Home Contests & Challenges Archives Game Art Workshops

Workshop #4 - Rufuss

Turned out to be alot harder to model this then i thought it would be but im pretty happy with the result.


awsm1.jpg

awsm02.jpg

Replies

  • SyncViewS
    Offline / Send Message
    SyncViewS polycounter lvl 13
    Hi Rufuss, and welcome to Polycount. You've made a good job and I got some points to push it a bit further if you like.

    ws004_Rufuss_01.jpg

    A. There is a smoothing error. Control the topology. You got there 6 edges converging on a single vertex and it can cause too much surface stretch. Add edges to avoid such poles.
    B. Check the hard curve from the reference, it is quite smooth until a kink, then goes straight over the trigger hole, and dies there.
    C. Check the Hole shape with the reference, should be a little more smooth and roundish.
    D. It seems too big and deep, also the lid on the left (top render) should be a little higher.
    E. Is it too deep? Not completely sure.
    F. It looks too big, deep and spherical. Should be little more squarish and the deepest inner part should lay flat. See the crappy section drawn.
    G. It misses the two holes with nuts to hold the adjustable "face pillow".
    H. It misses "a thing I don't know what it is supposed to be".

    If you want to refine a bit the shape it can turn out really well. Keep up!
  • Rufuss
    Awesome critique :) Im definitely gonna refine it.
    And about H i did add that thing but i forgot to unhide it for the render hehe
  • SyncViewS
    Offline / Send Message
    SyncViewS polycounter lvl 13
    Hey Rufuss, I forgot to mention I posted high resolution side views of the rifle in the Workshop #4 general thread (post 3). If you like, you can get them for easier reference matching.
  • Rufuss
  • SyncViewS
    Offline / Send Message
    SyncViewS polycounter lvl 13
    Very good enhancement! The socket for the loading lever seems still a bit too big, but you can tweak it as soon as you start modeling the body. Congratulation, it is much better!

    Now you need to model the opposite side, because it is slightly different: you don't have neither loading lever socket nor the long one for ammunition ejection.

    Keep up the good work!
  • Rufuss
    Theres still some small things to fix and i will probaly do the bipod aswell.

    awsm1.jpg

    awsm2.jpg

    awsm3.jpg

    awsm4.jpg
  • Herman the sherman
    Its looking good!
    but the clip looks a tad thin?
    if you look at this image:

    s_psg90t.jpg
  • SyncViewS
    Offline / Send Message
    SyncViewS polycounter lvl 13
    Hey Rufuss, very good work! It is clean and well proportioned. I seem to spot only some minor proportion issues with the scope. You may want to check references I added in this workshop general thread. The scope cone looks a bit off, I got the impression it is a bit too small compared to the back side. The two extremities shouldn't be bigger either. They have additional rubber caps on them, but are not part of the original bare scope. If you want to add them, you should add the caps too.

    The ammunition clip is definitely right, Herman's reference are probably of a different or older model. You can see the right references here: AW Series. Check the pdf brochures too.

    AW_series.jpg
  • Rufuss
    Yea i think your right about the scope.. I actually modeled the caps but i didint like how they looked from a fps view so i removed them. And i seem to have forgotten the end pieces were a part of the scope caps lol
  • Rufuss
    Been working on the low poly model, its far from finished though.
    wip2u.jpg
    wip3a.jpg
    I really dont have much experience doing low poly's, dealing with smoothing groups, uv's and baking. ive pretty much just modeled stuff..
    I've been doing small tests on parts of the model to figure out how it works and its been driving me :poly127: after digging trough some threads i rendered an object based normal instead of tangent, and it looked great :)
    There is still alot of errors in it and its on a pretty low res normal map but compared to my previous results it looks awesome.
    wipn.jpg

    I'm still pretty confused about smoothing groups and why the tangent based normal turned out so bad. It just seems like i would have to add a ton of geometry to get it as smooth as the object normal but maybe im just doing the smoothing groups wrong?
    wip5h.jpg
    So is object based normals the way to go for this and what would be the difference from using a good tangent map instead?
Sign In or Register to comment.