 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Guess who's the author of the SDL?
http://www.theinquirer.net/gb/inquirer/news/2007/11/30/first-inqpressions-intel-45nm
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
gregjohn <pte### [at] yahoo com> wrote:
> http://www.theinquirer.net/gb/inquirer/news/2007/11/30/first-inqpressions-intel-45nm
Does the povray standard benchmark really take less than 1 minute to
render in the most modern processors? That's quite fast.
In my computer it takes something a half an hour. (Although I think
I haven't tested with the latest version of the benchmark, though. I think
there are at least some changes with the media clouds?)
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>>
http://www.theinquirer.net/gb/inquirer/news/2007/11/30/first-inqpressions-intel-45nm
>
> Does the povray standard benchmark really take less than 1 minute to
> render in the most modern processors? That's quite fast.
>
> In my computer it takes something a half an hour. (Although I think
> I haven't tested with the latest version of the benchmark, though. I think
> there are at least some changes with the media clouds?)
I tested with the latest beta, and it takes almost exactly 5 minutes on my
laptop (dual-core 2 GHz). Under 1 minute for a newer 4-core 3.2 GHz machine
seems in the right ball-park.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp escribió:
> gregjohn <pte### [at] yahoocom> wrote:
>>
http://www.theinquirer.net/gb/inquirer/news/2007/11/30/first-inqpressions-intel-45nm
>
> Does the povray standard benchmark really take less than 1 minute to
> render in the most modern processors? That's quite fast.
>
> In my computer it takes something a half an hour. (Although I think
> I haven't tested with the latest version of the benchmark, though. I think
> there are at least some changes with the media clouds?)
>
From what I see on the *image*, there are apparently no clouds at all.
I know the 3.7 benchmark was modified, not the same as the 3.6
benchmark, but I don't know what were the changes specifically.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott wrote:
> Warp said:
>> Does the povray standard benchmark really take less than 1 minute to
>> render in the most modern processors? That's quite fast.
>>
>> In my computer it takes something a half an hour. (Although I think
>> I haven't tested with the latest version of the benchmark, though. I
>> think
>> there are at least some changes with the media clouds?)
>
> I tested with the latest beta, and it takes almost exactly 5 minutes on
> my laptop (dual-core 2 GHz). Under 1 minute for a newer 4-core 3.2 GHz
> machine seems in the right ball-park.
Hmm, I have a Pentium 2.66 GHz quad core here, and the standard
benchmark took 4 minutes, 20 seconds! Why would this be only slightly
faster that a 2 GHz duo-core machine?
I used version 3.7b23. Here are the messages:
----------------------------------------------------------------------------
Peak memory used: 8449408 bytes
Render Time:
Photon Time: 0 hours 0 minutes 1 seconds (1.938 seconds)
using 1 thread(s) with 1.937 CPU-seconds total
Radiosity Time: No radiosity
Trace Time: 0 hours 4 minutes 16 seconds (256.407 seconds)
using 4 thread(s) with 1022.327 CPU-seconds total
POV-Ray finished
-
CPU time used: kernel 0.27 seconds, user 1025.83 seconds, total 1026.09
seconds.
Elapsed time 260.08 seconds, CPU vs elapsed time ratio 3.95.
Render averaged 1007.94 PPS (255.48 PPS CPU time) over 262144 pixels.
----------------------------------------------------------------------------
Sam
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
stbenge wrote:
> scott wrote:
>> I tested with the latest beta, and it takes almost exactly 5 minutes
>> on my laptop (dual-core 2 GHz). Under 1 minute for a newer 4-core 3.2
>> GHz machine seems in the right ball-park.
>
> Hmm, I have a Pentium 2.66 GHz quad core here, and the standard
> benchmark took 4 minutes, 20 seconds! Why would this be only slightly
> faster that a 2 GHz duo-core machine?
Okay, I just tested it with 3.7-SSE2, and the render time was 3m 06s.
The messages:
----------------------------------------------------------------------------
Peak memory used: 9231564 bytes
Render Time:
Photon Time: 0 hours 0 minutes 1 seconds (1.203 seconds)
using 1 thread(s) with 1.203 CPU-seconds total
Radiosity Time: No radiosity
Trace Time: 0 hours 3 minutes 3 seconds (183.171 seconds)
using 4 thread(s) with 729.858 CPU-seconds total
POV-Ray finished
-
CPU time used: kernel 0.33 seconds, user 732.47 seconds, total 732.80
seconds.
Elapsed time 186.05 seconds, CPU vs elapsed time ratio 3.94.
Render averaged 1409.02 PPS (357.73 PPS CPU time) over 262144 pixels.
----------------------------------------------------------------------------
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp <war### [at] tagpovrayorg> wrote:
> In my computer it takes something a half an hour.
It seems that pov3.7 has currently a benchmark scene which is different
from the one in pov3.6, and which renders much faster (9 minutes in my
computer).
Given that this is the wrong direction for a benchmark, I have been
thinking about the idea of creating a completely new benchmark scene
for pov3.7 which would again keep rendering times comparable at a relatively
high resolution.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>>> I tested with the latest beta, and it takes almost exactly 5 minutes on
>>> my laptop (dual-core 2 GHz). Under 1 minute for a newer 4-core 3.2 GHz
>>> machine seems in the right ball-park.
>>
>> Hmm, I have a Pentium 2.66 GHz quad core here, and the standard benchmark
>> took 4 minutes, 20 seconds! Why would this be only slightly faster that a
>> 2 GHz duo-core machine?
>
> Okay, I just tested it with 3.7-SSE2, and the render time was 3m 06s.
Yeh, I used the SSE2 version, but I would have expected 4x 2.66 GHz to be
more than twice the speed of 2x 2.00 GHz...
It's an Intel Core2Duo T7200 btw.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott nous apporta ses lumieres en ce 2007/12/05 02:58:
>>>> I tested with the latest beta, and it takes almost exactly 5 minutes
>>>> on my laptop (dual-core 2 GHz). Under 1 minute for a newer 4-core
>>>> 3.2 GHz machine seems in the right ball-park.
>>>
>>> Hmm, I have a Pentium 2.66 GHz quad core here, and the standard
>>> benchmark took 4 minutes, 20 seconds! Why would this be only slightly
>>> faster that a 2 GHz duo-core machine?
>>
>> Okay, I just tested it with 3.7-SSE2, and the render time was 3m 06s.
>
> Yeh, I used the SSE2 version, but I would have expected 4x 2.66 GHz to
> be more than twice the speed of 2x 2.00 GHz...
>
> It's an Intel Core2Duo T7200 btw.
>
>
You have a bottle neck: Memory access. You also have some added overhead of
syncronizing the 4 cores.
--
Alain
-------------------------------------------------
You know you've been raytracing too long when you're starting to find these
quotes more unsettling than funny.
-- Alex McLeod a.k.a. Giant Robot Messiah
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
