How to deal with "BSP node with unbound voulume" error

Discussion in 'Mapping Questions & Discussion' started by hooodini, Nov 19, 2016.

  1. hooodini

    hooodini L1: Registered

    Messages:
    30
    Positive Ratings:
    0
    There are about 1000 of these in my map
     
  2. Vel0city

    aa Vel0city func_fish

    Messages:
    1,930
    Positive Ratings:
    1,540
    From the developer wiki:

    "This occurs when some part of your map is touching, or is out of the hammer map grid. Move away from the edges of the grid, then compile."

    It can also be caused by invalid solids. Go in Hammer, open your vmf, click Alt+P, and look what kinds of errors it spits out. If there are invalid brushes it'll tell you there and which one(s) is (are) invalid.
     
  3. hooodini

    hooodini L1: Registered

    Messages:
    30
    Positive Ratings:
    0
    there are no parts of the map that are invalid or outside of the map, if there were this problem would have already been fixed
     
  4. hooodini

    hooodini L1: Registered

    Messages:
    30
    Positive Ratings:
    0
    these are 4 compile logs from my map, take a look at them
     

    Attached Files:

  5. Muddy

    Server Staff Muddy Muddy

    Messages:
    2,387
    Positive Ratings:
    4,220
    I've had times before when Hammer has made a brush that extends all the way from the top of the grid to the bottom for no adequately explained reason, causing this error to occur. Check that that hasn't happened to you too.

    Otherwise, post the VMF here and someone will take a look at it
     
  6. hooodini

    hooodini L1: Registered

    Messages:
    30
    Positive Ratings:
    0
    that has happened before but i noticed it this time thee is literally nothing, take a look
     

    Attached Files:

  7. ibex

    aa ibex

    Messages:
    255
    Positive Ratings:
    369
    I have no clue how you managed, but I can't get this vmf to stop leaking, even with cordon. Okay wait, turned on all user visgroups and now it will stop leaking with cordon (that doesn't touch the edges of the hammer grid). I didn't know user visgroups could do that, but I don't use them very often.

    Also, when I toggled on cordon you had it sized to map extents, so I have a feeling you were compiling with cordon from the beginning and that was what was creating the unbound volume areas. Just going off the fact that you have a few holes in that box area leads me to believe you were compiling with cordon. I should note that for all the compiles I tried I didn't get any of the unbounded error messages. So either you've changed something about the file before uploading it or it was compiling with the cordon at map extents causing the errors.

    Beyond that problem, I think you shouldn't be mapping on the 1u grid size because that's just bad practice. And for the 16000u hole, do what you will, but I think it's pretty ridiculous distance. Also, what's with the grouping on the doomsday_event fortune teller rip? It doesn't respond to ungroup and every piece of the structure has one of those radial fade boundaries.
     
  8. hooodini

    hooodini L1: Registered

    Messages:
    30
    Positive Ratings:
    0
    Well I somehow fixed it and now I have a new error, parts of the map won't load at the same time and in the console it says the error
     
  9. hooodini

    hooodini L1: Registered

    Messages:
    30
    Positive Ratings:
    0
    I will re upload the VMF
     
  10. Gale

    Gale L1: Registered

    Messages:
    38
    Positive Ratings:
    13
    Fixed! Make sure you keep geometry 1024 units away from the edges of the grid to be safe. Anotherissue was an object that was a func_detail that was leaking, but was grouped together with 9 other objects a bunch of times so it just showed as "Group of 1 Object". I had to ungroup it 12 times to get to see it.
     

    Attached Files:

  11. hooodini

    hooodini L1: Registered

    Messages:
    30
    Positive Ratings:
    0