Arauna: Real-time Ray Tracing Benchmark - Post your Results

Post your Arauna Ray tracing CPU Results


This is carried over from THIS thread here http://www.xtremesystems.org/forums/...d.php?t=242204 ..

We decided to create a new thread for results of this Little Demo / Bench.
After seeing the above Time of Light Thread on XS, The man behind the Arauna Ray tracing engine has kindly made this little Benchmark up, that runs through a 3D map rendered using Ray Tracing on your CPU. This also runs correctly on all Modern CPU's.

So a big thanks to Jacco Bikker for the Bench, and to TalonMan for bring our attn to this neat renderer.

Please note at no time is the GPU used for this bench.

Araun Tech:
http://igad.nhtv.nl/~bikker/

"Arauna is a real-time ray tracer. Its primary goal is to render realistic 3D images of virtual worlds, using geometry, materials and lights"

Features:

* Real-time ray tracing of large triangle meshes (up to 2M per 1GB of memory)
* Full HDR pipeline with post-processing for HDR glow
* Recursive reflection and refraction
* Accurate shadows from an unlimited amount of point lights
* Texturing with bilinear filtering and normal mapping
* Support for dynamic objects
* Unlimited light sources (though the number of lights affecting one pixel should be limited)
* Full Phong shading model (ambient, diffuse, specular) with attenuation
* Hand-tuned SIMD code for extremely efficient ray tracing and shading
* Loads OBJ files for meshes and TGA files for textures

http://img268.imageshack.us/img268/9062/arauna.png http://img109.imageshack.us/img109/8638/capturecopy.png


DOWNLOAD:
Latest package HERE: http://igad.nhtv.nl/~bikker/files/arauna_bench.zip

INSTRUCTIONS:

1. Download package, and extract files into Arauna Folder.

2. Run Benchmark.exe, and ENSURE YOU DONT HIT ANT KEYS during the benchmark.

3. The program will take you through the map, and record Frames renderd. the loop takes about 2minutes, then will instantly start over.

At this point, the LAST CYCLE counter will freeze, and the FRAMES counter will reset and begin counting again. THE LAST CYCLE reading IS YOUR RESULT.
For best results however, you need to let it run through a 2nd Cycle, The next reading should always be higher. I have found 2 cycles is enough, further runs should have no increase in score.

4. POST A SCREENSHOT OF YOUR RESULT, And include System Details

* Pls State exact frequency during benchmark if using Turbo mode on your CPU!!

Further notes:

* This Benchmark is Multithreaded. It's been found to have good results up to 8 cores, but varying success beyond this. Please post your findings

New findings:
* Memory Bandwidth is VERY important, as is the whole Cache-memory controller susbsystem : Aside from aiming for low latency, high frequency:
*Core 2 Owners.. Push that FSB
*Athlon/ Phenom / Core ix Owners, push that CPU-NB aka UNcore!.

makes a bit difference and is crucial at high clockspeeds.

Get benching.. and also check out the Time of light Pinball Thread at the top of the page!

======================== ~ARAUNA XS RANKING~ ===================

OVERALL RANKING:

http://img683.imageshack.us/img683/4...ge001small.png#1 ---- 9056 - justin.kerr -------------Intel Core i7 980X ------- 4525Mhz
http://img32.imageshack.us/img32/321...e001silver.png#2 ---- 8849 - gosmeyer ------------Intel Core i7 980X ------- 4509Mhz4/8-- DDR3-1600 8-8-8-24

http://img33.imageshack.us/img33/3725/image00broze.png#3 ---- 8491 - Rubycon -------------Intel Core i7 980X ------- 4368Mhz4/8--



----- #4 ---- 7640 - STEvil ----------Intel Xeon ES Hex core-------- 4000Mhz -- 6/12-

-----#5------ 7095 - rge --------------Intel Core i7 950--------- 4807Mhz -- 4/8--

-----#6----- 6731 - -=DVS=- ---------Intel Core i7 920-------- 4299Mhz --

----- #7 ---- 6497 - Rubycon -------------Intel Xeon W3520--------- 4368Mhz -- 4/8-- DDR3-1866 7-7-7-18

----- #8 ---- 6460 - gmat -------------Intel Core i7 975EE------- 4142Mhz -- 4/8-- DDR3-2000 8-8-8-24 [/SIZE]

----- #9 ---- 6412 - chickenfeed ----------Intel Core i7 920---- 4000Mhz -- 4/8-- DDR3-1600 7-7-7-20

------#10---- 6329 - dengyong -------Intel Core i7 920--------- 4200Mhz

----- #11---- 6087 - Pyr0 -------------Intel Core i7 920--------- 3800Mhz -- 4/8-- DDR3-1810 8-8-8-24

----- #12 ---- 5926 - Lightman --------AMD Phenom II X6 1090T----4280Mhz -- 6/6-- DDR3-1584 7-7-7-19

----- #13---- 5830 - DAK1640 ----------Intel Core i7 860--------- 3899Mhz -- 4/8--

----- #14 ---- 5799 - Tiltevros ----------Intel Xeon E5520-------- 2400Mhz -- 8/16-- *Low (35%) CPU utilisation

----- #15 ---- 5349 - bot@xs -------------Intel Core i7 860------- 3601Mhz -- 4/8-- DDR3-1800 7-7-7-21

----- #16 ---- 4904 - STEvil -------------Intel Xeon E5450 x 2---- 3600Mhz -- 8/8--]

----- #17 ---- 4850 - Gaul --------------Intel Core 2 Q9550------- 4450Mhz -- 4/4--

----- #18 ---- 4387 - VoodooProphetII --AMD Phenom II X4 965-- 4013Mhz -- 4/4-- DDR3-1604 7-7-7-21

----- #19 ---- 4310 - Toysoldier -------Intel Core 2 Q9550------- 3910Mhz -- 4/4--

----- #20 --- 4302 - mAJORD ---------AMD Phenom II X4 965--- 4060Mhz -- 4/4-- DDR3-1598 7-7-7-18

----- #21 --- 4265 - poke349 ---------- Intel Xeon X5482 x 2---- 3191Mhz -- 8/8-- FBDDR2-800 5-5-5-18

----- #22---- 4146 - BeepBeep2 -------AMD Phenom II X4 965--- 4050Mhz -- 4/4-- DDR2-1200 5-5-5-15

----- #23---- 4000 - Particle ----------AMD Opteron 2427 x 2---- 2385Mhz -- 12/12-- *Low (55%) CPU utilisation

----- #24 --- 3934 - Jokester_wild ----Intel Core 2 Q6600------- 3639Mhz -- 4/4-- DDR2-1092 5-5-5-18

----- #25 --- 3934 - Talonman --------Intel Core 2 Q6600------- 3736Mhz -- 4/4-- *Low (70%) CPU utilisation

----- #26 --- 3839 - mAJORD ----------AMD Athlon II X4 620---- 3705Mhz -- 4/4--

----- #27 --- 3768 - Cheffy ------------AMD Phenom II X4 940--- 3640Mhz -- 4/4-- DDR2-1110 5-5-5-15

----- #28---- 3760 - mAJORD ---------Intel Core i5 750--------- 2800Mhz -- 4/4--

----- #29 --- 3717 - freeloader --------AMD Phenom II X4 965--- 3600Mhz -- 4/4-- DDR3-1072 6-6-6-30

----- #30 --- 3326 - Mechromancer ----AMD Phenom II X4 940-- 3200Mhz -- 4/4-- DDR2-1066 5-5-5-15

----- #31 --- 3091 - WeldZilla ----------Intel Core 2 E8600------ 4569Mhz -- 2/2-- DDR2-1066 5-5-5-6

----- #32 --- 2661- poke349 -----------Intel Core i7 720QM----- 1596Mhz -- 4/8-- DDR3-1154 6-7-7-20

----- #33 ---- 2385 - kl0012 -------------Intel Core i3 530--------- 3051Mhz -- 2/4-- DDR3-1396 9-9-9-24

----- #34 --- 2290 - Bobsama ---------Intel Pentium DC E6300-- 4000Mhz -- 2/2--

----- #35 --- 1410 - DAK1640 ----------Intel Core 2 P8600------ 2395Mhz -- 2/2--



------------CLASS RANKINGS - TOP 3:------------

DUAL & TRI CORE:

http://img683.imageshack.us/img683/4...ge001small.png#1 --- 3091 - WeldZilla ----------Intel Core 2 E8600------ 4569Mhz

----- #4 ---- 2385 - kl0012 -------------Intel Core i3 530--------- 3051Mhz

----- #2 --- 2290 - Bobsama ---------Intel Pentium DC E6300-- 4000Mhz

MULTI -SOCKET:

http://img683.imageshack.us/img683/4...ge001small.png#1 ---- 5799 - Tiltevros ----------Intel Xeon E5520-------- 2400Mhz

-------#2 --- 4904- STEvil ---------Intel Xeon E5450 x 2--- 3600Mhz

------ #3 --- 4265- poke349 ------------ Intel Xeon X5482 x 2-- 3191Mhz

Thanks! :)
I just wanted to post the screen shots in this thread:

Quote:

---

Originally Posted by rge

Im on windows 7 64 bit, no problems. Ran it on my core i7 at normal 4.6GHz, ran fine. GPU is silent during run not loaded at all, CPU is about 80% load.

Ran at 4.6GHz, this one scales better than pinball. 6518 at 4.6ghz
http://img34.imageshack.us/img34/5715/4600ar.jpg

btw the pinball game is not very linear with increasing mhz.
3 ghz 14 fps
3.8 17-18 fps
4ghz 17-18 fps
4.2 18-19
4.4 18-19
4.6 still 18-19

4.4 ht off 14-15

---

Quote:

---

Originally Posted by Pyr0

just tried the new bench in win7 x64: 6032 @ 3.8GHz
http://img96.imageshack.us/img96/1402/arauna.jpg

---

Mine (Talonman's):
http://img9.imageshack.us/img9/2674/abench.jpg

rge used a i7 950 @ 4600 with 4/8.

4000, Particle, Dual AMD Opteron 2427, 2380 MHz, 12/12

http://www.pcrpg.org/pics/computer/arauna.png

Note the low CPU utilization. I know that's a known issue, but I think a star or something would make it less embarrassing.

Updated, Thanks for the error corrections!

I'm commenting any resutls with less than 70% CPU utilisation.. Can anyone with less than that pls include Task manager in screenshot during a cycle.. :)

Thanks

3092, Mechromancer, Phenom II 940@3.2Ghz/2.4NB, 4/4, Hyperthreading off.

EDIT: Upon closer ispection, it would appear on the surface that this program is bandwidth sensitive (kind of expected really). I run 8GB@DDR2-800. mAJORD is likely at 1066+ or on DDR3. He pulls 300 frames more than I do in a cycle. I will clock back up to 1066 for a run. I keep my mem at 800mhz usually for F@H 24/7 SMP stability.

EDIT 2: I'm curious to know what a Core i5 or i7 will run at ~3.2Ghz if Turbo Boost and Hyperthreading are turned off. Can somebody give it a try?

Score 3531 CPU Utilization 74%
Phenom II 965 @ 3.6ghz
Windows 7 X64

http://img140.imageshack.us/img140/1402/arauna.th.jpg

http://img99.imageshack.us/img99/486...mapimage3w.png

4302 4.06Ghz

Pls not new comments in OP.. Very bandwidth hungry!, Mechromancer is on the money, and will see an improvement at 1066.
At the same time, CPU grunt and frequency counts.. you'll need to find a good balance for this bench, quite fun..

Updated OP and removed my stock phenom score.

-Edit. Will most likely include Memory specs in Ranking given the importance.. Pls post detials from now on.

Goodnight for now.

http://i132.photobucket.com/albums/q...ker/Arauna.png

Quote:

---

Originally Posted by Mechromancer

3092, Mechromancer, Phenom II 940@3.2Ghz/2.4NB, 4/4, Hyperthreading off.

EDIT 2: I'm curious to know what a Core i5 or i7 will run at ~3.2Ghz if Turbo Boost and Hyperthreading are turned off. Can somebody give it a try?

---

If I read your specs correctly you are at 3.2ghz, 2.4NB, 4/4, and mem 1066 at 5,5,5.

My i7 is at 3.2ghz, 2.4NB, 4/4 (ht off, turbo off), but had to run my mem tad slower at 960, 6,6,6 18, cant match yours exactly since my native ram is 1600 7,7,7, and bios wont allow cas 5 to even be set. I got 4118. If I run my mem a tad faster at 1200, I got 4229, and get ~same with both intel and amd benchmarks

http://img94.imageshack.us/img94/6742/3200armem6x.jpg

@RGE: I appreciate the test brother. Some of the specs in my sig are wrong. I'll update it tonight. I run my 8GB at 800mhz since I have all four mem slots populated and do heavy Folding@Home.

About the Arauna, I'm wondering if it is using all the SIMD extensions on AMD CPUs, like SSE3.

Quote:

---

Originally Posted by Mechromancer

About the Arauna, I'm wondering if it is using all the SIMD extensions on AMD CPUs, like SSE3.

---

Or it might not need it.

Though I don't know anything about this program myself, all I can say is that not everything will benefit from SSE3.

SSE2 has all the important stuff. SSE3 - SSE4 is merely the cherry on the pie.

SSE3 adds some useful instructions for doing complex number arithmetic.
SSSE3 and SSE4.1 adds instructions for video encoding.

SSE4a adds some instructions which aid in bit-stream shifting.
SSE4.2 adds hardware CRC support and a bunch of string crunching stuff...

7095 with i7 at 4.8ghz, water with 5C ambients
http://img51.imageshack.us/img51/170/4800ar7095.jpg

Updated..

go for it mechromancer.. 1066 shoudl see a noticable boost.

Another tip for all.. due to the less than 100% CPU usage, you can generally aim for higher clocks than usual :p

-Edit.. Wow nice RGE. did you play with RAm at all there or is it all frequency scaling

just my normal 8x ram multi and 7,7,7 21 timing...(ram speed I typically run is 8 x 192bclk or 1536 so did increase to 1672 just from increasing bclk from 192 to 210). I need an unlocked multi to up ram speed incrementally, my ram cant do 10x which would be 2090. uncore was at 18x.

6731
http://img97.imageshack.us/img97/9328/6731j.jpg

4005 pts on q9550 @ 3.6 GH


http://i48.tinypic.com/anc7ia.jpg

Phenom II 965BE @ 4050 Mhz, NB @ 2700, DDR2-1200 5-5-5-15 @ 1.9v :P This RAM tops out at 1210 at this voltage, it's OCZ Low Voltage Platinum (Blade) 1.8v DDR2-1150.

Score: 4146

http://img69.imageshack.us/img69/9784/aurana4146.jpg

Why, why on EARTH can't people make OS independent software?

Yes, I am pissed, missing so many good benches.. :(

What OS calmatory?

Some lower end action

3839 - Athlon II X4 620 3.7Ghz

http://img693.imageshack.us/img693/4...pimage4.th.png

Quote:

---

Originally Posted by mAJORD

What OS calmatory?

---

Linux

Some are working on it... ;)
http://forums.nvidia.com/index.php?showtopic=151677

FYI - I am still looking for input about the dual CPU'ed mobo's performance in this thread: (I would love to hear some opinions.)
http://www.xtremesystems.org/forums/...d.php?t=242754

FYI - An outstanding post by dast: (It has some CPU numbers that you might like to see.)
http://forums.nvidia.com/index.php?s...5&#entry978385
http://img526.imageshack.us/img526/5648/charty.jpg

With my memory at 1066 I get a much higher score of 3326:
http://i145.photobucket.com/albums/r...auna_bench.jpg

5141 on P55 i-860

The difference between mAJORD's score and mine is quite interesting. :)

DDR3-1600 CAS 7-7-7 is 3.57 % faster than DDR2-1200 CAS 5-5-5 in this quite memory intensive benchmark.

Shows how fast low latency DDR2 can be still :p

I should probably try for 1333 CAS 5 or something, but I don't think the RAM can do it

Score's 2290
http://valid.canardpc.com/cache/banner/938983.png
http://img42.imageshack.us/img42/5277/arauna2290wm.jpg

Pentium E6300 @ 400*10 = 4,000 MHz
Dual-Channel 6GB DDR2 800 CL5 5-5-15 2T
Radeon HD 3850 256MB (I don't think THAT will have any real affect)

Looking at the C2Q @ 4GHz's scores, I think a bit over half of that score is fine.

Memory at 1600 8-8-8-24-1T I tried to get the screen shot a peak load.
http://img683.imageshack.us/img683/631/arauna3421.jpg

i7-860 with tight timings 7-8-7-24

5830

Dell Latitude = 1401

Dragon backround with Intel CPU? Sounds fishy. XD

Quote:

---

Originally Posted by BeepBeep2

Dragon backround with Intel CPU? Sounds fishy. XD

---

Put up some results...:)

Quote:

---

Originally Posted by dengyong

Put up some results...:)

---

I'm confused. :confused: I already posted my result a while ago...maybe I should run Aurana on my netbook. :ROTF:

Quote:

---

Originally Posted by BeepBeep2

I'm confused. :confused: I already posted my result a while ago...maybe I should run Aurana on my netbook. :ROTF:

---

No I'm confused, I didn't read the whole thread. :doh:

Quote:

---

Originally Posted by mAJORD

#4 --- 6032 - Pyr0 -------------Intel Core i7 920--------- 3800Mhz -- 4/8-- DDR3-1600 8-8-8-24

---

just a quick post to say my mem is running DDR3-1810 9-9-9-24 :)

Update.. Thanks for the Dual core results.. ;)

Corrected, pyro. Cheers

not much difference at 8-8-8-24
6087
http://i2.sqi.sh/s_1/zm/ara.jpg

http://www.numberworld.org/screenie_..._1_10_2010.jpg

Turbo was never able to engage. So it's 1.6 GHz.

I might try my workstation later on, but it's gonna be tricky because it's 2000 miles away at home.
So I'll have to do it through remote desktop or team viewer. Not sure what kinda of overhead that will come with.

Here it is... 4265 frames

As you can see from Process Explorer... The CPU usage is very poor...
This was done through TeamViewer... I have no idea how much of an impact it has on this benchmark.

(click to enlarge)
http://www.numberworld.org/screenie_...2010_small.jpg

Phenom II 3.8ghz
Memory @ 6-6-6-30-24-1T
Score 3717.

Definitely likes tighter timings. :)



http://img263.imageshack.us/img263/5045/3717.th.png

Here's an idea: why don't you tier the ranks by core count (2-core, 4-core, 6-8 core, 12+ core)? Just an idea.

btw: There's two X5482s in my machine. Not one. :)

Quote:

---

Originally Posted by Bobsama

Here's an idea: why don't you tier the ranks by core count (2-core, 4-core, 6-8 core, 12+ core)? Just an idea.

---

Thanks for the suggestion. I've though about doing something like that,

With Hyperthreading , Tri-cores, cheap multi-core CPU's, low frequency multi socket server CPU's its become a bit blured as to how to "grade" them!

I was thinking two options:

1. Seperate Dual quad, Hex etc

2. A comment / recogniton of "Top Dual core" etc next to the relevent result.

Quote:

---

Originally Posted by poke349

Here it is... 4265 frames

As you can see from Process Explorer... The CPU usage is very poor...
This was done through TeamViewer... I have no idea how much of an impact it has on this benchmark.

---

Fits with my own experience. This thing seems to hate multi-socket. It's not unlike some other realtime renderers I've seen in that regard such as Cedega's software Direct3D 9 renderer (SwiftShader). It nearly dies on 12 cores and runs MUCH better with 6.

Quote:

---

Originally Posted by mAJORD

Thanks for the suggestion. I've though about doing something like that,

With Hyperthreading , Tri-cores, cheap multi-core CPU's, low frequency multi socket server CPU's its become a bit blured as to how to "grade" them!

I was thinking two options:

1. Seperate Dual quad, Hex etc

2. A comment / recogniton of "Top Dual core" etc next to the relevent result.

---

I like this idea. I'll finally be on top of SOMEthing this way. hehe

Quote:

---

Originally Posted by Particle

Fits with my own experience. This thing seems to hate multi-socket. It's not unlike some other realtime renderers I've seen in that regard such as Cedega's software Direct3D 9 renderer (SwiftShader). It nearly dies on 12 cores and runs MUCH better with 6.

---

A LOT of things hate multi-socket... lol :rofl:
You really need to write for them to get much out of dualies...

4904 on a dual E5450 @ 3.6 with 4x2gb FB-DIMM's @ 400mhz 5-5-5-x with i5000x chipset and snoop filter disabled.

65-80% thread utilization, 8 cores 8 threads, WinXP-64.

Up date !!!

q9550 @ 4.4 GH


http://i50.tinypic.com/m8yyc1.jpg

Quote:

---

Originally Posted by STEvil

4904 on a dual E5450 @ 3.6 with 4x2gb FB-DIMM's @ 400mhz 5-5-5-x with i5000x chipset and snoop filter disabled.

65-80% thread utilization, 8 cores 8 threads, WinXP-64.

---

I wonder if limiting to one stick per channel might give better timings or lower latency... (I've got 4 sticks per channel in my workstation...)

Not like I'm gonna try it though... since I don't have physical access to it while I'm at school, nor do I want to mess my ram config anymore...

UPdated OP..

Well done Gaul!, big freq for a 9550.

Added some Classes.. and jazzed things up a bit, get a bit of competition spirit going :P

Let me know if you think it's a poor choice..
eg, No class for Quads/Hex.. they have the biggest chance of being in the top 3 at this stage..

4387 - Phenom II 965 @ 4013.9mhz on TRUE 120 air

http://img37.imageshack.us/img37/6451/araurabench2.jpg

Quote:

---

Originally Posted by mAJORD

UPdated OP..

Well done Gaul!, big freq for a 9550.

---

thanx, MEGAHALEM rulez......for COOLER :up:

stock cooling stock settings only turbo mode on :/

Interesting error msgs there in the cmd window..

Still, you've got #1 multi socket, and a damn fine score for 2.4Ghz!

Quote:

---

Originally Posted by mAJORD

Interesting error msgs there in the cmd window..

---

That error will be because the 3D window is occluded by the task manager. It happens in every game where you Alt+Tab out of it or when you run it in windowed mode and partially cover it up, only applications usually don't bother reporting it.

update !!

Q9550 @ 4.45 GH


http://i47.tinypic.com/11ceb9v.jpg

thanks gaul!, is that score 4850?

yngdrw: veryt interesting.. thanks

Seems to be a decent score compared to others so far. 3957

http://i.imgur.com/sYXuo.jpg

Quote:

---

Originally Posted by mAJORD

Interesting error msgs there in the cmd window..

Still, you've got #1 multi socket, and a damn fine score for 2.4Ghz!

---

mAJORD: The bench from what I've found runs nomatter if it errors or not, I've gotten the same error's before and even screem corruption in the window :ROTF: but it keeps running and doesn't crash. ;) I guess as long as it still outputs a score it's valid. :p:

score: 3768 , x4 940be @ 3640mhz 1.4725v nb @2500mhz, ram @ 555mhz (1110) 5-5-5-15 2t

http://img229.imageshack.us/img229/7807/benche.jpg

and since that's as near as makes no difference identical to Jokester_wild for core speed and very close of ram some comparison can be made between Q6600 and x4 940br, as if there isn't enough in the world ;)

Quote:

---

Originally Posted by Cheffy

score: 3768 , x4 940be @ 3640mhz 1.4725v nb @2500mhz, ram @ 555mhz (1110) 5-5-5-15 2t
and since that's as near as makes no difference identical to Jokester_wild for core speed and very close of ram some comparison can be made between Q6600 and x4 940br, as if there isn't enough in the world ;)

---

Ah! very close system specs. Almost par. Pretty close results as well though.

A rather lame 2250 here.

http://i590.photobucket.com/albums/s...validation.jpg

The CPU usage never went up over 69% and I don't know whether CPU-Z is reporting the clockspeeds correctly as it is saying the cores are set at 800mhz when they should be at 2.2. I'm wondering if AMD cool 'n quiet is activated or comething like that but I didn't see anything like that when last I had a look in the motherboards bios.

I'm using a pair of Istanbul Opteron 2427's.

Core 2's can hang in but just barely.....3091

http://img.photobucket.com/albums/v2...la/arauna1.jpg

I will run it on my Q6600 also

WZ

Quote:

---

Originally Posted by WeldZilla

Core 2's can hang in but just barely.....

---

Cant read any numbers on the resized pics.

uh ....3091

WZ

UPdated!

Well done Weldzilla, #1 in Dual/tri ranking. Very good score for a dualcore. As mentioned can't read figures very well so correct me if I messed up any number.s

Chris_Redfield: You seem to have issues there for sure, you should be in the ballpark of Particles score of 4000 since he's running the same CPU.

Perhaps you have a Cool n Quiet issue?. Taking freq scaling into account The score is indicative of it running at less than half rated frequency, so it indeed may be runing at 800Mhz.
I've hesitated on posting your score incase you want to sort out the issue first :p

Quote:

---

Originally Posted by poke349

I wonder if limiting to one stick per channel might give better timings or lower latency... (I've got 4 sticks per channel in my workstation...)

Not like I'm gonna try it though... since I don't have physical access to it while I'm at school, nor do I want to mess my ram config anymore...

---

It definately will. More sticks on each channel adds latency. Search "How FB-DIMM's Work" on here or 2cpu.com.. or google ;)

edit - for the record I peak at 72% CPU Utilization and average 50-65% with the Dual E5450's. Testing my T7400 now.

1267 on a T7400ES @ stock 2.16ghz in a laptop lol :)

Ok, my first entry:

http://img695.imageshack.us/img695/9...26ghzmem16.jpg

Uploaded with ImageShack.us

First Phenom X6 here as well :D

Xeon 2.4ghz ES @ 4ghz (6 cores, 12 threads)
3x2gb DDR3 @ 9-9-9

7640.... :)

975EE @ 4.14GHz (reported CPUz is slightly off with the Classified) HT on, ram is Dominator GT @ 2000MHz timings in sig. Not done yet but holding on more OCing for now since it's crunching for WCG...

Score: 5843

http://img708.imageshack.us/img708/1...b284ghzmem.png

Uploaded with ImageShack.us

Slightly better run.

EDIT:

Score:5926

http://img27.imageshack.us/img27/537...b285ghzmem.png

Uploaded with ImageShack.us

At this rate I will crack 6000 by the end of week :D

Seems lower given clock speed and timings however other things are running on this system that I could not shut off during the tests!

Are you SURE the GPU has no influence?
I remember the REALSTORM raytrace bench definitely WAS impacted by GPU!

http://i157.photobucket.com/albums/t...f?t=1272926168

Guys, all this CPU power should be used on WCG for this week. No benching for this week! :yepp: :p:
Nice results though. Looking forward to more come 8th of may ;)

980X HT ON 4.22 GHz

I believe GPU is indeed making this run faster! It's 2000 pts faster than my W3520!

http://i157.photobucket.com/albums/t...f?t=1272937236

One more for the road :D 8849

http://img708.imageshack.us/img708/172/66311833.jpg

Score : 6412

http://img163.imageshack.us/img163/6...1280077.th.jpg

wow, a train of new results hehe, I'll update shortly.

There should be absolutely no GPU usage, the bench is very sensitive to RAM Bandwidth, Cache (L3 speed), (so those uncore/CPU-NB clocks) so perhaps your other system isn't as fast there..?

rig in sig, 24/7 settings
9056

ran it with HT off same settings otherwise, and got 8431
http://i18.photobucket.com/albums/b1...s3430/jjuj.png

5349 points
Attachment 103894

http://img168.imageshack.us/img168/1356/rttest.jpg

Whoa MajorD, you mixed me up with Pyr0:

Quote:

---

----- #8 ---- 6460 - gmat -------------Intel Core i7 920--------- 3800Mhz -- 4/8-- DDR3-1810 8-8-8-24

---

should be:

Quote:

---

----- #8 ---- 6460 - gmat -------------Intel Core i7 975EE------- 4142Mhz -- 4/8-- DDR3-2000 8-8-8-24

---

and i'll try to post a new bench next week after the WCG effort is over ;)

sorry gmat, fixed up..

So many results to post this update, got a bit muddled.

If I can ask a small favor of you guys, can you please format as per OP but without all the spacer lines would be really handy if you have time..

[ Score-Nick-CPU-Mhz-Core/thread- Ram details if you want ]

much appreciated..

Couple of comments.

Rubycon.. your 2000 extra pts came from the 2 extra cores! ;)

WOW @ Justin.Kerr . so close to 10K. congrats on a close 1st place there

and finally, thanks to Lightning man for the first Thuban results.. you can do 6k! ;)

The three computers I use the most:
Phenom X3
http://img.photobucket.com/albums/v6...h-PhenomX3.png
E5200
http://img.photobucket.com/albums/v6...ench-E5200.png
Core2Quad
http://img.photobucket.com/albums/v6...-Core2Quad.png

Nice tracer, runs smooth, load @80%.

Third run improved almost 1000 over second, that improved 500 over the first round. :rofl:
Third posted.
(Improvement second on third seems to be a glitch of the memory - Got errors soon after, reseated the memory, and now 3-500 points gain first to second. More bring no real gain) :shrug:

Run7 improved to 8951 :ROTF:



Did several runs now, using same setup, different settings.
Mem-tuning seems to be as relevant in improving scores, as high clocks.

Works nice, and almost clock-equal to Linx for stability testing.
Prime95 needs 1Mhz FSB less, to run stable with the same settings.
:clap::clap: Like it, this nifty little program :clap::clap:

Swapped mobo.
RexIII :up:- sofar on par with Classy:up:.

Again, great for quick stability testing. 1 click behind on LinX , 2 clicks behind on Prime95.

heres an x6 result.. for lightman :)

click for 1900x1200
http://i70.photobucket.com/albums/i8...una4340mhz.png

i must of missed something with this bench... if its ray tracing ...why is there so much aliasing... i dont get it...

doesnt ray tracing eradicate aliasing all together....???????????????

Quote:

---

Originally Posted by purecain

i must of missed something with this bench... if its ray tracing ...why is there so much aliasing... i dont get it...

doesnt ray tracing eradicate aliasing all together....???????????????

---

No, it doesn't ;) You need to supersample RT to get decent results.

Quote:

---

Originally Posted by purecain

i must of missed something with this bench... if its ray tracing ...why is there so much aliasing... i dont get it...

doesnt ray tracing eradicate aliasing all together....???????????????

---

no. ray tracing has the same problem because it determines visible surfaces in image precision (pixels) as opposed to object precision (primitives). from a mathematical view pixels are an x,y coordinate which are infinitesimally small but represent a finite area on a monitor. if you take one sample (a ray in this case) you are representing only one component of the information at that pixel.