On Wed, 01 Sep 1999 23:45:37 GMT, c.b### [at] cwcomnet (Cliff Bowman)
wrote:

>On Wed, 01 Sep 1999 06:11:34 GMT, NOS### [at] ibmnet (Thomas
>Baier) wrote:
>
>
>Double faces I think I know the reason for. A lot of modellers seem to
>get the normals *wrong* and so users sometimes make a copy of the
>model, flip the normals, and combine the two sets of meshes.

Right.

>
>Dirty, and probably works in many cases (i.e. patches holes) but time
>and memory are both wasted. I can't help but feel it'd cause some
>renderers to cough and splutter too  - though as I've not knowingly
>come across this yet I don't know how much POV "likes" or dislikes it.
>
>I'd expect there to be "coincident faces" (is that the term?) problems
>though.

I'm not really sure what you mean.

But one of the problems is that bordering faces do not use same the
vertices.

Anyway, you have to go around all problems to make a robust converter.

You have to build in some checks, which make the import slower.

-tb

Post a reply to this message

On Wed, 01 Sep 1999 23:45:37 GMT, Cliff Bowman wrote:

>Double faces I think I know the reason for. A lot of modellers seem to
>get the normals *wrong* and so users sometimes make a copy of the
>model, flip the normals, and combine the two sets of meshes.

I've actually seen this approach RECOMMENDED in an otherwise good
book.  Might have had something to do with the book being written
for an inferior modeller like MAX, though... <big grin for Lance>

Post a reply to this message

On 2 Sep 1999 09:22:17 -0400, par### [at] fwicom (Ron Parker) wrote:

>On Wed, 01 Sep 1999 23:45:37 GMT, Cliff Bowman wrote:
>
>>Double faces I think I know the reason for. A lot of modellers seem to
>>get the normals *wrong* and so users sometimes make a copy of the
>>model, flip the normals, and combine the two sets of meshes.
>
>I've actually seen this approach RECOMMENDED in an otherwise good
>book.  Might have had something to do with the book being written
>for an inferior modeller like MAX, though... <big grin for Lance>


????

Good grief.

And I'd just been working from conjecture... if such double faces WERE
imported into POV, how would it cope I wonder? I'll have to give it a
try. I can't help but think that with 2 triangles at the same point
POv would have difficulty deciding on which one to render at any given
X-Y co-ordinate - sometimes picking the one with the "correct" normal
and sometimes picking the other...



Cheers,

Cliff Bowman
Why not pay my 3D Dr Who site a visit at http://www.who3d.cwc.net/

Post a reply to this message

On Thu, 02 Sep 1999 07:01:36 GMT, NOS### [at] ibmnet (Thomas
Baier) wrote:

>On Wed, 01 Sep 1999 23:45:37 GMT, c.b### [at] cwcomnet (Cliff Bowman)
>wrote:
[snip]
>>I'd expect there to be "coincident faces" (is that the term?) problems
>>though.
>
>I'm not really sure what you mean.

If 2 faces using the same vertecices, but with flipped normals (to
avoid holes) were imported, I wonder if POV would suffer from it's
co-incident faces problem. Like I've seen it (admittedly in the past)
where two objects are unioned instead of merged and have x,y, and z
co-ordinates equal at some point(s) - let's say one red cube and one
blue one.

box <-1,-1,-1>,<0.5,0.5,1> RED
box <-0.5,-1,-1><1,1,1> BLUE

For any given point where the faces are identically positionsed (X >=
-0.5 and X <= 0.5) should POV render red or blue?

Poor wording but then I don't write well.

>But one of the problems is that bordering faces do not use same the
>vertices.

Argh. That could be a problem - bordering faces not using the same
vertices? I'll have to watch for that.

>Anyway, you have to go around all problems to make a robust converter.
>
>You have to build in some checks, which make the import slower.
>
Speed fortunately isn't my main concern :)


Cheers,

Cliff Bowman
Why not pay my 3D Dr Who site a visit at http://www.who3d.cwc.net/

Post a reply to this message

On Fri, 03 Sep 1999 21:20:21 GMT, Cliff Bowman wrote:
>And I'd just been working from conjecture... if such double faces WERE
>imported into POV, how would it cope I wonder? I'll have to give it a
>try. I can't help but think that with 2 triangles at the same point
>POv would have difficulty deciding on which one to render at any given
>X-Y co-ordinate - sometimes picking the one with the "correct" normal
>and sometimes picking the other...

It'd probably still work.  POV tends to flip the normal to the "correct"
orientation anyway, so both triangles would be essentially identical and
it wouldn't matter which of the two you hit.

Post a reply to this message

On 3 Sep 1999 17:29:39 -0400, par### [at] fwicom (Ron Parker) wrote:

>On Fri, 03 Sep 1999 21:20:21 GMT, Cliff Bowman wrote:
>>And I'd just been working from conjecture... if such double faces WERE
>>imported into POV, how would it cope I wonder? I'll have to give it a
>>try. I can't help but think that with 2 triangles at the same point
>>POv would have difficulty deciding on which one to render at any given
>>X-Y co-ordinate - sometimes picking the one with the "correct" normal
>>and sometimes picking the other...
>
>It'd probably still work.  POV tends to flip the normal to the "correct"
>orientation anyway, so both triangles would be essentially identical and
>it wouldn't matter which of the two you hit.

I've seen some pretty ugly, gappy .OBJ files imported with other tools
(OBJ2POV IIRC) where the triangles didn't get flipped - which is one
reason I've found announcements like SSS interesting. Figuring a way
to calculate normals for point son a mesh and comparing them to the
original to see if it's pointing the wrong way is something I'd like
to try (later). I'd rather not just recalculate normals in case there
are .OBJ files which actually *use* the normals they supply (for
bumps, dents, etc.).

I've got a good example of an OBJ file with many correct and many
incorrect normals. 3D Win seeems to do a good job of correcting those
facets when it exports to POV, but sadly not when it exports back to
OBJ format (no idea why).

I did try using the OpenGL version of 3D Win to see what it was
holding internally after importing an OBJ, but I can't get it not to
crash my PC :(

So - examining SSS and learning how to calculate normals seems like a
way to go :)


Cheers,

Cliff Bowman
Why not pay my 3D Dr Who site a visit at http://www.who3d.cwc.net/

Post a reply to this message

>I've got a good example of an OBJ file with many correct and many
>incorrect normals. 3D Win seeems to do a good job of correcting those
>facets when it exports to POV, but sadly not when it exports back to
>OBJ format (no idea why).
Hmm, be sure,  if you import a file, that the "unify" option is
switched on.

>
>I did try using the OpenGL version of 3D Win to see what it was
>holding internally after importing an OBJ, but I can't get it not to
>crash my PC :(
HW or SW OpenGL ?

Thomas

Post a reply to this message

On Mon, 06 Sep 1999 04:52:48 GMT, NOS### [at] ibmnet (Thomas
Baier) wrote:

>>I've got a good example of an OBJ file with many correct and many
>>incorrect normals. 3D Win seeems to do a good job of correcting those
>>facets when it exports to POV, but sadly not when it exports back to
>>OBJ format (no idea why).
>Hmm, be sure,  if you import a file, that the "unify" option is
>switched on.

Hmm.. tried that just now (while reading your message) and it hasn't
noticably helped - the .OBJ is still full of holes. Personally I blame
the modelling software for getting it wrong in the first place. You're
welcom to a copy of the Rex .OBJ file if you want to play with it
(although it's about 1.2MB unzipped).

>>
>>I did try using the OpenGL version of 3D Win to see what it was
>>holding internally after importing an OBJ, but I can't get it not to
>>crash my PC :(
>HW or SW OpenGL ?
>
Hardware - it's a TNT board made by Diamond. Let me check... Viper
V550 running under Windows98 SE, 96MB RAM, P 233 MMX (yes not even a
Pentium II) and far, far too much software on the hard drives :)


Cheers,

Cliff Bowman
Why not pay my 3D Dr Who site a visit at http://www.who3d.cwc.net/

Post a reply to this message

Hi,
>
>Hmm.. tried that just now (while reading your message) and it hasn't
>noticably helped - the .OBJ is still full of holes. Personally I blame
>the modelling software for getting it wrong in the first place. You're
>welcom to a copy of the Rex .OBJ file if you want to play with it
>(although it's about 1.2MB unzipped).
It would be nice to send me the TRex file. Perhaps i find some bugs
:_)

>
>>>
>>>I did try using the OpenGL version of 3D Win to see what it was
>>>holding internally after importing an OBJ, but I can't get it not to
>>>crash my PC :(
>>HW or SW OpenGL ?
>>
>Hardware - it's a TNT board made by Diamond. Let me check... Viper
>V550 running under Windows98 SE, 96MB RAM, P 233 MMX (yes not even a
>Pentium II) and far, far too much software on the hard drives :)
Hmm, i have tested 3DWinOGL with several GFX boards (but not TNT) .
Are you sure, that TNT renders into a window?

-tb

Post a reply to this message

On Wed, 08 Sep 1999 04:24:22 GMT, NOS### [at] ibmnet (Thomas
Baier) wrote:

>Hi,
>>
>>Hmm.. tried that just now (while reading your message) and it hasn't
>>noticably helped - the .OBJ is still full of holes. Personally I blame
>>the modelling software for getting it wrong in the first place. You're
>>welcom to a copy of the Rex .OBJ file if you want to play with it
>>(although it's about 1.2MB unzipped).
>It would be nice to send me the TRex file. Perhaps i find some bugs
>:_)

Always a possibility but I find it annoying, to say the least, to have
to write around other people's bugs. Why can't modelling programs "get
it right" in the first place? Then there'd be no need for these
shinnanegans...

Still, I'll e-mail a simple Rex .OBJ off to you (not the final quality
one I'd like to use, because that's big - but the one I've been using
for testing).

>>>>
>>>>I did try using the OpenGL version of 3D Win to see what it was
>>>>holding internally after importing an OBJ, but I can't get it not to
>>>>crash my PC :(
>>>HW or SW OpenGL ?
>>>
>>Hardware - it's a TNT board made by Diamond. Let me check... Viper
>>V550 running under Windows98 SE, 96MB RAM, P 233 MMX (yes not even a
>>Pentium II) and far, far too much software on the hard drives :)
>Hmm, i have tested 3DWinOGL with several GFX boards (but not TNT) .
>Are you sure, that TNT renders into a window?
>
Yup - sPatch, Unreal, 3D Labs X29 (an OpenGL "test" utility) all
render in a window (although Unreal's OpenGL seems decidely slow at
present - something they plan to address, I hear). Most other stuff I
tend to use fullscreen (actualy Unreal is Usually full screen but
they've got some bugs at the moment with D3D / TNT usage which makes
switching between windowed and full screen unreliable - no other App I
use has this problem - but their TNT D3D is still better than their
OpenGL support IMO).


Cheers,

Cliff Bowman
Why not pay my 3D Dr Who site a visit at http://www.who3d.cwc.net/

Post a reply to this message