 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Just rendered a complex scene under 3.7 beta 25 64-bit
under Vista Ultimate 64-bit under Boot Camp on a Mac Pro, 16 GB, 8-core 3.0
machine.
Render Time: 6' 36"
the same scene on Win XP SP2, Dell 2.2 GHz, 2 GB RAM:
3 hours 6 minutes!
best,
d
--
dhm### [at] comcast net
http://www.dennismiller.neu.edu
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Dennis Miller <dhm### [at] comcastnet> wrote:
> Just rendered a complex scene under 3.7 beta 25 64-bit
> under Vista Ultimate 64-bit under Boot Camp on a Mac Pro, 16 GB, 8-core 3.0
> machine.
> Render Time: 6' 36"
> the same scene on Win XP SP2, Dell 2.2 GHz, 2 GB RAM:
> 3 hours 6 minutes!
If my calculations are correct, that means that each core of the first
machine is over 3.5 faster than the processor in the Dell.
Are you certain all the settings were identical? Maybe those cores *are*
that much faster, but...
If everything is equal in both cases (including the resulting image),
then rather cool.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Don't underestimate the power of Warp...
>> 3 hours 6 minutes!
>
> If my calculations are correct, that means that each core of the first
> machine is over 3.5 faster than the processor in the Dell.
>
Maybe this Dell just ran out of RAM...?
Slawek
--
________
_/ __/ __/ Definitely. -- Dujour
\__ \__ \_______________________________________________________________
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp nous apporta ses lumieres en ce 2008/03/11 13:06:
> Dennis Miller <dhm### [at] comcastnet> wrote:
>> Just rendered a complex scene under 3.7 beta 25 64-bit
>> under Vista Ultimate 64-bit under Boot Camp on a Mac Pro, 16 GB, 8-core 3.0
>> machine.
>> Render Time: 6' 36"
>
>> the same scene on Win XP SP2, Dell 2.2 GHz, 2 GB RAM:
>> 3 hours 6 minutes!
>
> If my calculations are correct, that means that each core of the first
> machine is over 3.5 faster than the processor in the Dell.
>
> Are you certain all the settings were identical? Maybe those cores *are*
> that much faster, but...
>
> If everything is equal in both cases (including the resulting image),
> then rather cool.
>
If the slower machine ran out of RAM and started using the swap, then,that would
explain part of the difference.
A super machine with 64 cores available at 3 to 4 GHz, but only 1 GB of RAM can
be slower than a single core at 1 GHz with 16 GB...
--
Alain
-------------------------------------------------
Health is merely the slowest possible rate at which one can die.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Alain <ele### [at] netscapenet> wrote:
> If the slower machine ran out of RAM and started using the swap, then,that would
> explain part of the difference.
If that's the case, the comparison is completely moot. It's not comparing
CPU processing power side-to-side.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>Are you certain all the settings were identical?
Actually used the same ini file (1280 x 1024, AA 0.3)..
But I couldn't say if the Dell ran out of RAM... The scene is three
interlocked Julia fractals with some processor-intensive finishes
(reflection/refraction, et al).
Best,
d.
"Warp" <war### [at] tagpovrayorg> wrote in message
news:47d6bc13@news.povray.org...
> Dennis Miller <dhm### [at] comcastnet> wrote:
>> Just rendered a complex scene under 3.7 beta 25 64-bit
>> under Vista Ultimate 64-bit under Boot Camp on a Mac Pro, 16 GB, 8-core
>> 3.0
>> machine.
>> Render Time: 6' 36"
>
>> the same scene on Win XP SP2, Dell 2.2 GHz, 2 GB RAM:
>> 3 hours 6 minutes!
>
> If my calculations are correct, that means that each core of the first
> machine is over 3.5 faster than the processor in the Dell.
>
> Are you certain all the settings were identical? Maybe those cores *are*
> that much faster, but...
>
> If everything is equal in both cases (including the resulting image),
> then rather cool.
>
> --
> - Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp <war### [at] tagpovrayorg> wrote:
> Alain <ele### [at] netscapenet> wrote:
> > If the slower machine ran out of RAM and started using the swap, then,that would
> > explain part of the difference.
>
> If that's the case, the comparison is completely moot. It's not comparing
> CPU processing power side-to-side.
>
> --
> - Warp
For what it's worth, assuming no core-communications overhead (never a good idea
with multicore systems, but it makes things easier to figure) the 6min36sec on
the first system is roughly equivalent to 1h12min on the second system.
So yeah, it hit the swap pretty badly there (offhand, I'd guess the amount of
memory needed before it didn't hit swap at ~2.5GB, but that's just a
pulling-numbers-out-of-my-posterior guess, not a hard figure).
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
