|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
translate x*2 is the same as:
a. translate 2
b. translate 2*x
c. translate <2, 0, 0>
d. translate <2, 1, 1>
e. translate <4/2, 0, 0>
f. b, c and e
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Parsing time will be shorter if you use:
a. separate trasnformation operations - scale, rotate and translate
b. one transformation block - transform{scale x*n rotate y*m translate z*k}
c. matrix
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Gena <obu### [at] mailcom> wrote:
> Parsing time will be shorter if you use:
> a. separate trasnformation operations - scale, rotate and translate
> b. one transformation block - transform{scale x*n rotate y*m translate z*k}
> c. matrix
Have you actually measured which of those is fastest?
--
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3EDD4217.83B3A431@mail.com>, Gena <obu### [at] mailcom> wrote:
> Parsing time will be shorter if you use:
>
> a. separate trasnformation operations - scale, rotate and translate
> b. one transformation block - transform{scale x*n rotate y*m translate z*k}
> c. matrix
The answers are vaguely worded, so it is unclear what exactly they mean,
and the topic...there won't be enough of a difference to matter.
--
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Well, I don't want to post answers here :) I only measured difference between a
> and b.
> For b and c I rely on documentation - 6.3.4 Transform Identifiers (see bottom
> part).
In case of c) it would be more correct to say:
c. separate matrix for scale, rotate and translate.
Because I think one combined transformation matrix should have the same
performance as 'transform'. Please correct me if I'm wrong :)
Gena.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christopher James Huff wrote:
> The answers are vaguely worded, so it is unclear what exactly they mean,
Feel free to correct, add and modify everything :)
> and the topic...there won't be enough of a difference to matter.
Do you mean that just one category will be enough?
Gena.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Gena Obukhov <obu### [at] mailcom> wrote:
>> and the topic...there won't be enough of a difference to matter.
> Do you mean that just one category will be enough?
I think that he meant that the difference in parsing time between the
given transformation syntaxes are so minimal that in practice it isn't
a big deal. (And as we know, the difference in rendering time, which is
usually the most important issue, is non-existent.)
--
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
> I think that he meant that the difference in parsing time between the
> given transformation syntaxes are so minimal that in practice it isn't
> a big deal. (And as we know, the difference in rendering time, which is
> usually the most important issue, is non-existent.)
I posted some sources to p.b.s.-f. newsgroup (transform test). Try to
render that scene with two different macros: 'tomtree.inc' and 'tomtreet.inc'
The only difference between these two macros is 'transform'. The scene which
uses macro with 'transform' (tomtreet.inc) renders 6-7 times faster.
Gena.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 04 Jun 2003 18:56:49 -0700, Gena Obukhov <obu### [at] mailcom> wrote:
>> Well, I don't want to post answers here :) I only measured difference between a
>> and b.
>> For b and c I rely on documentation - 6.3.4 Transform Identifiers (see bottom
>> part).
>
>In case of c) it would be more correct to say:
>c. separate matrix for scale, rotate and translate.
>
> Because I think one combined transformation matrix should have the same
> performance as 'transform'. Please correct me if I'm wrong :)
Simple transformation written directly in modifiers, as
scale 2
translate x
means 4 tokens to be parsed, while enclosed by transform
transform{
scale 2
translate x
}
means 7 tokens to be parsed, while written as matrix
matrix <2,0,0,0,2,0,0,0,2,1,0,0>
means 26 tokens to be parsed. So all together difference might be not that
simple as it may look. They are rather other benefits from using one method or
another.
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3ede847e@news.povray.org> , Warp <war### [at] tagpovrayorg> wrote:
>> a. separate trasnformation operations - scale, rotate and translate
>> b. one transformation block - transform{scale x*n rotate y*m translate z*k}
>> c. matrix
>
> Have you actually measured which of those is fastest?
Does it matter? ;-) If you want, I can make any of these the fastest! ;-)
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
|
|
| |
| |
|
|
|
|
| |