Warp spake:

> Stefan Viljoen <spamnot@ <removethis>polard.com> wrote:
>> IMO get the fastest dual-core processor you can get. Beter yet, get a
>> quad motherboard (I think HP makes some nice ones) that will take four
>> processors. You can then theoretically render four times faster, but
>> you'll have to break your scene up into four segments. I have one
>> processor, and I
>> use  a "segments of two" approach when dual-core rendering - one virtual
>> CPU gets the "top" the other the "bottom". If one finishes before the
>> other, the remaining part is again divided by two, etc. You'll be able to
>> do this by fours if you get a quad based board.
> 
>   Please notice that if you do that, ie. render the scene in "four
>   segments",
> the amount of memory required will be four times as large. Often not very
> feasible. It's also troublesome.

I think I did mention that farther down my reply post. That might be a valid
reason to get as much memory as you can, as the poster seemed set to go
for.
 
>   You should know better than to suggest that, given that the new
> POV-Ray 3.7 can use multiple processors with basically no increased
> memory requirements.

Hmm - I did not know that and thus I stand corrected. 

Regards,
-- 
Stefan Viljoen
Software Support Technician / Programmer
Polar Design Solutions

Post a reply to this message

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mike Williams wrote:
> Wasn't it Thomas who wrote:
> 
>>4.  Any other advice that you think useful.
>>
> Be aware that POV-Ray uses mainly floating point arithmetic, but the
> vast majority of published processor speed information relates to the
> speed of integer arithmetic. 
> 
	Here's a pretty complete page of floating point benchmarks:
http://www.spec.org/cgi-bin/osgresults?conf=cfp2000;op=form

		Jerome
- --
******************************
*      Jerome M. Berger      *
*  mailto:jeb### [at] freefr   *
*  http://jeberger.free.fr/  *
******************************
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)

iD8DBQFD/Kyxd0kWM4JG3k8RAk6+AKCoSkp1mocfgj87eCdVrKi8wIPtxACgh6+v
EdzHmrNbVD+OETSRvpPPCTA=
=iaYv
-----END PGP SIGNATURE-----

Post a reply to this message

Stefan Viljoen <spamnot@<removethis>polard.com> wrote:

> I have one processor, and I
> use  a "segments of two" approach when dual-core rendering - one virtual
> CPU gets the "top" the other the "bottom". If one finishes before the
> other, the remaining part is again divided by two, etc.
>
 Can you explain that in more detail?  Sounds fascinating (and completely
unknown to me.) Thanks!

Ken

Post a reply to this message

"Thomas" <t_p### [at] hotmailcom> wrote:

> 4.  Any other advice that you think useful.
>
As mentioned, a fast bus speed will help. I recently bought an emachines
model T6412 (64-bit, single processor) machine, with a 1600MHZ front-side
bus, which is faster than many other (?) machines out there, that I've
seen.  Very good bang for the buck.

Ken

Post a reply to this message

Kenneth <kdw### [at] earthlinknet> wrote:
>  Can you explain that in more detail?  Sounds fascinating (and completely
> unknown to me.) Thanks!

  Just use POV-Ray 3.7 (when it comes out; if you are using Windows, you
can try the beta).

-- 
                                                          - Warp

Post a reply to this message

Warp <war### [at] tagpovrayorg> wrote:

>   Just use POV-Ray 3.7 (when it comes out; if you are using Windows, you
> can try the beta).
>
 I'm still trying to understand v3.6.1 ALPHA!  ;-)

Ken

Post a reply to this message

Kenneth <kdw### [at] earthlinknet> wrote:
> Warp <war### [at] tagpovrayorg> wrote:

> >   Just use POV-Ray 3.7 (when it comes out; if you are using Windows, you
> > can try the beta).
> >
>  I'm still trying to understand v3.6.1 ALPHA!  ;-)

  I didn't understand that.

-- 
                                                          - Warp

Post a reply to this message

Kenneth wrote:
> Warp <war### [at] tagpovrayorg> wrote:
> 
>>   Just use POV-Ray 3.7 (when it comes out; if you are using Windows, you
>> can try the beta).
>>
>  I'm still trying to understand v3.6.1 ALPHA!  ;-)
> 
> Ken
> 
> 

3.6.1a is not actually 3.6.1 Alpha, it's a bug-fix release for 3.6.1. 
So they could have called it 3.6.2, or 3.6.1.1, but went with 3.6.1a :)

...Chambers

Post a reply to this message

Kenneth spake:

> Stefan Viljoen <spamnot@<removethis>polard.com> wrote:
> 
>> I have one processor, and I
>> use  a "segments of two" approach when dual-core rendering - one virtual
>> CPU gets the "top" the other the "bottom". If one finishes before the
>> other, the remaining part is again divided by two, etc.
>>
>  Can you explain that in more detail?  Sounds fascinating (and completely
> unknown to me.) Thanks!

Tsk tsk Warp... he did ask, didn't he? And, as far as I know, 3.7 is not yet
available for Linux?

Hi Kenneth

For a detailed explanation, take a look at the image I submitted to Paul
Bourke at Swinburne univ, which was rendered on a "farm" of Alpha CPU based
systems, and the article he wrote about it:

http://astronomy.swin.edu.au/~pbourke/raytracing/parallel/

Basically, you use PovRay INI file statemens to restrict a certain instance
of PovRay to only rendering a part of the scene. With any multi-core or
multi-processor system you can make this as complicated as you want, I just
usually use horisontal segments, for simplicity. The idea is like any
co-operative effort - one core / CPU does one part of the image, another
does the next, etc. If one finishes before the other, the work assigned to
the remaining CPUs is redivided, and the "free" CPU again gets part of the
remaining work.

Get the idea? The floating point load is thus shared out among many
"workers", and, just like in a human team, it means the problem is solved
much more quickly. Each CPU / core does its part, and then everybody has to
work less to solve the remainder of the problem. 

Figuratively, the problem is decimated until it is solved, by assigning and
re-assigning CPUs / cores as the problem is reduced.

Thus, you can literally raytrace images in hours or days that would have
taken a uni-core or single CPU system weeks or months to finish.

For example, the image at the link above was attempted to be traced by me -
after about two weeks on my old 400 MHz PC about 10 lines of the 480 I
needed were done. I sent it to Paul at Swinburne University in Australia
and he did it on... I think it was 32 machines in about 10 or 12 hours.

Feel free to drop me a mail at spamnot@<removethis>polard.com if you want
some examples.

Kind regards,

Stefan Viljoen

Post a reply to this message

Stefan Viljoen <spamnot@ <removethis>polard.com> wrote:
> Tsk tsk Warp... he did ask, didn't he? And, as far as I know, 3.7 is not yet
> available for Linux?

  But what I am thinking is that the "manual parallelization" of pov3.6 is
not only a real memory hog (rendering takes as much memory as many instances
of POV-Ray you run) but is also quite cumbersome and difficult, requiring
third-party tools and whatnot. I was thinking that instead of teaching him
the really hard way of doing it, the better way would be to wait for pov3.7
which automatically uses all the available processors in the computer without
the user having to do anything special to achieve that, and additionally
POV-Ray will not take basically any extra memory because of that.

-- 
                                                          - Warp

Post a reply to this message