Best Of
Re: Sketchbook: Rima
I got quite frustrated doing and redoing the leg. I never quite get the knees right, and when they're wrong, it messes the whole thing up. I probably need to do some isolated studies of the knees by themselves. So, in order to be productive while being stuck on that, I did the arms in my free time today. Right hand side is the construction, left is once I'd gone over the edges to refine it a little.
Comparing their current state to the scan reference (left), they're not a million miles off. There's a lot of refinement to be done; angles to fix, highlights that need to be matches, lots of little tweaks. But it's not completely dogshit. I think I've become better at understanding whether or not it's turning out well relative to the stage it's at. That's something I've always struggled with, and is basically the reason I've had nothing to show despite how long I've been sculpting. I could only judge it relative to the target, rather than for its stage of the process, so it always looked like shit. It was like comparing a sketch to a final render. I suppose my brain just hates itself. Anyway, I've improved a little bit at that, I think, which is why I can look at it right now and see the stuff I need to fix as part of the process instead of as faults.
Next free time I get, I'm going to tackle the torso again. It's given me a lot of trouble with some of its eccentricities, but I feel like if I can get that working right, everything else should come together better.
This is definitely the workflow of a control freak.
Rima
3 ·
Re: The Bi-Monthly Environment Art Challenge | May - June (102)
@Fabi_G, looking nice!
The only issue I see with this method are little polygonal diamond shapes where the metal arcs are damaged: either it's the contrasting color in vertex color, or sharp edges, or both, but they look a bit too sharp. Especially compared to the similar shapes baked into the normal map.
I realize these are supposed to create interest in silhouette, so I wonder how to properly utilize them with a trim map like this. Simply make them smooth?
As per your suggestion, I shifted things around a bit to make the eye more prominent
There were some issues with the engine I was going to use - UE 4. I had to give up eventually - it crashes every 10 mins on the new system, (seemingly) at random, and I really don't want to re-install anything at the moment, so... Godot it is! Not that I was particularly invested in UE - I only rendered a model that one time in it and don't know anything basically. Plus Godot is much-much faster to launch.
Testing export/import: the textures are only some baked normals, very basic colors and simple curvature generators.
The only issue I see with this method are little polygonal diamond shapes where the metal arcs are damaged: either it's the contrasting color in vertex color, or sharp edges, or both, but they look a bit too sharp. Especially compared to the similar shapes baked into the normal map.
I realize these are supposed to create interest in silhouette, so I wonder how to properly utilize them with a trim map like this. Simply make them smooth?
As per your suggestion, I shifted things around a bit to make the eye more prominent
There were some issues with the engine I was going to use - UE 4. I had to give up eventually - it crashes every 10 mins on the new system, (seemingly) at random, and I really don't want to re-install anything at the moment, so... Godot it is! Not that I was particularly invested in UE - I only rendered a model that one time in it and don't know anything basically. Plus Godot is much-much faster to launch.
Testing export/import: the textures are only some baked normals, very basic colors and simple curvature generators.
stray
3 ·
Re: The Bi-Monthly Environment Art Challenge | May - June (102)
Hi! Did a blockout of the book with some test animation. In a game context, the book could have those states: interaction start/ open, turn page left/ right, interaction end/ close. If so, would need some turn page animation. Possibly a few more bones to have more control over the shape in various states (currently has clipping)?
model
model
Fabi_G
4 ·
Re: Handpainting Normal Maps pixel by pixel (100% Photoshop - 0% HighPoly)
Hi everyone,
I love to handpaint normal maps using just photoshop without doing any highpoly at all. I only do this when the polycount is kept quite low and the texture size is as small as possible.
When I work like this 90% of the work is done with the pencil tool in one pixel size (not the brush tool). The rest 10% of the work is done using any tool that photoshop can provide. I also have the "NVidia normal map" filter, but I barely use it. I only use it when I handpaint an organic "height map" and I whant its normal map, or when the UVs contains alot of diagonals. But when I use this filter I never use the resulting image, because with the texture sizes I'm working on are very small, the resulting image that the filter produces usually is quite blurry and messy. So, whenever I use the filter, I only use the resulting image as a reference to then paint my own normal map.
The images I'm going to show you are my own projects, they are not done for any actual game or anything. I just do them because I enjoy it so much. The first images I'm going to show are basicaly simple objects, and then I will show more complex objects at the end. The shaders I use for this are very simple, they only use difuse map, normal map and specular map. The screenshots are taken directly from 3dsMax, not using any game engine.
I hope you like it!
Bathroom:
Tools 1:
Tools 2:
Food 1:
Food 2:
Livingroom:
Scorpion Evo3 :
H&K MP5:
More guns:
I like guns, so I made a few. Polycounts and texture sizes are consistent with the rest of the objects.
Vehicles:
There is something I whant to say for the last object. I did this object and the normal map back in 2011. Since then I have improved alot, so today I would have done things quite diferent. I'm not particularly pround of the underneath part of it, but at that moment I was already too bored with the object, so I kept it as it is. Also, I never did the difuse texture, so I've created a very simple one just to show you how it could look (No specular map here). Still I think is good looking enough to show you.
RUX
4 ·
Re: Game Development Thread: Artists & Developers Who Consistently Share Valuable Insights
Simon Trümpler has some nice breakdowns of various game VFX, but also a ton of other resources!
https://simonschreibt.de/
https://simonschreibt.de/
Eric Chadwick
3 ·
Re: Sketchbook: Rima
Ahhh, I'm still suffering gravely from the curse of perfectionism. I ended up redoing the basic construction several times. I also procrastinated by playing around with my brushes more. I finally managed to make a spatula-type brush that feels just right to me, so I decided to play with some carving. I tried sculpting this leg as practice with a more architectural approach, and was able to see some of the more subtle plane changes and exaggerate them to block it in a bit better. I find this brush, and general approach, works a bit better for me; I'm not very good at efficient ZBrush where you use subdivision levels and smoothing a lot. Also because I'm a massive control freak.
It's not quite right yet, I would need to change some of the planes, fix some contours, but it's not complete dogshit.
Rima
4 ·
Re: Sketchbook: Frank Polygon
Some thoughts about learning, problem solving, and feedback.
Answer this question honestly when it feels like you're falling behind on a project or stuck on a particular problem: how much are you learning and how much are you getting done?
Each project has it's own unique challenges. Sometimes it takes more time to learn something new or build up the skills needed to complete something else. Everyone screws up from time to time and sometimes it takes more than a couple tries to get something done. So avoid making overly simplistic value judgements. Focus instead on improving your process without being overly critical of yourself.
Make samples. Compare different workflows. Evaluate each process based on the results. Take the time to figure out what works and what doesn't. Look for alternative solutions and specific things that could be improved. Talk to other artists and learn from them. It's often much easier to avoid problems and make sweeping changes earlier in a project so ask for help or feedback sooner rather than later.
When asking for help or feedback:
- State the problem clearly, provide examples of the goal.
- Look for existing answers, think about or ask how they could be applied.
- Explain what you have tried, what you think isn't working, what your next approach would be.
Evaluate and implement relevant feedback. Consider that multiple things can be true at the same time. Not everyone has the same goals. Not everyone has the same access to tools. Not everyone enjoys working one way or the other. More often then not there's no one simple answer to complex problems.
Most importantly: Be willing to start over or re-work things that need improvement.
Yes, finishing work is important. Even when it's not perfect but an important part of learning is identifying and fixing the workflow issues that cause repeat problems. Incremental improvement is great but sometimes it's just easier and quicker to start fresh, without having to deal with a bunch of accumulated compounded errors. So be open to trying different approaches and be willing to start over.
Also, avoid becoming so overly invested in any one particular approach or outcome that viable alternatives get dismissed out of hand. It's one thing to say you don't want to do something because you don't like the process or the way it looks and something else entirely to ignore it or say it can't be done because of some random reason or no reason at all.
If you don't like a particular workflow or are genuinely struggling with a particular process then just be honest about it.
There's no real shame in asking for help working through things to improve but there's certainly bound to be frustration in not learning or not improving and arguing about why it is the way it is. Sometimes it's uncomfortable to look at where things are at and working through that is an important part of personal and professional growth. So try to frame sharing work in progress and feedback as more about learning than anything else.
Feedback can also be a good indicator of when it's time to call something done. If the feedback is very broad or vague then that tends to correlate with things being further away from where they need to be. Especially if there's a long list of technical improvements or aesthetic suggestions. If the feedback is very concise and specific, focused on a few smaller details, then you know you're closing in on the desired quality target because you've resolved a lot of the potential issues and captured the look of what's in the concepts or references.
Recap: Ask for feedback early, ask for feedback often, and compare
feedback from multiple sources but make the creative decisions yourself
and avoid using one critique as a foil against the others. Ask
questions. Outline what you think the problem is, what you think the
possible solutions are, what you've tried, what works and what doesn't.
FrankPolygon
5 ·