In article <3905FCC7.A71197DD@ibm.net>, Andrew Clinton 
<cli### [at] ibmnet> wrote:

> Pov has lots of great features and creates some amazing images, but my
> biggest request for a future version would be to have a fast scanline
> renderer working alongside the raytracer, even if this renderer did not
> support many of the standard Pov features.  This would likely solve many
> of my preview/animation speed problems.
> 
> Of course, I have no idea how much work this would be for the Pov-Team!

POV-Scan? :-)
It would be a tremendous amount of work...first, you would have to 
tesselate all of the primitives. This is impossible for some of 
them...and you would have to add tesselation parameters to the language 
for all the objects. You would also have to make a completely new 
scan-line rendering engine, not much could be reused. And adding a 
rendering feature would be at least twice as much work...maintenance 
would be quite a bit more difficult.

It would be possible to make a raytracing "preview mode" with no 
lighting and simplified shading, and maybe simple depth cuing, maybe 
only rendering the bounding shapes, but this is really not very 
necessary with the quality level adjustment available.

-- 
Christopher James Huff - Personal e-mail: chr### [at] yahoocom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://chrishuff.dhs.org/
TAG Web page: http://tag.povray.org/

Post a reply to this message

Thank you for the insight... well it doesn't look like this will be happening
anytime soon.  Maybe something for Pov-ray version 10 :)

The quality or mosaic preview are nice but they still fall short when I need
both passable quality and rendering speed at the same time.  My current
project ( a physics motion simulation study) will probably take upwards of 1
week to render even with the best possible optimizations I can add to my code.

Andrew C


> POV-Scan? :-)
> It would be a tremendous amount of work...first, you would have to
> tesselate all of the primitives. This is impossible for some of
> them...and you would have to add tesselation parameters to the language
> for all the objects. You would also have to make a completely new
> scan-line rendering engine, not much could be reused. And adding a
> rendering feature would be at least twice as much work...maintenance
> would be quite a bit more difficult.
>
> It would be possible to make a raytracing "preview mode" with no
> lighting and simplified shading, and maybe simple depth cuing, maybe
> only rendering the bounding shapes, but this is really not very
> necessary with the quality level adjustment available.
>
> --
> Christopher James Huff - Personal e-mail: chr### [at] yahoocom
> TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
> Personal Web page: http://chrishuff.dhs.org/
> TAG Web page: http://tag.povray.org/

Post a reply to this message

you can't have a 6cpu PIII based system it's 1 2 4 8 not 6 it's called
semicircle multiprocessing 6 is not a power of 2 :)

Mike Weber wrote:

> Yeah, and I need a render box with 6 PIII 700Mhz CPU's to speed up my
> animations...
>
> --
> Mike
>
> wk: mik### [at] pyxiscom  www.pyxis.com
> hm: mwe### [at] sciticom  www.geocities.com/mikepweber
>
> "Bryan Valencia" <bry### [at] 209softwarecom> wrote in message
> news:3905DE93.A4642F36@209software.com...
> > Ok, there's official POV that can take a day to render a complex
> > scene...
> > ...and there's Mega POV which has lots of cool stuff, but takes a week
> > to render a complex scene.
> >
> > I need milli-POV which is optimized for speed and can render a complex
> > scene in 10 minutes.
> >
> >
> >
> > --
> > Bryan Valencia
> > Software Services - Making Windows Scream
> > http://www.209software.com
> > mailto:bry### [at] 209softwarecom
> >
> >

Post a reply to this message

In article <39062852.9E574715@ibm.net>, Andrew Clinton 
<cli### [at] ibmnet> wrote:

> Thank you for the insight... well it doesn't look like this will be 
> happening anytime soon.  Maybe something for Pov-ray version 10 :)

Do you think we will need one by then? Computers are getting faster, 
even in relation to rendering requirements.


> The quality or mosaic preview are nice but they still fall short when 
> I need both passable quality and rendering speed at the same time.  
> My current project ( a physics motion simulation study) will probably 
> take upwards of 1 week to render even with the best possible 
> optimizations I can add to my code.

What is your scene file composed of?!? Does it have a huge number of 
transparent surfaces, blurred reflections, radiosity with high recursion 
levels, etc?

-- 
Christopher James Huff - Personal e-mail: chr### [at] yahoocom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://chrishuff.dhs.org/
TAG Web page: http://tag.povray.org/

Post a reply to this message

Chris Huff wrote:
> 
> In article <39062852.9E574715@ibm.net>, Andrew Clinton
> <cli### [at] ibmnet> wrote:
> 
> > Thank you for the insight... well it doesn't look like this will be
> > happening anytime soon.  Maybe something for Pov-ray version 10 :)
> 
> Do you think we will need one by then? Computers are getting faster,
> even in relation to rendering requirements.
> 
> > The quality or mosaic preview are nice but they still fall short when
> > I need both passable quality and rendering speed at the same time.
> > My current project ( a physics motion simulation study) will probably
> > take upwards of 1 week to render even with the best possible
> > optimizations I can add to my code.
> 
> What is your scene file composed of?!? Does it have a huge number of
> transparent surfaces, blurred reflections, radiosity with high recursion
> levels, etc?

Indeed ! Perhaps Andrew is needlessly over complicating things.

-- 
Ken Tyler - 1400+ POV-Ray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/

Post a reply to this message

P.S. I would like to point out that this whole thread is off-topic for
this group. Text based discussions like this would be better discussed
in povray.general.

Post a reply to this message

Post a reply to this message

Attachments:
Download 'reply.gif' (5 KB)

Preview of image 'reply.gif'

Yeah, but the software gets bigger and slower as the PCS do, and we all add
more detail to our renders until it gets too slow.

I'd be interested to run whatever version of POV was out during the 80286
days on my current machine.  I know I'd lose a lot of features, but I'd just
like to see.

Chris Huff wrote:

> Do you think we will need one by then? Computers are getting faster,
> even in relation to rendering requirements.
>

--
Bryan Valencia
Software Services - Making Windows Scream
http://www.209software.com
mailto:bry### [at] 209softwarecom

Post a reply to this message

ROFL!

Peter Popov wrote:

>                 Name: reply.gif
>    reply.gif    Type: GIF Image (image/gif)
>             Encoding: base64

--
Bryan Valencia
Software Services - Making Windows Scream
http://www.209software.com
mailto:bry### [at] 209softwarecom

Post a reply to this message

Attachments:
Download 'us-ascii' (1 KB)

In article <39063634.A2080345@209software.com>, Bryan Valencia 
<bry### [at] 209softwarecom> wrote:

> Yeah, but the software gets bigger and slower as the PCS do, and we 
> all add more detail to our renders until it gets too slow.

But the available computing power is growing faster than the amount POV 
uses.(I'm not talking about other software, just the class of software 
POV-Ray is in.) You render typical scenes faster now than you would have 
done "back then", even when using the latest power-hungry features. 
There are no where near as many 5-day renders as there used to be, and 
most people consider 5 hours slow. Also, more efficient algorithms are 
being discovered("differential ray tracing", for example, could be a 
good replacement for the blurred reflection patch, and could be a lot 
faster.).


> I'd be interested to run whatever version of POV was out during the 
> 80286 days on my current machine.  I know I'd lose a lot of features, 
> but I'd just like to see.

It would run faster if you recompiled it with a recent compiler(there 
have been a lot of optimization improvements since then), and maybe a 
little faster even if you didn't.

-- 
Christopher James Huff - Personal e-mail: chr### [at] yahoocom
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Personal Web page: http://chrishuff.dhs.org/
TAG Web page: http://tag.povray.org/

Post a reply to this message