I'm wondering, given an "average" scene (whatever that means), what would
render faster:

2000*1500 with AA @ 0.1

 or

3000*2250 with no AA ?

Any thoughts?

--
Paul Vanukoff
van### [at] primenetcom

Post a reply to this message

The question is ambiguous. Define average?
If there is little colour variation - or the variation is smooth - few pixels
will cross the AA threshold and 1) will render faster. But a checker floor for
instance has a lot of abrupt colour changes, so it may take longer.

Margus

Paul Vanukoff wrote:
> 
> I'm wondering, given an "average" scene (whatever that means), what would
> render faster:
> 
> 2000*1500 with AA @ 0.1
> 
>  or
> 
> 3000*2250 with no AA ?
> 
> Any thoughts?
> 
> --
> Paul Vanukoff
> van### [at] primenetcom

Post a reply to this message

Paul Vanukoff <van### [at] primenetcom> wrote:
: I'm wondering, given an "average" scene (whatever that means), what would
: render faster:

: 2000*1500 with AA @ 0.1

:  or

: 3000*2250 with no AA ?

  It depends on how complex the scene is. If the antialiasing kicks in very
often, then it's slower of course.
  With 3000x2250 no AA pov has to shoot 6750000 rays from the camera.
  If we suppose that every 10th pixel has to be supersampled, at the
resolution 2000x1500 pov has to shoot about 5700000 rays (84% of the first
one). If every 5th pixel has to be supersampled, it has to shoot about
8400000 rays (1.2% of the first one).

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

Wow... who has a PIII 750 to render an image _that_ large??
"Paul Vanukoff" <van### [at] primenetcom> wrote in message
news:38ad4934@news.povray.org...
> I'm wondering, given an "average" scene (whatever that means), what would
> render faster:
>
> 2000*1500 with AA @ 0.1
>
>  or
>
> 3000*2250 with no AA ?
>
> Any thoughts?
>
> --
> Paul Vanukoff
> van### [at] primenetcom
>
>
>

Post a reply to this message

I'll go with the non-AA as being faster than with AA up until the image
resolution is doubled, in which case I'd say the AA could win the render-time
race when it's half the resolution of the non-AA one.  But then it is very
scene-dependant stuff as the others are saying and +a0.1 is mighty slow if
there's much of any sort of pixel to pixel variances.
If I can guess you're "average" scene is like the shapes.pov file used for
testing POV-Ray after install then non-AA almost always wins :-)
I tried renders of shapes.pov just now at 512x384 -a, and 320x240 +a0.1 and the
non-AA one took only about 91% of the time it took to do the AA one, relatively
close really.

Bob

"Paul Vanukoff" <van### [at] primenetcom> wrote in message
news:38ad4934@news.povray.org...
| I'm wondering, given an "average" scene (whatever that means), what would
| render faster:
|
| 2000*1500 with AA @ 0.1
|
|  or
|
| 3000*2250 with no AA ?
|
| Any thoughts?
|
| --
| Paul Vanukoff
| van### [at] primenetcom
|
|
|

Post a reply to this message

>Wow... who has a PIII 750 to render an image _that_ large??

I've rendered at 6400x6400 on my PII400. Sure it takes a while, but it
proves you don't need the latest equipment to do it.

Has anyone else tried Ofoto? (Do they deliver to Panama for the same price?)

Post a reply to this message

Nieminen Juha wrote:

> If every 5th pixel has to be supersampled, it has to shoot about
> 8400000 rays (1.2% of the first one).

1.2%? Wow, that's way faster! ;-)

--
___     ______________________________________________________
 | \     |_                 <dav### [at] faricynet> <ICQ 55354965>
 |_/avid |ontaine               http://www.faricy.net/~davidf/

"Sitting on a cornflake, waiting for the van to come" -Beatles

Post a reply to this message

David Fontaine <dav### [at] faricynet> wrote:
: 1.2%? Wow, that's way faster! ;-)

  Of course I meant 120%...

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message