[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/3/ - 3DCG


View post   

File: 2.08 MB, 1021x1324, e45p8rqtacn81.png [View same] [iqdb] [saucenao] [google]
941062 No.941062 [Reply] [Original]

I'm back and I'm bored. Last thread ended up having lots of industry talk, this time I'd like to critique people's art a bit more and give anyone who wants it some honest pointers.
Keep in mind my specialty is environments, but I can try giving decent advice in other fields... mostly.

>Q: Why didn't you just reply to the /wip/ thread instead of making a whole new thing?
Not everyone who posts on /wip/ is looking for unprompted feedback, and 90% of it would be telling coomers to watch an anatomical sculpting tutorial.
>Q: What's the most important piece of advice you'd give any beginner?
Find a better place than /3/ to get your art advice from. But since you're already here, let's get started.

>> No.941067

>>941062
Do people in the industry act the same way they do here?

>> No.941069

>>941067
Give me some examples of how people act here and I'll answer that.

>> No.941070
File: 1.74 MB, 1248x1324, standardized.png [View same] [iqdb] [saucenao] [google]
941070

How does it feel to be part of the post-GTC23 Nvidia Omniverse?

>> No.941072

>>941070
I think it falls outside the scope of just this industry.
Either AI will stop progressing and stay roughly where it is now (this isn't as smoothbrain of a take as you think it is), or it'll keep improving, and you and me will quickly have way bigger problems to think about than CG artists losing their jobs.

Personally I see it as a major democratization of creative tools, and as one step closer to the final "coom button".

>> No.941075

>>941069
I guess I was referring to all of the negativity, misery, egotism, elitism, arguing, and complaining about how much they hate actually doing 3dcg or learning.
If I were to get a job doing 3dcg, would the work environment be filled with the types of people that are found here?

>> No.941076

>>941062
how much money do zbrush developers spend to shit on blender sculpting mode?

>> No.941077

>>941075
>negativity, misery, egotism, elitism, arguing, and complaining about how much they hate actually doing 3dcg
Oh, wait, what? I thought you were asking about people being retards or weirdoes. Nah, there is none of that, this is about the aspect in which this board is the furthest away from the industry lol. It's the industry, people want to be there. If they didn't enjoy it they would have left long ago or never made it in the first place.
Work conditions can be another thing but it's not like most of the people who rant here know about those

>> No.941078

>>941077
What the fuck, I keep forgetting this board doesn't have spoilers. Why is that even an option in the first place?

>> No.941079

>>941076
>how much money do zbrush developers spend to shit on blender sculpting mode?
Is it actually comparable? Blender is genuine shit for working with high density models.

>> No.941083

>>941075
not OP but lmao get out of your bubble for once. You seem like the type of person who has never had a job before who thinks everybody is out there stabbing you at the back while taking offense at the slightest criticism against you kek.

>>941062
I am assuming you work in the West, what's a realistic timeline to six fig territory and do people tend to switch to tech at some point?

>> No.941084

>>941083
>You seem like the type of person [...] blahblah weird projection lmao kek
You okay man?
>I am assuming you work in the West, what's a realistic timeline to six fig territory and do people tend to switch to tech at some point?
It depends as much where you work as how senior you are. From observation, in North America/Europe, a good senior from a high paying company makes on average roughly twice as much as a junior from a low paying company in the same area. But there's also massive differences, salaries in California are roughly three-four times what they are in England and I'm not even kidding.
Nobody besides Riot pays juniors six figures, but you can be a broke junior working at Treyarch barely making rent on $70k/yr. Realistically, your best chance of making fast money currently is to have enough years of experience and skill for a high paying studio to hire you remote.

>> No.941086

>>941079
This fine sexy woman has a different outlook on the matter

https://www.youtube.com/watch?v=0ZvyqOAXMZI

>> No.941088

I read the pt.1 thread OP and it was good shit.
Now I'm having existential crisis about the software I'm using. Currently using 3Ds Max (cracked) as as an intermediate artist and I'm wondering if it's worth it in the long run to switch to blender and focus primarily on it. It feels like I'm missing out on a lot. I'm currently focusing on hard surface but I'm going into archivis as well for school. So what do you recommend, OP?

>> No.941089

>>941088
>Currently using 3Ds Max (cracked) as as an intermediate artist and I'm wondering if it's worth it in the long run to switch to blender and focus primarily on it. It feels like I'm missing out on a lot. I'm currently focusing on hard surface but I'm going into archivis as well for school. So what do you recommend, OP?
Hmm, I don't know. That's actually a good question. Blender is definitely superior to Max in terms of speed for hard surface, and has access to more interesting plugins for modeling like MACH1N3 tools, HOps/Boxcutter, etc. Max is also serviceable, but I don't remember anything in it actually making the modeling particularly worthwhile (besides curves, curves are so much better in Max than Blender).

For Archviz, that's a different situation, and I'm not that qualified to advise since I don't dabble in Archviz myself. I remember Max being *the* standard in Archviz about 5 years ago, and it still outshined Blender in terms of plugins until very recently. But now, a lot of the stuff that used to be great in Max, like Railclone and ForestPack, is now available in Blender thanks to equivalent plugins, or their features being achievable with the new geo nodes. Aside from archviz-specific data I/O (which I've got no clue about, but *hope* somebody could've gotten working in Blender by now), it's 100% viable for archviz projects, and is starting to see more and more use industry-wide.

That said, the archviz industry seems to move slower than games/vfx, and I've seen them be fucking dinosaurs about some things. I dunno if some studios have started to switch to Blender or simply tolerate it, but with the number of clients you get it seems like the kind of industry where it'd pay off to be organized studio-wide, and Max would still be the standard simply cause the studios are already on it and switching costs valuable time.

>> No.941090

>>941089
>Switching costs valuable time
This also means that you might eventually see some studios switch to Blender because it's free, and that fact alone makes the switch eventually worth it at equivalent features in face of Autodesk's yearly fee.

>> No.941091

>>941086
I will report on this later, I'm high as shit right now.

>> No.941119

>>941090
the cost of autodesk licenses is like a rounding error on the budgets of hollywood studios and big architecture firms. the budget for that puss in boots movie was $130 million dollars, dude.

>> No.941122
File: 36 KB, 647x651, refencepic.jpg [View same] [iqdb] [saucenao] [google]
941122

Hey quick question, do you know what the colors for the rig controllers mean? I always see it in videos and pictures and i'm working on one but i don't know if what i'm doing is correct.

>> No.941128

>>941086
She has a deep voice. I also like the special olympics segment right at the beginning. I believe her.

>> No.941137
File: 60 KB, 663x422, fruitcake.jpg [View same] [iqdb] [saucenao] [google]
941137

This explains a lot.

>> No.941138

rank the following by their average iq:
>modeller
>environment artist
>texture artist
>lighter
>animator
>rigger
>fx td
>pipeline td

>> No.941139

>>941089
Did you see the KeyHydra plug in for Max? I feel like it's an appropriate replacement for HardOps/Boxcutter and it really speeds up the workflow in Max. I can't live without it now.

>> No.941141

>>941119
>the budget for that puss in boots movie was $130 million dollars
While Autodesk licenses aren't bank-breaking for studios, I think they're nothing to scoff at, either.
For a 50-seat VFX studio, $2000 yearly license x 50 seats = a 100k/year operating expense, times as many years as you operate

Or if you don't like that example, another one: $2000/yr x 5 years = 10k per employee per five-year period.
If it costs you less than 10k in loss of man-hours to make the switch (that's a month of wages if we're being extremely generous), switching becomes profitable after a five-year period.

There's plenty of finer details to argue (and it's a completely useless discussion), but I hope this shows the numbers are of *some* consideration to many studios.

>> No.941142

>>941122
>do you know what the colors for the rig controllers mean?
Red for Right, bLue for Left?
Other than that I dunno if the rest is standard, it looks like random utility colors. Yellow and cyan are common colors for active/inactive elements in various DCCs, green seems to be used to call attention to certain important controls of the rig here though that's definitely not a ubiquitous standard.
I'm no rigger but I don't think the industry has gravitated towards a single standard yet. If you grab enough pictures of various rigs, you should be able to compare them all and extract whatever standards you can from that, though (like the red and blue convention)

>> No.941143

>>941138
>rank the following by their average iq:
They're all equal, being retarded is a lifestyle rather than a specialty.

>> No.941144

>>941139
>Did you see the KeyHydra plug in for Max?
I did not know about that! That's super cool, good to know there's a Boxcutter equivalent in Max now.

>> No.941147

>>941086
I finally gave this a watch, and to my surprise, it seems like all of the feedback pertains to the rendering and camera, not to the sculpting process itself:
>You can sculpt in a real 3d viewport (zbrush has funky perspective which can mess with artists)
>You can place shadow-casting lights on your model (genuinely great for gauging form and something missing from zbrush)
>Accurate corneal reflections (nice QoL but not that impactful)
>Multiple live cameras at once (nice QoL, more impactful)
These features are objectively nice, yeah. They're things I would want, and things all of my character artist friends have lamented the absence of in Zbrush. I don't know if Blender sculpting is at feature parity in terms of brushes, tools, etc., but if it is then you can (and should) use it.
One thing I'd be a bit wary of is performance. Loading a million+ poly mesh on a normal computer in Blender is a fucking SLOG. That said, I tried sculpting on a heavy mesh just now and the sculpting mode remains perfectly responsive.

Heck, if I can get my environment sculpting tools inside of Blender and the performance isn't too painful, I might just drop Zbrush myself.

>> No.941152

Do you have any tips for edgeflow and topology, especially for game art?

As far as my understanding goes, you can remove edges as long as the silohuette isn't affected and gradients don't occur in a low poly mesh.

>> No.941154

What is better, pure subd, booleans with tris/Ngons or booleans+subd cleanup? I keep seeing box cutter and hardops shilled on YouTube but the tutorials mostly deal with ngons while arrimus and elementza mostly favour pure subd.

>> No.941155

>>941152
>Do you have any tips for edgeflow and topology, especially for game art?
If you're doing non-subd, non-deforming game art, edgeflow does not matter and topology becomes about keeping as many faces as possible as n-gons to make things easy to select and work with.
My best advice for topology is to watch other people model, it's a collection of dozens of small situational tricks you pick up over time.

>> No.941163

>>941154
>What is better, pure subd, booleans with tris/Ngons or booleans+subd cleanup?
They all have their pros and cons:

Subd
Pros:
>Nice looking results
>Great for soft blended surfaces
>Lightweight HP model
>Can texture the HP directly
Cons:
>Slow
>Destructive
>Takes longer to learn
>Fundamentally boolean-averse, which is painful since most manmade objects feature subtractive or additive elements
>Can't blend complex shapes without lots of bullshit
>Topo gets exponentially more tedious with rising model complexity
>Making a separate LP can be tedious

Midpoly
Pros:
>Fast
>Goes into the game directly, WYSIWYG
>Option to use nondestructive operations like bools & bevels
>Can be combined with trim sheets, decals, plugs, etc. for even MOAR SPEED
Cons:
>Edges/fillets lack G2 continuity (only autists will care but it does cause a "CG look")
>Meshes can end up with lots of surface area which is a problem for unique texturing
>Higher polycounts than baked assets

Dynamesh + bake
Pros:
>Fast
>Basically all the same pros as a midpoly workflow
>Reuse the midpoly with minor cleanup as your LP model, basically you get your LP for free
Cons:
>Highpoly mesh is unusable for anything besides baking
>Bespoke control over microbevels needs to be handled manually and is often skipped in favor of a uniform bevel width (per-material is fine)
>Still no G2 edge continuity

CAD
Pros:
>Fast
>Easy to learn
>Instinctively close to real-world manufacturing techniques
>Easy to do complex blended surfaces you'd never see done in another program
Cons:
>Programs suck for game workflows (Plasticity will hopefully change that)
>Generated LP geometry can suck, though Moi and Plasticity seem to have decent exporters
>Can do G2 edge continuity, but it isn't worth the effort

The way you listed your stuff is weird, no hard-surface artist would consider something like "pure subd vs. booleans + subd cleanup". You just block out your mesh using bools where they help and do a subd pass after.

>> No.941164

>>941154
>>941163
>Okay, but, which one should I be using?
All of them have their place, and each is capable of carrying out 80% of the same modeling tasks. The last 20% is little pros and cons that make each method better or less suited to your specific situation.

For example, if you wanted to model some scifi armor for a game like Halo, you'd start with a remeshed sculpt, then probably start working on it like a regular subd model because that's the path of least resistance. But if you had to do dozens of cheap hard surface props, midpoly would have tons of benefits. Want to make a set of props sharing similar design features? Midpoly with a trim sheet will get you there fastest. Wanna concept something wild as fast as possible? Go with CAD. Wanna have a final game asset with lots of geo for something like UE5 + Nanite? Pick your workflow, but you can't use dynamesh if you wanna have an easy time texturing it.

>> No.941169

>>941075
I don't work in the "industry" but dude, you need to not think in this "crab in a bucket" mentality. 4chinz is that last place you should be looking for advice. This is literally a propaganda website.. But the mentality and negativity sticks across all of the boards.

>> No.941173

thanks for doing these threads last one was good.

i'm 4 months out of school - did a diploma course from one of those seemingly reputable rookies schools. been consistently looking for junior environment work and applying at various studios both local and remote. I'm more interested in advertising and motion graphics but it seems like you need to be in a city like LA, NY or London to really get into that industry. I'm in a more game and film focused city and have no interest/portfolio for film.

Should I have gotten a job by now? or does that mean my reel is shit and I need to hit the drawing board again for a bit. most people I trust with feedback such as former teachers have told me it's at least passable enough for junior work but i'm feeling doubtful lately

Also I'm in the process of learning houdini on my own, it'll take a while to be able to fully integrate into my workflow but I want to keep bolstering my portfolio with my personal work, eventually adding more procedurally based assets/projects, is that a good way to improve my chances of finding work while continuing to learn?

>> No.941186

Was the software you worked with so drenched in plugins that it didn't resemble vanilla Max or Maya? Did you get to work with any proprietary tech during a learning period before the project? Did you get any plugins custom made for yourself?

>> No.941187
File: 209 KB, 900x770, 1451605946863.jpg [View same] [iqdb] [saucenao] [google]
941187

>>941062
>Ask an industry pro anything
Is dot got ogay?

>> No.941188

>>941173
>Should I have gotten a job by now?
Nah, not having a job 4 months after graduation isn't unusual, but if you've applied to lots of junior positions and nobody got back to you that's a sign you're not very good and could fix some things in your portfolio. Post something good from your portfolio and I can give you a much better estimate of how you're doing, though.
>learning houdini on my own
This is good, yeah. Houdini is useful for many things and few people know it, so it's always in high demand. If you're a junior with no skills and start doing Houdini you might end up being shoehorned as the "Houdini guy" and not be given many chances to learn other things, though, so make sure you follow your passions too.
>I'm more interested in advertising and motion graphics but it seems like you need to be in a city like LA, NY or London
What, they don't do remote for those jobs? Mograph and advertising are big industries, it seems bizarre they wouldn't be found across the world.

>> No.941189

>>941173
Also
>environment work
>reel
Please tell me it's not an actual reel and you have an Artstation like every normal person for the past 8 years.

>> No.941191

>>941189
NTA but does it help to be picky with what you upload onto your Art Station portfolio or do people find more reception through playing a Pareto numbers games even if it's with a second account?

>> No.941192

>>941186
>Was the software you worked with so drenched in plugins that it didn't resemble vanilla Max or Maya?
I never saw anything modded tf out, mostly lots of shitty barebone plugins trying to replicate base functionalities of other programs (like weighted normals in max lmao)
>Did you get to work with any proprietary tech during a learning period before the project?
The project *is* the learning period. (Your managers are obviously aware of this and don't expect you to perform at full speed on the first day, don't stress.)
>Did you get any plugins custom made for yourself?
I was never special enough to deserve it, but after a while I got fed up and started making my own tools. Once you become able to create your own tools and workflows it opens up vastly more possibilities in art production, and unlocking those skills ia as easy as spending a couple months learning something like Houdini, Python, or editor scripting in your off time.
I heavily recommend it even for people who don't have an affinity for programming or math, I've been able to survive on Youtube tutorials and elementary school-level math knowledge for years now.

>> No.941193

>>941187
>Ask an industry pro anything
Okay fine. Ask an industry pro almost anything (I don't know about Godot.)

>> No.941194

>>941191
>does it help to be picky with what you upload onto your Art Station portfolio
Yes, absolutely. The worst sin fresh graduates make is uploading a school portfolio of 1 vehicle, 1 environment, 1 character, 1 animation... all executed at the lowest level of mastery.

Nobody is hiring "junior generalists" (you might as well hire your cousin), so if a studio is looking for a junior *specialist*, they're only going to judge you by what's relevant. Take a studio looking to hire character artists: they're gonna look at your portfolio, see one shitty character, immediately realize you've got a total experience of three weeks in sculpting, and you're not going to get the job. That's obviously a problem.

>playing a Pareto numbers games even if it's with a second account?
This is some next level shit, I've honestly rarely heard an idea this goofy (and by goofy I mean fucking terrible)

>> No.941195

>>941192
What did you work with for retopology? If quad draw then how did you not go insane putting up with this?

https://forums.autodesk.com/t5/maya-modeling/quad-draw-still-not-a-professional-retopology-tool/td-p/7834400

>> No.941196

>>941194
>This is some next level shit, I've honestly rarely heard an idea this goofy (and by goofy I mean fucking terrible)
It's a 3D market place strategy since industry pros who shift gear find themselves not even selling half of their previously made digital assets. So I guess you just take down old cruder work and replace it with specific specialized pieces that show off your mastery with a clean wireframe. Is there some sort of sought after minimum like 3 or 6 pieces? What would you say is a Maximum limit?

>> No.941198

>>941195
>What did you work with for retopology?
Nothing. I might've used a couple shots of zremesher on things like blankets, but I'm not a character artist so it's usually automated retopo for me.

>> No.941200

>>941196
How is a second account gonna help you? Why not upload all your best work in one place? Are you trying to maintain multiple personas? I don't understand.

>> No.941205
File: 439 KB, 1497x863, jung station.jpg [View same] [iqdb] [saucenao] [google]
941205

>>941200
>Are you trying to maintain multiple personas?
Hey that's pretty good.
>I don't understand.
Digital identity hygiene comes in handy. More freedom to experiment and then wipe the slate clean. If one considers themselves an idiosyncratic artist as well, being confined to industrial cog status with their work attached to a single public identity would be a bit suffocating.

>> No.941208

is there age discrimination in the industry? i'm way better at programming than any 3d shit, but it seems like in programming after a certain age you are expected to go into management and never code again. would it be ridiculous to switch to 3d stuff to try to stay creative? maybe if i bring some scripting or shader skills to it?

>> No.941223

>>941208
I'm no OP but I'm the author of popular artwork around here such as >>939054

>is there age discrimination in the industry?
Yes. In the sense that you have better chance of entering the industry as minor than as an artist. Maybe OP can confirm.

>> No.941228

>>941062
what part of Houdini makes 6 figures (large scale sims?) and will it be replaced by AI?

>> No.941230

>>941228
Again not OP but maybe I can help.

>what part of Houdini makes 6 figures
The part that's not you and will never be you.
>will it be replaced by AI
No. Because AI is a novelty item and it has already jumped the shark. Again. Like VR. They'll re-launch it in 20 years or so.

>> No.941234

>>941230
i was listening to the 2016 book "competing against luck" by clayton christensen recently, and in it there's a chapter where he talks about self-driving cars. it's funny because self-driving cars are still in exactly the same place now as they were seven or eight years ago. remember in 2016 when andrew yang ran for president on the premise that truck drivers would all be unemployed in 5 years? don't hear much about that anymore. now all i hear about is how we're going to have electric semi trucks, but i have yet to see one of those either.

>> No.941241

>>941234
One of my all time favorite corporate announcements was that of a startup company that claimed that thanks to their proprietary technology, they were able to move ships using only wind power.
I'm not kidding. I've heard it recently and I think that if phoenicians lawyers were still alive today, they would have had something to say to them.

But there are a variety of choices. Just off the top of my head:
. return to the moon and/or go to mars
. fusion power is 5 years ahead
. cure for [insert something incurable]
. the C language is obsolete
. robot waifus
. linux on the desktop

You get the idea.

>> No.941257

>>941230
>No. Because AI is a novelty item and it has already jumped the shark. Again. Like VR. They'll re-launch it in 20 years or so.
Mad cope based on zero under up to date research.

>> No.941277

>>941208
>is there age discrimination in the industry?
Hmm, I'm tempted to say no. It's an industry that evolves quickly and people who don't keep up do fall off, but the best artists are almost all old guys so people are pretty trusting of seniors (sometimes even too much).

>>941223
>you have better chance of entering the industry as minor than as an artist. Maybe OP can confirm.
Idk, personally I've seen more /3/ posters get booted for harassment than I've seen women.

>> No.941278

>>941228
>what part of Houdini makes 6 figures
In games at least, all of it. Every project is gonna have different Houdini needs, and the work can be varied enough that it would be hard to recommend a specialty (you do end up specializing, but all specialties are broadly equally useful.)
In film this is obviously different. I don't have the pulse on the film industry but I don't figure Houdini sim monkeys get paid as much if their job is just to run the same Pyro sims over and over for Explosion #13617. Don't quote me on it, though.

If you're curious, look up Houdini production breakdowns on youtube. You'll get a good idea of the kind of stuff people make with it in the industry. Dumb rule, but the cooler it looks, the harder it is to execute, and the more expensive potential clients for the feature seem to be (Apple vs Indian mobile gaming), the more money it's likely to bring in.

>> No.941279

>>941062
Do companies get angry when you pirate their products? Or is it more of a
>well, it's not good that you're doing this
>but if you get hooked on our software now, you're not gonna be using anything else in the future
>in said theoretical future, you'll become a customer paying a subscription, and that's good
?

>> No.941280

>>941230
>AI is a novelty item and it has already jumped the shark
Hard disagree with this, AI is incredibly useful in a myriad of ways. I'm convinced (for rational reasons, not out of a hopeful asspull) that it is the biggest paradigm shift in the world of human creative endeavors since the invention of drawing and writing.
Completely unrelated to that, I use it a lot, and it has made my work better, faster, and more fun.

>> No.941281 [DELETED] 

>>941234
OP here, even if AI stopped dead in its tracks today, it would still be a gamechanger.

>> No.941283

>>941279
>Do companies get angry when you pirate their products?
I saw a (no kidding) police raid in India over pirated licenses that resulted in 2 deaths back in 2017-18. So some people definitely don't mess around.
I think it is more like music lawsuits in the early 2000s (Napster, etc.) where the chances of getting caught are low, but companies will make an example of you if you get caught. Some companies have small departments dedicated to rooting out pirated licenses. The Foundry (Modo, Nuke) for example used to be notoriously aggressive about it.
The severity of the response obviously depends on the case, if you get caught pirating as a sole individual and don't have a business to speak of, they'll probably just tell you to buy a license. If you run a vfx sweatshop with 50 pirated copies of Max, that's a different story.
Also, please don't get caught handing pirated material to your clients as a freelancer. It happens surprisingly often, you can get caught, everyone will be annoyed.

>> No.941284

>>941280
You have your point of view and it's a valid point of view, so I'm not going to argue.
Just a funny detail. I've followed /g/ for a while and the people there who play with AI call themselves "prompters".
I can see that one day companies will have job openings for "senior prompters". Beyond that I don't know.

>> No.941285

>>941234
OP here, even if AI tech stopped dead in its tracks today, I would still consider it a gamechanger as-is.

>> No.941286

>>941284
OP here again.
Pro-AI people have this weirdly hopeful view that artists will stick around and enter a beautiful collaboration with AIs, but I don't see how that's realistic. What's more likely is we'll get a period of artist-AI collaboration, until we eventually figure out ways to move the remaining parts of the creative process over to AI, at which point AI is going to be a straight up better creator than humans. Humans will still be able to mess around with art, but it won't be comparable and will be more like your aunt in a wine & paint class. Would you rather play a videogame made by a world-class developer, or by someone who sucks at it?

Why do I say this?
I'm not a /g/ guy, but I'm an art guy, and I can tell you everything that makes art pleasing to humans can be summed up in rules and laws. Our entire pursuit of art, starting from the first cave paintings, has been about discovering and perfecting those rules. Lots of people don't realize this, so they tell themselves that art has "soul", something intangible that makes the work of great artists magically better, whereas the truth is those artists have simply studied a ton and applied certain rules to achieve desired effects.

AI can do that.

I'm not worried at all, though. If AI tech keeps progressing on a crazy incline, there's basically two outcomes: one is a global war for resources, the other is post-scarcity. (The third option is nothing happens and we all stay here.) If I get nuked in a war there's nothing I can do about it, and if I don't, well, we're gonna have AIs able to create videogames custom-tailored to our personal fantasies, and I don't give a shit about losing my job if that's what I get in return.

>> No.941287

>>941188
>>941189
I have my work on artstation too but job applications tend to ask for both that and a reel.

I'm 100% willing to be the houdini guy if i can get to that level, I've worked a lot in substance designer and proceduralism is one of my favourite things in 3D so far

at the small risk of self doxxing here's some of my recent work, the "ruins" piece was my final for school and near everything is from scratch made over several months and the rest is smaller projects i've done since then.

https://www.veed.io/view/8d8f959d-d113-4d48-896b-35e4d893be0c?sharingWidget=true&panel=share

>> No.941288

>>941287
excuse the poor upload quality of this veed.io thing, obviously it's all way cleaner in my actual portfolio

>> No.941289

>>941287
amazing stuff dude, well done. gives a "tech demo for a new game engine" vibe. hope you make it!

>> No.941291

>>941287
>>941062
thoughts on the impact AI tools like midjourney will have on your profession?

>> No.941293

>>941287
>https://www.veed.io/view/8d8f959d-d113-4d48-896b-35e4d893be0c?sharingWidget=true&panel=share
Dude, that's actually some really nice work for a junior. I expected way lower quality than that.

Frankly, the only real issue I can see is that you have a 60% archviz reel with only one game environment. That said that stuff *shouldn't* matter for games, but I don't know if some recruiters are stupid and are going "well, he can model a chair, but can he model a barrel?? let's skip him"

This is above the quality I'd expect for a junior (which isn't to say it has veteran-levels of polish, but it's above what most people would consider "hireable").

>> No.941294

>>941287
>>941293 (cont.)
How many places have you applied to? What sort of job postings? Did you manage to get interviews? It's bizarre you couldn't find work, though if you say your applications *demand* a reel, so are you applying on the archviz side of things more than games? I know the archviz market can be more competitive because it has a lower bar for entry and is very saturated.

Between your ability to put together a decent junior showcase and your interest/willingness to learn proceduralism, you should have a good career.

>> No.941295

>>941286
You know that according to music theory, strictly speaking, there's only one good song and it's Wonderwall by Oasis.
There's something intangible (or not yet tangible) about art that goes beyond optimization rules. There has to be.
In my opinion, AI hasn't proven that it understands "it" any better than a humans can.

>> No.941297

>>941289
>>941293
>>941294
Thanks for the kind words!

>archviz
Yeah there was one promising archviz job that opened up before I'd done anything besides the game environment piece that i wanted to apply to but didn't have the portfolio for. I gave myself two weeks to crank out as much adjacent work as possible, never heard back but on the bright side it was all compiled in ue5 so in a way applies to games.

Most other listings I've applied to have been game environment jobs but admittedly the vast majority have been intermediate level at big studios like EA, the coalition, etc since there just doesn't seem to be that many junior positions that get advertised. Sadly I haven't gotten a single interview yet, told myself it was because the industry slows down in the winter but i guess it should've picked up since then.

I'm at a bit of a loss, i guess my only real option is to keep applying and doing personal work - probably gonna do a more sci-fi hard surface type environment next to show more breadth but god damn i'm so sick of working meaningless restaurant and labour jobs I just want to start making money doing what I love.

Really appreciate the input though

>> No.941298

>>941291
>thoughts on the impact AI tools like midjourney will have on your profession?
I have lots of thoughts

I'm sad that AI immediately became a "sensitive topic" in the industry. I get people's fears, but it's sad so many have very visceral and reactionary opinions towards it.

A lot of people don't seem to use AI when they could and should. I don't know if these people are aware of what AI can do and choose to ignore it, but if so, I don't get it.

One of my friends had to work with a concept artist so bad he started feeding his concepts into AI to fix them. We don't understand why the studio still pays for this person's work.

Right now, the biggest appeal of image AIs like stable diffusion is lightning-fast concept generation. It's so good at giving you what you want and mashing things together. It's also great at generating quick textures for visdev.

I can't wait to see AIs enhance and eventually replace current photoreal texturing methods. The need for increasing asset fidelity has slowed down game development cycles considerably, this is a chance to take some of that time back.

AI in its current state is "dumb" in certain ways. It doesn't understand composition (the details are good but most broad-level compositions are highly incoherent) and it doesn't have a good structural/metaphysical/whatchamacallit understanding of the things it creates. These issues are fixable (I don't know how tf MJ does it) but in the meantime it makes AI a bit hard to use if you're not a decent concept designer yourself.

AI can and should be used to train your own art. Sketch something, feed it in an AI, see a better version of your sketch. Copy it. Good, you now know ways to make your sketch better, and you don't even need an AI for the next one. I've been teaching myself how to draw with AI and it is CRAZY GOOD for that. It's like having the best art mentor on the planet on call 24/7.

AI will massively speed up the development process of solo devs and small teams.

>> No.941299

>>941298
Wolfram says that AI is going to stick around as a human-computer interface and I believe that.

>> No.941315

>>941299
i agree with that. it will finally make conversational interfaces usable unlike siri and the other shit we've had so far.

>> No.941334

>>941277
>I've seen more /3/ posters get booted for harassment
It happened to me in the early 2000. 4chan didn't exist yet. I was working as programmer at "relatively big company".
One day I noticed another programmer was wearing mascara and I said something about it. I don't remember what I said but he reported me immediately to management.
I didn't get fired that day because I was the leader in a project but everybody made sure to send me the message that I was done there.
A week or so later I told everybody to suck my cock, left "relatively big company", went to work in a factory and never ever looked back.
Fuck "the industry". I'm glad I left it and I wish I never touched it.

>> No.941350

>>941163
So if i'm using the dynamesh + bake method, do I have to use live booleans in zbrush or can i just boolean in my preferred software then import into zbrush?

>> No.942000
File: 16 KB, 450x450, 1665412094297843.jpg [View same] [iqdb] [saucenao] [google]
942000

>>941086
>This fine sexy woman
>>941128
>She

That's a faggot man