![](/i/fill.gif) |
![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Kenneth wrote:
>
> Here's one, although it may not be the one iceqb refers to.
>
> http://www.neoseeker.com/Articles/Hardware/Reviews/amd64_3400/8.html
>
> I very recently bought an AMD 64 machine (an emachines T6412), so I'm also
> curious about this "optimization" question.
To make it short that article is really crap:
- It starts with not mentioning the POV-Ray version used.
- The description of raytracing is completely wrong.
- It does not mention what scene and render parameters are used (the
text indicates it is the builtin benchmark but it does not really say that)
- the argumentation on optimization is really just "reading tea leaves",
you should not try to conclude anything from this.
You can find some well executed benchmarks on:
http://pov4grasp.free.fr/articles/fastpov1/
This includes an Opteron 246 as the only 64bit machine but the results
are quite representative for other AMD64 processors. It also uses the
currently available official POV-Ray version. What's missing there of
course are the 64bit Intel procs.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.imagico.de/ (Last updated 31 Oct. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
"Kenneth" <kdw### [at] earthlink net> wrote:
> Here's one, although it may not be the one iceqb refers to.
>
> http://www.neoseeker.com/Articles/Hardware/Reviews/amd64_3400/8.html
>
> I very recently bought an AMD 64 machine (an emachines T6412), so I'm also
> curious about this "optimization" question.
>
> From all I've read so far, an AMD-based machine is faster at POV rendering
> than an Intel P4 equivalent. Good news for me!
>
> Ken
I know where I found it now: PC-Magazine (May-issue)! They used two
reference systems, only different by motherboard (one Intel, one VIA, if I
can remember correctly). They used POV-Ray 3.7 beta. I don't think they
used the official benchmark, because here are the results (a bit
approximated, because I seem to have misplaced it)
P4 D about 1700 seconds
AMD 64 about 1500 seconds
There were some dual-core tests too! If someone can verify these results,
please do, because I'm absolutely not sure.
PS: sorry for late respons, off for the weekend.
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
On Sun, 20 Nov 2005 16:42:10 EST, "iceqb" <nomail@nomail> wrote:
> I know where I found it now: PC-Magazine (May-issue)!
> ...
> They used POV-Ray 3.7 beta.
May? They had to write in April or earlier then. Were they using first beta
which IIRC was far from typical POV-Ray rendering (many features not
implemented, experimental algorithms, experimental optimisatons, etc.)?
ABX
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
ABX <abx### [at] abx art pl> wrote:
> May? They had to write in April or earlier then. Were they using first beta
> which IIRC was far from typical POV-Ray rendering (many features not
> implemented, experimental algorithms, experimental optimisatons, etc.)?
They don't mention anything more than v3.7 beta (nothing further). I don't
even know why they used POV at that moment, it just amazed me to see it.
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Christoph Hormann <chr### [at] gmx de> wrote:
> - It starts with not mentioning the POV-Ray version used.
They could have well used the exact same official 32-bit binary (which is
more or less intel-optimized) in both systems. Would not be a rather fair
comparison...
--
- Warp
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
iceqb <web.437b73f68869d1ae28323aad0@news.povray.org> Wednesday 16 of
November 2005 19:01
As a side not, last time I built Pov for 32 bit system, but using k8
instructions and optimization, in GCC 3.4.x or 4.0.x was it, there was
noticible speed improvment, but the resulting images didnt match normal k7
mode... hmm I have to repeat that test more carefully one day.
--
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
> Interesting, do you recall where you read that?
Here are the exact results:
Testsystems Athlon 64-system Pentium 4-system
Motherboard/Chipset Asus A8V-E Deluxe, VIA K8T890 Intel D925XECV2
Memory 2x 512 Mb Corsair DDR400 op 2x 512 Mb Corsair
twin2X 5300C4Pro
CAS 2-2-2-5 DDRII/533 op CAS
3-3-3-8
Graphicscard nVidia GeForce 6800 GT PCIe idem
Hard Disk Seagate Baracuda 160 GB idem (with NCQ)
7200 tpm, SATA
Optical drive Atapi-DVD-rom idem
Sound Creative Sound Blaster Audigy idem
OS Windows XP Professional 64-bit Edition (build 1289)
64 bit results
POV-Ray, 64-bits POV-Ray,
32-bits
Pentium 4 660 (3,6 GHz) 1774 s 1432 s
Ahlon 64 4000+ 1274 s 1607 s
results
POV-Ray Rendertest
Pentium 4 EE (3,73 GHz) 1380 s
Pentium 4 660 (3,6 GHz) 1430 s
Pentium 4 EE (3,46 GHz) 1390 s
Pentium 4 570 (3,8 GHz) 1342 s
Pentium 4 560 (3,6 GHz) 1416 s
Athlon 64 3800+ 1603 s
Athlon 64 4000+ 1607 s
Athlon 64 FX-55 1481 s
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |