I guess. After 16.5 there was a huge wave of newcomers watering down the forums. Everyone in the Houdini community is feeling it, the software's becoming the hot new meme among the Maya/Max/Blender brainlets (jk all these tools are great). Since 17 will probably give us even more tools for non-procedural modelling, game development and other quality of life features it's safe to assume that it's gonna pour more gasoline on the fire.
>>
>>643466
That's obviously a byproduct of making the software more welcoming and giving it more attractive features, but it doesn't detract from the power that Houdini offers. Even if Houdini becomes the leading DCC of choice, you will still be able to do all the hardcore shit you want, and that will set you apart from the majority of the userbase.

Regarding the forums, you have a point. I'm more optimistic, though. Unless there's a massive influx of lazy, uneducated people (not possible with the current Houdini's learning curve, I believe), the community that currently fuels them will stay.
>>
>>643466
so the only thing tat's missing is a procedural pr0n node
>>
>>643536
soon™ there are people who already work on procedural orifices next step is algorithmically describing the actual process
>>
File: index.jpg (5 KB, 167x121)
5 KB
5 KB JPG
>>643541
>>
https://youtu.be/MIcUW9QFMLE

Did Houdini 17 just btfo Marvelous Designer?
>>
File: forRealNigga.jpg (22 KB, 480x480)
22 KB
22 KB JPG
>Watching a Houdini tutorial
>There's a siren blaring in the background of the video
>>
>>643561
GPU accelerated multi-layer cloth solver.

Yep, it looks like it.
>>
Hey guys does this looks to you like cgi or drone footage?
>>
>>643561
Holy shit, everything in that looks incredible. Maybe it's time to switch.
>>643573
I wouldn't exactly say though, even if the cloth sim is far superior I personally find the 2D pattern-based workflow of Marvelous to be its strongest suite. It's what makes MD incredibly easy and convenient to use. Unless Houdini implements a similarly simple interface for their clothing stuff (or better yet: a bridge between MD and Houdini to port patterns and avatars for simulation in Houdini), I don't think Marvelous is out of the game yet.
>>
>>643561
This shit looks BALLER! Although not as whiz-bang as some of the other features shown, the "straighten UV islands" thing is going to really come in handy.
>>
>>643448
>>643451
>>643466
>>643473
This is an actual issue, happens with a lot of software (even worse for software that used to be paid and then went "free"). The forums get flooded with retards asking retarded questions, the devs can't keep up, and solutions get diluted, in the end your actual question that would have gotten a dev answer a month earlier gets ignored (or worse, some brainlet mongoloid tries to hijack your thread with a retarded question).

This happened to Unreal 4 when they removed the fee. Immediately underage and third world hordes swarmed the forums and wrecked the place.
>>
>>643575
It looks like satellite data extrapolated to 3D.
>>
>>635812
Wait so what happens if you use a morph target ona rigged model that changes proportions? That just doesnt work with houdini? Jesus christ.
>>
>>635812
>Just so people can deal with a "more intuitive" concept like a bone instead of an oriented point in 3d space.
Man that's really weird, considering I've heard that Houdini already treats everything as point in 3d space, and houdini adepts are pretty fanatical about pushing that
Why would they choose bones? Joints-as-points would logically fit into "houdini way" as far as I understand it
>>
>>643561
new unity plugin? fuck I have so much to learn

Houdini
Unity
Resolve15
Fusion 9
Substance Painter
Substance Designer
zBrush or Coat3D
Python
Vex
C#

anal fuck me up my ass
>>
>>643641
>>643644
Is there a knowledgeable Houdini rigger out there to comment on this? I'm interested to hear both sides of the argument.
>>
>>643671
man up bruh if you learn all this stuff you will become a god walking amongst mere mortals
>>
File: output.jpg (789 KB, 1024x1024)
789 KB
789 KB JPG
I wanted to get a better understanding of VOPs, so I rebuilt the Principled Shader using the Classic Shader interface. It still lacks a couple of pieces, but it's 90 % done.
>>
> Spend a day learning about how to make HDAs and use them in Houdini Engine
> Carefully rebuild one of my fap setups to turn it into something portable and interoperable with Maya
> Load plugin, fail
> wtf, freaking out
> Assume something is wrong with my complex setup, try to fix it, no help
> Try to build a simple HDA that generates a tetrahedron
> No output
> Open UE, load Houdini plugin
> Works perfectly
> Maya 2018.4 completely broke Houdini Engine

https://forums.autodesk.com/t5/maya-programming/array-attr-inside-a-compound-are-not-evaluated-maya-2018-update/td-p/8233305

Makes me happier I'm moving to Houdini.
>>
>>643851
I wanted to try Houdini Engine for the first time the other day. Fiddled with it for half an hour trying to get it to work until I finally googled "houdini engine maya 2018.4" and discovered that same thread. Good job Autodesk. I mean bugs are forgivable and expected, but why do users have to wait months between point releases when SideFX is able to push daily bug fixes.

We just started using Maya at work (after years of LW) and I'm really surprised at how janky and unpolished the program is. I was definitely expecting more from the industry standard. If Autodesk and SideFX each maintain their current pace of development I can see Houdini taking a good chunk of market share from Maya in the next few years.
>>
SideFX are also pretty good at viral marketing it seems
>>
>>643875
Imagine if the Blender Foundation had half the money available for marketing at SideFX. In a couple of years, Houdini couldn't compete.
>>
>>643875
>people are praising something i don't like, must be the jews at work
Fuckin' Jews, amiright?
>>
>>643892
Who said I don't like houdini? Im currently learning it
it's just all my friends in this sphere and you guys suddenly started talking about it
>>
>>643893
We've always been talking about it, probably so have your friends. You're just becoming aware of it, broski. Plus, with 17 on the horizon talks about Houdini's obviously gonna be more prominent.
>>
>>
>>
>>643990
re: Hydra.

https://www.youtube.com/watch?v=JmH4KYcmHOo
>>
>>643990
>>643991
>>644024

Help a brainlet out. What is the implication of a 3d package having USD/Hydra support? Would it be treated as another import/export format or something like an alternative scene format?
>>
>>644051
The way I understand (and expect) it to be is a replacement of the current viewport. Hydra is an OpenGL renderer intended to work as the direct interface between the artist and the scene. In the Pixar video, they demonstrate it in their own Presto software, how it helps the artists have a better idea of what the scene is going to end up looking like. And at a whopping 60 fps for all the detail involved (ambient occlusion, depth of field, caustics...).

USD deals with scene representation. My understanding of it is pretty limited though. Some core benefits seem to be a much increased performance, improved team workflow by allowing incremental changes in scenes using layers/branches, and working as an exchange format for geo/animation/shading (one step over Alembic, if you wish).
>>
>>644077
Interesting, thank you. I guess we'll know more in a couple of days.
>>
>>644024
damn that's awesome
>>
Is Houdini's work flow similar to Substance Designer's and would be easy to get into if you know it well enough?
>>
>>644268
If you're good at Substance Designer you're more likely to be good at Houdini since it shares a lot of thought processes. Knowledge of Substance Designer won't particularly help you get into Houdini much easier since there's very little overlap in the actual nodes or data models.
>>
>>643888
Viral marketing is low budget marketing you fucken pleb. If the Blender Foundation could create videos showcasing things as good as Houdini they would get the exact same attention
>>
Launch Presentation - Houdini 17 Banshee:

https://vimeo.com/293116223
>>
>>644303
Something tells me blender can't really compete with houdini in these vids, houdini specializes on the stuff after all, they can easily find a senior vfx artist work made in their package to show
>>
>>644428
Good stuff.

>There is one particular feature that we didn't cover that I wanna single out, and that's the integration of nvida's OptiX denoiser into Mantra IPR and batch rendering. Now that's a very cool feature and definitely useful. But the reason I wanted to bring it up, is that it holds a much greater significance to us, and hopefully to you in the future, as it is an early foreshadowing of a major new architecture that we are very busy developing in RnD these days. You've heard of it, possibly as LOPs, or Project Solaris. It's an architecture that goes straight to the heart of lighting and rendering, and in one big swoop, takes Houdini's proceduralism, it combines it with GPU-accelerated workflows across the board, dedicated artist workflows, and it weaves all of that into a live and native USD core. But that's a story for another day...

Kind of sounds like Houdini will be treading into Clarisse/Katana territory, no?
>>
>>643466
It's happening.
>>
>>644511
>implying
>>
>>644490
It can't. A lot of what makes Houdini so powerful is the workflow, not any particular feature. Until Blender catches up in workflow it will never get anywhere near Houdini, nor will anything else.
>>
>>644522
Does it need to adopt that workflow tho?
>>
Why is Houdini's interface on mac os fucking tiny?
iMac 27 incher
>>
>>644530
Aging Qt 5.6 basically, I'm not sure whether Houdini 17 is upgrading to Qt 5.12 but that should improve things

http://www.vfxplatform.com/
>>
>>644524
To get on Houdini's level? absolutely!
>>
>hair card generator
WANT
>>
>>644530
Probably no high-DPI support. Retina screens are considered high-DPI and apps that support high-DPI will look normal on them, however apps that don't will have severe scaling issues (too small like you described). Either wait until high-DPI gets supported or look around for some UI scaling function.
>>
>>644530
Maybe it's Mac-specific, but if not, have you tried the High DPI setting?
>>
I fucking love Houdini.
>>
>>644719
>>644726
already did that

I guess I have to squint then
>>
>>644830
File a bug, then. If they find the same issue it'll probably be solved in a few days.
>>
17 W H E N
>>
>>644876
October 10.
>>
>>644830
Try putting this line into your houdini.env file:

HOUDINI_UISCALE = 150

Adjust the number to whatever looks good. You can also adjust UI font size independently:
https://www.sidefx.com/forum/topic/47950/?page=1#post-216211
>>
>>644882
nice
I wonder will it let me export cards from apprentice
>>
>>644947
Sure why not?
Apprentice does write obj files so it shouldn't be a problem.
>>
File: cat and fish.gif (2.8 MB, 300x365)
2.8 MB
2.8 MB GIF
>>644950
NOICE

I'm currently learning Houdini because I saw an asset that makes cards out of grooms, and it turned out that getting grooms themselves done in Houdini is much faster, easier and comfy than setting up xgen, but the cardmaker asset is pretty limited so I thought "hey why don't they have some similar stuff in house?"
...two days later I was reading some cg news channel and *BAM* they were working on that all along!
>>
https://vimeo.com/293870919

this is best type of autism
>>
>>645237
At first I thought, well, they must've implemented an assembly interpreter in VEX, but then... woah.
>>
Apple wins again, unfortunately. This, and the reliance on OptiX for denoising (understandable, though, given the lack of competition), leads me to assume that SideFX does not intend to stay fully vendor-independent any longer.
>>
>>645241
if AMD's behaviour in the gpu render space is anything to go by, they will shoulder most of the blame.

i expect to see rtx in viewports before cuda over opencl in solvers though.

maybe intel's new gpu efforts will sav..... hahahhahahhahah
>>
File: metal.png (32 KB, 1548x364)
32 KB
32 KB PNG
>>645241
Not so fast.
I sincerely doubt any of the major DCC apps are getting ported to Metal. Apple is pretty much a dead-end for 3D work.
>>
>>645257
Phew, I'm glad I misread too much into his previous post.
>>
>>645241
who the fuck does 3D work on Apple, anyway?
>>
File: h17.png (61 KB, 1186x541)
61 KB
61 KB PNG
It's up homies

https://www.sidefx.com/download/
>>
>>645268
let's hope X-force doesn't fuck up again
>>
>>645276
Why not use Apprentice?
>>
>>645302
>Apprentice
restricted and limited BS version
>>
>>645302
because

here be pirates

I really don't want to learn houdini but it's force is beckoning me like a slut to a dick
>>
>>645312
only on render size.
>>
>>645302
>>645327
How about not being able to export as FBX?
>>
>>645353
And Alembic. Valid point. But you have Houdini Indie for that. If you are making money, that is.
>>
>>645268
HELL YEAH
>>
Introduction to vellum.

https://vimeo.com/294379827
https://vimeo.com/294375836
https://vimeo.com/294374382
https://vimeo.com/294388853
https://vimeo.com/294385572

Have fun with it boys.
>>
>>645367
god damn man
is marvelous finally topped?
>>
>>645370
I honestly don't know. I am barely able to create a t-shirt in MD and a complete nob when it comes to Houdini. (I can use the shelf tools...yeah!)
If MD doesn't move forward it will be over-rolled by Houdini sooner or later, that's for sure.
>>
>>645370
See the latter part of >>643601.
I don't get why people keep saying "marvelous is dead!" whenever advanced cloth dynamics like these pop up. MD isn't about its cloth sim, it's about its ease of usage. Marvelous is super easy to use and has an incredibly gentle learning curve compared to other programs. Its 2D pattern window makes making clothing ridiculously simple once you get the hang of things.
>>
File: Untitled.webm (1.12 MB, 1280x720)
1.12 MB
1.12 MB WEBM
>>
>>645414
I don't know what I'm looking at.
>>
>>645414
I knew where this would go after the first few frames
>>
>>645418
Welcome to the club
>>
>>645418
You'll understand at the end...
>>
I'm working on crowd sims in Houdini and I want to bring them into Maya to render with Arnold. It's proving to be harder than I thought. How the hell do I export as .ASS?
>>
>>645537
Just use HtoA, it's a huge pain in the ass exporting crowd sims to maya
>>
>>645537
Why not use crowd sims plugin for Maya?
There are several to choose from.
>>
>>645559
Not sure if serious or memeing...
>>
>>645586
I am serious but i also don't know shit about crowd sims.
What is the problem with Miarmy or Goleam?
>>
>>645586
Golaem? Miarmy? These do their job in many studios.
>>
>>645594
>>645593
I've used both but prefer Houdini, and for this project I will have to run the simulation on another pc without either of those plugins.

>>645543
I thought as much, the cracked version of HtoA from CGP is a bitch to get working. I'll give it another try.
>>
Playing with the vellum solver.
>>
hey dudes i need a little help with summin:

gfycat /IckyRawIndianhare

this is a very basic popnet driven by a pyrosim
i've just got a trail connected to the output and it behaves nicely until it absolutely freaks the fuck out

additionally if i try and add a collider object to the pop sim, the trail sop starts including points from it initially (and eventually freaks the fuck out)

anyone know why this is happening?
>>
>>645952
oh and the trail is connecting as polygons

alternatively, is there a better way to do particle trails?
>>
>>645938
>and they say this will not absolutely btfo marvelous
>>
>>645987
While it's better than marvelous, it won't btfo it. Think about the kind of people who use marvelous. You think they could learn Houdini in a short enough amount of time for them to justify dropping marvelous?
>>
>>646007
they will if they want to keep their jobs
>>
>>645987
>>646008
It's been said like 4 fucking times in this thread already, it won't because the use cases are entirely different. People aren't using MD for cloth simulations, goddammit. Its main usage is to easily make clothing. Look at all the testimonies on MD's site. It's all about how "MD's helped [game] make stunning clothing!", not "MD's made my tarps and tents look so much better!". Until Houdini implements a similarly easy-to-use UI as MD's 2D pattern window, there's little reason to switch over, even if the end product looks better, because it just won't be as efficient.
>>
You know you can use both, right? I think that's the best way now: design in MD, simulate in Houdini.

At least, until someone makes an HDA mimicking all of MD's patterning/stitching/draping niceties.





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.