Figured I would take a stab at splash damages new art test. Here is a link
http://www.splashdamage.com/images/content/arttest_concept03.jpg
Last couple of nights Ive been chipping away at this bad boy, Currently This is the HP model and I'm looking for any kind of feedback I can get before I move onto the LP model. Thanks -BW
UPDATE: (compare lower on page)
Replies
Unless I'm completely wrong and they just want a nice high to low polygon bake based on accuracy. But that seems a bit flat and not really testing your creativity, understanding of scaling and been able to reference and create an object with purpose from nothing.
I can only assume Another test they are also trying to achieve is how well you use your texture space, as generally an object of this scale would not be uniquely unwrapped.
Its a really testing challenge if you look past its very primitive detailed surface and more about its scale and how you would fit it into a console budget.
It looks very Brink-ish, so I'm guessing they'd want some saturated colours on it too :P
As for the mesh itself, everything except the triangles (already discussed) and the middle section look spot on. In the concept the middle section (where the two "barrels" meet) seems to have a steeper incline towards the middle and the edge where the incline starts is a bit softer in the concept. Looking forward to your bake man, liking this a lot so far
IchII3d - I guess I see both sides of the coin, the only reason I guess I'm taking the side of 1 for 1 is that the concept art piece seems to have a large amount of smaller details as appose to the old splash damage art test that was very loose and not detailed. Now I going to be adding in some higher frequency detail nrms in tandem with the texture which should help give a better sense of scale.
sketch81- I think its just the darkness of the render but I also see what your talking about, I think im going to lift up that pannel alittle also, seems to be sitting a little low
Skamberin - I see what you mean about the incline, Ill have to soften that up a little thanks for the feedback.
Overall I think its nearly spot on! I took a look at your work and its all looking fantastic! I can't wait to see how this looks as a game asset!
Here is just a quick update, I made a comparison pic to show some of the things I changed
1. New floater triangle to get better shape and better edge control
2.Fix the spacing on the bottom container (sits a little higher now)
3. Added in more edge detail in the front panel to make it more accurate and less sloppy
4. Adjusted the top silhouette of the pillar to be a little more straight
5. New hard edge rounded floating triangle
6. adjusted angle of inside slope slightly
--And of course just a few other mirror adjustments
All critiques and feedback are welcome and encouraged . Thanks guys for all the feedback the first time.
just a few minor things i would add in/change/adjust.
love it dude! looking forward to seeing more!
The two "shelf" type shapes, the one sticking out on the bottom and the one on top, are still a little off proportionally, you can paste the render over the concept in photoshop to see.
We've seen a lot of it lately since PhilipK and 3DMotive did their awesome tuts on it, but the first thing I thought of when I looked at the concept was this:
http://www.brameulaers.com/cyber.html
What you've got is fantastic, and a great representation of the concept, but in terms of demonstrating your ability, I'd have said going with a modular workflow might show of your ability to tackle this kind of thing in a more versatile way.
What I mean by that is if you make this asset the way you have, then you've got one asset, textured in one way. Whereas if you did it in a way like in the link I just posted, you could have almost limitless possibilities in how you could construct your buildings/crates/whatever they are, and from a in-house, deadlines, maximum work minimum effort point of view for an environment artist, I'd have thought that was an appealing trait.
Looks great though, and sorry for going off on a tangent!