Yeah, it looks really good, and like the way you present it on your website
But as burtonyang said, 41 k tris seams like way to much even for a fps weapon.
No critics on the concept and textures
41000 tris? Woa macarony, that's way too much, still an amazing result.
There's a lot of detail on it, almost too much, gribble sometimes get in the way of design and make things look noisy, which I think that's what happened here.
I'm also not a fan of that weird brown, I'm not even sure what kind of material I'm looking at, is it wood? Is it plastic?
Color, texture and material wise could do with some extra work. I recommend you start studding materials, what kind of diferent metals are there, and how does it react in diferent light. Paint metal, brushed metal, etc. Study this things, and it will pay off in your next work.
It will also help you bringing more material variety on your works. Here's an example of material variety:
You see some sort of rubber material, at least 2 diferent kinds of metal surfaces, plastic and some ware here and there.
You've modeled a massive amount of small details that would easily be done with the normal map, a bunch of stuff that is intersected into other areas and will never be seen(and likely wastes a lot of texture memory). There are also just some really questionable areas, where you have a lot more detail than you need, but at the end of the day with 41K you still have areas that look jagged and lowpoly.
In addition to all of that, the design just feels cluttered, you could get rid of a lot of the detail here and have a more cohesive design. Stylistically its a bit of a mash-up, like random bits from a variety of sources crammed together.
I really think it would benefit from a more refined concept, and a more realistic target polycount rage, so you would be forced to design and model more efficiently.
It's a portfolio piece - poly count shouldn't dictate as much as it might in production. Sure it's important, but what's more important is the wow factor, which you've definitely nailed.
It's a portfolio piece - poly count shouldn't dictate as much as it might in production. Sure it's important, but what's more important is the wow factor, which you've definitely nailed.
No, this is BS. A portfolio must show art skills as well as technical skills. As soon as you see the wires and realize this asset is 41K, when it should be 10-15K any appreciation for how cool the thing is goes out the window to an employer.
Kinda going along with what everybody else said...
40k tris is a lot, but looking at your silhouette, I expected less than 15k. A large portion of your geometry is used in areas where no silhouette is needed, like many of the recesses and smaller details.
I think the design is nice, and you have a shit ton of detail, which I think is cool..maybe overkill, but I like it.
OK as for 41 k triangles
i don't work in the game industry yet. so I just do what i want thus far, learning on my own. its 41K polies for BOTH sides of teh gun/scope/clip fully mirrored from left to right. If i planned it to be fps only model in a particular game engine with solved animations, i could delete many faces from the unseen side. but i rendered it in Marmoset, so i just kind of planned to cover all bases in terms of shots and angles. It is also 3 separate objects with 3 separate materials (diff, spec, norm)
they are the:
gun: 15,455 tris unmirrored 34,858 tris mirrored as i have it now 2048 textures
scope: 3,374 tris 2048 textures
clip: 1,732 tris 2048 textures
so idk what your company does for all those issues, but i can down-res harder and keep detail if necessary, this is just personal work so i go all-out with what i like, (i can down res textures too, its just faster to baked everything at 2k)
gsokol - its supposed to be a fps gun, and for that reason I feel that the polies are necessary. The gun is like 30% of your screen all the time moving and twisting and lighing up from every angle. It needs to be interesting.. and thx for the crit. im pretty proud of the original design work and aggressive silhouette
[HP] yea i concur. With this project I was attempting new workflows on z brush material bakes and just material differentiation in general. The brown is plastic, not wood. but i struggled getting it to not look like stucco for a while . But as you can see i made a novel attempt at a dark metal, light metal, rubber turned out pretty good, and the hard plastics are ok using this workflow. that brown plastic is meh at this point. next project ill be hoping to perfect these mats and more - Furhtermore, using this technique... after solving the materials on the gun how i wanted them, I fully textured the scope in 15 minutes, and the clip in 5. (after the material bakes) im onto something here.
its mostly drawn from http://www.zbrushcentral.com/showthread.php?74301-Re-Angel
and http://www.pixologic.com/zclassroom/homeroom/tutorial.php?lesson=joseph
thx for the crit - ps yea those Brink weapons are shweet. - i was using them as ref for specs and metals... but if you look at them one after another they are rly rly rly similar and kind of simple material wise.. and shape wise.. very Line and Cylindery through and through. its like theyre all "good guy" guns but the bad guys spray paint all over thiers. w/e though its cool stuff tho; i have that game and the guns are the best part
which reminds me - I need to apply to 3 Point.
Earthquake - Yea it is a lot of detail. I like detail, and I figure for a next-next gen fps shooter, mo4r deets and mo4r polys...but there are def open spaces that balance. I even have material shifts and very deliberate shape changes and line breaks and implied lines to move the eye around. It is for sure a mash up. There are many elements of different real guns that are just straight up lifted and mashed in. Early critiques from my rl friends called it "Super Deluxe", "Technical" and "Futuristic", so I pushed towards these design trends. If you visit the my website, (linked in the original post) you can see the iterative design process that I used and kinda how it came to be. Anyways yea go check out the site if you to want to investigate - the only part that wont be seen is the butt, which is lower poly
and
Amadreaus - yea man. I agree with you. You get exactly what I was going for with this.. and OH dont worry there will be more projects. And in the future there will be more polies in games anyways. and def more wow factor. amiright?
i don't post much, but i hope this response cleared up some misconceptions
thx again for the critique!
A. Few studios do the delete-half-the-model thing anymore, this was common with much older games where 2000 tris was a big deal. For animation flexibility, this is rarely done.
B. If your model used 41K well, it wouldn't be an issue, as it stands it is modeled very inefficiently with loads of needless detail. This isn't a matter of "could this run in a game" its just a matter of workmanship. If you built an awesome 40K asset that really put all of those tris to good use, it would be very impressive. Again an asset like this could be done with 10-15K tris.
So its not simply a matter of someone thinking "41K, thats way too much!", but an honest and extremely relevant critique to the execution and efficiency of the model.
As you do not work in the industry, I would suggest you take this information to heart and not just brush it off as something that isn't important. Referring to feedback as "misconceptions" doesn't give off the best impression.
PS: Thanks for the comments on our Brink assets, those guns were 5000-9000-ish, with the chaingun being about 12,000 tris - just for reference. =P
"but I hope this response cleared up some misconceptions" I don't think think it was any misconceptions, it's still to highpoly, I just see a lot of excuses in you post.
You should invest more time into a concept phase, even if you don't draw it out.
The current design is very noisy and colors and different materials feel all randomly applied too.
Its like you took 5 different weapon photos and just overlayed them and modeled that.
You do seem to have the modelling skills and would benefit a lot by spending some time designing cool overall shapes with a clearly readable main body for the gun, rather than letting all the random protrusion design the gun for you.
Agree about the design. Chuckled a little bit when i looked at it. It does look like 4 different guns mashed together. Haveing 3 maps for this is a bit silly. A map for the clip? Really? It makes sence though considering how many polygons your working with. You do one thing wrong and the rest just follows along. If you dropped kicked the tri count i bet you could get it on one map and have better resolution with it too.
And Earthquake is right. You need to imporve your use of polygons. Its all over the place at the moment.
I like your spec map, though the metals should be brighter. Everything is a bit monotone at the moment brightness wise.
41K is fine, let's just say he's thinking ahead of his time : ) Though i do agree that the poly distribution is a bit all over the place.
Cool gun, as others have said it's a bit busy which distracts the eye from the actual design.
The sillouette seems to work though, and i personally have no issue with the colour. The problem with the material for me comes with the definition of it. It seems fairly mat and stained and bumpy, which makes it feel like it's made from clay instead of metal/paint.
Loving the texture on the nozzle though, nice chunky metal.
I like the design a lot. I think it would be cool if he kept the 41k tri version, and also made a 4-5k version too. Maybe market himself as an artist that can do art for cinematics as well as gameplay.
Most of the bad tri concentrations are just little knobs on the side of the gun, so aren't really a big deal to fix.
Replies
But as burtonyang said, 41 k tris seams like way to much even for a fps weapon.
No critics on the concept and textures
There's a lot of detail on it, almost too much, gribble sometimes get in the way of design and make things look noisy, which I think that's what happened here.
I'm also not a fan of that weird brown, I'm not even sure what kind of material I'm looking at, is it wood? Is it plastic?
Color, texture and material wise could do with some extra work. I recommend you start studding materials, what kind of diferent metals are there, and how does it react in diferent light. Paint metal, brushed metal, etc. Study this things, and it will pay off in your next work.
It will also help you bringing more material variety on your works. Here's an example of material variety:
You see some sort of rubber material, at least 2 diferent kinds of metal surfaces, plastic and some ware here and there.
You've modeled a massive amount of small details that would easily be done with the normal map, a bunch of stuff that is intersected into other areas and will never be seen(and likely wastes a lot of texture memory). There are also just some really questionable areas, where you have a lot more detail than you need, but at the end of the day with 41K you still have areas that look jagged and lowpoly.
In addition to all of that, the design just feels cluttered, you could get rid of a lot of the detail here and have a more cohesive design. Stylistically its a bit of a mash-up, like random bits from a variety of sources crammed together.
I really think it would benefit from a more refined concept, and a more realistic target polycount rage, so you would be forced to design and model more efficiently.
No, this is BS. A portfolio must show art skills as well as technical skills. As soon as you see the wires and realize this asset is 41K, when it should be 10-15K any appreciation for how cool the thing is goes out the window to an employer.
40k tris is a lot, but looking at your silhouette, I expected less than 15k. A large portion of your geometry is used in areas where no silhouette is needed, like many of the recesses and smaller details.
I think the design is nice, and you have a shit ton of detail, which I think is cool..maybe overkill, but I like it.
i don't work in the game industry yet. so I just do what i want thus far, learning on my own. its 41K polies for BOTH sides of teh gun/scope/clip fully mirrored from left to right. If i planned it to be fps only model in a particular game engine with solved animations, i could delete many faces from the unseen side. but i rendered it in Marmoset, so i just kind of planned to cover all bases in terms of shots and angles. It is also 3 separate objects with 3 separate materials (diff, spec, norm)
they are the:
gun: 15,455 tris unmirrored 34,858 tris mirrored as i have it now 2048 textures
scope: 3,374 tris 2048 textures
clip: 1,732 tris 2048 textures
so idk what your company does for all those issues, but i can down-res harder and keep detail if necessary, this is just personal work so i go all-out with what i like, (i can down res textures too, its just faster to baked everything at 2k)
gsokol - its supposed to be a fps gun, and for that reason I feel that the polies are necessary. The gun is like 30% of your screen all the time moving and twisting and lighing up from every angle. It needs to be interesting.. and thx for the crit. im pretty proud of the original design work
[HP] yea i concur. With this project I was attempting new workflows on z brush material bakes and just material differentiation in general. The brown is plastic, not wood. but i struggled getting it to not look like stucco for a while
its mostly drawn from
http://www.zbrushcentral.com/showthread.php?74301-Re-Angel
and
http://www.pixologic.com/zclassroom/homeroom/tutorial.php?lesson=joseph
thx for the crit - ps yea those Brink weapons are shweet. - i was using them as ref for specs and metals... but if you look at them one after another they are rly rly rly similar and kind of simple material wise.. and shape wise.. very Line and Cylindery through and through. its like theyre all "good guy" guns but the bad guys spray paint all over thiers. w/e though its cool stuff tho; i have that game and the guns are the best part
which reminds me - I need to apply to 3 Point.
Earthquake - Yea it is a lot of detail. I like detail, and I figure for a next-next gen fps shooter, mo4r deets and mo4r polys...but there are def open spaces that balance. I even have material shifts and very deliberate shape changes and line breaks and implied lines to move the eye around. It is for sure a mash up. There are many elements of different real guns that are just straight up lifted and mashed in. Early critiques from my rl friends called it "Super Deluxe", "Technical" and "Futuristic", so I pushed towards these design trends. If you visit the my website, (linked in the original post) you can see the iterative design process that I used and kinda how it came to be. Anyways yea go check out the site if you to want to investigate - the only part that wont be seen is the butt, which is lower poly
and
Amadreaus - yea man. I agree with you. You get exactly what I was going for with this.. and OH dont worry there will be more projects. And in the future there will be more polies in games anyways. and def more wow factor. amiright?
i don't post much, but i hope this response cleared up some misconceptions
thx again for the critique!
B. If your model used 41K well, it wouldn't be an issue, as it stands it is modeled very inefficiently with loads of needless detail. This isn't a matter of "could this run in a game" its just a matter of workmanship. If you built an awesome 40K asset that really put all of those tris to good use, it would be very impressive. Again an asset like this could be done with 10-15K tris.
So its not simply a matter of someone thinking "41K, thats way too much!", but an honest and extremely relevant critique to the execution and efficiency of the model.
As you do not work in the industry, I would suggest you take this information to heart and not just brush it off as something that isn't important. Referring to feedback as "misconceptions" doesn't give off the best impression.
PS: Thanks for the comments on our Brink assets, those guns were 5000-9000-ish, with the chaingun being about 12,000 tris - just for reference. =P
The current design is very noisy and colors and different materials feel all randomly applied too.
Its like you took 5 different weapon photos and just overlayed them and modeled that.
You do seem to have the modelling skills and would benefit a lot by spending some time designing cool overall shapes with a clearly readable main body for the gun, rather than letting all the random protrusion design the gun for you.
And Earthquake is right. You need to imporve your use of polygons. Its all over the place at the moment.
I like your spec map, though the metals should be brighter. Everything is a bit monotone at the moment brightness wise.
Cool gun, as others have said it's a bit busy which distracts the eye from the actual design.
The sillouette seems to work though, and i personally have no issue with the colour. The problem with the material for me comes with the definition of it. It seems fairly mat and stained and bumpy, which makes it feel like it's made from clay instead of metal/paint.
Loving the texture on the nozzle though, nice chunky metal.
Most of the bad tri concentrations are just little knobs on the side of the gun, so aren't really a big deal to fix.