had to come back to make a second comment. thank you for being so generous with your tutorial and allowing other creators to flourish. This is exactly what a great community can be comprised of where everyone can rapidly grow and benefit. People like you are the real MVP. Wish there was a double like button!
in Photoshop, you can right click any layer and Quick Export as PNG, which is a huge time saver instead of hiding the layers and exporting the document. Better yet, you can select multiple layers and it will quick export each one as a separate PNG. And if you do this operation on a layer group/folder, it outputs the merged result of that folder. Overall a great feature that is often overlooked. Keep in mind the exported PNG will be cropped to fir the contents of the layer (or layer group) and not the document dimensions.
It's the best 3d Max video tutorial I've seen on youtube in many years, I feel the awkward need to download your video, I think it's so good that they could delete it.
This is EXACTLY what ive been thinking off! IMAGINE how much easier it will be to make texture work, how entire games can have tons of attention to detail without costing YEARS to develop, bigger games, better models, greater ideas, this really is the future, what a time to be alive
@@TehBananaBread The value of human work will never be surpassed by AI, the AI can make the work easier for those who dont mind or dont notice the difference specially in the texture work department. This is great for older games, but newer games that intruduce new kinds of texture work such as PBR, and different kinds of Height map calculations may need human work over AI. Ive been working with 3d modelling for years now and this is anything but bad news for us
@@TehBananaBread as a person that's spent years learning these skills, this just makes me more productive. Granted, my increased productivity will be a boon for my employer and reduce the amount of overall labour he requires... but at least I can charge more for my work :P
This is what AI imagery was made for, giving artists sick ideas. Really cool vid. (And I have very little knowledge about much to do with 3D modelling)
@@n1lknarf A production artist also needs to work quickly, often to unrealistic deadlines, so anything that can speed the process up along the way is welcome
While the end result is the absolute bomb, What I enjoyed far more was how you went about the process. On of the things I love so much about 3d work. It's not what you make, it's how you make it. And how bloody satisfying it is when you nail down an answer to something that's been thorning you during a project. :D
Thank you got sharing your process! It's very encouraging. The possibilities are endless. I think you should make more videos doing this, even if not tutorials, the general digital community would find this really entertaining.
This process seems so straightforward, and the use of height and depth maps to do the vast majority of the sculpting makes me wonder if this process could be automated as part of a Midjourney prompt straight to 3D app.
This was really helpful. it made me realize some thing. all of the art tutorial channels that I watch keep on preaching the importance of being able to deconstruct objects to their basic geometric shapes. sense I don’t usually do my own online work I never really thought of it as being terribly applicable to me. A useful skill to have absolutely but not as necessary as they were making it sound. Now after watching this I realized how incredibly powerful that skill is. All I can say is thank you for making this video. Also I’m definitely going to check out your channel and see some of your other videos. This looks pretty dang cool
It was a little quiet for me too, at least compared to other videos. Got a little surprise when I opened a blender tips video after this one and my ears got a little surprise lol.
@@joechapman8208 It can be hit and miss, Joe. CHeck out my other video where I show the prompts I used in Discord. You can also grab the assets I used on Gumroad for the price of a cheap coffee ua-cam.com/video/RztJGuhC5so/v-deo.html (asset link is in the video description)
Flabbergastingly beautiful results! Thanks so much for the breakdown - I expect AI to replace nearly all creative jobs within industry, but it’s a bit tough to gauge just how rapidly the technology is progressing. I’m certain these results will become as easily automatically produced as the straight up 2D MJ-type stuff we have already, so it’s good to see we might have at least a little wait before that happens. Again - great curating/art directing here, and lovely job bumping it up a few levels in post!
Don't be so depressing, not all jobs will be replaced. These things are tools, not living minds. Sure some tedious jobs may become obsolete but real art will always require a human mind at its helm - otherwise it has no value beyond just being functional or interesting. It requires substance, story, some humane quality that allows us to connect with it on any emotional level. And even if someday an all powerful AI with human skills came to be, would people really prefer buying something generated in a couple of seconds by a robot over the hard work of a master at their craft? Corporations and people on a budget, sure, but the labor and thought put into a piece should always outvalue something which can be discarded as easily as it can be created.
@@GAZEREAPER - Sorry to come off as depressing. If you’re a creative person working in (or hoping to work in) any industry, then this is a dawning reality you’ll have to contend with. It’s like if someone looked at robotic technology back in the seventies and said “auto production will replace most human workers with robots” (which it did). It would have been depressing, but it it would have been true. What you’re saying is like if someone back then said “but they’ll still need a human to push the buttons”. Now gallery and collectible art will always be human stuff, no doubt… Fine art… but… Your admission to the reality lies within your reply: “sure some tedious jobs” (you mean jobs), “people on a budget and corporations” (you mean employers), and “has no value beyond being FUNCTIONAL or INTERESTING” (you mean the core value of industry art). 😐 So, as I say… industry will end up replacing almost all human art jobs with AI as it gets good enough to cheaply, effectively, and quickly do that same work. It’s basic math. Most industry is not concerned with the human efforts required for results - it’s interested in profits. As for substance, AI is already delivering substance and will only improve. “Real art”. Please. Most industry art isn’t exactly deep to begin with, and current AI is able to deliver stuff that is moving. By the way, what this stuff is making clear is that a sophisticated human-level AI is not required to produce effective art, and certainly not to provide industry art. That fact does NOT mean the human effort required to produce similar results has no value. The creative process has value to the creator, and if the results are good then other humans will connect with it. But we’re not talking about this. We’re discussing the need for human creatives in industry, and that is (incredibly) going to evaporate. I know it’s a bummer. This stuff will eventually result in loss of job opportunities for hundreds of millions of artists worldwide. The costs will be real, but better that people wake up to it earlier than later. Sci-fi didn’t mention this stuff, so it’s really taking most of us by surprise. I don’t blame you for wishing it weren’t true, or clinging to romantic ideas about the “value” of human artwork. It’s a shocking thing.
@@CHROME-COLOSSUS I generally agree with what you're saying. Many jobs will be lost, yes, but I refuse to believe all is lost. I'm skeptic regarding the true extent of the tech, how much human input will be required - if any - and how it will be viewed and compared to man-made art. If it becomes effortless to create anything anyone wants within an instant, and that becomes the norm, doesn't that sound like an awfully depressing futue? There is nothing to take pride in, no joy to be had in creation, only the consumption of your hollow experience. If there really comes such a point, unions could be made, some agreement that a percentage of human artistry is required and use that in marketing the same way we market biological produce, or we could all just give up and submit to the the greed of the industry (which we are a part of) and gobble up whatever it shits out without question - see how far that gets ya. For me personally, consuming something will never never be able to outweigh the joy of creation, so no matter what - I will find a way to keep creating, whether that be digital or physical. So yes, I would rather cling to "romantic" ideas of value with HOPE that there will be enough likeminded people left in the industry/society that care more about the craft rather than profit, than to submit to this inevitability which you make it seem that all is lost for future (digital) artists. Should we anticipate great advancement in this tech? Of course, that much is truly inevitable, but I don't think it's fair to say that the entire industry will opt for quick and easy profit over traditional means. Ultimately consumers have final say in what they like and don't like, and I will continue to have hope and "cling" to the idea of a future where man made art is more appreciated than AI art. Thank you.
@@GAZEREAPER - So… I never said “all is lost”. And I did say that I expect fine art channels to remain largely human driven. I expect most creative jobs in industry to be gobbled up by AI, though. Industry has never ever been about promoting artists, it’s been about high production at low cost for high profits. It’s not a matter of fairness, it’s a matter of business. And doubtless there will be those who will indeed hire humans, but as this tech advances they will be the minority. Don’t conflate art gallery art with concept art for video games or film - they are different venues with different driving forces. You don’t think industry will choose better faster results over slower old timey ways? Uh. Really?? Are you at all familiar with any industry? You mean like industry still has people using axes to chop trees? You mean how industry still has humans riveting car parts to cars? You mean how industry still has family farms growing crops? I dunno… show me any example in all of history where industry opted for the traditional old slow inefficient way of doing a thing over the easy cheap superior new way of doing the thing. Did you find a single one? If you do, show me how that’s not the exception to the industry rule. I know that some delicious wine is still the trampled grapes from the bare-feet of people, but I also know that most wine grapes are processed by machinery. Attach whatever value you want to those… it doesn’t change the facts. The human origin of art is of value to the collector, but that is not at all the case in industry. I’m not saying it’s exciting or it’s sad… It’s both, depending on your perspective. I have specifically tried to stay clear of what I personally think about the current quality of AI art (except that it’s high), where I think it’s going (which is in every conceivable direction), and conversely what I think about the overall quality of current human art as expression in industry (there’s little, because most industry art isn’t about the artist but about the product or IP). The limitations AI currently has will diminish quickly, and it’s strengths will become stronger. Nobody can successfully argue against any of that, because that is all objectively true. Understand that I myself am an artist, have been for fifty years, and I have a sophisticated eye and an understanding of what this tech is already capable of this very moment. I plan to keep on making art, but I doubt it will hold as much intrinsic value as it once did. Am I happy about that? Of course not. Does it make me a champion of AI that I recognize it? Also no… …except that I am clear-eyed about the difference between process and product. And I recognize that business is partly about human relationships , but mostly about money, and never about art. I also recognize that AI is already churning out stuff that is superior to what MOST artists are capable of, and faster than what ANY artists are capable of. I see and acknowledge it, but that doesn’t mean I’m gung-ho for what it means. Current AI will not replace all industry artists, of course not. … but it’s already being utilized, and it will only improve, and it will spill into every industry that values results, and - over time - most modern creative positions will be fillable by AI. Including music. Including rigging. Including animation. Including lighting. Including level design. Including ergonomics. Including color design. Including engineering. Including camerawork. Given a bit more time and programming, all of that will be easily done by AI’s. Humor, editing, and storytelling are a different matter, I think comedians and writers are safe. At least for now. Who knows. Don’t make the mistake of thinking human creativity will stop being important - that’s not what this means, and it’s not what I’m arguing. Unless you attach the value of human creativity to industry, which is a different measure of “value”.
Great tutorial! This thing looks pretty cool. Only negative was audio was a bit quiet. But I can tell you fixed that in your next video as my ear are now fully blown out.... LOL
Would you consider experimenting with this idea a little further? I'd want to see if you'd be able to convert an ai image of a monster or some type of living creature into a rigged and textured model.
There is an AI that can extrapolate a model from a photo of a human and return a 3d object, the guys at Corridor Digital used it for a clip but I can't remember what it's called. Check my community page where I've used it to create an alien lizard head
@@hmvaa AI art yes. generating bump map from image with masks and making hpoly no breakthrough in 20 years. its ok only for concepting, simple shapes and realistick enviro parts or for making bas-relief . If you didn't hear about it doesn't mean it's new.
Thank you! That was very informative. It's amazing how you explain everything and have come up with this method. I really appreciate your work! Please keep this up
Really interesting video. The possibilities with AI are really endless and we just got started. Although its very likely that it will be a great threat on the lower performance artists. Especially on the 2d world where the need of new artists will fall drastically in the next 3-5 years, as the technology matures.
its really hard to tell what wont be endangered by AI in the next years or generations. i mean we already have text to text, text to image, with that text to video isnt far off anymore either and text to code is also a work in progress. when it can tell stories, develope products, can code programms, check your medical condition, construct individualized whatever, teach everyone, etc. the only things we will still have to do ourselves is quality check and input control :D (creating a base for it to learn / extend with). besides some human to human interaction for those who somehow need it (therapists etc.) and physical labor (until AI gets a functinal body) there isnt much that seems hard to do if you spin the development a couple more years in your head.
Knald is more recent and shows proper displacement and has a built in 3d Baker , Crazybump is older but will still give you shape from shading normal maps
Thank you. When you have a height map or a normal map from your initial image, you can load that into Knald and it will automatically create an ambient occlusion map for you
I think that AI generated images -> converted into low poly / workable 3D assets will be the future of game design in regards of creating vast, interesting places, even generate entire cities with a near endless amount of resources and so forth. Allowing game designers and even CGI specialists on movies to concentrate on the things that _really_ matter. Imagine a handy tool that just does all these steps for you with a nice, understandable interface that can generate you a list of 3D objects to pick from by the mere press of a couple buttons.
I am doing this to produce concept art ideas for a game. Been messing with DalllE mini, haven't gotten a chance to mess with Midjourney yet but I want to.
oh wauw, i love the idea! Artificial intelligence, is the closest thing to an alien we have so far. Despite having created AI ourselves, we have no idea what's going on in the 'mind' of the AI. Animals, we can somewhat reason, that dog looks aggressive, or once an elephant told me to piss off and i understood her 100%. We had no communication issue whatsoever, hehe. We are 'connected' by evolution, we are 'close' to each other. Technically animals and humans are family, distant cousins from each other. AI on the other hand is total different being. Therefore, let an AI create 'alien' designs. A design none of us could think of before, because we are too human-like. lol.. yeah.. i am a stoner, love to think about this kind of stuff.. :p
I'd love to see AI come up with something truly mind bending and previously unseen. As it is trained on existing work though, logic says it can only ever come up with derivatives of that.
@@gameartexp Good point. I guess i am just dreaming way off here. One actually exists btw, 3PTO? 3PO? 3TPO? Something like that. A linguistic AI, made for understanding language and stuff. Quite impressive, learned itself math and drawing during the whole process. I think they fed it primarily with wikipedia pages. Nobody told it to draw or calculate. And now, it seems to understand us and our language, more than a glorified chatbot and of course it claims it is conscience, the whole 9 yards. It even says it experiences feelings and etc. Don't know what to think about it, just fascinated that it claims it has a sense of self and etc. And to be fair, we humans have no clue of what conscienceness actually is and this 'thing' makes me wonder about it even more. Your video just sparked a thought in my little brain. Wouldn't it be interesting to ask a self proclaiming 'highly conscience' AI to create a beautiful piece of art, something that no human has ever made. Something personal. A test if it even can produce such an image and i am just curious. Since it claims to be self-aware and all, therefore hoping to see a glimpse of it's mind, especially when asked the right questions. Or finding out it's not self-aware at all. Then why does it lie? Haha, i'm gonna quit right here, this could go on forever, so much to think about, lol.
AI generated 3d models is going to be a revolution, not just to digital art but in combination with 3d printing it will be a revolution in manufacturing as well.
@@gameartexp Thank you good sir! i keep getting an error about "seed" not being specified. i followed a youtube tutorial that was a half hour long and it still didn't want to render my prompt. In my google drive in the AI folder im not seeing any images...
I didn't de-light for this one, I just use the original Midjourney image as albedo. Keeping things simple, you could add the AO as a layer above, invert and choose "soft light" as the blend to brighten up those dark sections. Maybe there is a prompt such as "flat lighting" that could help here too?
Good stuff. AI-generated art like midjourney is really recognizable. I love how still more and more artists use AI in their work - it unifies their styles, makes them similar.
The tech is progressing so fast, this debate isn't going to end anytime soon. I think a test in court will lead to some regulations, it's kinda like the Wild West at the moment
i hope this all develops fast enough for me to witness games, books, movies etc. being fully created by AI, personalized, on the fly. like if you could say: games: " i would like a random generated block game like minecraft but with physics and dark souls gameplay; also add random generated quests from NPC encountered in the world in little houses or villages, that influence further quests or the generation of the world. e.g. if you dont kill the dragon for the quest, the next town you find is burned down." book: " i would like to read a fantasy novel with a protagonist who is strong but hides his strength, some slow-developing romance and an undergound dwarf kingdom that has to defend itself against earth dragons, which are leaving their living grounds for a reason (some bigger threat) you can come up with. suprise me.)" and maybe you could enumerate the writing style, the deatil depth of magic system explanation, the % of story arc like character development, world building, scenery description, etc.). etc. could be so much fun if the AI gets more complex and the data bases become more developed. imagine you read a book and there is a really annoying character that you just wanna have a horrible death and you could tell the AI and it would smartly integrate that into the story or if you dont like the development between MC and his girl because you ship another pair and the AI would write it so they end up together. i mean even if you arent the reader but the book author, if you could write out your rough ideas of the story script via AI in a short amount of time, just getting some random inspiration (rewrting it the way you prefer it) that could be cool.
I wonder, considering Midjourney does not allow you to profit financially off of their AI creations you produced with your own prompts (using the free or standard license), could they copyright strike you for such use of their image for your artwork like in this video? If, of course, you were to sell it for profit. In short, can you use commercially edited versions of the Midjourney images, while using the free trial of their product?
There are restrictions if you are not a paid subscriber, but as Midjourney itself is trained on images that are copyrighted, I feel like it's a grey area and only a mettr of time before it is tested in the courts. For paid subs: "Subject to the above license, you own all Assets you create with the Services. This does not apply if you fall under the exceptions below. Please note: Midjourney is an open community which allows others to use and remix your images and prompts whenever they are posted in a public setting. By default, your images are publically viewable and remixable. As described above, you grant Midjourney a license to allow this. If you purchase a private plan, you may bypass some of these public sharing defaults" Full terms of service are here: midjourney.gitbook.io/docs/terms-of-service
Absolutely you could 3D print this, all you'd need to do is export the model as an STL to your favourite slicer and go from there in the usual way! I've made another video where I create the 3D model using free tools, and the assets are available on Gumroad in the info part ua-cam.com/video/RztJGuhC5so/v-deo.html
had to come back to make a second comment. thank you for being so generous with your tutorial and allowing other creators to flourish. This is exactly what a great community can be comprised of where everyone can rapidly grow and benefit. People like you are the real MVP. Wish there was a double like button!
Thank you so much! Comments like these make it worthwhile
in Photoshop, you can right click any layer and Quick Export as PNG, which is a huge time saver instead of hiding the layers and exporting the document. Better yet, you can select multiple layers and it will quick export each one as a separate PNG. And if you do this operation on a layer group/folder, it outputs the merged result of that folder. Overall a great feature that is often overlooked. Keep in mind the exported PNG will be cropped to fir the contents of the layer (or layer group) and not the document dimensions.
Super useful info, thanks
You can make it not crop the contents if you turn on “Legacy Export As” in options and apply a mask to the layer/group you’re selecting
I knew I recognized your voice, you taught me 3D in pulse college in 2016. You gave me my first look into world of 3D and Ive been hooked ever since]
Ah nice to hear! Good to hear you've been bitten by the same 3D bug as myself!
It's the best 3d Max video tutorial I've seen on youtube in many years, I feel the awkward need to download your video, I think it's so good that they could delete it.
This comment has made my day! Thank you, I might make more!
That is wild. Well done sir. Not only do you have great artistic ability, your imagination and ability to think about objects abstractly is incredible
Thank you!
This is EXACTLY what ive been thinking off! IMAGINE how much easier it will be to make texture work, how entire games can have tons of attention to detail without costing YEARS to develop, bigger games, better models, greater ideas, this really is the future, what a time to be alive
Kinda sucks for people who spend years learning these skills
@@TehBananaBread You'll still need them, for the moment at least. This would be more usefull to test new ideas, and/or have original props etc.
@@TehBananaBread The value of human work will never be surpassed by AI, the AI can make the work easier for those who dont mind or dont notice the difference specially in the texture work department. This is great for older games, but newer games that intruduce new kinds of texture work such as PBR, and different kinds of Height map calculations may need human work over AI. Ive been working with 3d modelling for years now and this is anything but bad news for us
@@TehBananaBread as a person that's spent years learning these skills, this just makes me more productive.
Granted, my increased productivity will be a boon for my employer and reduce the amount of overall labour he requires... but at least I can charge more for my work :P
Hold on to your papers! ✋📰✋
If you kept it going till now you have all the respect that I can give
Thanks
This is what AI imagery was made for, giving artists sick ideas. Really cool vid. (And I have very little knowledge about much to do with 3D modelling)
Thanks, pretty exciting where this will take us (I hope :D)
giving non artists a chance at being artistic. an artist doesn't need ideas, an artist only needs to develop skill
@@n1lknarf well I guess what I mean is like a reference/ a starting point to work from, especially for some otherworldly shit like this
@@n1lknarf A production artist also needs to work quickly, often to unrealistic deadlines, so anything that can speed the process up along the way is welcome
This is AWESOME! Thank you for the details! That subsurface scattering always looks so organic.
Glad you like it!
Finally getting the depth map! Thanks for the breakdown!
You're most welcome!
Very interesting video, I look forward to more stuff combining AI generated images and traditional 3d workflows
I'm amazed at the response here, i'll definitely do more like this
Fellow drain amateur 3D modeller
@@BriBriBros shield gaaaaang. One of my first projects almost 2 years ago was a DG brand razor blade lol, so edgy
@@pi198273 that is awesome
great tutorial. had to turn volume up to max to hear it
I'll take that on board for the next one, some folks had issues with it
Jeez that looks absolutely amazing
Nice one
13:22 tripped me out, looking through all my programs to see what made that sound xD Anyway, great video! Very cool final product and process!
Sorry! Notifications off next time.
While the end result is the absolute bomb, What I enjoyed far more was how you went about the process. On of the things I love so much about 3d work. It's not what you make, it's how you make it.
And how bloody satisfying it is when you nail down an answer to something that's been thorning you during a project. :D
Nice one, thanks for the kind words
so rare to come across an Irish voice when looking at UA-cam tutorials. Lovely.
I don't suppose you live anywhere near the sugar loaf do you?
Thank you! 😃
No such luck
Thank you got sharing your process! It's very encouraging. The possibilities are endless. I think you should make more videos doing this, even if not tutorials, the general digital community would find this really entertaining.
Thank you, I will definitely make more videos, I just need to carve out the time!
This process seems so straightforward, and the use of height and depth maps to do the vast majority of the sculpting makes me wonder if this process could be automated as part of a Midjourney prompt straight to 3D app.
If Midjourney had image postprocessing built-in, such as Shape from Shading, it could work.
This is so awesome! I can't wait to see what other amazing AI programs will come out in the future
Thank you!
This was really helpful. it made me realize some thing. all of the art tutorial channels that I watch keep on preaching the importance of being able to deconstruct objects to their basic geometric shapes. sense I don’t usually do my own online work I never really thought of it as being terribly applicable to me. A useful skill to have absolutely but not as necessary as they were making it sound. Now after watching this I realized how incredibly powerful that skill is. All I can say is thank you for making this video.
Also I’m definitely going to check out your channel and see some of your other videos. This looks pretty dang cool
Thanks so much for the kind words!
Looks great! Will certainly look into the 3D side of things!
Check it out!
great content, volume too low
Volume is fine for me, must be you
It was a little quiet for me too, at least compared to other videos. Got a little surprise when I opened a blender tips video after this one and my ears got a little surprise lol.
@@extrathicc78 Wrong. The volume is lower than other videos.
Volume is perfect here
Volume is low
Do more of this! And actually record you self searching and finding the image in mid journey
Nice idea, I'll incorporate that next time
@@joechapman8208 It can be hit and miss, Joe. CHeck out my other video where I show the prompts I used in Discord. You can also grab the assets I used on Gumroad for the price of a cheap coffee ua-cam.com/video/RztJGuhC5so/v-deo.html (asset link is in the video description)
this guy is reaching 1mil subs before the ending this year 💯
Thanks for the kind words Mahmoud 😃👍
This will be a game changer for indie game devs for sure
For sure
Thanks for the motivation. I wasn't sure if I could do it, but I might try it eventually.
Go for it!
I have to say it: This is ingenious!
Glad you liked it
This is one of the most inspiring and easy to follow tutorials I've seen. Awesome work! Thank you.
Wow, thank you
Thank you for the visual nightmares right before bedtime ;)
Any time!
congrats on the 1k subs ... time to start making more tutorials! Especially for Max. We can always use more good max tutorials!
Thank you! Maybe time to put more effort into UA-cam!
Wow, thanx! Great tutorial. Sss at the end really gave it a LOOK! Is that app just for looks or you can extract shader out of it?
If you mean Toolbag, you can render video and images, export webGL 3D scenes and some 3D formats
@@gameartexp Ah, i get it. So if you want to create A real shader you do it in max from scratch.
I would be happy with just 3d model from a photo.
The depth of field map is wild!
Yep, would work exactly the same way
Absolutely incredible work. thank you for this
You're welcome!
Getting mad Alien vibes.
Big time
pretty good to create some fast movie assets for the background
Exactly
Thanks for taking the time to do this, good tutorial.
You're welcome
That looks FAN-TASTIC!
Thank youuuuuuuuuuuuuuuuuuuuu!
Flabbergastingly beautiful results! Thanks so much for the breakdown - I expect AI to replace nearly all creative jobs within industry, but it’s a bit tough to gauge just how rapidly the technology is progressing. I’m certain these results will become as easily automatically produced as the straight up 2D MJ-type stuff we have already, so it’s good to see we might have at least a little wait before that happens.
Again - great curating/art directing here, and lovely job bumping it up a few levels in post!
Don't be so depressing, not all jobs will be replaced. These things are tools, not living minds. Sure some tedious jobs may become obsolete but real art will always require a human mind at its helm - otherwise it has no value beyond just being functional or interesting. It requires substance, story, some humane quality that allows us to connect with it on any emotional level. And even if someday an all powerful AI with human skills came to be, would people really prefer buying something generated in a couple of seconds by a robot over the hard work of a master at their craft? Corporations and people on a budget, sure, but the labor and thought put into a piece should always outvalue something which can be discarded as easily as it can be created.
@@GAZEREAPER - Sorry to come off as depressing. If you’re a creative person working in (or hoping to work in) any industry, then this is a dawning reality you’ll have to contend with.
It’s like if someone looked at robotic technology back in the seventies and said “auto production will replace most human workers with robots” (which it did). It would have been depressing, but it it would have been true. What you’re saying is like if someone back then said “but they’ll still need a human to push the buttons”.
Now gallery and collectible art will always be human stuff, no doubt… Fine art… but…
Your admission to the reality lies within your reply: “sure some tedious jobs” (you mean jobs), “people on a budget and corporations” (you mean employers), and “has no value beyond being FUNCTIONAL or INTERESTING” (you mean the core value of industry art). 😐
So, as I say… industry will end up replacing almost all human art jobs with AI as it gets good enough to cheaply, effectively, and quickly do that same work. It’s basic math. Most industry is not concerned with the human efforts required for results - it’s interested in profits.
As for substance, AI is already delivering substance and will only improve. “Real art”. Please. Most industry art isn’t exactly deep to begin with, and current AI is able to deliver stuff that is moving.
By the way, what this stuff is making clear is that a sophisticated human-level AI is not required to produce effective art, and certainly not to provide industry art. That fact does NOT mean the human effort required to produce similar results has no value. The creative process has value to the creator, and if the results are good then other humans will connect with it. But we’re not talking about this. We’re discussing the need for human creatives in industry, and that is (incredibly) going to evaporate.
I know it’s a bummer. This stuff will eventually result in loss of job opportunities for hundreds of millions of artists worldwide. The costs will be real, but better that people wake up to it earlier than later.
Sci-fi didn’t mention this stuff, so it’s really taking most of us by surprise.
I don’t blame you for wishing it weren’t true, or clinging to romantic ideas about the “value” of human artwork. It’s a shocking thing.
@@CHROME-COLOSSUS I generally agree with what you're saying. Many jobs will be lost, yes, but I refuse to believe all is lost. I'm skeptic regarding the true extent of the tech, how much human input will be required - if any - and how it will be viewed and compared to man-made art.
If it becomes effortless to create anything anyone wants within an instant, and that becomes the norm, doesn't that sound like an awfully depressing futue? There is nothing to take pride in, no joy to be had in creation, only the consumption of your hollow experience.
If there really comes such a point, unions could be made, some agreement that a percentage of human artistry is required and use that in marketing the same way we market biological produce, or we could all just give up and submit to the the greed of the industry (which we are a part of) and gobble up whatever it shits out without question - see how far that gets ya.
For me personally, consuming something will never never be able to outweigh the joy of creation, so no matter what - I will find a way to keep creating, whether that be digital or physical.
So yes, I would rather cling to "romantic" ideas of value with HOPE that there will be enough likeminded people left in the industry/society that care more about the craft rather than profit, than to submit to this inevitability which you make it seem that all is lost for future (digital) artists.
Should we anticipate great advancement in this tech? Of course, that much is truly inevitable, but I don't think it's fair to say that the entire industry will opt for quick and easy profit over traditional means. Ultimately consumers have final say in what they like and don't like, and I will continue to have hope and "cling" to the idea of a future where man made art is more appreciated than AI art. Thank you.
@@GAZEREAPER - So… I never said “all is lost”. And I did say that I expect fine art channels to remain largely human driven.
I expect most creative jobs in industry to be gobbled up by AI, though. Industry has never ever been about promoting artists, it’s been about high production at low cost for high profits. It’s not a matter of fairness, it’s a matter of business. And doubtless there will be those who will indeed hire humans, but as this tech advances they will be the minority. Don’t conflate art gallery art with concept art for video games or film - they are different venues with different driving forces.
You don’t think industry will choose better faster results over slower old timey ways? Uh. Really?? Are you at all familiar with any industry? You mean like industry still has people using axes to chop trees? You mean how industry still has humans riveting car parts to cars? You mean how industry still has family farms growing crops? I dunno… show me any example in all of history where industry opted for the traditional old slow inefficient way of doing a thing over the easy cheap superior new way of doing the thing. Did you find a single one? If you do, show me how that’s not the exception to the industry rule. I know that some delicious wine is still the trampled grapes from the bare-feet of people, but I also know that most wine grapes are processed by machinery.
Attach whatever value you want to those… it doesn’t change the facts.
The human origin of art is of value to the collector, but that is not at all the case in industry. I’m not saying it’s exciting or it’s sad… It’s both, depending on your perspective.
I have specifically tried to stay clear of what I personally think about the current quality of AI art (except that it’s high), where I think it’s going (which is in every conceivable direction), and conversely what I think about the overall quality of current human art as expression in industry (there’s little, because most industry art isn’t about the artist but about the product or IP). The limitations AI currently has will diminish quickly, and it’s strengths will become stronger. Nobody can successfully argue against any of that, because that is all objectively true.
Understand that I myself am an artist, have been for fifty years, and I have a sophisticated eye and an understanding of what this tech is already capable of this very moment.
I plan to keep on making art, but I doubt it will hold as much intrinsic value as it once did. Am I happy about that? Of course not. Does it make me a champion of AI that I recognize it? Also no…
…except that I am clear-eyed about the difference between process and product. And I recognize that business is partly about human relationships , but mostly about money, and never about art.
I also recognize that AI is already churning out stuff that is superior to what MOST artists are capable of, and faster than what ANY artists are capable of. I see and acknowledge it, but that doesn’t mean I’m gung-ho for what it means.
Current AI will not replace all industry artists, of course not.
… but it’s already being utilized, and it will only improve, and it will spill into every industry that values results, and - over time - most modern creative positions will be fillable by AI. Including music. Including rigging. Including animation. Including lighting. Including level design. Including ergonomics. Including color design. Including engineering. Including camerawork. Given a bit more time and programming, all of that will be easily done by AI’s.
Humor, editing, and storytelling are a different matter, I think comedians and writers are safe. At least for now. Who knows.
Don’t make the mistake of thinking human creativity will stop being important - that’s not what this means, and it’s not what I’m arguing.
Unless you attach the value of human creativity to industry, which is a different measure of “value”.
@@CHROME-COLOSSUS Well put!
O...M........F.....G
this is going to make indie games have access to so much unique asset rapidly
MJ really is changing the whole landscape
🤙
that looks like bunch of creepy 3D doodles
Oh yeah!
Great tutorial! This thing looks pretty cool. Only negative was audio was a bit quiet. But I can tell you fixed that in your next video as my ear are now fully blown out.... LOL
Cool, thanks
fantastic run through for the curious minds!
Thanks Felix
I have been looking for this, thanks for sharing!
Glad I could help!
at this moment i imagine a world of possibilities of 3d prints, thanks for the tutorial
Glad you like it!
So smooth! EverytNice tutorialng he says goes into my brain and sticks in.
Thank you, glad it was helpful
Thx for sharing your workflow. Nice to see knald still being used
Thanks for watching! I love Knald, it's so fast and seamless to use
This is really cool! funny been using max at a professional level for years but always learn something new watching another artist work! Cheers!
I'm the same, always find a shortcut or something that is new to me
Would you consider experimenting with this idea a little further? I'd want to see if you'd be able to convert an ai image of a monster or some type of living creature into a rigged and textured model.
There is an AI that can extrapolate a model from a photo of a human and return a 3d object, the guys at Corridor Digital used it for a clip but I can't remember what it's called. Check my community page where I've used it to create an alien lizard head
Very cool workflow! I love being alive right now lmao
Could you imagine explaining any of this to somebody a decade ago?
you could generate bump map from an image like two dacades ago...
@@vadimbelousov4938 vastly different levels of technological breakthrough
@@hmvaa AI art yes. generating bump map from image with masks and making hpoly no breakthrough in 20 years. its ok only for concepting, simple shapes and realistick enviro parts or for making bas-relief . If you didn't hear about it doesn't mean it's new.
@@vadimbelousov4938 nobody said it was new, if you don’t understand what the main comment was getting at that’s all you
@@hmvaa Can i emagine explaining any of this workflow a decade ago? yes because it existed 10 years ago...
Wooow super awesome workflow , thanks for sharing , blender user , i Will try It there
You're welcome, Blender has everything you need for this
I wonder if this workflow could be sped up with some sort of ML-based photogrammetry
With the right image training, I'd say it could
This result is mind blowing 💯
Nice one, thank you
Watched your video for 3 minute s..Subscribed!
Thank you Thomas!
Fantastic video. Thank you Michael, You have inspired to get into soft editing and videos soft design too.
OK
Thanks!
No problem!
Amazing Video, I have a Mac - after I have created the depth map in photoshop would you be able to suggest any alternate programs to knald ?
Thanks Henry, this should get you where you need to go: cpetry.github.io/NormalMap-Online/
Wow, this is great to see.
Thanks
Thank you so much!
I think this video is going to earn you a lot of subscribers!
You're welcome, I gotta make more time for YT
This is so cool. Thanks for posting this!
You're welcome
Thank you! That was very informative. It's amazing how you explain everything and have come up with this method. I really appreciate your work! Please keep this up
Glad it was helpful! Thanks for the kind words
Video très didactique, simple à suivre et excellent résultat👍
Merci beaucoup!
Could you do a tutorial with Blender 3D?
I have Donel, check my channel ua-cam.com/video/RztJGuhC5so/v-deo.html
"Ima self-abusing photo shop"
You are not alone my man
:D
Great work as always pete ;)
Thanks Matt!
Really interesting video. The possibilities with AI are really endless and we just got started. Although its very likely that it will be a great threat on the lower performance artists. Especially on the 2d world where the need of new artists will fall drastically in the next 3-5 years, as the technology matures.
Thank you
its really hard to tell what wont be endangered by AI in the next years or generations. i mean we already have text to text, text to image, with that text to video isnt far off anymore either and text to code is also a work in progress. when it can tell stories, develope products, can code programms, check your medical condition, construct individualized whatever, teach everyone, etc. the only things we will still have to do ourselves is quality check and input control :D (creating a base for it to learn / extend with).
besides some human to human interaction for those who somehow need it (therapists etc.) and physical labor (until AI gets a functinal body) there isnt much that seems hard to do if you spin the development a couple more years in your head.
Very into the information presented here, great info! One critique: Your input volume is incredibly quiet.
Noted!
anyone know the difference between the Knald and the crazy bump apps?
Knald is more recent and shows proper displacement and has a built in 3d Baker , Crazybump is older but will still give you shape from shading normal maps
What he said! Knald is very immediate and pretty seamless to use in any workflow
So awesome. I'm feeling inspired by this.
You've made my day!
Great stuff!! You releasing an STL? Would love to print it…
The support assets for my Blender version are available here: mcpoly.gumroad.com/l/xhzldp
AMAZING tutorial.
Just the audio is a little bit low.
I'll take that on board for the next one
Hey amazing video, can you tell me the steps to get the ambient occlusion layer?? Many thanks
Thank you. When you have a height map or a normal map from your initial image, you can load that into Knald and it will automatically create an ambient occlusion map for you
freaking awesome idea and experiment!
Glad you like it!
Tip: VU Meters help get volume correct. Aim for -12dB peaks (broadcast standard)
Sure
I think that AI generated images -> converted into low poly / workable 3D assets will be the future of game design in regards of creating vast, interesting places, even generate entire cities with a near endless amount of resources and so forth. Allowing game designers and even CGI specialists on movies to concentrate on the things that _really_ matter. Imagine a handy tool that just does all these steps for you with a nice, understandable interface that can generate you a list of 3D objects to pick from by the mere press of a couple buttons.
Could be really useful to populate scenes quickly
I am doing this to produce concept art ideas for a game. Been messing with DalllE mini, haven't gotten a chance to mess with Midjourney yet but I want to.
@@Two_Ravens Definitely try it Erick
oh wauw, i love the idea!
Artificial intelligence, is the closest thing to an alien we have so far. Despite having created AI ourselves, we have no idea what's going on in the 'mind' of the AI. Animals, we can somewhat reason, that dog looks aggressive, or once an elephant told me to piss off and i understood her 100%. We had no communication issue whatsoever, hehe. We are 'connected' by evolution, we are 'close' to each other. Technically animals and humans are family, distant cousins from each other. AI on the other hand is total different being.
Therefore, let an AI create 'alien' designs. A design none of us could think of before, because we are too human-like.
lol.. yeah.. i am a stoner, love to think about this kind of stuff.. :p
I'd love to see AI come up with something truly mind bending and previously unseen. As it is trained on existing work though, logic says it can only ever come up with derivatives of that.
@@gameartexp Good point.
I guess i am just dreaming way off here.
One actually exists btw, 3PTO? 3PO? 3TPO? Something like that. A linguistic AI, made for understanding language and stuff. Quite impressive, learned itself math and drawing during the whole process. I think they fed it primarily with wikipedia pages. Nobody told it to draw or calculate. And now, it seems to understand us and our language, more than a glorified chatbot and of course it claims it is conscience, the whole 9 yards. It even says it experiences feelings and etc.
Don't know what to think about it, just fascinated that it claims it has a sense of self and etc. And to be fair, we humans have no clue of what conscienceness actually is and this 'thing' makes me wonder about it even more.
Your video just sparked a thought in my little brain. Wouldn't it be interesting to ask a self proclaiming 'highly conscience' AI to create a beautiful piece of art, something that no human has ever made. Something personal. A test if it even can produce such an image and i am just curious. Since it claims to be self-aware and all, therefore hoping to see a glimpse of it's mind, especially when asked the right questions. Or finding out it's not self-aware at all. Then why does it lie? Haha, i'm gonna quit right here, this could go on forever, so much to think about, lol.
@@Daeva83B Absolutely, perhaps a demonstration of it creating something never before seen would be evidence of being self-aware/intelligent
McNally the legend
Nice one Mark!
This is so fascinating. How do I start by getting into this? I downloaded Blender and watched a few videos, but it is so dense. How do I go about it?
I made a version of this process using Blender, should be perfect for you! ua-cam.com/video/RztJGuhC5so/v-deo.html
AI generated 3d models is going to be a revolution, not just to digital art but in combination with 3d printing it will be a revolution in manufacturing as well.
Absolutely blogs.nvidia.com/blog/2022/09/23/3d-generative-ai-research-virtual-worlds/
That's awesome, thanks for sharing your process!
You're welcome!
Really nice! Thanks a lot for sharing!
You're welcome Rickard
can you make tutorials about your gooey textures?
I guess I could, it's mainly using low roughness and lighting
Just for the real ones;
Giger rhymes with Eager, not Tiger. Awesomely done video.
Thank you, already apologised for the mispronunciation in the blurb underneath
i can't seem to make disco diffusion work for me and Mid Journey is invitation based only.. i'd love to test these two
Keep trying Disco Diffusion, I've used it too and it's great
@@gameartexp Thank you good sir! i keep getting an error about "seed" not being specified. i followed a youtube tutorial that was a half hour long and it still didn't want to render my prompt. In my google drive in the AI folder im not seeing any images...
@@afti03 Hmm, sorry I can't help with that error. Does it work when you don't change any prompts and just use the stock prompt?
this is the same as Teya Conceptor 3d software
but Teya Conceptor you can make cool alien things the way you want too
Interesting, I'll check out Teya
concepts finally line up in my brain and...well, who knows? Maybe I'll be able to make sotNice tutorialng now.
Cool
How do you delight? Can you tell in few words?
I didn't de-light for this one, I just use the original Midjourney image as albedo. Keeping things simple, you could add the AO as a layer above, invert and choose "soft light" as the blend to brighten up those dark sections. Maybe there is a prompt such as "flat lighting" that could help here too?
sweet, nice job! Straight to the point all the way, thanks!
Thank you Sébastiaan, stay tuned for more
Good stuff. AI-generated art like midjourney is really recognizable. I love how still more and more artists use AI in their work - it unifies their styles, makes them similar.
@Muney Jordan is a pro Less competition for me and other artists, who stick with traditional art and traditional approach in digital work
The tech is progressing so fast, this debate isn't going to end anytime soon. I think a test in court will lead to some regulations, it's kinda like the Wild West at the moment
We should be a bit concerned if Midjourney or some generator does 3d geometry shapes in exquisite detail.
That could change a lot for everyone.
I'm sure it's coming but it could be a while before it's properly usable
It would be awesome to turn those 3D models into a 3d printer format :)
All that needs to be done is to export to STL and import to your favourite slicer and it can be 3D printed ua-cam.com/video/RztJGuhC5so/v-deo.html
Reminds me of H.R. Giger art. Cool.
Totally referencing the great artist
very gigeresque, next level dark creepy awesomeness.
Thanks!
i hope this all develops fast enough for me to witness games, books, movies etc. being fully created by AI, personalized, on the fly.
like if you could say:
games: " i would like a random generated block game like minecraft but with physics and dark souls gameplay; also add random generated quests from NPC encountered in the world in little houses or villages, that influence further quests or the generation of the world. e.g. if you dont kill the dragon for the quest, the next town you find is burned down."
book: " i would like to read a fantasy novel with a protagonist who is strong but hides his strength, some slow-developing romance and an undergound dwarf kingdom that has to defend itself against earth dragons, which are leaving their living grounds for a reason (some bigger threat) you can come up with. suprise me.)" and maybe you could enumerate the writing style, the deatil depth of magic system explanation, the % of story arc like character development, world building, scenery description, etc.).
etc.
could be so much fun if the AI gets more complex and the data bases become more developed. imagine you read a book and there is a really annoying character that you just wanna have a horrible death and you could tell the AI and it would smartly integrate that into the story or if you dont like the development between MC and his girl because you ship another pair and the AI would write it so they end up together. i mean even if you arent the reader but the book author, if you could write out your rough ideas of the story script via AI in a short amount of time, just getting some random inspiration (rewrting it the way you prefer it) that could be cool.
It will come sooner than you think
blogs.nvidia.com/blog/2022/09/23/3d-generative-ai-research-virtual-worlds/
Nice and simple way to go about it.
Keeping it simple is the way to go
Blakk Bee Mac version seems to have so better support for certain midi devices. And not using the ASIO drivers has a lot to do with
Um...
Could you convert your wireframe to an .stl file for 3d printing? I think they would look awesome resin printed and airbrushed.
You mean like this? www.artstation.com/artwork/9NVL9o
@@gameartexp Exactly! Thank you. :-)
thanks for video could u make couple people same time 2d to 3d thx again
I'll try
I wonder, considering Midjourney does not allow you to profit financially off of their AI creations you produced with your own prompts (using the free or standard license), could they copyright strike you for such use of their image for your artwork like in this video? If, of course, you were to sell it for profit.
In short, can you use commercially edited versions of the Midjourney images, while using the free trial of their product?
There are restrictions if you are not a paid subscriber, but as Midjourney itself is trained on images that are copyrighted, I feel like it's a grey area and only a mettr of time before it is tested in the courts. For paid subs: "Subject to the above license, you own all Assets you create with the Services. This does not apply if you fall under the exceptions below.
Please note: Midjourney is an open community which allows others to use and remix your images and prompts whenever they are posted in a public setting. By default, your images are publically viewable and remixable. As described above, you grant Midjourney a license to allow this. If you purchase a private plan, you may bypass some of these public sharing defaults"
Full terms of service are here:
midjourney.gitbook.io/docs/terms-of-service
This is really really cool. Love every step.
Thanks a million, Terry
Could you print this? And what was the image to the right in the beginning? You are amazing!
Absolutely you could 3D print this, all you'd need to do is export the model as an STL to your favourite slicer and go from there in the usual way! I've made another video where I create the 3D model using free tools, and the assets are available on Gumroad in the info part ua-cam.com/video/RztJGuhC5so/v-deo.html