That looks like a fun gnarly character with a lot of detail, but especially the beauty shots are really hard to read. The metal arm basically disappears into the body in the first shot, even without squinting. Perhaps you wanted to show the skulls, but having a stronger light from the left/back/top and less from the front might help to set the arm apart a bit more and unify some details in that shot. The shading of the blackened metal swallows a lot of light and form, too, but a different light setup might remedy that. Or perhaps play with the pose a bit.
The sculpt of the body and the skin details looks really great. Minor nitpick: The index finger is extended the most, but the other sinews on the back of the hand are more prominent. Perhaps that's the perspective, though.
The eye sockets of the skulls are pretty shallow, which makes them look a tad cartoony/fake, but that might be a stylistic decision. The skulls are fairly high resolution compared to other parts like the muscles of the thighs or the knees. I assume you kept the separate teeth so you could make variations quickly? Since you opted for relatively fresh, still wet blood, it might help to keep the bone itself less shiny as a contrast. However, I would have stuck closer to the concept in that area anyway. While having the belly as an active furnace to burn the skulls like in the concept doesn't make too much sense at first sight, it's still a great visual and the glowy bits help to set that area apart. It would also get rid of the strong internal value contrast in that area and having the skulls at least partly burned and turned to grey/wite ash would add to all of the aforementioned. The pointy coals(?) you added make the area more busy from most angles. They work the best in your first shot (which is overall the strongest), as they form some kind of spikey frame.
This is likely a matter of taste and probably not really beneficial, but I would have relaxed the topology a bit in places to have a more even flow. Where it matters is for example around the brachioradialis area that could be smoother if the polygon density was evened out a bit. You could still keep your overall edge flow. Also, the transition to the traps(?) in the front doesn't follow the shapes very well and looks a bit messy for that reason, even considering that this area probably doesn't deform a lot.
I don't think that's actually a pigeon head in the concept, e.g. the "beak" is just another part of the neck collar in my interpetration, but perhaps you simply liked that idea. In some of your shots, the reflections of two rectangle lights create the illusion of oversized, cross-eyed "Donald Duck" eyes, though, if you know what I mean. Took me a while to realize that was an illusion.
Overall, very strong organic sculpting work, but if you want to invest some more time into the shading or at least lighting (and a dedicated light setup for each shot), that would be very beneficial, in my opinion.
That looks like a fun gnarly character with a lot of detail, but especially the beauty shots are really hard to read. The metal arm basically disappears into the body in the first shot, even without squinting. Perhaps you wanted to show the skulls, but having a stronger light from the left/back/top and less from the front might help to set the arm apart a bit more and unify some details in that shot. The shading of the blackened metal swallows a lot of light and form, too, but a different light setup might remedy that. Or perhaps play with the pose a bit.
The sculpt of the body and the skin details looks really great. Minor nitpick: The index finger is extended the most, but the other sinews on the back of the hand are more prominent. Perhaps that's the perspective, though.
The eye sockets of the skulls are pretty shallow, which makes them look a tad cartoony/fake, but that might be a stylistic decision. The skulls are fairly high resolution compared to other parts like the muscles of the thighs or the knees. I assume you kept the separate teeth so you could make variations quickly? Since you opted for relatively fresh, still wet blood, it might help to keep the bone itself less shiny as a contrast. However, I would have stuck closer to the concept in that area anyway. While having the belly as an active furnace to burn the skulls like in the concept doesn't make too much sense at first sight, it's still a great visual and the glowy bits help to set that area apart. It would also get rid of the strong internal value contrast in that area and having the skulls at least partly burned and turned to grey/wite ash would add to all of the aforementioned. The pointy coals(?) you added make the area more busy from most angles. They work the best in your first shot (which is overall the strongest), as they form some kind of spikey frame.
This is likely a matter of taste and probably not really beneficial, but I would have relaxed the topology a bit in places to have a more even flow. Where it matters is for example around the brachialis area that could be smoother if the polygon density was evened out a bit. You could still keep your overall edge flow. Also, the transition to the traps(?) in the front doesn't follow the shapes very well and looks a bit messy for that reason, even considering that this area probably doesn't deform a lot.
I don't think that's actually a pigeon head in the concept, e.g. the "beak" is just another part of the neck collar in my interpetration, but perhaps you simply liked that idea. In some of your shots, the reflections of two rectangle lights create the illusion of oversized, cross-eyed "Donald Duck" eyes, though, if you know what I mean. Took me a while to realize that was an illusion.
Overall, very strong organic sculpting work, but if you want to invest some more time into the shading or at least lighting (and a dedicated light setup for each shot), that would be very beneficial, in my opinion.
This is great critique @Noren, thanks a lot for taking your time to type it out.
You are right, as the beauty shots were the parts I struggled with the most, with many iterations being made. It really was hard to pose and light the model while maintaining a consistent and natural look across all materials. I did use the same light rig for all angles in the first 3 shots, and I wasn't sure if it would be a point of critique, so thanks a lot for confirming it. I will try to get it fixed either tonight or tomorrow.
The original concept indeed has no pigeon head or any other bird design whatsoever. At first, it was quite hard for me to figure out how the head was exactly drawn, and at some point, I imagined an eye and a beak, so I thought it would be a fun idea to give the character an ornamental helmet, and I liked the contrast, a goofy, small head with googly eyes, and viscera. I hope it came out as interesting!
Regarding the skulls, I assigned a higher texel density as I wanted the area to be a highlight. I know it's not an industry standard to push it to that level of difference, but I was eager to detail them. I did keep the teeth separate to make quick variations, but as you can see, they all have the same albedo value, so it doesn't make much sense. This is something I did not notice before finishing most renders.
I wondered if I should remove the heavy close-ups of the skin and leave only the skulls. Maybe it wasn't such a good idea to include those, as there's an obvious difference in detail between both materials, as you pointed out. Regarding the skull's eye sockets, I was planning to push them with AO, but it seems it wasn't enough. This went over my head completely, so thanks for the feedback.
I agree with your point regarding the edge flow. I will take great care of it for my next project.
Finally, regarding what you mentioned about the traps, could you give me more insight? I couldn't quite figure out what you meant exactly, and I've found all your feedback extremely valuable, so I'm very interested.
I did not mention all the things you pointed out, but all of it was very observant, and I'm keeping them all in mind.
This was indeed an extremely enjoyable creature to develop, and it's part of an effort to assemble a character portfolio capable of meeting industry expectations. I’m currently looking for opportunities in the field, and while I haven’t worked in a studio environment yet, this project and the two additional characters I’m planning over the next months represent my attempt to bridge that gap. Your critique helps tremendously, so if there’s anything else you think could improve it (or future pieces) to make it more studio-relevant, I’d be grateful to hear it.
I meant brachioradialis, not brachialis up there, my bad. I updated it.
Regarding the skulls, I was actually talking about the mesh resolution. Should have made that clearer. It's not uncommon to give some areas more resolution (both mesh and texture) if it's clear they'll be seen from a closer distance or have more details. In this case, although it's good that it happened intentionally, it's unfortunately not immediately apparent why it happened, and it might be seen as a mistake. Especially regarding the mesh, there are some areas that are simply too low (another example would be the neck guard), while in the skull area, there seems to be some pretty wasteful automated quad retopo going on in a couple of places.
This might be somewhat controversial, but as soon as everything was in its place, you probably could have used an automated decimation to reduce the whole stomach area (as in the contents of the cauldron), ideally something where you can set a maximum edge length so it's still somewhat deformable overall. You'd have to rebake at least the normals, though.
The skin and muscle are among the best parts of the model. You are playing it pretty loose in terms of anatomy, but it has a great organic feel to it. To be honest, I didn't even notice the lower texture resolution. So no, do not remove the skin closeups, at least not hand, knee and back of the arm. The next two (upper body front and back) don't add much and the sculpt of the back looks much better in this case.
Heavy AO for the eye sockets would help if you removed the specular as well, else it looks like they were painted black or at the very least filled with dirt. But black eye sockets (when parts of the inside should be lit) look a bit cartoony either way. Looking at it again, they don't seem to be that shallow after all, so if you had actually modelled the details of the inside, it would probably look better if you removed the heavy shadowing instead. But again, that might also be a style choice
Edit: And it's probably best to say this directly, I'm afraid the skull area is among the weaker parts of the model at the moment, at least if you ask me, so I'd feature it a bit less. Going from the top, the first skull closeup is almost completely illegible. I suspect you want to show your material work, but it took me a while to identify the bottom side of the skull. The second one (to the right of it) is good. The three a bit further down are more or less redundant. I'd probably show the left or right one. They do look mighty gnarly, though, I give you that. And now that I've looked at the close-ups, I did notice that they are extremely sharp after all, maybe even too much in some areas where it gets noisy. The cloth looks good. Some of the holes and tears look like they might have been added after the cloth was simulated, as the folds don't seem to react to them, but I might be wrong.
Either way, overall this isn't a bad effort at all, so keep up the good work and I'll be looking forward to your next artworks!
Replies
Perhaps you wanted to show the skulls, but having a stronger light from the left/back/top and less from the front might help to set the arm apart a bit more and unify some details in that shot. The shading of the blackened metal swallows a lot of light and form, too, but a different light setup might remedy that. Or perhaps play with the pose a bit.
The sculpt of the body and the skin details looks really great. Minor nitpick: The index finger is extended the most, but the other sinews on the back of the hand are more prominent. Perhaps that's the perspective, though.
The eye sockets of the skulls are pretty shallow, which makes them look a tad cartoony/fake, but that might be a stylistic decision. The skulls are fairly high resolution compared to other parts like the muscles of the thighs or the knees.
I assume you kept the separate teeth so you could make variations quickly?
Since you opted for relatively fresh, still wet blood, it might help to keep the bone itself less shiny as a contrast. However, I would have stuck closer to the concept in that area anyway. While having the belly as an active furnace to burn the skulls like in the concept doesn't make too much sense at first sight, it's still a great visual and the glowy bits help to set that area apart. It would also get rid of the strong internal value contrast in that area and having the skulls at least partly burned and turned to grey/wite ash would add to all of the aforementioned.
The pointy coals(?) you added make the area more busy from most angles. They work the best in your first shot (which is overall the strongest), as they form some kind of spikey frame.
This is likely a matter of taste and probably not really beneficial, but I would have relaxed the topology a bit in places to have a more even flow. Where it matters is for example around the brachioradialis area that could be smoother if the polygon density was evened out a bit. You could still keep your overall edge flow.
Also, the transition to the traps(?) in the front doesn't follow the shapes very well and looks a bit messy for that reason, even considering that this area probably doesn't deform a lot.
I don't think that's actually a pigeon head in the concept, e.g. the "beak" is just another part of the neck collar in my interpetration, but perhaps you simply liked that idea. In some of your shots, the reflections of two rectangle lights create the illusion of oversized, cross-eyed "Donald Duck" eyes, though, if you know what I mean. Took me a while to realize that was an illusion.
Overall, very strong organic sculpting work, but if you want to invest some more time into the shading or at least lighting (and a dedicated light setup for each shot), that would be very beneficial, in my opinion.
You are right, as the beauty shots were the parts I struggled with the most, with many iterations being made. It really was hard to pose and light the model while maintaining a consistent and natural look across all materials. I did use the same light rig for all angles in the first 3 shots, and I wasn't sure if it would be a point of critique, so thanks a lot for confirming it. I will try to get it fixed either tonight or tomorrow.
The original concept indeed has no pigeon head or any other bird design whatsoever. At first, it was quite hard for me to figure out how the head was exactly drawn, and at some point, I imagined an eye and a beak, so I thought it would be a fun idea to give the character an ornamental helmet, and I liked the contrast, a goofy, small head with googly eyes, and viscera. I hope it came out as interesting!
Regarding the skulls, I assigned a higher texel density as I wanted the area to be a highlight. I know it's not an industry standard to push it to that level of difference, but I was eager to detail them. I did keep the teeth separate to make quick variations, but as you can see, they all have the same albedo value, so it doesn't make much sense. This is something I did not notice before finishing most renders.
I wondered if I should remove the heavy close-ups of the skin and leave only the skulls. Maybe it wasn't such a good idea to include those, as there's an obvious difference in detail between both materials, as you pointed out. Regarding the skull's eye sockets, I was planning to push them with AO, but it seems it wasn't enough. This went over my head completely, so thanks for the feedback.
I agree with your point regarding the edge flow. I will take great care of it for my next project.
Finally, regarding what you mentioned about the traps, could you give me more insight? I couldn't quite figure out what you meant exactly, and I've found all your feedback extremely valuable, so I'm very interested.
I did not mention all the things you pointed out, but all of it was very observant, and I'm keeping them all in mind.
This was indeed an extremely enjoyable creature to develop, and it's part of an effort to assemble a character portfolio capable of meeting industry expectations. I’m currently looking for opportunities in the field, and while I haven’t worked in a studio environment yet, this project and the two additional characters I’m planning over the next months represent my attempt to bridge that gap. Your critique helps tremendously, so if there’s anything else you think could improve it (or future pieces) to make it more studio-relevant, I’d be grateful to hear it.
Regarding the skulls, I was actually talking about the mesh resolution. Should have made that clearer.
It's not uncommon to give some areas more resolution (both mesh and texture) if it's clear they'll be seen from a closer distance or have more details.
In this case, although it's good that it happened intentionally, it's unfortunately not immediately apparent why it happened, and it might be seen as a mistake. Especially regarding the mesh, there are some areas that are simply too low (another example would be the neck guard), while in the skull area, there seems to be some pretty wasteful automated quad retopo going on in a couple of places.
This might be somewhat controversial, but as soon as everything was in its place, you probably could have used an automated decimation to reduce the whole stomach area (as in the contents of the cauldron), ideally something where you can set a maximum edge length so it's still somewhat deformable overall. You'd have to rebake at least the normals, though.
The skin and muscle are among the best parts of the model. You are playing it pretty loose in terms of anatomy, but it has a great organic feel to it. To be honest, I didn't even notice the lower texture resolution. So no, do not remove the skin closeups, at least not hand, knee and back of the arm. The next two (upper body front and back) don't add much and the sculpt of the back looks much better in this case.
Heavy AO for the eye sockets would help if you removed the specular as well, else it looks like they were painted black or at the very least filled with dirt. But black eye sockets (when parts of the inside should be lit) look a bit cartoony either way. Looking at it again, they don't seem to be that shallow after all, so if you had actually modelled the details of the inside, it would probably look better if you removed the heavy shadowing instead. But again, that might also be a style choice
Edit: And it's probably best to say this directly, I'm afraid the skull area is among the weaker parts of the model at the moment, at least if you ask me, so I'd feature it a bit less. Going from the top, the first skull closeup is almost completely illegible. I suspect you want to show your material work, but it took me a while to identify the bottom side of the skull. The second one (to the right of it) is good. The three a bit further down are more or less redundant. I'd probably show the left or right one.
They do look mighty gnarly, though, I give you that. And now that I've looked at the close-ups, I did notice that they are extremely sharp after all, maybe even too much in some areas where it gets noisy.
The cloth looks good. Some of the holes and tears look like they might have been added after the cloth was simulated, as the folds don't seem to react to them, but I might be wrong.
Either way, overall this isn't a bad effort at all, so keep up the good work and I'll be looking forward to your next artworks!