|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Mon, 03 Mar 2003 10:07:19 +0100, ABX <abx### [at] abxartpl> wrote:
> On Fri, 28 Feb 2003 16:06:35 -0500, Christopher James Huff
><cja### [at] earthlinknet> wrote:
>> > megapov.exe +Iglowlits.pov +GAoutput.txt +W640 +H480 -A -J0.0 -FN -D
>>
>> glowlits.pov is not a benchmark scene, and is nothing like typical
>> scenes. I would suggest using benchmark.pov.
>
> I was not concerned much on choosing the right sphere because I considered it as
> begining for further tests with various conditions. The longer test with
> benchmark.pov I leaved for weekend. Now I have results:
>
> app. | compiler | GUI | P-II | PIII |
> -----------+-------------+-----+------+------+
> povray 3.5 | intel | win | 2740 | 815 |
> megapov | djgpp(gcc) | dos | 2865 | 852 |
...
did you use profiling with ICC?
it makes povray run about 15-25% faster (forgot the exact figures).
for gcc profiling does not seem to help much with povray.
I use these...
-tpp6 -O3 -xi -restrict -long_double -ipo -ipo_obj -unroll
-prof_dir /tmp/iccprof -pc80
-prof_genx and -prof_use are the profiling options,
docs tell how to use 'em.
--
Safari - y7p### [at] sneakemailcom
"Talk is cheap. Show me the code." - Linus Torvalds
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <ou566vgg0gqih2vvlsovillshbf09t7r4g@4ax.com> , ABX
<abx### [at] abxartpl> wrote:
> app. | compiler | GUI | P-II | PIII |
> -----------+-------------+-----+------+------+
> povray 3.5 | intel | win | 2740 | 815 |
> megapov | djgpp(gcc) | dos | 2865 | 852 |
> megapov | mingw(gcc) | con | 2909 | 872 |
> megapov | open watcom | con | 4094 | 1234 |
> megapov | borland 5.5 | win | 4407 | 1332 |
> megapov | dmc | con | 4907 | 1454 |
IIRC, Intel offers a trial version download of ICC, and the Linux version is
free (I think). So you might want to see if it isn't only a problem of bad
compilers...
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
ABX wrote:
>
> app. | compiler | GUI | P-II | PIII |
> -----------+-------------+-----+------+------+
> povray 3.5 | intel | win | 2740 | 815 |
> megapov | djgpp(gcc) | dos | 2865 | 852 |
> megapov | mingw(gcc) | con | 2909 | 872 |
> megapov | open watcom | con | 4094 | 1234 |
> megapov | borland 5.5 | win | 4407 | 1332 |
> megapov | dmc | con | 4907 | 1454 |
>
That list looks quite realistic now. I miss the MSVC megapov verstion
though.
Christoph
--
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 28 Feb. 2003 _____./\/^>_*_<^\/\.______
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thorsten Froehlich wrote:
> In article <ou566vgg0gqih2vvlsovillshbf09t7r4g@4ax.com> , ABX
> <abx### [at] abxartpl> wrote:
>
>
>>app. | compiler | GUI | P-II | PIII |
>>-----------+-------------+-----+------+------+
>>povray 3.5 | intel | win | 2740 | 815 |
>>megapov | djgpp(gcc) | dos | 2865 | 852 |
>>megapov | mingw(gcc) | con | 2909 | 872 |
>>megapov | open watcom | con | 4094 | 1234 |
>>megapov | borland 5.5 | win | 4407 | 1332 |
>>megapov | dmc | con | 4907 | 1454 |
>
>
> IIRC, Intel offers a trial version download of ICC, and the Linux version is
> free (I think). So you might want to see if it isn't only a problem of bad
> compilers...
http://www.intel.com/software/products/compilers/ --> All version are
with a price tag. There is the evaluation version, though, for thirty days.
-Nathan
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 3 Mar 2003 18:04:28 -0500, Safari <y7p### [at] sneakemailcom> wrote:
> did you use profiling with ICC?
No. I used official binaries. I do not think I can use profiling over ready
binaries. I need recompile them with profiling settings which changes sense of
testing official binaries.
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nathan Letwory wrote:
> Thorsten Froehlich wrote:
>> IIRC, Intel offers a trial version download of ICC, and the Linux version
>> is
>> free (I think). So you might want to see if it isn't only a problem of
>> bad compilers...
>
> http://www.intel.com/software/products/compilers/ --> All version are
> with a price tag. There is the evaluation version, though, for thirty
> days.
If you look at the evaluation version for Linux, you discover the free
unsupported version for uncommercial use:
http://www.intel.com/software/products/compilers/clin/noncom.htm
I agree however that Intel has gone a good job in hiding this gift.
Greetings
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thomas Willhalm wrote:
> Nathan Letwory wrote:
>
>
>>Thorsten Froehlich wrote:
>>
>>>IIRC, Intel offers a trial version download of ICC, and the Linux version
>>>is
>>>free (I think). So you might want to see if it isn't only a problem of
>>>bad compilers...
>>
>>http://www.intel.com/software/products/compilers/ --> All version are
>>with a price tag. There is the evaluation version, though, for thirty
>>days.
>
>
> If you look at the evaluation version for Linux, you discover the free
> unsupported version for uncommercial use:
> http://www.intel.com/software/products/compilers/clin/noncom.htm
> I agree however that Intel has gone a good job in hiding this gift.
>
> Greetings
> Thomas
Great, thanks. I thought I had browsed all the links, but apparently not :)
-Nathan
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 04 Mar 2003 09:30:02 +0100, "Thorsten Froehlich" <tho### [at] trfde>
wrote:
> IIRC, Intel offers a trial version download of ICC, and the Linux version is
> free (I think). So you might want to see if it isn't only a problem of bad
> compilers...
I think there is a payment for full Linux version but I have Intel compiler in
plans. But having two active children can't test more than one compiler per
month. I leaved those with trials for the end because evaluation times cause I
should have testing enviroment ready and tested then.
BTW, Thorsten this is off topic but that ASCII table I made in this thread
doesn't work on www. Horizontal line is visible when you quoted me at
http://news.povray.org/3e64640e%241%40news.povray.org
but disappear in my originall post
http://news.povray.org/ou566vgg0gqih2vvlsovillshbf09t7r4g%404ax.com
That's not important but can be unexpected. As well as why dot is entity.
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3e64693b$1@news.povray.org> , Nathan Letwory
<jes### [at] KINGKONG-letworynet> wrote:
> http://www.intel.com/software/products/compilers/ --> All version are
> with a price tag. There is the evaluation version, though, for thirty days.
Of course! Seems like Intel doesn't advertise it, but for POV-Ray is is
acceptable:
<http://www.intel.com/software/products/compilers/clin/noncom.htm>
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <u0t86v4o0t05ed6p0n0nqmkecn05vdu50n@4ax.com> , ABX
<abx### [at] abxartpl> wrote:
> BTW, Thorsten this is off topic but that ASCII table I made in this thread
> doesn't work on www. Horizontal line is visible when you quoted me at
> http://news.povray.org/3e64640e%241%40news.povray.org
> but disappear in my originall post
> http://news.povray.org/ou566vgg0gqih2vvlsovillshbf09t7r4g%404ax.com
> That's not important but can be unexpected. As well as why dot is entity.
Yes, that is normal. It is mistaken for an attachment separator in munpack
:-(
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |