POV-Ray : Newsgroups : povray.beta-test : Priority troubles Server Time
2 Nov 2024 11:25:05 EDT (-0400)
  Priority troubles (Message 1 to 3 of 3)  
From: Leroy Whetstone
Subject: Priority troubles
Date: 15 Nov 2001 23:51:17
Message: <3BF4B750.39EA34A7@joplin.com>
I have POV3.5 beta7 for windows under WIN98 running on an Athlon 1.2ghz with 128
meg ram

The renders priority when set to highest or higher renders slower than when it
it set to normal.
I found this out by running a simple scene just one light, a plane and a sphere
with the output_to_file=off.
Then I ran a few more such quick programs that where a little more complicated
with the same results.
All those 'quick programs' ran under 2 minutes.

I thought that someone else had posted on this subject, if so it's lost to me.


--
Have Fun!


Post a reply to this message

From: Leroy Whetstone
Subject: Re: Priority troubles
Date: 18 Nov 2001 00:37:09
Message: <3BF7650C.EE69008F@joplin.com>
Here is a list of time trials

lowest   41s  ---  3.26  ---  9.26
lower    41s  ---  3.26  ---  9.17
normal  40s  ---  3.26  ---  9.26
higher   42s  ---  3.25  ---  9.19
highest  41s  ---  3.24  ---  9.26


Have Fun!


Post a reply to this message

From: Adrien Beau
Subject: Re: Priority troubles
Date: 18 Nov 2001 04:55:54
Message: <3BF785BB.70877850@free.fr>
Leroy Whetstone wrote:
> 
> I have POV3.5 beta7 for windows under WIN98 running
> on an Athlon 1.2ghz with 128 meg ram
> 
> The renders priority when set to highest or higher
> renders slower than when it it set to normal.

How much slower?

Leroy Whetstone wrote:
> 
> Here is a list of time trials
> 
> lowest   41s  ---  3.26  ---  9.26
> lower    41s  ---  3.26  ---  9.17
> normal  40s  ---  3.26  ---  9.26
> higher   42s  ---  3.25  ---  9.19
> highest  41s  ---  3.24  ---  9.26

Not slower. A one or two seconds difference in a render
is not very significant. There are lots of things Windows
can do in the background that can make you lose one
second in a particular render.

> I found this out by running a simple scene just one
> light, a plane and a sphere with the output_to_file=off.
> 
> Then I ran a few more such quick programs that where a
> little more complicated with the same results.
> 
> All those 'quick programs' ran under 2 minutes.

Using "quick programs" to measure render speed differences
is a bad idea. As I said, Windows can introduce random
micro-pauses that can amount to a small difference. But
when compared to the time your renders take, the difference
is not so small.

Why don't you try to render a batch of several of POV 3.5
example files. Try to reach at least an half hour (or the
whole night?) and see if you still have "highest priority"
renders slower than "normal priority" renders.

-- 
Adrien Beau   adr### [at] freefr   http://adrien.beau.free.fr/


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.