Julia Set Example/Benchmark

Index > Windows > Julia Set Example/Benchmark

Goto page 1, 2, 3, 4 Next

Author

Thread

kempis

Joined: 12 Jun 2008
Posts: 49

kempis 16 Dec 2008, 14:25

Hello

I have written this program
a Set Julia animation
maybe it's interesting for benchmark or just for fun

This is new version of JuliaSSE

Description:	JuliaSSE-1 (modified)	Download
Filename:	JuliaSSE-1.zip
Filesize:	7.05 KB
Downloaded:	317 Time(s)

Last edited by kempis on 26 Dec 2008, 23:48; edited 1 time in total

16 Dec 2008, 14:25

MHajduk

Joined: 30 Mar 2006
Posts: 6115
Location: Poland

MHajduk 16 Dec 2008, 14:49

Nice program. Smile

It could be also used as a cool screensaver. Wink

I have only one suggestion: maybe it would be good to add some hotkey for program termination?

16 Dec 2008, 14:49

shoorick

Joined: 25 Feb 2005
Posts: 1614
Location: Ukraine

shoorick 16 Dec 2008, 15:10

16 Dec 2008, 15:10

kempis

Joined: 12 Jun 2008
Posts: 49

kempis 16 Dec 2008, 15:12

MHajduk wrote:

Nice program. It could be also used as a cool screensaver.

And your computer will explode because of the heat. Wink

MHajduk wrote:

I have only one suggestion: maybe it would be good to add some hotkey for program termination?

Thank's for the suggestion.
But it's actually for benchmark.

Maybe I will write the screensaver version with low CPU workload an keypressed-termination. Very Happy

16 Dec 2008, 15:12

DJ Mauretto

Joined: 14 Mar 2007
Posts: 464
Location: Rome,Italy

DJ Mauretto 16 Dec 2008, 15:22

I had to restart my computer to exit from your program, and then windows xp automatically deleted your program.
Rolling Eyes

_________________
Nil Volentibus Arduum Razz

16 Dec 2008, 15:22

asmfan

Joined: 11 Aug 2006
Posts: 392
Location: Russian

asmfan 16 Dec 2008, 15:45

Cute, but almost restarted too but took patience while pressing different ctr+alt+del which didn't work)) and waited till the end. Need some improvements or precaution of system unavailability for some considerable time.

16 Dec 2008, 15:45

LocoDelAssembly
Your code has a bug

Joined: 06 May 2005
Posts: 4624
Location: Argentina

LocoDelAssembly 16 Dec 2008, 16:02

I liked it very much, thank you for sharing it!

Quote:

---------------------------
Julia SSE Benchmark
---------------------------
AMD Athlon(tm) 64 Processor 3200+

Number of processors: 1

Framerate: 23.529053 fps

16 Dec 2008, 16:02

bitRAKE

Joined: 21 Jul 2003
Posts: 4073
Location: vpcmpistri

bitRAKE 17 Dec 2008, 03:02

Very entertaining for a benchmark. Very Happy

Quote:

---------------------------
Julia SSE Benchmark
---------------------------
Intel(R) Xeon(R) CPU L5410 @ 2.33GHz

Number of processors: 8

Framerate: 256.385803 fps

...unrolling the loop didn't seem to have an impact. (set unrolling to 320 actually got slower)

A fractal naïve lady friend said, "Wow, it looks like someone threw a tornado into a hall of mirrors - kind of like life." Very Happy

17 Dec 2008, 03:02

kas

Joined: 16 Jan 2008
Posts: 36
Location: UK

kas 17 Dec 2008, 07:01

Very beautiful visual, kempis...

Quote:

---------------------------
Julia SSE Benchmark
---------------------------
AMD Phenom(tm) 9500 Quad-Core Processor

Number of processors: 4

Framerate: 123.840332 fps

Kas.

17 Dec 2008, 07:01

kempis

Joined: 12 Jun 2008
Posts: 49

kempis 17 Dec 2008, 13:26

Thank's for trying it.

DJ Mauretto wrote:

I had to restart my computer to exit from your program, and then windows xp automatically deleted your program.

I am sorry for that.
Just be patient to wait 2^LOG2_TICK_MAX milliseconds.

17 Dec 2008, 13:26

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20445
Location: In your JS exploiting you and your system

revolution 17 Dec 2008, 13:31

kempis wrote:

,,, 2^LOG2_TICK_MAX ,,,

I thought 2^log2(something) = something.

So, does that mean that 2^LOG2_TICK_MAX = TICK_MAX?

And what is TICK_MAX?

17 Dec 2008, 13:31

kempis

Joined: 12 Jun 2008
Posts: 49

kempis 17 Dec 2008, 13:31

bitRAKE wrote:

...unrolling the loop didn't seem to have an impact. (set unrolling to 320 actually got slower)

So I default it to 1.

17 Dec 2008, 13:31

kempis

Joined: 12 Jun 2008
Posts: 49

kempis 17 Dec 2008, 13:39

revolution wrote:

I thought 2^log2(something) = something.

So, does that mean that 2^LOG2_TICK_MAX = TICK_MAX?

And what is TICK_MAX?

Yes, it is TICK_MAX
TICK_MAX is the maximum tick (milliseconds) that the program runs.
Just try TICK_MAX=16,15,14, etc...(don't too big).

17 Dec 2008, 13:39

kempis

Joined: 12 Jun 2008
Posts: 49

kempis 18 Dec 2008, 05:47

Quote:

---------------------------
Julia SSE Benchmark
---------------------------
Intel(R) Core(TM)2 CPU T5300 @ 1.73GHz
Number of processors: 2
Framerate: 60.241699 fps

Quote:

---------------------------
Julia SSE Benchmark
---------------------------
Intel(R) Pentium(R) 4 CPU 1.60GHz
Number of processors: 1
Framerate: 16.044617

18 Dec 2008, 05:47

Picnic

Joined: 05 May 2007
Posts: 1403
Location: Piraeus, Greece

Picnic 18 Dec 2008, 11:31

I reset my pc the first time, second time i read the posts and just wait Very Happy

Very nice kempis!

Quote:

---------------------------
Julia SSE Benchmark
---------------------------
Intel(R) Pentium(R) 4 CPU 3.00GHz
Number of processors: 2
Framerate: 40.176392

_________________
Hobby BASIC Interpreter

18 Dec 2008, 11:31

Kuemmel

Joined: 30 Jan 2006
Posts: 200
Location: Stuttgart, Germany

Kuemmel 21 Dec 2008, 21:54

Hi Kempis !

Nice ! As far as I see in the code you detect the number of logical cores of a processor by with the CPUID, or am I wrong ? And then you set up as many threads as there are cores.

Is that method reliable for any Intel/AMD whatever cores ? In that count are there also the Hyperthreading processors counted correctly ? So 4 core HT enabled processor should get 8 Threads ?

Would that work for example with a Dual - CPU System ?

I think I could add that to my Mandelbrot bench, too...

21 Dec 2008, 21:54

asmfan

Joined: 11 Aug 2006
Posts: 392
Location: Russian

asmfan 22 Dec 2008, 08:38

Quote:

As far as I see in the code you detect the number of logical cores of a processor by with the CPUID

Where do you see this? Only brand string & sse2 support there. No cores & threads per core detected.

22 Dec 2008, 08:38

Kuemmel

Joined: 30 Jan 2006
Posts: 200
Location: Stuttgart, Germany

Kuemmel 22 Dec 2008, 10:50

asmfan wrote:

Where do you see this? Only brand string & sse2 support there. No cores & threads per core detected.

Ups...I just saw that in the result the number of cores (and may be virtual cores also) is displayed and thought that is done with the CPUID, as I know only that method, didn't look that close in the code...or is it done otherwise...got to look again or Kempis can tell...

22 Dec 2008, 10:50

asmfan

Joined: 11 Aug 2006
Posts: 392
Location: Russian

asmfan 22 Dec 2008, 11:08

Intel & AMD have good specs on CPUID: cores & threads, caches and other interesting stuff.
[AMD CPUID Specification] - 25481.pdf
[Intel® 64 and IA-32 Architectures Software Developer’s Manual Volume 2A - Instruction Set Reference, A-M] - 253666.pdf

22 Dec 2008, 11:08

Kuemmel

Joined: 30 Jan 2006
Posts: 200
Location: Stuttgart, Germany

Kuemmel 22 Dec 2008, 11:20

Hm, thanks for the Links...I looked again at Kempis code but didn't understand it fully...it seems there are Threads created in the ".thread_init_loop" and the number of threads stored in "[.thread_count]"...what is also printed later as the "Number of Processors"...how does that loop work/or how does it know when not to create more threads !?

22 Dec 2008, 11:20

Goto page 1, 2, 3, 4 Next

< Last Thread | Next Thread >

Forum Rules:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum