3D Berserk armor

puella

Berserk forever
Boodongi, a Korean fan of Berserk, made the Berserk armor with 3DS Max. He said it took him 10 hours and he is working on a video these days.

berserker3d1.jpg


berserker3d2.jpg


berserker3d3.jpg
 

Vaxillus

The one and only severed head
Not quite Grorum quality (see his/her thread) but pretty nice still, especially for the amount of time it took. The textures are awesome though the eyes need color.
 

Majin_Tenshi

The can opener went bye-bye...
it looks like it was texture mapped with scans from the manga, could ya ask if I'm right? This is a great way to make frames to insert into the manga or to make an animated manga sort of thing, but for fitting it in a real environment Grorum much higher poly version is better.

This one does appear to be a low poly model, depending on texture mapping to give it its shape. I've never used this type of texture mapping myself. I know how to do a flowy cape in Lightwave 3d.... It'd screw up his texture map though.

I'm not sure about the size of the DS, it is very poorly lit or at a bad angle in every frame, so it could be perfect.
 
Yes, in low polygon, maps is above 75% of the final look.
Anyway, The model is not bad, but if I can crit' this work, I would say neither the cape or the head are following the Miura style.
If I were the artist, I would use more polygons to detail important parts, like the head, or adding more faces on the arms/legs armors, which could look more 'lizard scales' instead of 'posed plates'.
Sure the sword is too small, original is about 1.5 units longer. (Or I need glasses :badbone:)
About the texture itself, I'm pretty sure the artist did it alone, with a good software like painter or photoshop, graphic tablet and, of course, 2 hands, it's just a question of time to do it.

Indeed, I think it's a good model, we can tell it's the berserker when we look at it. If anybody knows the polycount, it would be interesting.
 

puella

Berserk forever
Boodongi said he used lowpoly because he wanted to make it like a game character. He unmapped and made a mapping source for himself afrer modeling in 3DS Max. He did scan rendering(basic rendering). He used around 2,100 polygons. And he regretted that he didn't make much effort rendering. If he did it, he thinks it wouldn't miss some details.
He wants to make a video of Guts in the Berserk's armor fighting an apostle. But the problem is that he should also make an apostle.
 

Voyevoda

Breathe and Start
First post here. Hello everyone!

And to get back on topic:
That's ten hours well spent I'd say. I really like the strong blacks in the texture map and it is nice material overall.

I too think it seems the textures were created to present the model as if it was taken directly from a berserk manga episode cell, which is a very inspiring thing to do.
I'm a bit dissapointed  how fast people abandoned the old way of texturing and started raving on and on about how everyone should just move on to specular, parallax/normal and diffusion mapping etc. "skinning" is (or rather, it's becoming) the new old school way of going on about things and it is very refreshing to see as opposed to just gazing at PS script generated patterns all day.

However, if this is to be for a video I think the creator should really think about dropping the heavy highlights in the colour map and use a specular map. This, because of the specular offset, will make the animation seem more dynamic and more alive, which would definetly help this piece in motion.
Now that I think about it, you could actually use that colour map as a specular map with no changes whatsoever made. This is all not neccessary, mind you. It is merely my personal oppinion and, after all, it is the animation that makes or breaks a video the most.

Now on to the model it self. I can't really comment all that much on this aspect since I have no wireframe renders to go on (so I can comment on polygon distribution) and the contruction looks very solid and accurate if I recall the berserker armour correctly. If I absolutely have to point something out, then it would be that the armoured gloves do look a bit tacked on to the arm guards. However, this all comes down to how the model is rigged and how the vertices are weighted, I guess.

Overall, this is (as I have already mentioned but it deserves to be reiterated) refreshing and solid work.

Would be very cool to see this fellow animated of course.

Oh, and which apostle(/s) will Guts be fighting in the video?  :guts:

- CaSan
 
 

CnC

Ad Oculos
Voyevoda said:

First of all, welcome to the boards!

Secondly, the "old way" of texturing, which I assuming you're referring to UV mapping has never been more alive. The advent of technologies such as HDR (high dynamic range) lighting, real time specular lighting, or other (albeit still limited) procedural methods of mapping have made games look cool, but the basics of texturing via UVs remain the same.

Keep in mind, this is a low poly character, i.e for games. Now every game engine has its own limitations or set of rules to follow. Some games don't allow for multiple maps per character (however I will admit most recent games do). If he's designing this character for a game that doesn't support specular maps, then he's going to have to paint the specular highlights right onto the color map or leave the specular highlights to be procedural.
However I will agree with you that if the game engine supports some form of specularity, than it would be best not to paint said highlights onto the color map.
Also, if it does support specular maps, the usage of the color map for the specular is not really a good idea. Sure, it can be a basis on which to build a better specular map but even in a case where the majority of the color map is black and white values (like this armor) the specular map would probably be better off being something besides the already existant color map. Of course, this is the artist's decision, I've seen it done both ways.

To critique his mesh at this point is pointless (as you've stated) as we don't have a wireframe view to look at. However I still think he sword is too small.

Thirdly, welcome to the board!
 

Voyevoda

Breathe and Start
CnC said:
First of all, welcome to the boards!

Thank you!


CnC said:
Secondly, the "old way" of texturing, which I assuming you're referring to UV mapping

I guess that the way I phrased it in my post was a bit sloppy and for that I apologize.
UV mapping is part of the procedure to acquire the coordinates for all different maps that you apply to meshes, but that is not what I was getting at (frankly, it's only getting better/easier now with the tools 3dsmax8 and Modo have introduced). What I wanted to say is that the old method of basically putting all content of the texture into the inital colour map (the procedure of using several map channels isn't all that new, but as you said, it hasn't been supported by game engines for that long).



CnC said:
Keep in mind, this is a low poly character, i.e for games.  Now every game engine has its own limitations or set of rules to follow.  Some games don't allow for multiple maps per character (however I will admit most recent games do).  If he's designing this character for a game that doesn't support specular maps, then he's going to have to paint the specular highlights right onto the color map or leave the specular highlights to be procedural. 
However I will agree with you that if the game engine supports some form of specularity, than it would be best not to paint said highlights onto the color map.
Also, if it does support specular maps, the usage of the color map for the specular is not really a good idea.  Sure, it can be a basis on which to build a better specular map but even in a case where the majority of the color map is black and white values (like this armor) the specular map would probably be better off being something besides the already existant color map.  Of course, this is the artist's decision, I've seen it done both ways.

I wasn't aware that the video the creator is making is to be rendered in a game. I guess I read through this thread a bit hastily. I'm very aware of a game engines limitations on what is and what is not possible (though UE3.0's proving to be quite interesting as far introducing current core technology to DirectX goes at least, we just got to see what the SIGGRAPH boys and girls have to say about silhouette masking). Once again I have to apologize for that oversight. I also agree with you that using the current texture map as a base for a possible future specular map is not a great idea. I just suggested it as a possibility (e.g. if the author wanted to just save some time and is willing to sacrifice some quality, if the time schedule is a problem that is).

I'd also say the sword is indeed a tad on the small side  :serpico:

Thirdly (!?) from me as well, thank you for the warm welcome to your great forums!
 
Well Boodongi, that rendering is alot better than I could every do! :puck:

Welcome to the board CaSan!!!!!!!!!!!!! :zodd:

- C
 
Top Bottom