POV-Ray : Newsgroups : povray.beta-test : Code gets the processor too hot? Server Time
31 Oct 2024 19:25:26 EDT (-0400)
  Code gets the processor too hot? (Message 12 to 21 of 21)  
<<< Previous 10 Messages Goto Initial 10 Messages
From: clipka
Subject: Re: Code gets the processor too hot?
Date: 2 Apr 2013 14:38:10
Message: <515b2592@news.povray.org>
Am 02.04.2013 19:20, schrieb Ger:

>> Note that POV-Ray doesn't collapse nested differences into a single one
>> (although it would probably be a good idea).
>
> I have looked at the source code to find out as to what might possibly cause
> this, but it's several miles beyond my knowledge of code. (I don't go much
> further then "hello world")

Well, the thing that causes that nested differences aren't collapsed 
into a single one is the /absence/ of code to effect the collapsing, so 
no surprise you didn't find anything :-P

It probably would have to go in the parsing code for CSG objects.


Post a reply to this message

From: Ger
Subject: Re: Code gets the processor too hot?
Date: 2 Apr 2013 17:22:20
Message: <515b4c0c$1@news.povray.org>
clipka wrote:

> 
> Well, the thing that causes that nested differences aren't collapsed
> into a single one is the /absence/ of code to effect the collapsing, so
> no surprise you didn't find anything :-P
> 
> It probably would have to go in the parsing code for CSG objects.

Well, I'm not going to try and patch the povray sources. That's beyond my 
level. :)
What I did however is store the differenced spheres in a text file and later 
#include that. Doing it this way you don't have the nested differences 
anymore and that has a huge impact on the render time. 7 seconds against 
1:36 minutes.

As it turns out (from what I can determine) povray spends a lot of time 
handling the nested differences.

-- 
Ger


Post a reply to this message

From: Kenneth
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 13:50:01
Message: <web.515c6b1217e3da99c2d977c20@news.povray.org>
clipka <ano### [at] anonymousorg> wrote:

>
> A well-designed computer should be ok even at full throttle. However,
> not all computers are well-designed with respect to thermal management,
> and POV-Ray has a habit of making the CPU sweat like no other application.

Yep, I've certainly noticed that; my old emachines single-core AMD64
Athlon-based computer kicks into 'high gear' on some of my scenes. (I can even
hear a subtle high-pitched whine--not from the cooling fan AFAIK, but, I'm
guessing, from the CPU(?).)

> ... AMD CPUs have a reputation for failing
> catastrophically and suffering permanent damage.

Yikes! Didn't know that. But it seems that my system's fan throttles well
enough; happy to say that I've never had such a failure. I *do* clean the
machine's innards from time to time, though.


Post a reply to this message

From: clipka
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 13:53:38
Message: <515c6ca2@news.povray.org>
Am 03.04.2013 19:48, schrieb Kenneth:
> clipka <ano### [at] anonymousorg> wrote:
>
>>
>> A well-designed computer should be ok even at full throttle. However,
>> not all computers are well-designed with respect to thermal management,
>> and POV-Ray has a habit of making the CPU sweat like no other application.
>
> Yep, I've certainly noticed that; my old emachines single-core AMD64
> Athlon-based computer kicks into 'high gear' on some of my scenes. (I can even
> hear a subtle high-pitched whine--not from the cooling fan AFAIK, but, I'm
> guessing, from the CPU(?).)

Yeah, that's the CPU crying for mercy ;-)


Post a reply to this message

From: Kenneth
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 14:15:00
Message: <web.515c6ef417e3da99c2d977c20@news.povray.org>
Ger <ger### [at] gmailcom> wrote:
>
> What I did however is store the differenced spheres in a text file and later
> #include that. Doing it this way you don't have the nested differences
> anymore and that has a huge impact on the render time. 7 seconds against
> 1:36 minutes.

Really?! That's quite fascinating; I've never tried such a trick. Being able to
shave off that much time is a method I need to investigate--especially for
animation!


Post a reply to this message

From: Kenneth
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 14:25:00
Message: <web.515c730817e3da99c2d977c20@news.povray.org>
Warp <war### [at] tagpovrayorg> wrote:

>
> - The CPU might not be getting enough ventilation because of too much dust
>   on its heatsink. You should check that.
>

Great advice. It's amazing how much dust accumulates in a heatsink--like
smothering the CPU in a blanket.


Post a reply to this message

From: Christian Froeschlin
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 14:26:22
Message: <515c744e$1@news.povray.org>
Kenneth wrote:

> Yep, I've certainly noticed that; my old emachines single-core AMD64
> Athlon-based computer kicks into 'high gear' on some of my scenes. (I can even
> hear a subtle high-pitched whine--not from the cooling fan AFAIK, but, I'm
> guessing, from the CPU(?).)

must be your CPU kicking up to 20 KHz turbo mode ;)


Post a reply to this message

From: Kenneth
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 15:45:00
Message: <web.515c85f917e3da99c2d977c20@news.povray.org>
Christian Froeschlin <chr### [at] chrfrde> wrote:
> Kenneth wrote:
>
> > Yep, I've certainly noticed that; my old emachines single-core AMD64
> > Athlon-based computer kicks into 'high gear' on some of my scenes. (I can even
> > hear a subtle high-pitched whine--not from the cooling fan AFAIK, but, I'm
> > guessing, from the CPU(?).)
>
> must be your CPU kicking up to 20 KHz turbo mode ;)

Ha!  No no, 30KHz  :-P


Post a reply to this message

From: Ger
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 15:58:26
Message: <515c89e2$1@news.povray.org>
Kenneth wrote:

> Ger <ger### [at] gmailcom> wrote:
>>
>> What I did however is store the differenced spheres in a text file and
>> later
>> #include that. Doing it this way you don't have the nested differences
>> anymore and that has a huge impact on the render time. 7 seconds against
>> 1:36 minutes.
> 
> Really?! That's quite fascinating; I've never tried such a trick. Being
> able to shave off that much time is a method I need to
> investigate--especially for animation!
Yes. I have tried it multiple times and the results are quite astonishing. 
You have to keep in mind however that this is done on a barebones scene. 
Just one main sphere with a bunch of differenced spheres, no AA, simple 
light_source and so on. Whatever else you put in your scene does not profit 
from this trick. Having said that, what I tend to do is when I have objects 
that are the result of time consuming calculations I try to always store 
those in external text files and later #include those files.

-- 
Ger


Post a reply to this message

From: Ger
Subject: Re: Code gets the processor too hot?
Date: 3 Apr 2013 15:59:46
Message: <515c8a32@news.povray.org>
Kenneth wrote:

> Warp <war### [at] tagpovrayorg> wrote:
> 
>>
>> - The CPU might not be getting enough ventilation because of too much
>> dust
>>   on its heatsink. You should check that.
>>
> 
> Great advice. It's amazing how much dust accumulates in a heatsink--like
> smothering the CPU in a blanket.

That's not the case here. I have seen too many PC's burn out over the years. 

-- 
Ger


Post a reply to this message

<<< Previous 10 Messages Goto Initial 10 Messages

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