Results 1 to 10 of 10

Thread: Pinholes in liquids & glass

  1. #1
    Super Member sublimationman's Avatar
    Join Date
    Mar 2007
    Location
    Carlsbad, CA
    Posts
    517

    Pinholes in liquids & glass

    So in the attached image you can see little pinhole speckles near the top of the tea. I have never had this problem before and I have made quite a few liquids in glass before.

    The glass is sub-d and I selected the inside poly's up to the liquid level, copy, paste to a new layer, select top points, make poly, then cut and paste the top poly (Thanks Swampy for this trick).

    This is how I have been doing it lately and it's been working fine.

    Now both the glass and liquid are using dielectric node. If I make the liquid a solid color it still happens. If I select the liquid object and size it to 99.99% size then the pinholes disappear but then the refraction is way off because the liquid is not touching the glass.

    Now on this project I just used Photoshop to fix the final render. The image attached has been blown up a bit to help see the pinholes.

    I am done with this project but would like to understand what might have caused this so I can avoid it in the future.

    Any ideas?
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	pinholes.jpg 
Views:	292 
Size:	48.6 KB 
ID:	100298  

  2. #2
    I'm quite newbie on this glass and liquid matter, but i remember zardoz talking of a way of doing this properly.

    Make it a single mesh. Just the top of the liquid's surface is a separared mesh but the glass and filled liquid are the same mesh, only that the filled liquid polygons are flipped.

    http://forums.newtek.com/showthread....69#post1019669

    Anyway an example
    Click image for larger version. 

Name:	glass.png 
Views:	185 
Size:	553.1 KB 
ID:	100302
    glass.lwo

    I don't know how much IOR the Glass+Liquid surface should have, I just put the same as the liquid 1.333.

    Cheers
    Last edited by probiner; 12-16-2011 at 10:09 PM.

  3. #3
    I noticed there is something wrong with the previous setup (Single Mesh Flipped-1st) since there is no glass on the filled liquid.

    examples order and layer is the same. 1st is like my first post, just for back comparison.

    The 3rd one is like you have, I guess.
    The 2nd one is still Single Mesh, but double sided, using a Material Switch and Material Mix to control it.

    Results seem preety much the same (Absorption looks a little different). And you are dealing with a single polygon surface and not 2 overlayed.

    Cheers

    glass.lwo
    Click image for larger version. 

Name:	glass2.png 
Views:	212 
Size:	1.21 MB 
ID:	100321
    Last edited by probiner; 12-17-2011 at 12:12 PM.

  4. #4
    borkalork BORKALORK! biliousfrog's Avatar
    Join Date
    Dec 2003
    Location
    Lowestoft, UK
    Posts
    2,488
    You definitely need a slight seperation, it's caused by the polygons overlapping. The refraction shouldn't be off because it will treat the glass and liquid as seperate objects (as it should)...if anything, the refraction is probably wrong as-is.

    Obviously it all depends on the look you want, whether or not it's 100% accurate is irrelevant. Probiner's examples look very nice.

  5. #5
    Registered User
    Join Date
    Jan 2011
    Location
    near a beach
    Posts
    35
    Interesting topic...

    Rendered the glass.lwo with everything on.

    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	glasses.png 
Views:	608 
Size:	346.9 KB 
ID:	100415  

  6. #6
    CGI_Artist can you share the scene? =) Would like to make a render of it too since you have a nice reflection ceeling going on there =P

    Cheers

  7. #7
    Registered User
    Join Date
    Jan 2011
    Location
    near a beach
    Posts
    35
    sure thing.
    Attached Files Attached Files

  8. #8
    Thanks =) I requested because I noticed it might show more differences than my test, and it does.
    Just changed the camera to ortographic, change the GI to work better with transparency and activated Adaptive Sample (though, it still looks noisy).

    So, I guess Double Sided setup is no good unless I am using the wrong setting for the "null" Material on the double sided side (where is mixed with the Liquid)



    Quote Originally Posted by biliousfrog View Post
    You definitely need a slight seperation, it's caused by the polygons overlapping. The refraction shouldn't be off because it will treat the glass and liquid as seperate objects (as it should)...if anything, the refraction is probably wrong as-is.
    Well the strange thing is while I was making these test objects I had renders with the pinholes artifacts, but then afterwards they vanished :S

    I dont know if a slight separation would be good, I tryed to scale the liquid mesh a bit and results were bad =/




    So, for the opening post pinholes issue?
    Is it badly calculated normals? Give "Normal Corrector" plugin a try or the good old Unweld/Triple/Merge in Modeler to recalculate the normals
    Ray Precision and Cutoff settings?

    Cheers
    Last edited by probiner; 12-21-2011 at 04:36 PM.

  9. #9
    Super Member dblincoe's Avatar
    Join Date
    Aug 2004
    Location
    Springfield, Missouri
    Posts
    818
    Could be sugar?

    Seriously, is it a Vstack issue?

  10. #10
    Here's what the pinhole thing might be. I encountered something a little like it recently. As Biliousfrog mentioned, it may be polygons overlapping, and it may be caused by one object having it's subdivision levels set different to the other, causing some intersection. I'm not saying that is absolutely the cause, it's just one possibility.

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
  •