On Tue, 02 Apr 2002 14:43:07 +0200, "Thorsten Froehlich" <tho### [at] trfde>
wrote:
> Warp asked to "Please go through the officially unconfirmed bugs", not through
> all bugs...                                     ^^

Yes, I noticed this. I walked through all becouse of many resons:
- new beta
- many instances of povray started so best free short time to fight with bugs
- before confirmation of unconfirmed bugs I wanted to refresh current not
working but confirmed list to verify later if there could be something related
in unconfirmed list
- walking through fixed list sometimes shows that something is not fixed
- it not hurt anybody probably
- errare humanum est

ABX

Post a reply to this message

Skiba <abx### [at] babilonorg>  wrote:

>>* Media problem
>>  (A visible whitish line appears at the edge of a transparent container
>>  box for media. Needs confirmation.)
>>  Reported in:
>> Media visible container line in front of object
>> http://news.povray.org/3bf45a3d@news.povray.org
>
> I can confirm appearance of bright line. I have changed camera location in
> example to location <1+clock, 1.7, -5> and started animation.
> I have rendered with +FN with standard settings 800x600, AA 0.3.

I tried it and at least don't see anything unexpected.  Could you post an
image, please?

>> * Prism transformation problem
>>   (Conic prism illumination is not correct after certain rotations. Bug
>>   status unclear.)
>
> Note I have repeated this behaviour with matrix also (coded rotation)
> http://news.povray.org/d1oh6u821648rvs7frgmt4cl7u41s6cmmo%404ax.com
>
>>   (Note: It has been reported that this bug would not happen in POV-Ray 3.1.
>>     This needs verification.)
>
> What about recompilation of 3.1 with the same compiler and settings as current
> betas to verify compiler issues ?

So you have tried it in 3.1 and it doesn't appear?  If it doesn't happen in
3.1, could you post an image of the result in 3.5, please?

>>* Shadow rendering problem
>
> At first I think originall report needs some changes. First instead
> multiplying coordinates of camera probably scale sc should be applied at end.
> Second light_source should be also scaled with sc to recreate the same exact
> world. Anyway I think it is EPSILON problem.

So you are suggesting it is only an odd camera and light position together
with an overall very "small" scene?  Does it go away when scaling everything
by 10 or 100?

>> * Crash during radiosity pretrace
>>  (Needs confirmation.)
>
> since example crashed after 7 hours on Athlon 1 GHz, 512 Mb RAM
> I'm probably not the best person to check this with poor PII 233 MHz, 128 MB

Which means it can be close to anything.  Maybe it was just an overheated
processor :-)

    Thorsten

____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde

Visit POV-Ray on the web: http://mac.povray.org

Post a reply to this message

>> * Slowdown using +a +am1 +ua (job000191)
>>  (Using +ua and antialiasing with method 1 causes abnormal slowdown.)

> can't recreate
> any reference to original script ?

  It seems that it happens (only?) with media. Make a clear sphere with
some media inside and render with those settings.
  I'll add a note.

-- 
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}//  - Warp -

Post a reply to this message

On Tue, 02 Apr 2002 15:32:03 +0200, "Thorsten Froehlich" <tho### [at] trfde>
wrote:
> > >* Shadow rendering problem
> >
> > At first I think originall report needs some changes. First instead
> > multiplying coordinates of camera probably scale sc should be applied at end.
> > Second light_source should be also scaled with sc to recreate the same exact
> > world. Anyway I think it is EPSILON problem.
>
> So you are suggesting it is only an odd camera and light position together
> with an overall very "small" scene?

I suggest that after intersection test for 'screen ray' interesection test for
'shadow ray' returns point so close to original intersection point that distance
between them is lower than some epsilon and therefore is considered as the same
surface and not produce shadow.

> Does it go away when scaling everything
> by 10 or 100?

Yes I tested, It appear only for small values, not for large.
Try below animation with +KFF20

#declare sc=10^div(frame_number-9,2);

#if(mod(frame_number-1,2)=0)
  merge {
#else
  union {
#end
box{<-.5,0.8,-.5>,<.5,1.2,.5> pigment{rgb 1} scale sc }
box{<-.5,0,-.5>,<.5,.4,.5> pigment{rgb 1} scale sc }
box{<-.5,0,-20>,<20,-.2,20> pigment{rgb 1} scale sc }
}

background{rgb<1,0,0>}

light_source{<-200,150,100> rgb 1 scale sc}

camera{
  location<2,3,-4>
  look_at<-200,-400,1000>
  scale sc
}

ABX

Post a reply to this message

On Tue, 02 Apr 2002 15:32:03 +0200, "Thorsten Froehlich" <tho### [at] trfde>
wrote:
> > >* Media problem
> I tried it and at least don't see anything unexpected.  Could you post an
> image, please?
> > > * Prism transformation problem
> So you have tried it in 3.1 and it doesn't appear?  If it doesn't happen in
> 3.1, could you post an image of the result in 3.5, please?

all above posted to povray.beta-test.binaries

ABX

Post a reply to this message

> * Image_maps using png images are gamma-corrected (when they shouldn't)

Note that everything works fine when assumed_gamma (of any value) is
specified in the scene. The problem is only when no assumed_gamma is
specified. I don't think the solution is to not gamma-correct them at all.

Anders

Post a reply to this message