DarkMatter22 posted at 1:19 PM Tue, 10 September 2024 - #4489187

I’m still stuck trying to make a cube that doesn’t look like a sad marshmallow.

Subdivision modeling gone wrong? Might need hold/supporting edges in three directions instead of just one?

As Lobo said, if you wanna throw up some pictures, and what software you're using someone around here can probably point you in the right direction.

As a character artist? Probably not yet. It looks like you've generally got the technical understanding of the software/workflow for creating a character, but you need to develop some of your "artistic" fundamentals further, proportions and anatomy most notably. If that's something you want to pursue, I'd recommend getting away from characters covered entirely in body armour, baggy clothing, helmets, gloves and pouches for bit, and posting some figure studies some where that you can get a feedback/revision loop.

Your props and vehicles look like they might be able to get you in the door somewhere. There's a few instances of wonky geometry/flow/density in some ares of the wireframes, but nothing that can't be cleaned up/corrected easily, and everything else about them seems pretty good.

Two I cooked up over the last week or two. Generally I don't texture/render in Blender, and to date I've hadn't actually used 2.92  [the newest version that is still supported on my PC] for anything beyond setting it up ages ago,because I was quicker/more comfortable with 2.66 and 2.75 for most of what I use Blender for, so these took quite a bit longer than I anticipated, aw well.

The Scrabble one was rendered with Eevee, cause I didn't realize THAT was the default render engine [until I called it done, and was half way thru making the neon one]. Who makes a realtime engine the default? Renders sure were snappy tho! Some of the texturing and lighting issues I fought with on that one were because of the engine. The neon one was rendered with Cycles.

Happy year of the Dragon!

snopcyzee posted at 6:24 AM Mon, 5 February 2024 - #4481262

Everything else on the list looks simple enough to do in about an hour a day.  A screwdriver and screw?  Really? 

Put your polygons where you mouth is then.

You pick the item, we'll get a WIP thread going, and we'll both model it. Proper poly modeling tho, box modeled, lofted, extrusion, whatever, but none of that zbrush/etc tom foolery. If you're gonna throw shade at a decade old thread, we're kicking it old school.

Yes ... I know I just challenged what is most likely a bot to a model-off ... but whatever.

I think what you're looking for, would have been in the Texturing forum before it was sent to the nether, if it was anywhere. Would have been nice if those forums could have been archived rather than . . . disappeared.

I don't think moving materials between different software gets a whole lot of attention because the scenarios where you'd be doing that most frequently would be using  UVs/image-texture maps, so the information is "baked" into the image, its not very common to move "procedural" or shader-based* textures between different software packages, there's just too many differences [well, render engines technically] for that, so non-image based texturing tends to be very software dependant.

The general concepts of texturing hold true across all software packages. For example, if you're making a glass surface, you're going to want to apply a "fresnel" effect. How you apply that is going to vary from software to software, and there's usually going to be a few options that each come different with pros/cons in every software packaged. Same goes for everything else. Marble? You're going to want some sub-surface scattering on there. How you get that is going to vary from one engine to another, and there's likely to be a couple options. It comes down to learning the general concepts, and then figuring out how to apply them in your software.

Usually artists using multiple tools aren't using them for the same purpose, ie, they're not rendering in half a dozen different programs. They might be modeling in one, painting textures in another, rigging in something else, and rendering in a fourth or something.

*Some software packages/render engines do support certain shader-types, say, Render-Man type shaders for example.

. . . well, at least the alliteration in the first line is something that resembles sense then?

Sure is a stretch, but, what can you really say about [********]? You kinda have to just pull words outta the air and pretend they're relevant i guess. I can only imagine whoever was tasked with writing that hated their job that day.

I know you said you're not going to read any responses, just drop an argument then run away which is kinda weak sauce, I could make some comment about how that might make it look like you know your argument holds limited merit or something, I'm not going to [or I just did?] but that's okay, I'm going to take the time to reply. You've made some silly points, and you've spent nearly half your post contradicting your own argument.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

 i don't like the quotes in the title around "art", because they're the same as they were when other breakthroughs had been made.

That's okay, you don't have to like the quotes, but they're appropriate at this point in time. As it stands now AI "art" is not recognized as such. It is not eligible for copyright [nor is a photo taken by a monkey but that's neither here nor there].

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

hand painting artists said that as photography came up, since a click on a button could immortalize an image they'd need hours/days/week to paint by hand. or anyone doing hand made art as computers came up, just 2D graphics/photo retouch. and then again when 3D came up. and inside the 3d community, you have the hardcore creators, and then the lesser users of premade assets... etc etc. so now it's AI, and all the previous artists are again jumping up and down, because they had to do their art the harder way than the newcomers.

Sure, there's always going to be push back from the old guard against new technologies, but there's a key difference between all those arts that came before and AI "art" now, and that's that they all require a skill set developed by human beings and imparted to the work. Sure, in each case, perhaps the amount of work required changes, but each new medium still requires skill developed by humans.

We'll circle back to this, because you've referenced skill multiple times throughout your post.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

well, as long as people can spread feces and splatter random paint on a canvas, or stick some weird materials together as an "installation" (with questionable results, not talking about occasional great works), then give it an awesome trendy name and call it "art", and worse, sell it for abominable amounts while people starve and animals die - i don't think we should start judging what's "art" at all, because then it becomes complicated and a lot of hypocrisy is shown.

It is probably relatively pointless to rehash "what is art" at this juncture that horse have been tenderized long enough. I'm not entirely sure where you think the hypocrisy comes into it, but there is indeed some rather "meh" art out there, "art for art's sake" and a lot of stuff that is only art in the "eye of the beholder" so to speak. I'm not sure any of that bolsters your argument tho.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

 same for 3d stuff now here on rendo more particularly - as long as people can and do render boring and often offensive clichee chick nudes with poor skills, or feel the need to upload every single render they've made (so since they do, why shouldn't others..hmm?), or take random photos of every stone, tree, sky bit, dog etc while walking to work and expect from everyone to call it "art" on the same level as people who have created beautiful inspiring hard-worked-on 3D works, i don't think we can and should judge people who create with AI.

We're back to "skill". You say we shouldn't be judging people who use AI, but you've just judged the the ever-loving-crap out of I can't even guess what percent of the users of this site.

Nobody is calling the random picture of a stone someone took on their walk to work and posted on 'Rosty the next "Lunch atop a Skyscraper". They're not calling it that for a myriad of reasons, not least of which being the lack of skill shown taking the photo, lack of composition, poor lighting, lack of focus, mundane/trite subject matter etc, etc.

I like to think most of the users have that self awareness tho, and are enable to self critique enough to realize they're not creating a masterpiece with every image. That they can look at their photo, and photos from other users in the gallery and see that their photo might be better than some, and worse than others.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

either we only accept high quality self-made work which basically means suffer, getting frustrated, starting over and over again until you've reached a point and learn to manipulate a brush and color or clay etc, and maybe but really maybe also extremely magnificent photography showing some grand skills and subjects, and maybe amazing 3d art that is innovative and truly beautiful/esthetical...

If they take enough pictures of rocks on their way to work tho, and post them, accept and implement feedback, continue to iterate, that's the point innit? Develop skill over time, and then maybe they'll take that picture someday that IS something.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

... or we are inclusive of all kind of manifestations of art and skill levels and amount of work involved, no judging - which means people who create with AI are part of it. period.

You've mentioned skill several times and seem to be dividing the images posted on this site by the users skill, where's the skill in AI art? Learning what prompts give certain results? Can you ever really get "better" at AI "art"? If an experienced AI "artist" sits down at the computer next to someone that's never used AI to generate an image before, and they both type the same words into the same engine is there not an equal chance of it spitting out the same image? I know the chance of it generating the SAME image is borderline non-existent but that's not really the point.

Is the more experienced user going to generate the better image [all else being equal]? Is the more experienced user going to make a better choice in lighting or composition or selective focus? No, they're not, they can't. AI "art" as it stands now doesn't allow the individual that generates the image to impart those learned skills that are [generally considered] fundamental aspects of what art is onto the image creation process.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

as to where AI learning materials have been "stolen", that is another discussion. involving real technical knowledge of how AI works and has been learning, and ethical choices in art and life (every artist complaining about how their livelihood is threatened should also be vegan or even fructivore, otherwise they are actively supporting the killing of other lives! not just livelihoods. and same for their clothing etc, only ethical - but do they all?)

Good grief, that's a silly extreme. If I had to make one guarantee, it would be that you won't see the sun set tomorrow without having benefited from something else dying. There is no life without death. You seem to be dug in at animals, but you're okay with the slaughter of living plants? You're okay with plants? What about insects? Insects probably died in the production of those plants/fuits and vegetables. Forget your "ethical" clothing, something died for it, its only "ethical" so far as you can bury your head in the sand.

You might even have stepped on a dozen bugs on your walk to work, while you were distracted by someone taking pictures of stones. That's life. Nothing lives without something else dying. Let's call the "lives vs livelihood" argument the poppycock it is. If you want to wax philosophical about the ethics of existing, have at it, but its not relevant in the least here.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

and how every art form learner has basically been copying others (would beethoven and monet complain if they were stilll alive?

If someone was skirting the confines of the law, you bet they would! You know what, some of the "greats" did in their time.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

how many musicians start by covering others or use pre-made sound elements? do you brain bleach every time you look at a render or any art form just in case your brain would unconsciously get inspired and copy parts of it in your work?).

Interestingly enough there's a whole system in place where performing covers of a piece of music actually gives money to the original recording artist in the form of royalties? [Is it a fair amount, probably not, music hasn't been a terribly fair industry for a long long time.] And pre-made sound elements [or samples], those are usually a bought-and-paid for elements right? Kinda like the stuff in the marketplace here. So again, original creator, benefits, blah-blah blah. When those elements aren't legally sourced [or exceed the length legal sample length], or someone benefits from covers without credit/royalties ... then the [copyright] law gets involved.

Brain bleach? I mean, maybe after reading something really dumb on the internet, but otherwise, no.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

i do NOT agree about how nobody was asked before the AI techs started scraping images and texts all over the place, this was wrong, and should have been done differently from the start.

This, I think, is the core of what a lot of the problem is no? A lot [most? all?] of the AI engines have been trained in a less than ethical manner. I don't think you'd see nearly as much of a hard-line push back were that not the case. People would still be calling it out for what it is, and they'd be rightfully [in my opinion] disappointed when an AI image is posted next to theirs without being identified as such, but the fact the AI engines scraped huge amounts of work without consent, is a very serious issue.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

 on the other hand, everyone should realize that once something is made public, and especially when made accessible on the internet, it's prone to get used by others, copyright here or not, so you want it safe, leave it in a safe, shit happens. anyway, that ship has sailed, and the databanks have been made. should the new technology get boosted back into the stone age now?

Copyright or not? Protection when you make something public is the whole point of copyright. It gives you legal recourse when someone uses your work without permission. Just because something is made public shouldn't mean its free-use or tough-cookies to the creator.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

anyway, that ship has sailed, and the databanks have been made. should the new technology get boosted back into the stone age now?

I don't see any reason AI engines trained in a less than ethical manner shouldn't be wiped. I kinda think that could be a starting point to fair compromise. They're not a living thing after all, and they can be trained again, the right way in an ethical manner. Or can they?

I imagine it would be a LOT harder to re-train them "in the light of day". Be a mighty tall order to get artists to allow it. Unfortunately you're right, those worms may not be going back in that can.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

so why not embrace this new form of creation?

You'll note in one of my previous posts, if you took the time to read anything before your diatribe-and-dash, I said there is probably some merit in AI generated images and that it will probably find a use for concept/idea generation before being refined by an actual artist for production. I have a hard time seeing a scenario where something considered "final" could be completely AI generated when there's so little control over the actual output.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

 at least when no (big) money is been done by totally cloning someone's work. reminder: an art style is legally not copyright-able. only precise works.

Curious. Why does [big] money have any impact on what's right or wrong? Before you were concerned with animals lives to sustain other lives, but now your concern is right and wrong when money [or should I say livelihood] is involved? Ethics, only when money is involved? Seems counter intuitive to what is "ethical".

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

 there is some/a huge lot work to do about the legal aspects, the use, etc. and ways to make AI access impossible if you don't want your stuff to be scraped should be implemented. but compromises should be made on all sides i think, somehow.

What compromise should artists be making? Should they be forced [I mean ... "compromise"] to allow a limited forfeiture of copyright to train AI? That doesn't seem a very fair compromise

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

so, i will not have someone tell me that some images i really loved making in AI, sometimes quick but sometimes with effort, and that i consider beautiful and pleasant to look at for others, are basically always worth less than any other 3d render or photo or whatever independently of the effort made by the other person and the resulting creation. nope. never.

I don't think anyone is discounting the joy generating the image might have brought you, or your satisfaction with the end result, I'm certainly not, I've said as much prior. I think we are just asking that you have a little self-awareness, and recognize the difference between generating an image with AI, and an image that was created through the development of skills [whatever stage those skills may be at] through hard work over time.

You wanted compromise, asking that you don't roll up in here drop an AI generated image and proclaim, "now I'm as good as the best artist!", seems like a fair place to start.
You can ignore anyone that tells you it is worth less than an original work, when someone else uses it for something, don't complain, after all you did make it public, and you did so knowing you don't have copyright protection.

I wonder how fast knickers would twist if artists were to start creating work based off imagery generated by users of AI. That work might be eligible for copyright? Its created based on an image that isn't copyrighted, so I'm not sure you could call it derivative, and it would be created with direct input and control by a humans . . . that could be an interesting wrinkle.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

and as insults go, according to the mention above, i should feel just the same then, when i spent hours/days on a daz scene, trying to create something pleasant or interesting to look at, and someone else is shoving some quickly or poorly made discutable stuff and is considered just the same... because so, basically i can produce any crap i want, as long as it's not AI then? ok, duly noted. lol. interesting point of view, worth studying.

And we are back at skill, and judging others . You clearly wouldn't consider "a poorly made discutable" the same as something you "spent hours/days on" given the derision in your post, why do you think others would consider it the "same"? Do you think the creator of said "discutable" considers it the "same"? Do you think someone that looks [insert your choice of "best"] image on the site, and [insert your choice of "worst"] image on the site side by side thinks to themselves, "yep, same!"? I have a fairly bleak outlook these days, and even I still [usually] give [most] people that much credit at least.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

also, some people have opted for AI instead of 2D/3D or other forms of art because they have physical limitations, or technical limitations, and their own story. but i guess anti-AI fanatics don't give a s***, as long as their narrative doesn't allow for this much consideration.

Anti-AI fanatics? Narrative? Persecution complex?

I can't say i've seen anyone belittle someone for using AI because of a disabiltiy. I'm sure its happened in some back-water recesses of the web, people suck and this is the internet, that's how it rolls unfortunately, but in any sensible debate on the topic? No.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

AI is a threat in life, that's a reality - because PEOPLE use it for fakes etc. the people are the problem, not the medium. it just helps. ban Ai and they will find other ways to manipulate people, they've done it for ages with tv. besides, y'all most probably have AI in your phone, and/or smart objects and houses. who doesn't use google or other search engines? online clouds? online subscription apps? everything today is a threat in a way. the whole problem are dishonest people, autocrats, etc. they just have better methods with AI. i don't agree with a lot of things. but AI art for fun like here is by far not the worst threat.

I mean, sure, people are the worst, I'm not going to disagree with your there. I'm just not sure I see how people being awful has any relevance on the merits of AI generated imagery. You're right tho, there are bigger problems in the world than someone generating AI "art". Wonder if it would be as much fun on "ethically-trained" AI engines tho.

Seems a lot of AI engines limit the types of imagery you can generate, isn't foundational censorship like that somewhat counter intuitive to the freedom of expression that is art? ... Anyway!

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

anyway.

so, i am not subscribing to this thread. i did for another and wasted a lot of time. i just said my opinion, i'm not arguing. which means i'm not following nor reading any answers. because basically, those against Ai art won't budge an inch, and those who support AI share some/all of my thoughts already and i don't have to inform/try to win them.

Really, what was the point of all this? If you were only interested in an echo chamber surely you could have found it elsewhere? If you were only upset by the use of "art" surely you could have moved on, or at least spent less time on a more concise post.

Half your post argues that AI "art" is good because it eliminates the element of skill making everyone equal, while simultaneously deriding people who post images that aren't equal to you.

maneki_neko posted at 11:16 PM Sat, 9 September 2023 - #4474387

bye, i won't be seeing you here cuz i'm not following, nor in paradise, cuz i'm an AI art supporter so i'm going to hell ^^.

Viva la Skynet ... I guess?

Gosh, they don't make replying in-line easy around here, hopefully this isn't scrambled.

Ehhh ... not a fan.

I just skimmed thru it, but it looks like most of the creasing/weighting could have easily been sorted by an extra loop, and that would make the mesh universal to all software. Using creasing/edge weighting you're left with a mesh that's software specific [possibly even version specific], and you've only saved maybe a couple hundred polygons, if that? I'm sure there are situations where it has its merits, but I'd rather see proper edge flow without the crutch.

keppel posted at 12:23 AM Sun, 30 July 2023 - #4471479

Whilst reading your replies it occurred to me that I might be able to do a series of front ortho renders of the column.  Render, rotate, render, rotate etc, Then take the image sequence and stitch them into one image and use that image to generate, normal and displacement maps for a new mesh.    Thanks again for the help. 

Ooof! Good grief don't do that! Please! I know I said not to discount brute force, but that's not what I meant!

You're far better off baking a projection from one mesh to another mesh, even if you use a basic 24 sided cylinder, than you would be trying to stitch a series of images together. You wouldn't necessarily even need to "flatten" the cylinder after that. UV map it, which will given you a flattened/"unrolled" image when you bake it. Then create a flat plane, that shares the same UV space/image.

Yup, what @LuxXeon said is probably the logical way of doing it. Bake/project the high poly to a displacement, apply it to a lower poly mesh, and unroll that mesh.

Without seeing the mesh [how much relief there is in the carving/sculpting], I wouldn't discount pure brute force. Volume/area selection tools, and going in small increments, it'll take you a bit of time but its not difficult, just tedious [you don't have to be perfect, the process is usually somewhat forgiving]. In the days before modern UV unwrapping tools that was sometimes the only way to do it!

Or as @Lobo3433 mentions, you could just rig it. A ring of bones, rotate them until they're straight, and save the transformed mesh.

kenmo posted at 4:37 PM Thu, 20 July 2023 - #4470677

heddheld posted at 8:17 AM Tue, 15 January 2013 - #4031332

theres even a jobs thread  on BA

but another one you may like to look at is http://www.cgsociety.org/

its blender forum isnt as busy but has some great artists there

CG Society appears to be down or no longer available.

Sure does!

I would assume that's a temporary thing tho. Sever failure/migration/upgrade, DDOS/hack, something. I just received their standard newsletter this afternoon. I can't imagine, even tho CG Society is a shell of what it once was, it'd be going offline anytime soon, at least not without more fanfare.

Actually, now it appears to be back up.

Crystal46 posted at 12:42 PM Thu, 18 May 2023 - #4465700

I know this comment is going to be misunderstood but I am going to say it anyway. This image is something I made after studying AI for about 2 weeks.

I always wanted to be a 3D artist but could never get the hang of 3D.  I have been loving and admiring all of you who can create amazing 3D since 2007 when I joined Renderosity.  Fast forward to 2023, and I can create art that rivals any 3D artist with a few clicks of a mouse because of AI. I realize that AI still has a long go. For example, AI has mastered the human face but still makes horrible hands and extra body parts sometimes on the subject.  With 3D, the artist controls every detail of the image.  With AI, you are at the mercy of the computer.  You have to keep entering prompts over and over until you are satisfied.  Writing good prompts is the skill.  With that said, I love AI because now I can play with you guys on your level. 

I'll do my best to be polite.

The lighting and overall aesthetic are nice. It would make a solid pre-vis, style guide, or concept image.That said ...

"...mastered the human face..." has it tho? The dude in the grey suit [left side of the picture on the monitor] ... is wearing his face backwards! Backwards! Ooph!

Also peculiar the display doesn't fill the monitor. At first I was willing to pass it off as a "slideshow" type app, but you know what, there's an awful lot of docked icons at the bottom for that.

There are somewhere between, 200 and far-too-many-to-count-keys on that keyboard, thereisnospacebartho.

The left edge of the desktop deteriorates into ... well I'm not even sure, we'll call it a touch of surrealism. The right edge of the desktop is a bit wonky as well. Its a mix of solid and hollow that probably couldn't work in the real world.

The back of the left side of the desk where the drawers meet the wall isn't correct. At the bottom where you have the light and the floor board, that indicates a fairly sizable gap between desk and wall, at the top however, the tightness of the shadow indicates a very narrow gap.

You've got two handles on one drawer on the right side. The drawers on the left side have a clear protrusion from the front face on the left, but are flush at the right edge.

Reflections on the desktop seem alright, except the apple logo, and the black label on the one candle, that reflection clearly isn't correct.

The reflections in the picture on the wall shows a double pillar in the window, but there's only a single pillar. Its reading as "this is supposed to be a reflection, play along please" rather than an actual representation of the "space" its reflecting.

The whole desk in general is pretty odd, there's little to no signs of or way it could possibly have been assembled [short of superglue and will power?].

Reflections on the desktop seem alright, except the apple logo, and the black label on the one candle, that reflection isn't correct. The reflections in the picture on the wall shows a double pillar in the window, but there's only a single pillar. It generally reads as "this is supposed to be a reflection, play along please" rather than an actual representation of the "space" its reflecting.

That candle on the right side is doing all kinds of strange, we'll throw that in the surrealism category.

Some of those flowers are pretty suspect. Left bouquet, there are at least two that are pretty indistinct "swirls".

The chair is missing a leg. That should be a five leg base [based on the positions of the four in the image], and while you could make the argument that the fifth would be [mostly or entirely] occluded by the center post, cause, you'd be right it probably would be, you should still see the caster and its interaction [shadow/reflection] with the floor.

Those are the things that jump out to me at first glance. We'll ignore some of the general amorphous ambiguity of some of the "pictures" in the image and chalk that up to being them "abstract" or visual representations of "lorem ipsum".

"Fast forward to 2023, and I can create art that rivals any 3D artist with a few clicks of a mouse because of AI." Honestly and simply put ... no you can't. I feel like that statement is an over reach at best, as you admit, "... AI still has a long [way to] go." At worse, its borderline demeaning and offensive [I'm sure you didn't intend it as such].

Replacing your arm with an air cannon, doesn't make you a major league pitcher. Personally, I see more merit in what you might consider "a lesser quality image" that someone who is struggling with/learning the process might produce than hitting a "make art button".

If you want to continue down the AI road, and there's nothing wrong with that if that's what makes you happy, I would suggest a bit more effort/focus on post work. At least some of the things I listed, would be pretty minor/easy things to clean up in post, would elevate the image quality significantly, and make it less obviously AI generated.

Concerns, other than an active thread disappearing with the Lightwave forum? :(

Its possible I'm just overlooking it., but can we at least have an archive function, a way to search the threads in these now deleted sections? There's a lot of knowledge and information that's now gone.

Edited to add -  A little heads up would have been nice to have. I knew there were discussions about the possibility of some forums being axed, but there probably should have been a post, in those forums, maybe a week or so before they actually got the axe?

benwhe posted at 10:46 PM Sat, 25 March 2023 - #4459598

For AI: I would like end user to be able to use simple script commands to tell the model what to do. Like go there and pick up that. I remember hearing about such an existing program but it had primitive stick-figure models.

AI step up: I would like end user to be able to give model some traits and goals. Then see how it interacts with enviroment and other models. I remember hearing about such an existing program.

I think you're describing video games. Good news then, there's not really a reason to be using AI to "drive" the model, the end user doesn't need to use anything as cumbersome as a script to make a character walk, we solved that eons ago; arrows keys or buttons on a controller can do that, and another button push to pick up an object. ;)

So like ... the Sims?



You can ignore just about everything I said in previous posts, none of that is relevant anymore, its no longer even a question. You're not getting ANYWHERE near a realistic model/animation that's going to fool anyone. Look at gameplay [not cut-scenes or pre-rendered video, not that that's fooling anyone either, but actual gameplay] footage from something like Horizon: Forbidden West, or CyberPunk 2077, or God of War 5, those are probably about the peak of realism in real-time graphics right now. They look great, for what they are, but they're not fooling anyone. Bear in mind, those were done with very large teams of artists.

Almost all the stuff I talked about above is exclusive to offline render scenarios, relies on dozens and dozens of hours of render time per frame [after hours and hours and hours of physics calculations], then a hefty amount of time compositing multiple renders, lighting passes, subsurface scattering, occlusion, reflections, specular, etc. If you're thinking about real-time graphics, which you are if you want a user to be able to give a command and watch it executed ["now", not two months after they input the command and have completely forgotten what command they even issued] then the bar for "realism" has to drop, like through the floor, through the floor's floor, and into a very VERY deep crater in the basement, in fact, put cinder blocks shoes on your expectations, and drop them in the ocean.



It sounds like what you're really after is a simulation, that has hyper-realistic graphics? You're going to be handicapped in what you can do by modern computing power. The more advanced the AI in your simulation,and the more "actors" in your simulation, the more you'll have to compromise on the graphics. The more complex the environment they're interacting with, the more you'll have to compromise, etc etc. That's likely the reason what you're recalling had stick-figures [well that, and it was probably done by one person/programmer without help of any artists as an experiment where graphics weren't important].