The BRAWL² Tournament Challenge has been announced!
It starts May 12, and ends Sept 12. Let's see what you got!
https://polycount.com/discussion/237047/the-brawl²-tournament
It starts May 12, and ends Sept 12. Let's see what you got!
https://polycount.com/discussion/237047/the-brawl²-tournament
Watch tower in the forest (UE5.4)
greentooth
This project is based on Jourdan Tuffan's Esbatuan Architecture. I made
this in order to become more comfortable working with a trim workflow
and trying to reduce the amount of hand-textured assets in a scene. I
heavily relied on Silke Van Der Smissen's dissertation on material
layers and highly recommend it for anyone who is curious about it.
Some material breakdowns and videos can be found at my artstation : https://www.artstation.com/artwork/a0Z0P9
Substance Designer, Maya, Photoshop and Zbrush for the props.
Replies
Don't stop now. You're at 70% - this could be mind-blowing if taken to 90%+.
Being smart and not spending a lot of time I think you can tweak a few things:
1. Composition
Add a foreground and background to create some real depth and separation.
2. Color
You don't have any complimentary color here. Splashes of pink/purple bushes and/or trees would create some juicy color contrast!
3. Lighting
Adding the sun in the back with some subtle lens effects could be a cheat that would go far. Or maybe even a rainbow like this Witcher shot. (little particles of flaying grass/leaves in the foreground are also a simple way to create dynamism and depth)
Don't stop keep going! You got this 💪
Sorry I couldn't resist doing a little paintover:
Mostly it's
1. Raising the landscape in the back to add a background
2. Adding complimentary color flowers for a foreground
3. Moving sun into the frame for more dramatic lighting.
Just add a some foreground leaf/insect particles to create a bit of motion and you'll have a cinematic level-up for just a little work.
@MBauer17 thank you for your question. If you're referring to my signature — I'm not an AI promoter. I'm a founder of an AI company.
I'm also an art director who's been making 3D art for the last 20 years, building titles like Uncharted and The Last of Us, all while constantly working to better the lives of artists. I've taught art on every continent, including organizing trips to Africa to help build a 3D animation industry for people entirely devoid of the opportunities we have here.
I've been co-running the Art Direction Summit at GDC for 11 years, creating cutting-edge education for all of us in the industry. I've released free educational DVDs, mentored with Sony Talent League and organized IGDA roundtables and meetups for this very Polycount community.
And I'm also the only AI CEO to advise on the side of artists in their court case against Stable Diffusion.
My credentials are readily available, and if you have any issues with my feedback, I’m more than happy to explain it.
That being said, the sentiment you shared is important, and I deeply understand your concerns. And amidst all of this I’d invite you to consider: what does it say about us as artists if we’d rather remain worse at our art than entertain the tools that might help us improve?
I want to assure you - as the person building this technology - that at no point is your work being fed into any database or used by anyone else.
As an artist first, and someone who’s personally experienced the pain of having their work used without permission, I’m just as frustrated by that as you are.
And as someone who’s dedicated their life to empowering artists, it probably frustrates me even more that this kind of misuse has become the default expectation around technology.
I’ve never participated in that — and never will.
For transparency, the sources and citations for any feedback I give are clearly included in every post.
@Eric Chadwick
thank you so much for the shout out bud! Hope you've been well 🤍
None of that would change the quality of the feedback, of course. It might change the appreciation of it and that might seem a bit unfair, but the latter will certainly change a lot if people feel they have been misled in some way.
I for one see no reason to assume any of those images were used for training. (Provided you control every aspect of the process. I would not trust e.g. a large corporation like Adobe even if I checked all the boxes that supposedly prevent that and even though I generally believe they are more ethical than others.) But the whole AI thing is a mess and people are on edge. And if they've already felt deceived once, you basically have become their Adobe in this context.
Obviously, we give up some control over everything we upload to the internet, but people do have a right to stay "worse" by rejecting AI tools. That doesn't mean they have the right to "shut down opportunities for others", of course, but I also don't think that anything like that was the intention, here. From their perspective, you took their image, uploaded it to an AI tool, something they themselves might never voluntarily agree to, and then basically all bets are off.
That being said, the accusation as phrased in the other thread is obviously not ok and probably quite upsetting (so I read some sentences of your reply with that in mind), I just wanted to lay out why/how it happened as I understand it.
*In my opinion, there's a very slippery slope to content (model) creation, here, but that's probably something for the general AI thread.
That said, the insinuation that anyone’s work was taken advantage of is entirely not okay. And it’s important to draw that line very clearly.
Beyond that, I don't believe anyone here wants to be in the business of policing how people arrive at feedback or advice. This is a place for learning. Good feedback is very rare. And is quite literally the reason we are all here. And I stand by the fact that the feedback provided was very thoughtful and highly constructive.
I’m here to help artists grow, connect with art fundamentals, be gainfully employed and to celebrate their growth.
And I hope you and @MBauer17 can be too.
However, I feel there still seems to be some kind of disconnect regarding the notion of policing or denying people feedback in some way. Being more transparent wouldn't prevent you from giving feedback, after all. Asking for consent regarding AI might, but then that's their decision/loss so to speak.
I take your point, and I understand where you’re coming from - but I’m not sure I agree.
The closest analogy I can think of is asking for someone’s permission to use photobashing in a paintover.
This is a platform built for receiving feedback. And since we’ve already established that no one’s art is being taken advantage of, it doesn’t seem fair - or aligned with the purpose of this place - to expect those volunteering their time and expertise to first ask permission for how someone likes to receive it.
If valid feedback is being offered, I don’t believe adding more friction to its delivery benefits us as a community.
The difference between photobashing and generative fill is that we have a fairly good idea what happens in the first process (plus, it tends to be more obvious), while in the second case, the image might be uploaded to a third party, at least in parts. So far you haven't really answered what happened to those pictures in the process of creating those overpaints. That's one aspect of it.
The other is that I simply want to know, and I can't imagine I'm the only one. I don't know if this simply was overshadowed by the other point, but currently it seems to me like you don't even want to proactively disclose when you used AI or that you are promoting your own reference tool. I know you see your tool as giving back to the community and naturally you believe in it and want to see people using it for the benefit of us all, but it's still advertisement.
I don't mind the fact that you promote it, I just ask you to be open about it. For me, that's a no-brainer, while I feel a strange kind of resistance from your side even regarding this point.
Anyway, Eric just posted in the general AI thread, so I 'll continue there.