POV-Ray : Newsgroups : povray.binaries.images : warm_up redo Server Time
28 Mar 2024 22:23:38 EDT (-0400)
  warm_up redo (Message 6 to 15 of 15)  
<<< Previous 5 Messages Goto Initial 10 Messages
From: Norbert Kern
Subject: Re: warm_up redo
Date: 26 Dec 2019 03:00:01
Message: <web.5e0467f62d7f15a0ab765f000@news.povray.org>
"Thorsten Froehlich" <nomail@nomail> wrote:

> How long does it take to render today?

A nearly identical 3.5 version of the old image takes 349 s trace time with the
same settings (800*600 AA0.1 /3.0 GHz/16 threads/64 GB system). In 2001 a brand
new 1.4 GHz Athlon C / 1 GB system needed 101 h 18 m trace time.

So my current and slightly outdated system is 1045 times faster - astonishing...


Regards
Norbert


Post a reply to this message

From: Dick Balaska
Subject: Re: warm_up redo
Date: 26 Dec 2019 11:21:20
Message: <5e04de00$1@news.povray.org>
Am 12/25/19 6:14 AM, also sprach Norbert Kern:

Ouch.  I think a mosquito just bit me.

-- 
dik
Rendered 49,882,521,600 of 49,882,521,600 pixels (100%)


Post a reply to this message

From: Thomas de Groot
Subject: Re: warm_up redo
Date: 27 Dec 2019 02:40:01
Message: <5e05b551@news.povray.org>
Op 25/12/2019 om 15:57 schreef Norbert Kern:
> Thomas de Groot <tho### [at] degrootorg> wrote:
> 
>>> Media is faked, because I rendered an image from light position, made an
>>> height_field, filled it with emitting media and added it via a separate render.
> 
>> *That* is a beautiful piece of work indeed! Interesting to compare with
>> the original. Very, very well done. You are the unsurpassed master of
>> these kind of scenes. I am not sure I understand entirely how you
>> achieved your media here, though.
>>
>> --
>> Thomas
> 
> 
> Thank you very much!
> 
> I simply made an black-white image of the trees alone and used it as an
> height_field filled with emitting media after orienting it towards the
> light_source.
> Despite using emitting media only it is slow when incorporated in the original
> scene, so it had to be rendered separatedly and combined in an extra scene.
> 
> 
> Norbert
> 
> 

Smart! I shall have to try this (one day...).

-- 
Thomas


Post a reply to this message

From: BayashiPascal
Subject: Re: warm_up redo
Date: 29 Dec 2019 09:10:00
Message: <web.5e08b3302d7f15a0b56e457a0@news.povray.org>
Very well done. I like it !
Especially the trick about the "fake" media. Really clever...


Thomas de Groot <tho### [at] degrootorg> wrote:
> Op 25/12/2019 om 15:57 schreef Norbert Kern:
> > Thomas de Groot <tho### [at] degrootorg> wrote:
> >
> >>> Media is faked, because I rendered an image from light position, made an
> >>> height_field, filled it with emitting media and added it via a separate render.
> >
> >> *That* is a beautiful piece of work indeed! Interesting to compare with
> >> the original. Very, very well done. You are the unsurpassed master of
> >> these kind of scenes. I am not sure I understand entirely how you
> >> achieved your media here, though.
> >>
> >> --
> >> Thomas
> >
> >
> > Thank you very much!
> >
> > I simply made an black-white image of the trees alone and used it as an
> > height_field filled with emitting media after orienting it towards the
> > light_source.
> > Despite using emitting media only it is slow when incorporated in the original
> > scene, so it had to be rendered separatedly and combined in an extra scene.
> >
> >
> > Norbert
> >
> >
>
> Smart! I shall have to try this (one day...).
>
> --
> Thomas


Post a reply to this message

From: Warren
Subject: Re: warm_up redo
Date: 30 Dec 2019 15:55:02
Message: <web.5e0a63a42d7f15a05a5c2bce0@news.povray.org>
"Norbert Kern" <nor### [at] t-onlinede> wrote:
> I redid my 18 years old image, since it hasn't aged well.
>
> Back then it was impossible to optimize textures, lighting and so on. Render
> speed was 1.2 pps and memory need was at the limit.
>
> So I recently changed most plants, lighting, added a media effect and the water
> flows now in different directions...
>
> Media is faked, because I rendered an image from light position, made an
> height_field, filled it with emitting media and added it via a separate render.
>
>
> Merry x-mas to all
> Norbert Kern

That's a really nice image. A paradise for butterflies indeed. It reminds me of
the 'marais ( =swamp ) Poitevin' in west of France.


Post a reply to this message

From: Norbert Kern
Subject: Re: warm_up redo
Date: 31 Dec 2019 10:55:00
Message: <web.5e0b6ed52d7f15a0ab765f000@news.povray.org>
"Warren" <nomail@nomail> wrote:

> That's a really nice image. A paradise for butterflies indeed. It reminds me of
> the 'marais ( =swamp ) Poitevin' in west of France.


Thank you for pointing to marais poitevin.

Norbert


Post a reply to this message

From: Jörg "Yadgar" Bleimann
Subject: Re: warm_up redo
Date: 7 Jan 2020 18:52:05
Message: <5e1519a5$1@news.povray.org>
Hi(gh)!

On 25.12.19 12:14, Norbert Kern wrote:
> I redid my 18 years old image, since it hasn't aged well.
> 
> Back then it was impossible to optimize textures, lighting and so on. Render
> speed was 1.2 pps and memory need was at the limit.
> 
> So I recently changed most plants, lighting, added a media effect and the water
> flows now in different directions...
> 
> Media is faked, because I rendered an image from light position, made an
> height_field, filled it with emitting media and added it via a separate render.

Stunning! I would like to (like I already did in back 2001, when I saw 
the last version) see an olive-skinned mustached "swamp gypsy" with 
waist-long black hair wading through the water...

See you in Khyberspace!

Yadgar


Post a reply to this message

From: Jörg "Yadgar" Bleimann
Subject: Re: warm_up redo
Date: 7 Jan 2020 19:05:38
Message: <5e151cd2$1@news.povray.org>
Hi(gh)!

On 26.12.19 08:57, Norbert Kern wrote:

> A nearly identical 3.5 version of the old image takes 349 s trace time with the
> same settings (800*600 AA0.1 /3.0 GHz/16 threads/64 GB system). In 2001 a brand
> new 1.4 GHz Athlon C / 1 GB system needed 101 h 18 m trace time.
> 
> So my current and slightly outdated system is 1045 times faster - astonishing...

Too bad that I lost my first Moray-powered POV-Ray 2.2 scenes long 
ago... would be very intriguing to compare rendering times on my current 
AMD FX-6200 six-core machine with those on the Cyrix 80486 40DX back in 
1996!

But... 64 GB? Wow... and I thought, at 24 GB, I was "over-motorized"! If 
I only could afford my own scaled-down liquid salt moderated thorium 
fission reactor on the balcony, I also would go for a such powerful machine!

See you in Khyberspace!

Yadgar


Post a reply to this message

From: Paolo Gibellini
Subject: Re: warm_up redo
Date: 8 Jan 2020 04:16:25
Message: <5e159de9$1@news.povray.org>
Norbert Kern wrote on 25/12/2019 12:14:
> I redid my 18 years old image, since it hasn't aged well.
> 
> Back then it was impossible to optimize textures, lighting and so on. Render
> speed was 1.2 pps and memory need was at the limit.
> 
> So I recently changed most plants, lighting, added a media effect and the water
> flows now in different directions...
> 
> Media is faked, because I rendered an image from light position, made an
> height_field, filled it with emitting media and added it via a separate render.
> 
> 
> Merry x-mas to all
> Norbert Kern
> 

An incredible image.
A very credible image.

Paolo


Post a reply to this message

From: William F Pokorny
Subject: Re: warm_up redo
Date: 1 Mar 2020 09:07:31
Message: <5e5bc1a3$1@news.povray.org>
On 12/25/19 9:57 AM, Norbert Kern wrote:
> Thomas de Groot <tho### [at] degrootorg> wrote:
> 
...
> 
> I simply made an black-white image of the trees alone and used it as an
> height_field filled with emitting media after orienting it towards the
> light_source.
> Despite using emitting media only it is slow when incorporated in the original
> scene, so it had to be rendered separatedly and combined in an extra scene.
> 

Amazing image - then and again now.

I found myself this morning pondering - not for the first time since 
your post - why the emitting media was so slow you couldn't use it in 
scene.

Radiosity?

Are the heightfield and scene shapes getting tangled causing messier 
media intervals in the full scene?

Do you have thoughts as to why? Guess, I expected such a technique to 
more or less be a direct adder compute wise in scene.

Bill P.


Post a reply to this message

<<< Previous 5 Messages Goto Initial 10 Messages

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