Page 1 of 2 12 LastLast
Results 1 to 15 of 24

Thread: Please HELP with a simple subpatch problem

  1. #1

    Please HELP with a simple subpatch problem

    Maybe not so simple? I have this curved surface (both ways) (1) and want to insert front and back mic holes (2). But, for the life of me I can't get this to work (3). Any help is much desired!

    Here's the file:
    http://dl.dropboxusercontent.com/u/378843DontMove/subpatchTest.lwo

    And a pic showing the problem:

  2. #2
    For a start I would delete the panels you have done and replace them with the original single polygon, and then increase the resolution (poly-count) of the object to give yourself more to work with. I'm amazed you got what you did out of a single polygon face. Why skimp on polygons since it is a detailed section which justifies a heavier mesh?

  3. #3

  4. #4
    It would seem one could do this in less than the patch resolution you show. FWIW, I ended up just freezed and boolean subtracting, but I'd rather have done it in Sub-D. Is the only way to accomplish this by increasing the base mesh resolution? Seems like I've seen car poly cages which are about this rough which have as fine a detail in the meshes.

  5. #5
    Axes grinder- Dongle #99
    Join Date
    Jul 2003
    Location
    Seattle
    Posts
    14,726
    Quote Originally Posted by djwaterman View Post
    Click image for larger version. 

Name:	Mesh.jpg 
Views:	95 
Size:	697.0 KB 
ID:	124453

    And object;

    subpatchTest2.lwo
    surely that's wasteful? A lot of those polys are doing very little.

    Of course, it's not that it's wasteful per se, all these bits are recycled, but the polyoverkill would make it harder to edit.

    (polyoverkill is too much too type: can we standardize on 'POK'?)
    They only call it 'class warfare' when we fight back.
    Praise to Buddha! #resist
    Chard's Credo-"Documentation is PART of the Interface"
    Film the cops. Always FILM THE COPS. Use this app.

  6. #6
    It's not wasteful, lose the idea that low poly count is a worthwhile goal for a quality model. Check production quality models and see that they rarely try to cut corners (If you know what the final shape is you wouldn't be editing it after having got the correct shape first, obviously you'd get the overall shape before upping the resolution). It's more important to get a nice looking mesh than trying to be clever or tricky with a low poly count, particularly on non organic product design type objects, there's really nothing to be gained by keeping poly count low, not now days.

  7. #7
    Axes grinder- Dongle #99
    Join Date
    Jul 2003
    Location
    Seattle
    Posts
    14,726
    My concern would be editing this object, that's just a whole lot of points to wrangle.
    They only call it 'class warfare' when we fight back.
    Praise to Buddha! #resist
    Chard's Credo-"Documentation is PART of the Interface"
    Film the cops. Always FILM THE COPS. Use this app.

  8. #8
    Quantum Mechanic danielkaiser's Avatar
    Join Date
    Feb 2003
    Location
    Riverside CA
    Posts
    1,056
    Look at your point statistics theres alot going wrong there and it has nothing to do with mesh density.
    Daniel Kaiser

    "A mistake is always forgivable, rarely excusable and never acceptable." Robert Fripp

    AMD Phenom II x6 3.2
    Win 7 Pro x64
    8gig Ram
    Nvidia 640 GT 2Gb
    LW 11 x64

  9. #9
    Hi Daniel,

    I always look at the polygon statistics for two and one point polygons. I'm not sure how to correctly interpret point statistics other then if there are single points connected to more than five edges-- it seems like it's a problem. Can you help me to understand what I should be looking for?

  10. #10
    Super Member
    Join Date
    Aug 2006
    Location
    S
    Posts
    3,511
    Quote Originally Posted by jeric_synergy View Post
    My concern would be editing this object, that's just a whole lot of points to wrangle.
    You can easily strip things back with bandglue and unify. Djw is right, this is a preference issue and even when you want low poly sometimes you get better results by working backwatds.

  11. #11
    Super Member
    Join Date
    Aug 2006
    Location
    S
    Posts
    3,511
    Quote Originally Posted by chippwalters View Post
    Hi Daniel,

    I always look at the polygon statistics for two and one point polygons. I'm not sure how to correctly interpret point statistics other then if there are single points connected to more than five edges-- it seems like it's a problem. Can you help me to understand what I should be looking for?
    Single points with more than 5 edges are a problem if the subd mode is "subpatch" and the polygons are five sided. e.g. if you "make pole" an octagonal disc the single point will have an array of 8 triangles around it, no problem

  12. #12
    Here is what is happening:

    Click image for larger version. 

Name:	SubpatchTest1.jpg 
Views:	59 
Size:	106.8 KB 
ID:	124462

    When it is subpatched it...

    Click image for larger version. 

Name:	SubpatchTest2.jpg 
Views:	67 
Size:	91.2 KB 
ID:	124463

    An economical "stepped down" version of something that works:

    Click image for larger version. 

Name:	SubpatchTest3.jpg 
Views:	70 
Size:	106.7 KB 
ID:	124464

    The "dense mesh" argument centers around the trade off between economy and control and a smoother mesh. The denser it is the less movement you will have and in turn less bumpyness due to detail.

    It is something you determine on an object by object basis. More organic objects of course have more leeway.

    Flat surfaces allow you to step down more steeply

    Planned bumps (knuckles on a finger for example) can be a great place to step down.

    In general try to be as economical as possible as even with today's computers and ram, things can crawl to a halt with a large scene full of objects.

  13. #13
    "Flat surfaces allow you to step down more steeply."
    Great advice.

    Thanks Richard! Exactly what I was looking for! Can you post the UN sub-patched (Tab off) version of those images? Just so I can see exactly how the topology looks? Also, were you able to view this in specular shaded mode to see if there were any surface aberrations or wrinkles? Even if there were, I suspect by fine-tuning the Z values of the coordinates it could be made smooth, though it may take a bit of time.

    Much appreciated!

  14. #14
    Axes grinder- Dongle #99
    Join Date
    Jul 2003
    Location
    Seattle
    Posts
    14,726

    Red face

    2nd on the "unTabbed" (ie polygonal) request: I hang my head, but it's just so much easier to understand in poly mode.

    I'm always stunned by Richard and JoePoe's ability to simplify a subd hairball. Wizards!
    They only call it 'class warfare' when we fight back.
    Praise to Buddha! #resist
    Chard's Credo-"Documentation is PART of the Interface"
    Film the cops. Always FILM THE COPS. Use this app.

  15. #15
    Click image for larger version. 

Name:	SubpatchTest4.jpg 
Views:	54 
Size:	102.0 KB 
ID:	124465

    Yeah I usually do that, but forgot.

    I always work with subpatches turned off for what it is worth.

    And this mesh is a mess. I mean if you were looking for a hard surface smooth mesh you could subdivide it more first. (so you would not have to step down) I just wanted to show the basic poly flow.

    PS: Not a big fan of endless tweaking on a mesh that has been stepped down. Do it a little but if it is causing issues give yourself more geometry.
    Last edited by Surrealist.; 09-24-2014 at 10:36 AM.

Page 1 of 2 12 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •