#declare Arrows2 = union {
#declare Num = 3000;
#declare K = 0;
#while (K < Num )
cylinder {<0.5,0,0>, <1,0,0>, 0.05
rotate <0,360*32*K/Num, 0>
Spline_Trans(KNOT_SplineC, K/Num, y, 0.000001, 0.00001)
pigment { color rgb <0.2,0.8,0.3>  } finish {KFIN} }
#declare K = K + 1;
#end
}

That I get an error is not strange, since I make parse errors all the time.
The strange part is that if I reduce "Num" to 1221 or smaller the error
goes away.  At 1222 or greater the error returns.  Or, on the other hand,
if I leave "Num" at 3000 and get rid of the "rotate" the error goes away.
But if I even include "rotate <0, 0, 0>" the error returns.
Or, if, instead of using Spline_Trans, I just translate along the spline
(leaving "Num" at 3000, and leaving the rotation in), the error goes away.
So I don't think there's anything wrong with my spline (it's a big ugly one
I imported from KnotPlot), and I can't figure why the "rotate" before
orienting via Spline_Trans works for larger increments, but not smaller
ones....  Also, changing the last two parameters on Spline_Trans seems to
have no effect on the error.

Any ideas?

Post a reply to this message

"Dave Matthews" <dav### [at] mnwestedu> wrote:
> Any ideas?

Here's a complete scene example:  The problem in this one starts at _Num =
607 (and higher):

#declare KNOT_SplineC =
  spline {
    cubic_spline
    -.25, <0,0,-5>
    0.00, <5,0,0>
    0.25, <0,5,0>
    0.50, <-5,0,0>
    0.75, <0,0,-5>
    1.00, <0,-5,0>
    1.25, <0,0,-5>
  }

#include "transforms.inc"

#declare Arrows = union {
#local _Num = 607;
#local _K = 0;
#while (_K < _Num )
sphere {0, 0.5 translate KNOT_SplineC(_K/_Num)
pigment { color rgb <1,0.8,0.3>  } }
cylinder {<0.5,0,0>, <1,0,0>, 0.05 rotate <0, 0, 360*72*_K/_Num>
Spline_Trans(KNOT_SplineC, _K/_Num, y, 0.001, 0.001)
pigment { color rgb <0.2,0.8,0.3>  } }
#local _K = _K + 1;
#end

}

object { Arrows }

camera { location -15*z look_at 0 }

light_source { -300*z rgb 1 }

/////////////////////////////////////////////////

Strangely, with the spheres removed, _Num can go as high as 2144 before the
error is encountered.  I can't see why that should happen, at all....

////////////////////////////////////////

Now, to make matters even weirder, eliminate the spheres, and replace the
_Num in the denominator of the rotation angle with (_Num + 0), or (_Num +
anything else I tried), and the error begins with _Num = 144.

Post a reply to this message

"Dave Matthews" <dav### [at] mnwestedu> wrote:
> "Dave Matthews" <dav### [at] mnwestedu> wrote:
> > Any ideas?
>
Oh, this was happening with version 3.6.0, so I downloaded 3.6.1b just to be
sure, and see the same behavior.  (POVRay for Windows)

Post a reply to this message

Dave Matthews wrote:
> "Dave Matthews" <dav### [at] mnwestedu> wrote:
>> "Dave Matthews" <dav### [at] mnwestedu> wrote:
>>> Any ideas?
> Oh, this was happening with version 3.6.0, so I downloaded 3.6.1b just to be
> sure, and see the same behavior.  (POVRay for Windows)

My suggestion is that you try with POV-Ray v3.5 and see if the error is 
still there.

-- 
Tor Olav
http://subcube.com

Post a reply to this message

Just thought I'd let you know I had exactly the same error a while ago while
using Spline_Trans. Although an experienced POV user this was the first
time I ever had an error that I couldn't find the cause for. Eventually I
gave up using it.
Like Dave I was using it in a loop for an animation & the error appeared
when the animation reached a particular frame. Changing or removing bits of
code that had no apparent connection (like removing your spheres) caused
the error to appear at a different frame no (or clk value).
This was the weirdest POV problem I've ever encountered. I was using POV
3.6.1. win xp


M.E.

Post a reply to this message

Tor Olav Kristensen <tor### [at] TOBEREMOVEDgmailcom> wrote:
> My suggestion is that you try with POV-Ray v3.5 and see if the error is
> still there.
>

Thanks, Tor Olav.  I just checked it.  It works fine with v3.5.  Now I think
it's something funny about changes to function calls with 3.6, since I also
had a POVRay graphing calculator that worked fine with v3.5, and I had to
modify to get it to work with v3.6.  I just assumed (with the earlier
glitch) that I had used some non-standard language that luckily worked with
v3.5, but no longer with v3.6.  With this problem, I don't think I did
anything bad, so maybe it's a bug (?)  Should I report it?  (I'm always a
bit scared to do that....)

Dave Matthews

Post a reply to this message

"Mike E." <nomail@nomail> wrote:
> Just thought I'd let you know I had exactly the same error a while ago while
> using Spline_Trans.

That's reassuring -- or maybe it isn't.  If it was just something I was
doing wrong, someone could tell me how to fix it.  As it is, someone else
will have to investigate.

Anyway (see above), if you really want to do whatever you were doing,
version 3.5 seems not to have the problem.

Dave Matthews

Post a reply to this message

Dave Matthews wrote:
> Tor Olav Kristensen <tor### [at] TOBEREMOVEDgmailcom> wrote:
> 
>>My suggestion is that you try with POV-Ray v3.5 and see if the error is
>>still there.
>>
> 
> 
> Thanks, Tor Olav.  I just checked it.  It works fine with v3.5.  Now I think
> it's something funny about changes to function calls with 3.6, since I also
> had a POVRay graphing calculator that worked fine with v3.5, and I had to
> modify to get it to work with v3.6.  I just assumed (with the earlier
> glitch) that I had used some non-standard language that luckily worked with
> v3.5, but no longer with v3.6.  With this problem, I don't think I did
> anything bad, so maybe it's a bug (?)  Should I report it?  (I'm always a
> bit scared to do that....)

Dave and Mike:

IIRC I have seen that error message many times too, while working with 
functions in POV-Ray v3.6 (both Windows and Linux versions). But I have 
never been able to make a minimal scene file that demonstrates this 
error clearly. (Commenting out, removing or adding code makes the parser 
complain at wildly different places in my code. And adding or removing 
semicolons after function and/or array declarations makes the parsing 
errors jump around too.)

I'm not sure, but I think that every time I have tried the to parse the 
problematic files in a v3.5 version, the problem has disappeared.

So no I have skimmed through news://povray.bugreports and 
http://www.povray.org/beta/ trying to find any known errors that relate 
to this, but I have not been able to find any.

I suspect this to be a bug (maybe more than 1 ?), so yes I think that 
somehow this should be reported.

-- 
Tor Olav
http://subcube.com

Post a reply to this message

> So no I have skimmed through news://povray.bugreports and 
> http://www.povray.org/beta/ trying to find any known errors that relate 
> to this, but I have not been able to find any.
> 
> I suspect this to be a bug (maybe more than 1 ?), so yes I think that 
> somehow this should be reported.

I reported it once to p.general, back in June 2006:

http://news.povray.org/povray.general/thread/%3C40e065f5$1@news.povray.org%3E/?ttop=214222&toff=700

Complete with minimal scene, etc. I still think that that's the way to 
go with bug reports (first posting to p.general), but nobody of the team 
ever answered to that thread. Could've just slipped through their 
attention, though.


HTH,
Florian

Post a reply to this message

Florian Brucker wrote:
...
>> I suspect this to be a bug (maybe more than 1 ?), so yes I think that 
>> somehow this should be reported.
> 
> I reported it once to p.general, back in June 2006:

Hrmmm... Are you sure your time machine was set to 2006 and not 2004 ?

=)

-- 
Tor Olav
http://subcube.com

Post a reply to this message