The BRAWL² Tournament Challenge has been announced!
It starts May 12, and ends Oct 17. Let's see what you got!
https://polycount.com/discussion/237047/the-brawl²-tournament
It starts May 12, and ends Oct 17. Let's see what you got!
https://polycount.com/discussion/237047/the-brawl²-tournament
Best Of
Re: portfolio critique needed!
kosolapchik said:Ah, I see. Some of the cylinders could be straighter. ThanksFabi_G said:
Some borders seem almost aligned with the pixel raster, but not quite, many others seems more relaxed (didn't mark all islands in question, but hopefully it's clear. I find tube-like/ cylindrical shapes and strips are very fitting for this). Perhaps it comes across as nitpicky,but I think even when work-samples don't have LODs, it doesn't take much time during unwrapping and allows to show a habit of making assets "future-proof".
Here is a cool thread by @Neox on the topic: https://polycount.com/discussion/235489/uv-straightening-and-its-perks-on-lod-generation, with other users chiming with methods for other modeling apps.
This is why I like polycount!
A bit confused about when the undistorted ratio is more preferable over the straighten lines, though!
it depends on the final use really. the higher the texture resolution, the less this matters. if you dont have lods, go wild, make banafingers and curvy shapes.
distortion is a topic, 3d painting and projection solves this to
some extend. But as shown in my thread, squaring for the sake of square
shapes and not looking into how much it distorts, is also not the way
to go. It is always a balance between many factors.
With 3d painting not a lot of this matters anyways. that being said,
painting a straight line in 2d will always be simpler than brushing it
over a 3d curved surface.
say you wanna add stitching to that belt marked here
it can be a 2 click operation, or a lot more complex to solve, and no matter how good the tools become it will stay more complex than 2 clicks in 2d space. it will look less clean and pixelated depending on the final resolution and how close you actually can get. To me personally this is always time well spent, because it is time you save yourself or possibly others down the road in texturing, or lodding, or skinning at times using UVs to select the needed verts is also simpler than doing this in 3d.
Always keep in mind that the current step is likely not the last step, try to keep all steps of the production in mind at any step of the way.
Neox
1 ·
Re: The Bi-Monthly Environment Art Challenge | January - February (94)
Great work so far everyone! Quick update from me, things have been going slowly due to an illness the other week but I'm almost done with blockout out the main shapes - more of a mid poly at this point. Here's a quick render inspired by the orange renders @Hymil made.
I realised after posting this that the charging contact is facing the wrong direction and has since been fixed)
ATaylor96
7 ·
Re: Kevin Johnstone has passed
I'm happy to host them on my web server too if needed.
metalliandy
2 ·
Re: Modeling a bunch of VHS & Music Player props
I tried making a test trim sheet earlier, it seems to work okay with a few floating polygons for the trim details. Very low poly.Fabi_G said:Hi! Perhaps create an atlas of the front/ controls and planar project UVs. Extruded geometry used to detail closeup meshes could be mapped to some generic texture patches and removed for distance meshes. There was this thread recently, illustrating this on a flare gun handle.
However it only has AO, Cavity and Normal map data. So I assume it's not good, since I can't change the colors or text of the buttons etc?
But if I do a trim sheet with already textured buttons, numbers, letters etc then I can't really change it on the fly in engine?
Asim7
1 ·
Re: Kevin Johnstone has passed
I have found a few - I can work with @ScoobyDoofus and maybe @Eric Chadwick ? To get them hosted somewhere for dl/preservation. Vic20 and MacNulty SDK for sure so far. Still digging around and organizing things.
gorecap
4 ·
Re: Looking for Honest Feedback on My Stylized Game Environment!
Hi, cool project!
I would say if the goal is to support a moba/ rts style gameplay, the environment has to much contrast overall. Ideally have some units/ heroes in there throughout the creation process, so you can balance everything in a way so the gameplay elements read well. Perhaps you can add some parameters to your materials so you can dial it in in-engine. Generally vertical gradients can help to make things more readable (for example make trees stand out from ground).
Regarding style, I believe it currently looks more bumpy pbr than hand-painted, with some textures looking like sourced from photos.
The grass being distributes so evenly introduces noise/ contrast, especially with the colors being so different to the underlying ground texture. I would try having some groups of gras in select places. Particular where assets like rocks intersect with the ground, so the intersection appears detailed.
The paths currently seem backward to me: I would expect stone slates being placed in the middle of the path, where's frequently traffic. Perhaps look into mixing a mask into the blend (heightmap?) to achieve a more interesting/ varied transitions, currently the blends borders look repetitive.
For the final presentation, I would add some beauty shots without any OS UI. I don't know if you did this, but through-out the process, it can help to take screenshots and paint on top to quickly explore options.
Keep it up!
I would say if the goal is to support a moba/ rts style gameplay, the environment has to much contrast overall. Ideally have some units/ heroes in there throughout the creation process, so you can balance everything in a way so the gameplay elements read well. Perhaps you can add some parameters to your materials so you can dial it in in-engine. Generally vertical gradients can help to make things more readable (for example make trees stand out from ground).
Regarding style, I believe it currently looks more bumpy pbr than hand-painted, with some textures looking like sourced from photos.
The grass being distributes so evenly introduces noise/ contrast, especially with the colors being so different to the underlying ground texture. I would try having some groups of gras in select places. Particular where assets like rocks intersect with the ground, so the intersection appears detailed.
The paths currently seem backward to me: I would expect stone slates being placed in the middle of the path, where's frequently traffic. Perhaps look into mixing a mask into the blend (heightmap?) to achieve a more interesting/ varied transitions, currently the blends borders look repetitive.
For the final presentation, I would add some beauty shots without any OS UI. I don't know if you did this, but through-out the process, it can help to take screenshots and paint on top to quickly explore options.
Keep it up!
Fabi_G
2 ·
Re: portfolio critique needed!
Sure I was referring to the UVs of the latest character:
Some borders seem almost aligned with the pixel raster, but not quite, many others seems more relaxed (didn't mark all islands in question, but hopefully it's clear. I find tube-like/ cylindrical shapes and strips are very fitting for this). Perhaps it comes across as nitpicky ,but I think even when work-samples don't have LODs, it doesn't take much time during unwrapping and allows to show a habit of making assets "future-proof".
Here is a cool thread by @Neox on the topic: https://polycount.com/discussion/235489/uv-straightening-and-its-perks-on-lod-generation, with other users chiming with methods for other modeling apps.
Some borders seem almost aligned with the pixel raster, but not quite, many others seems more relaxed (didn't mark all islands in question, but hopefully it's clear. I find tube-like/ cylindrical shapes and strips are very fitting for this). Perhaps it comes across as nitpicky
,but I think even when work-samples don't have LODs, it doesn't take much time during unwrapping and allows to show a habit of making assets "future-proof".Here is a cool thread by @Neox on the topic: https://polycount.com/discussion/235489/uv-straightening-and-its-perks-on-lod-generation, with other users chiming with methods for other modeling apps.
Fabi_G
2 ·
Re: portfolio critique needed!
Yeah letting go of old projects is pretty difficult. Fortunately ArtStation lets you hide projects so you don't have to perma delete them or something. It may be difficult to land a job atm, it's difficult for everybody, but you have solid stuff in there. Good luck!
Ashervisalis
1 ·