![](/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) |
That is a pretty decent system.
How much did it set you back?
ian
"H. Karsten" <h-karsten()web.de> wrote in message
news:web.477b84569d2ee969ead2f34a0@news.povray.org...
> Hi, we use PovRay for a film-documentary (underwater scenes).
>
> The implementation of the hyper-treading in povray would speed up our
> rendering
> about 5GHz.
> We using this machine with 8 cores, 16 with hyper-treading:
> http://supermicro.com/products/system/4U/8044/SYS-8044T-8R.cfm
> Are there any planes to build in hyper-treading into the render process in
> PovRay?
>
> Ps.: My job is to visualise the diving from whales, using so called
> "d-tag"
> data. (These are real data recorded from real wales)
> To make a complete automated process from reading the data to rendering
> the
> image, I've created a way to use PovRay with "bones". Means that PovRay it
> self
> edit the geometric data (vertices and normals) of the whale while
> rendering. If
> there is any interesting in it, I would like to share the trick, and the
> program I've wrote for.
>
>
>
>
>
>
>
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) |
Hi Warp.
I've made a benchmark test with PovRay Beta (windows version ,"Run Benchmark,
All CPUs" under render menu)
average 1590.86 PPS with HT over 262144 Pixels
average 2336.98 PPS without HT over 262144 Pixels
Is it possible, that the maximum CPU-count in PovRay is 8?
Even if PovRay only take 8 Cores instead of 16 (with HT) the average PPS would
not reach its count while HT is turned of.
Any idea, what this could be?
HK
Warp <war### [at] tag povray org> wrote:
> H. Karsten <h-karsten()web.de> wrote:
> > Are there any planes to build in hyper-treading into the render process in
> > PovRay?
>
> No. The closest you get is the multithreading support in POV-Ray 3.7.
>
> According to my own experiments hyperthreading can speed up slightly
> a multithreading application like pov3.7.
>
> --
> - 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) |
Hi ian,
since one year I have this machine. Don't know, what it costs to day, I've paid
about 9800 Euro for it (Each processor costs about 1000 euros, and the board
with the Chassis was 3000)
The machine is fast, but loud!! I have about 10 meters cable on it for the
monitor and access it by using synergy from my laptop. There is also a little
"IPMI" device in it which lets me access the bios and powercontrollers, using
TCP.
Very funny!
"[GDS|Entropy]" <gds### [at] hotmail com> wrote:
> That is a pretty decent system.
> How much did it set you back?
>
> ian
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) |
H. Karsten nous apporta ses lumieres en ce 2008/01/07 08:27:
> Hi ian,
>
> since one year I have this machine. Don't know, what it costs to day, I've paid
> about 9800 Euro for it (Each processor costs about 1000 euros, and the board
> with the Chassis was 3000)
>
> The machine is fast, but loud!! I have about 10 meters cable on it for the
> monitor and access it by using synergy from my laptop. There is also a little
> "IPMI" device in it which lets me access the bios and powercontrollers, using
> TCP.
> Very funny!
>
> "[GDS|Entropy]" <gds### [at] hotmail com> wrote:
>> That is a pretty decent system.
>> How much did it set you back?
>>
>> ian
>
>
>
Ever tought of water cooling? No or much less noisy fans.
--
Alain
-------------------------------------------------
Please hassle me, I thrive on stress.
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) |
Alain <ele### [at] netscape net> wrote:
> >
> Ever tought of water cooling? No or much less noisy fans.
>
> --
> Alain
> -------------------------------------------------
> Please hassle me, I thrive on stress.
What? Wake up before you post! Maybe some root tea. Hassle Hassle
:)
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) |
alphaQuad nous apporta ses lumieres en ce 2008/01/08 11:38:
> Alain <ele### [at] netscape net> wrote:
>
>> Ever tought of water cooling? No or much less noisy fans.
>>
>> --
>> Alain
>> -------------------------------------------------
>> Please hassle me, I thrive on stress.
>
> What? Wake up before you post! Maybe some root tea. Hassle Hassle
> :)
>
>
It's about the quote or me reply?
--
Alain
-------------------------------------------------
You know you've been raytracing too long when your personal correspondence to
friends starts out with #Dear Linda =
Ken Tyler
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) |
> Ever tought of water cooling? No or much less noisy fans.
Yes! But not for long: The fans are cooling lots of additional parts and I have
lots of trouble with my last water cooling system. If I would have such problems
in this machine, that would be catastrophic!
But I'm experimenting with oil cooling by using silicon oil on a test computer.
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) |
H. Karsten wrote:
> I've made a benchmark test with PovRay Beta (windows version ,"Run Benchmark,
> All CPUs" under render menu)
>
> average 1590.86 PPS with HT over 262144 Pixels
> average 2336.98 PPS without HT over 262144 Pixels
Have you actually timed this? Rather than depending on the PPS count?
My experience shows that HT can add as much as 20% to rendering speed.
-- Chris
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) |
9 800 Euros = 14 414.82 U.S. dollars
Holy crap. No wonder though, and a relatively good value as well...
I have looked into a system with specs similar to:
4 quad core xeons,
RAID 10k rpm SAS drives,
dual SLI nvidia geforce 768mb 8800s,
32gb ddr3 RAM
But the closest thing I could get to that (but obviously did not) was from
Dell (boo..hiss) and was over $25K.
I'll just wait until my AmEx Black Card gets here and buy a Cray XT-4 with
320 processors.....
Which will happen just about as soon as NEVER!! ;-p
Damn I do wish though...minimum mandatory $250k per year spending...
Ian
"H. Karsten" <h-karsten()web.de> wrote in message
news:web.478228dcf997677801ffd4b0@news.povray.org...
> Hi ian,
>
> since one year I have this machine. Don't know, what it costs to day, I've
> paid
> about 9800 Euro for it (Each processor costs about 1000 euros, and the
> board
> with the Chassis was 3000)
>
> The machine is fast, but loud!! I have about 10 meters cable on it for the
> monitor and access it by using synergy from my laptop. There is also a
> little
> "IPMI" device in it which lets me access the bios and powercontrollers,
> using
> TCP.
> Very funny!
>
> "[GDS|Entropy]" <gds### [at] hotmail com> wrote:
>> That is a pretty decent system.
>> How much did it set you back?
>>
>> ian
>
>
>
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) |
> Have you actually timed this? Rather than depending on the PPS count?
>
> My experience shows that HT can add as much as 20% to rendering speed.
>
> -- Chris
yes, it was the PPS-count. I have made a lot of experience by using hyper
threading and it was all the same: If the software does not support HT in the
code, its always slower.
For example: When I'm using the Indigo renderer with HT, the result is more
grainy (same render time as without). And Indigo does also not support HT.
Sorry for the long delay
Holger
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |