"SamuelT" <STB### [at] aolcom> wrote in message news:39517770.34EF822F@aol.com...
| I honestly can't find why the code does that. Nice landscape, BTW. The
clouds look
| improved from the last desert islands posting.

Thanks... and they look even better at 320X240, ha ha.  Not nearly as fast a
render as yours were either.
That CSG problem is a strange one, I don't recall ever seeing that kind of
thing go wrong like that.  It seems simple enough so I'll probably figure a
work-around at the very least.

Bob

Post a reply to this message

Did you try to remove auto-bouding? and vista buffers?
(just a _very_ wild and uneducated guess - gut feeling. My rational brain
doesn't believe this either, but trying should take very little time)




Philippe

PS: I always wanted to make a dragonfly anim... maybe later (as usual)

Post a reply to this message

On Wed, 21 Jun 2000 17:29:10 -0500, Bob Hughes wrote...
> Posting this for the reason of showing a CSG-gone-wrong situation.
> In the zoomed view of the dragonfly you will be able to see how a separate
> cylinder doesn't mix with the rest of the object at all, it actually removes
> (or more precisely, makes invisible) the parts behind it.
> All I could figure is that maybe one of the new MegaPov features is somehow
> acting upon it, such as the 'projected_through' or 'no_image', something
> anyhow; although perhaps nothing to do with it.  I put 'sturm' in the 'blob'
> which is the part behind the cylinder.  No CSG 'difference' is being done with
> the blob itself, only 'union'.  Here's that part of the pov script rendered
> under MegaPov 0.5a:

What have you got max_trace set to?

That's about the only thing I can think of which might cause that sort of 
effect, especially if you've got the bg colour set to white...

Bye for now,
     Jamie.

Post a reply to this message

"Philippe Debar" <phi### [at] hotmailcom> wrote in message
news:39520043@news.povray.org...
| Did you try to remove auto-bouding? and vista buffers?
| (just a _very_ wild and uneducated guess - gut feeling. My rational brain
| doesn't believe this either, but trying should take very little time)
|
| PS: I always wanted to make a dragonfly anim... maybe later (as usual)

Thanks but nothing changed it.  I even turned the light buffer off and each on
it's own.
Wish you luck on animating one.  I think I'll drop the blob and go with
sphere_sweep since I'm using MegaPov anyway.

Bob

Post a reply to this message

"Jamie Davison" <jam### [at] dh70qdu-netcom> wrote in message
news:MPG### [at] newsstmuccom...
|
| What have you got max_trace set to?
|
| That's about the only thing I can think of which might cause that sort of
| effect, especially if you've got the bg colour set to white...

No, max_trace_level is at 12 so I can see through the layered clouds, of which
there are only 5.  The background is a ordinary sphere with hollow on for the
fog but I've even removed all those leaving just the dragonfly CSG.  It's
definitely something odd anyhow.
Thanks for the try though.

Bob

Post a reply to this message

I've found replacing union with merge helps and yet does not completely cure
the problem of the disappearing blob behind the transparent cylinder.

Bob

Post a reply to this message

"Bob Hughes" <per### [at] aolcom?subject=PoV-News:> wrote in message
news:39535923@news.povray.org...
> I've found replacing union with merge helps and yet does not completely
cure
> the problem of the disappearing blob behind the transparent cylinder.
>
> Bob
>

I take it that max_trace has been ruled out?

Post a reply to this message

"Tom Melly" <tom### [at] tomandluf9couk> wrote in message
news:39535af8$1@news.povray.org...
|
| I take it that max_trace has been ruled out?

Yes  :-(

Post a reply to this message

Bob Hughes <per### [at] aolcom?subject=PoV-News:> wrote in message
news:39514220@news.povray.org...
> Posting this for the reason of showing a CSG-gone-wrong situation.
> In the zoomed view of the dragonfly you will be able to see how a
separate
> cylinder doesn't mix with the rest of the object at all, it actually
removes
> (or more precisely, makes invisible) the parts behind it.

The only time I had a similar problem it was solved by raising
max_intersections.
Check your stats for I_stack overflow.


--
Alf

http://www.peake42.freeserve.co.uk/
http://ourworld.compuserve.com/homepages/Alf_Peake/

Post a reply to this message

"Alf Peake" <alf### [at] peake42freeservecouk> wrote in message
news:3954cfab@news.povray.org...
|
| Bob Hughes <per### [at] aolcom?subject=PoV-News:> wrote in message
| news:39514220@news.povray.org...
| > Posting this for the reason of showing a CSG-gone-wrong situation.
| > In the zoomed view of the dragonfly you will be able to see how a
| separate
| > cylinder doesn't mix with the rest of the object at all, it actually
| removes
| > (or more precisely, makes invisible) the parts behind it.
|
| The only time I had a similar problem it was solved by raising
| max_intersections.
| Check your stats for I_stack overflow.

I wasn't getting a I_stack overflow, tried raising max_intersections per your
suggestion anyway.
Thing is, I know what the problem is, or at least related to anyhow.  It's the
very small scales I had been using.  I've known that 'sphere' will begin to
deteriorate in strange ways at a scale of about >0.02 unit from what I had
learned, which is nearly the starting diameters of the blob components.  Then
I further scaled down to 0.0025.  What fooled me was that it rendered properly
outside of the CSG so I thought I was safe.  Fact is, in using what I'd call
imprecise decimal numbers you always risk a degradation, especially of
'sphere' objects and quite obviously other primitives too ('cone' seems very
sturdy however).  Keep in mind this is CPU and OS dependant stuff from what I
understand.
So, with the blob components being at around 0.02 or greater in diameter the
problem vanishes, if you'll pardon the pun.
Thanks for trying, you and the others.  It was pretty stupid of me to struggle
over this with what I already understood.

Bob

Post a reply to this message