Topic: JRMark -- JRiver's Benchmark Tool (Read 219785 times)

glynor · « **Reply #50 on:** October 15, 2009, 01:09:29 pm »

Quote from: Listener on October 15, 2009, 10:50:10 am

If you are using MC on an Atom based PC, post your benchmark results and comment on how MC works on that PC.

Ask and ye shall receive...

ASUS EeePC 1000H (Intel Atom N270 @ 1.6GHz, 2GB RAM):

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 22.997 seconds
   Single-threaded floating point math... 13.751 seconds
   Multi-threaded integer math... 25.094 seconds
   Multi-threaded mixed math... 14.782 seconds
Score: 248

Running 'Image' benchmark...
   Image creation / destruction... 7.259 seconds
   Flood filling... 6.386 seconds
   Direct copying... 5.024 seconds
   Small renders... 16.392 seconds
   Bilinear rendering... 17.601 seconds
   Bicubic rendering... 17.377 seconds
Score: 314

Running 'Database' benchmark...
   Create database... 4.734 seconds
   Populate database... 15.007 seconds
   Save database... 0.805 seconds
   Reload database... 0.330 seconds
   Search database... 17.370 seconds
   Sort database... 17.020 seconds
   Group database... 17.239 seconds
Score: 297

JRMark (version 14.0.83): 286

MC runs pretty well on it. Thumbnailing can be very slow and can really bog the system down. With CoreAVC installed it'll play pretty much anything I throw at it (including my 720p MKVs). Without CoreAVC, decoding H264 compressed video is a bit tough, and anything in HD resolutions is impossible. XviD compressed video and audio work fine. MC itself is amazingly speedy, even with my full network server's library loaded. Theater View works fine (I do turn the quality setting down all the way), and the Standard View UI is still responsive. Searches and heavily filtered views (that need to do lots of searching) are much slower than on my godlike systems at home, but still quite usable.

EDIT: Added additional details later because typing on the 1000H is annoying (due to the right shift key placement).

EDIT Again: I feel bad... Even my NETBOOK beat poor Rick's score...

glynor · « **Reply #51 on:** October 15, 2009, 01:20:40 pm »

Quote from: glynor on October 15, 2009, 01:09:29 pm

ASUS EeePC 1000H (Intel Atom N270 @ 1.6GHz, 2GB RAM):

Running 'Math' benchmark...
Single-threaded integer math... 22.997 seconds
Single-threaded floating point math... 13.751 seconds
Multi-threaded integer math... 25.094 seconds
Multi-threaded mixed math... 14.782 seconds
Score: 248

Apparently the benchmark is quite sensitive (at least in the multi-threaded metric) to hyperthreading. The N270 is a single-core CPU with hyper-threading enabled. Since it is single-core, you'd expect to see much higher multi-threaded times than you do here, based on what Matt described earlier. It looks like hyperthreading actually saves the Atom a bit in this test and allows it to behave like an "almost dual-core".

glynor · « **Reply #52 on:** October 15, 2009, 02:13:00 pm »

I figured why not... Here's my work PC (from my desk at my office). It is...

CPU: Intel Pentium D @ 3.2GHz (Presler, dual-core northburst)
RAM: 2GB DDR2 @ 333.3MHz (5-5-5-15)
Motherboard: Intel D975XBX
GPU: AMD X1900XT 512MB

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 13.901 seconds
   Single-threaded floating point math... 8.097 seconds
   Multi-threaded integer math... 12.821 seconds
   Multi-threaded mixed math... 7.521 seconds
Score: 449

Running 'Image' benchmark...
   Image creation / destruction... 4.322 seconds
   Flood filling... 3.830 seconds
   Direct copying... 3.246 seconds
   Small renders... 7.721 seconds
   Bilinear rendering... 6.396 seconds
   Bicubic rendering... 5.820 seconds
Score: 702

Running 'Database' benchmark...
   Create database... 2.090 seconds
   Populate database... 7.686 seconds
   Save database... 0.355 seconds
   Reload database... 0.236 seconds
   Search database... 6.469 seconds
   Sort database... 6.476 seconds
   Group database... 5.783 seconds
Score: 739

JRMark (version 14.0.83): 630

Listener · « **Reply #53 on:** October 15, 2009, 02:56:33 pm »

Thanks, glynor for the Atom info.

Benchmark results for a 2.8 GHz P4 laptop with 512 MB ram and AGP graphics.

Running 'Math' benchmark...
Single-threaded integer math... 13.613 seconds
Single-threaded floating point math... 6.203 seconds
Multi-threaded integer math... 28.393 seconds
Multi-threaded mixed math... 17.990 seconds
Score: 287

Running 'Image' benchmark...
Image creation / destruction... 4.676 seconds
Flood filling... 20.076 seconds
Direct copying... 19.509 seconds
Small renders... 14.352 seconds
Bilinear rendering... 20.572 seconds
Bicubic rendering... 25.963 seconds
Score: 209

Running 'Database' benchmark...
Create database... 3.237 seconds
Populate database... 11.952 seconds
Save database... 3.062 seconds
Reload database... 0.732 seconds
Search database... 13.073 seconds
Sort database... 9.186 seconds
Group database... 14.927 seconds
Score: 383

JRMark (version 14.0.83): 293

JimH · « **Reply #54 on:** October 17, 2009, 09:46:49 am »

Here's my living room PC. It's an i7 at 2.67 Ghz.

The first time I tested, I got around 1200, which was slower than my Dual Core machine. I found that I had it on a power saving setting.

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 4.742 seconds
Single-threaded floating point math... 2.675 seconds
Multi-threaded integer math... 1.231 seconds
Multi-threaded mixed math... 0.909 seconds
Score: 1988

Running 'Image' benchmark...
Image creation / destruction... 0.973 seconds
Flood filling... 0.668 seconds
Direct copying... 0.767 seconds
Small renders... 2.963 seconds
Bilinear rendering... 1.415 seconds
Bicubic rendering... 1.390 seconds
Score: 2691

Running 'Database' benchmark...
Create database... 0.603 seconds
Populate database... 2.165 seconds
Save database... 0.185 seconds
Reload database... 0.084 seconds
Search database... 2.837 seconds
Sort database... 2.964 seconds
Group database... 2.540 seconds
Score: 1890

JRMark (version 14.0.84): 2190

JimH · « **Reply #55 on:** October 17, 2009, 10:09:16 am »

I re-ran my office machine, which scored 1684 originally. I set the power to high performance in Windows.

Running 'Math' benchmark...
Single-threaded integer math... 4.449 seconds
Single-threaded floating point math... 2.634 seconds
Multi-threaded integer math... 2.316 seconds
Multi-threaded mixed math... 1.415 seconds
Score: 1757

Running 'Image' benchmark...
Image creation / destruction... 1.559 seconds
Flood filling... 1.637 seconds
Direct copying... 2.054 seconds
Small renders... 3.206 seconds
Bilinear rendering... 1.547 seconds
Bicubic rendering... 1.559 seconds
Score: 1903

Running 'Database' benchmark...
Create database... 0.685 seconds
Populate database... 2.633 seconds
Save database... 0.200 seconds
Reload database... 0.090 seconds
Search database... 3.403 seconds
Sort database... 3.286 seconds
Group database... 2.939 seconds
Score: 1624

JRMark (version 14.0.84): 1761

neFAST · « **Reply #56 on:** October 17, 2009, 11:02:38 am »

This is what I get on my one and only computer (an old but faithful P4) :

Running 'Math' benchmark...
   Single-threaded integer math... 24,483 seconds
   Single-threaded floating point math... 13,630 seconds
   Multi-threaded integer math... 15,581 seconds
   Multi-threaded mixed math... 9,200 seconds
Score: 302

Running 'Image' benchmark...
   Image creation / destruction... 16,132 seconds
   Flood filling... 11,738 seconds
   Direct copying... 9,402 seconds
   Small renders... 18,842 seconds
   Bilinear rendering... 12,960 seconds
   Bicubic rendering... 16,419 seconds
Score: 257

Running 'Database' benchmark...
   Create database... 4,900 seconds
   Populate database... 13,331 seconds
   Save database... 1,484 seconds
   Reload database... 1,150 seconds
   Search database... 20,512 seconds
   Sort database... 12,090 seconds
   Group database... 17,079 seconds
Score: 305

JRMark (version 14.0.84): 288

So should I trash my comp?

JimH · « **Reply #57 on:** October 17, 2009, 11:09:55 am »

It would be interesting to see some comparisons of before and after disabling a virus checker.

neFAST · « **Reply #58 on:** October 17, 2009, 11:21:33 am »

It's even worse (except for Math) on my second try with nod32 disabled!
=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 17,146 seconds
   Single-threaded floating point math... 8,446 seconds
   Multi-threaded integer math... 15,666 seconds
   Multi-threaded mixed math... 8,617 seconds
Score: 381

Running 'Image' benchmark...
   Image creation / destruction... 16,880 seconds
   Flood filling... 12,514 seconds
   Direct copying... 9,329 seconds
   Small renders... 24,526 seconds
   Bilinear rendering... 12,832 seconds
   Bicubic rendering... 16,423 seconds
Score: 238

Running 'Database' benchmark...
   Create database... 5,705 seconds
   Populate database... 20,071 seconds
   Save database... 1,404 seconds
   Reload database... 0,534 seconds
   Search database... 16,776 seconds
   Sort database... 18,465 seconds
   Group database... 20,690 seconds
Score: 257

JRMark (version 14.0.84): 292

rick.ca · « **Reply #59 on:** October 17, 2009, 11:25:21 am »

Quote from: JimH on October 14, 2009, 05:13:40 pm

STANDINGS:
Best: glynor JRMark (version 14.0.83): 2867
Worst: rick.ca JRMark (version 14.0.83): 264

I tested again with fewer things running and scored 291. I wouldn't want anyone to think my record is completely out of reach. Now I await glynor's advice on overclocking and liquid nitrogen cooling systems.

KingSparta · « **Reply #60 on:** October 17, 2009, 03:42:58 pm »

so whats better a low number or a higher number?

I am thinking higher is better?

JimH · « **Reply #61 on:** October 17, 2009, 03:57:25 pm »

Higher is better. It's a linear scale, so a machine with a score of 2000 would be twice as fast as a machine with a score of 1000.

tonytoronto · « **Reply #62 on:** October 17, 2009, 04:12:27 pm »

I guess ain't to bad

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 4.198 seconds
Single-threaded floating point math... 2.372 seconds
Multi-threaded integer math... 2.392 seconds
Multi-threaded mixed math... 1.359 seconds
Score: 1841

Running 'Image' benchmark...
Image creation / destruction... 0.980 seconds
Flood filling... 0.676 seconds
Direct copying... 0.960 seconds
Small renders... 2.648 seconds
Bilinear rendering... 1.414 seconds
Bicubic rendering... 1.382 seconds
Score: 2730

Running 'Database' benchmark...
Create database... 0.539 seconds
Populate database... 1.880 seconds
Save database... 0.152 seconds
Reload database... 0.063 seconds
Search database... 2.442 seconds
Sort database... 2.623 seconds
Group database... 2.310 seconds
Score: 2148

JRMark (version 14.0.84): 2240

KingSparta · « **Reply #63 on:** October 17, 2009, 07:19:39 pm »

This Is A Cheep Dell Vostro 220 (About 6 Months Old)

Microsoft Windows Vista 6.0 Service Pack 2 (Build 6002)
Intel Pentium III 2610 MHz MMX / Memory: Total - 2060 MB, Free - 679 MB

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 4.508 seconds
Single-threaded floating point math... 2.663 seconds
Multi-threaded integer math... 4.483 seconds
Multi-threaded mixed math... 2.674 seconds
Score: 1326

Running 'Image' benchmark...
Image creation / destruction... 2.081 seconds
Flood filling... 2.404 seconds
Direct copying... 2.508 seconds
Small renders... 3.267 seconds
Bilinear rendering... 3.157 seconds
Bicubic rendering... 3.087 seconds
Score: 1333

Running 'Database' benchmark...
Create database... 0.718 seconds
Populate database... 2.742 seconds
Save database... 0.224 seconds
Reload database... 0.149 seconds
Search database... 3.478 seconds
Sort database... 3.357 seconds
Group database... 2.983 seconds
Score: 1575

JRMark (version 14.0.84): 1411

morrison · « **Reply #64 on:** October 18, 2009, 04:33:52 pm »

Sorry for the long post (sometimes missing function to "cut" on the forum), but it's possible will be interesting for someone. I did test a variety of my home platforms

"Net Top" Asrock ION 330 (Nvidia ION, Intel Atom 330 Dualcore 1,6GHz, 2GB RAM, Win7 x32)
// My new kitchen pc (replaces of my old custom miniitx box with celeron and SiS video - we want full HD and Theater View without delays )). This PC can play any HD-formats with DXVA or CUDA decoders with 40-60% loads CPU, but JRMark result right - work with the database here rather slow

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 22,342 seconds
    Single-threaded floating point math... 13,542 seconds
    Multi-threaded integer math... 12,853 seconds
    Multi-threaded mixed math... 7,630 seconds
Score: 337

Running 'Image' benchmark...
    Image creation / destruction... 8,444 seconds
    Flood filling... 5,531 seconds
    Direct copying... 4,518 seconds
    Small renders... 15,589 seconds
    Bilinear rendering... 9,042 seconds
    Bicubic rendering... 8,960 seconds
Score: 422

Running 'Database' benchmark...
    Create database... 4,112 seconds
    Populate database... 12,851 seconds
    Save database... 0,802 seconds
    Reload database... 0,356 seconds
    Search database... 16,486 seconds
    Sort database... 15,364 seconds
    Group database... 12,813 seconds
Score: 342

JRMark (version 14.0.84): 367

UMPC Samsung Q1Ultra (Celeron A110 800MHz, 1GB RAM, Win7 x32, "high perfomance" power profile)
// our family UMPC, possible t-remote device (after add client control feature in MC), bathroom-pc with kensington audio dock ), traveling companion, "mobile home TV" ).
246 JRMark scores

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 29,071 seconds
    Single-threaded floating point math... 11,124 seconds
    Multi-threaded integer math... 57,696 seconds
    Multi-threaded mixed math... 21,895 seconds
Score: 159

Running 'Image' benchmark...
    Image creation / destruction... 8,319 seconds
    Flood filling... 7,104 seconds
    Direct copying... 9,155 seconds
    Small renders... 18,647 seconds
    Bilinear rendering... 32,007 seconds
    Bicubic rendering... 31,776 seconds
Score: 206

Running 'Database' benchmark...
    Create database... 3,377 seconds
    Populate database... 10,517 seconds
    Save database... 0,844 seconds
    Reload database... 0,340 seconds
    Search database... 15,504 seconds
    Sort database... 14,146 seconds
    Group database... 12,661 seconds
Score: 375

JRMark (version 14.0.84): 246

UMPC Sony Vaio UX180 (Core Solo 1,2 GHz, 512MB RAM, Win7 x32 (latest days, later downgrade back to XP image), "high perfomance" power profile)
// another possible t-remote device and nice bathroom-pc (good quality of internal soundcard). Also used with USB DAC/Amp and headphones for leasent hi-def music over wi-fi with no problem.

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 13,551 seconds
    Single-threaded floating point math... 8,793 seconds
    Multi-threaded integer math... 27,315 seconds
    Multi-threaded mixed math... 17,666 seconds
Score: 282

Running 'Image' benchmark...
    Image creation / destruction... 6,992 seconds
    Flood filling... 4,822 seconds
    Direct copying... 33,945 seconds
    Small renders... 14,925 seconds
    Bilinear rendering... 27,058 seconds
    Bicubic rendering... 28,118 seconds
Score: 190

Running 'Database' benchmark...
    Create database... 3,709 seconds
    Populate database... 10,662 seconds
    Save database... 1,340 seconds
    Reload database... 0,568 seconds
    Search database... 13,863 seconds
    Sort database... 14,171 seconds
    Group database... 12,441 seconds
Score: 379

JRMark (version 14.0.84): 284

Sony Vaio SZ650 (Core2Duo 7500 2,2 GHz, 4GB RAM, Win7 x64, "power managment off" power profile)
// laptop, my main personal computer

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 6,587 seconds
    Single-threaded floating point math... 4,400 seconds
    Multi-threaded integer math... 7,136 seconds
    Multi-threaded mixed math... 4,428 seconds
Score: 843

Running 'Image' benchmark...
    Image creation / destruction... 2,863 seconds
    Flood filling... 3,605 seconds
    Direct copying... 3,197 seconds
    Small renders... 5,790 seconds
    Bilinear rendering... 6,123 seconds
    Bicubic rendering... 7,157 seconds
Score: 766

Running 'Database' benchmark...
    Create database... 1,811 seconds
    Populate database... 7,561 seconds
    Save database... 1,213 seconds
    Reload database... 0,756 seconds
    Search database... 9,221 seconds
    Sort database... 8,133 seconds
    Group database... 6,980 seconds
Score: 603

JRMark (version 14.0.84): 737

Custom server (Core Duo 6420 2,13GHz, 4GB RAM, Win2003Std x32 with RAM-drive for DB files) // My primary DB server

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 5,329 seconds
    Single-threaded floating point math... 3,314 seconds
    Multi-threaded integer math... 5,331 seconds
    Multi-threaded mixed math... 3,345 seconds
Score: 1097

Running 'Image' benchmark...
    Image creation / destruction... 2,291 seconds
    Flood filling... 2,229 seconds
    Direct copying... 2,532 seconds
    Small renders... 4,288 seconds
    Bilinear rendering... 4,080 seconds
    Bicubic rendering... 4,011 seconds
Score: 1132

Running 'Database' benchmark...
    Create database... 0,860 seconds
    Populate database... 3,453 seconds
    Save database... 0,249 seconds
    Reload database... 0,102 seconds
    Search database... 4,203 seconds
    Sort database... 4,147 seconds
    Group database... 4,510 seconds
Score: 1227

JRMark (version 14.0.84): 1152

craigmcg · « **Reply #65 on:** November 01, 2009, 03:28:19 pm »

Out of sheer curiosity I ran the benchmark test on the clunky little laptop (Celeron 300 running XP, no antivirus) I have been using in my dining room to check online weather and play FLAC/MP3 files on a daily basis. MC14.0.90 does run ASIO through the Indigo DJ> stereo just fine using the library on my MP3 server while checkng the weather.

I believe that I now have the "floormark".

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 76.760 seconds
   Single-threaded floating point math... 29.512 seconds
   Multi-threaded integer math... 153.634 seconds
   Multi-threaded mixed math... 59.136 seconds
Score: 60

Running 'Image' benchmark...
   Image creation / destruction... 26.721 seconds
   Flood filling... 51.918 seconds
   Direct copying... 67.378 seconds
   Small renders... 55.157 seconds
   Bilinear rendering... 107.433 seconds
   Bicubic rendering... 98.577 seconds
Score: 54

Running 'Database' benchmark...
   Create database... 13.592 seconds
   Populate database... 43.820 seconds
   Save database... 4.142 seconds
   Reload database... 1.806 seconds
   Search database... 71.174 seconds
   Sort database... 49.663 seconds
   Group database... 44.746 seconds
Score: 94

JRMark (version 14.0.90): 69

JimH · « **Reply #66 on:** November 01, 2009, 04:05:24 pm »

Awesome! A new caboose.

rick.ca · « **Reply #67 on:** November 01, 2009, 04:47:18 pm »

Oh, well. It was humbling while it lasted.

glynor · « **Reply #68 on:** November 02, 2009, 11:05:13 am »

It's tough when you even lose at losing.

rick.ca · « **Reply #69 on:** November 02, 2009, 01:24:16 pm »

There goes my hope you might explain how to under-clock a processor.

eba · « **Reply #70 on:** November 02, 2009, 02:08:15 pm »

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 5.798 seconds
Single-threaded floating point math... 3.660 seconds
Multi-threaded integer math... 11.482 seconds
Multi-threaded mixed math... 9.520 seconds
Score: 624

Running 'Image' benchmark...
Image creation / destruction... 2.927 seconds
Flood filling... 4.135 seconds
Direct copying... 4.287 seconds
Small renders... 6.305 seconds
Bilinear rendering... 11.417 seconds
Bicubic rendering... 11.803 seconds
Score: 538

Running 'Database' benchmark...
Create database... 1.553 seconds
Populate database... 4.623 seconds
Save database... 0.376 seconds
Reload database... 0.173 seconds
Search database... 6.117 seconds
Sort database... 6.942 seconds
Group database... 5.893 seconds
Score: 837

JRMark (version 14.0.90): 666

I have the devil's computer.

glynor · « **Reply #71 on:** November 02, 2009, 02:11:08 pm »

Quote from: eba on November 02, 2009, 02:08:15 pm

JRMark (version 14.0.90): 666

I have the devil's computer.

Ha ha ha!!

Quote

And that no man might buy or sell,
save he that had the benchmark,
or the name of the beast,
or the number of his name.

JimH · « **Reply #72 on:** November 16, 2009, 07:02:39 am »

Cray has a new model we should test:

http://www.top500.org/lists/2009/11/press-release

globetrotters1 · « **Reply #73 on:** November 20, 2009, 02:11:05 pm »

Here is my benchmark test with version 14.0.101:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 4.886 seconds
Single-threaded floating point math... 2.756 seconds
Multi-threaded integer math... 2.498 seconds
Multi-threaded mixed math... 1.404 seconds
Score: 1646

Running 'Image' benchmark...
Image creation / destruction... 1.252 seconds
Flood filling... 0.565 seconds
Direct copying... 0.704 seconds
Small renders... 3.090 seconds
Bilinear rendering... 1.435 seconds
Bicubic rendering... 1.406 seconds
Score: 2603

Running 'Database' benchmark...
Create database... 0.589 seconds
Populate database... 2.210 seconds
Save database... 0.178 seconds
Reload database... 0.071 seconds
Search database... 2.857 seconds
Sort database... 3.063 seconds
Group database... 2.596 seconds
Score: 1859

JRMark (version 14.0.101): 2036

This is with an Asus P7P55D Deluxe m/b, i5 750 CPU, 4GB memory, MC database has 300'000 entries right now

boydn · « **Reply #74 on:** November 27, 2009, 02:46:57 am »

Basement HTPC
Core2 Duo 2.93 4 GB ATI Radeon HD 4600

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 5.268 seconds
   Single-threaded floating point math... 3.135 seconds
   Multi-threaded integer math... 5.249 seconds
   Multi-threaded mixed math... 3.147 seconds
Score: 1131

Running 'Image' benchmark...
   Image creation / destruction... 1.865 seconds
   Flood filling... 2.105 seconds
   Direct copying... 2.314 seconds
   Small renders... 3.804 seconds
   Bilinear rendering... 3.557 seconds
   Bicubic rendering... 3.540 seconds
Score: 1280

Running 'Database' benchmark...
   Create database... 0.941 seconds
   Populate database... 3.288 seconds
   Save database... 0.244 seconds
   Reload database... 0.105 seconds
   Search database... 3.982 seconds
   Sort database... 3.900 seconds
   Group database... 3.579 seconds
Score: 1341

JRMark (version 14.0.94): 1251

________________________________

Living Room PC
Core2 Duo 3.0 2GB ATI Radeon HD 3400

=== Running Benchmarks (please do not interrupt) === Running 'Math' benchmark...

   Single-threaded integer math... 3.960 seconds
   Single-threaded floating point math... 2.339 seconds
   Multi-threaded integer math... 3.969 seconds
   Multi-threaded mixed math... 2.358 seconds

Score: 1505

Running 'Image' benchmark...

   Image creation / destruction... 1.702 seconds
   Flood filling... 1.936 seconds
   Direct copying... 2.148 seconds
   Small renders... 2.842 seconds
   Bilinear rendering... 2.693 seconds
   Bicubic rendering... 2.712 seconds

Score: 1568

Running 'Database' benchmark...

   Create database... 0.659 seconds
   Populate database... 2.405 seconds
   Save database... 0.194 seconds
   Reload database... 0.084 seconds
   Search database... 2.983 seconds
   Sort database... 2.956 seconds
   Group database... 2.781 seconds

Score: 1782

JRMark (version 14.0.94): 1618

Main Computer
Core 2 2.13 4gb ATI Radeon HD 4350

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...

   Single-threaded integer math... 5.283 seconds
   Single-threaded floating point math... 3.292 seconds
   Multi-threaded integer math... 5.295 seconds
   Multi-threaded mixed math... 3.314 seconds

Score: 1106

Running 'Image' benchmark...

   Image creation / destruction... 2.123 seconds
   Flood filling... 2.145 seconds
   Direct copying... 2.604 seconds
   Small renders... 4.322 seconds
   Bilinear rendering... 4.053 seconds
   Bicubic rendering... 3.951 seconds

Score: 1146

Running 'Database' benchmark...

   Create database... 1.065 seconds
   Populate database... 4.020 seconds
   Save database... 0.486 seconds
   Reload database... 0.113 seconds
   Search database... 4.162 seconds
   Sort database... 4.412 seconds
   Group database... 3.982 seconds

Score: 1179

JRMark (version 14.0.94): 1143

Magic_Randy · « **Reply #75 on:** December 05, 2009, 09:03:17 pm »

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 4.903 seconds
Single-threaded floating point math... 3.071 seconds
Multi-threaded integer math... 5.181 seconds
Multi-threaded mixed math... 3.242 seconds
Score: 1159

Running 'Image' benchmark...
Image creation / destruction... 2.172 seconds
Flood filling... 3.800 seconds
Direct copying... 3.304 seconds
Small renders... 4.244 seconds
Bilinear rendering... 4.041 seconds
Bicubic rendering... 3.901 seconds
Score: 1025

Running 'Database' benchmark...
Create database... 0.943 seconds
Populate database... 3.289 seconds
Save database... 0.266 seconds
Reload database... 0.151 seconds
Search database... 4.023 seconds
Sort database... 4.837 seconds
Group database... 3.950 seconds
Score: 1232

JRMark (version 14.0.108): 1138

p7389 · « **Reply #76 on:** December 05, 2009, 11:47:52 pm »

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 3,962 seconds
Single-threaded floating point math... 2,233 seconds
Multi-threaded integer math... 2,063 seconds
Multi-threaded mixed math... 1,185 seconds
Score: 2012

Running 'Image' benchmark...
Image creation / destruction... 0,904 seconds
Flood filling... 0,559 seconds
Direct copying... 0,695 seconds
Small renders... 2,498 seconds
Bilinear rendering... 1,171 seconds
Bicubic rendering... 1,153 seconds
Score: 3152

Running 'Database' benchmark...
Create database... 0,550 seconds
Populate database... 2,211 seconds
Save database... 0,288 seconds
Reload database... 0,056 seconds
Search database... 2,313 seconds
Sort database... 2,597 seconds
Group database... 2,171 seconds
Score: 2111

JRMark (version 14.0.106): 2425

glynor · « **Reply #77 on:** December 07, 2009, 11:17:18 am »

Quote from: p7389 on December 05, 2009, 11:47:52 pm

JRMark (version 14.0.106): 2425

Nice one. That must be a Nehalem machine. What are you running there? Core i7 8x0 chip, I assume, but it could be a 9-series one or even an overclocked i5 (but not as overclocked as mine).

p7389 · « **Reply #78 on:** December 07, 2009, 11:46:50 am »

Quote from: glynor on December 07, 2009, 11:17:18 am

Nice one. That must be a Nehalem machine. What are you running there? Core i7 8x0 chip, I assume, but it could be a 9-series one or even an overclocked i5 (but not as overclocked as mine).

I intend to try for 3,6-4,0 GHz at some point. Don't foresee any problems, just can't be bothered with it right now. It's sufficiently speedy at its current frequency...

I guess this is in the thread already, but what are you running at?

glynor · « **Reply #79 on:** December 07, 2009, 12:21:02 pm »

Quote from: p7389 on December 07, 2009, 11:46:50 am

[img]I intend to try for 3,6-4,0 GHz at some point. Don't foresee any problems, just can't be bothered with it right now. It's sufficiently speedy at its current frequency...

I guess this is in the thread already, but what are you running at?

I have my Core i5 750 @ 4.0GHz stable. Only thing I had to do was turn off the Turbo mode, and drop the QPI link speed down to the lower setting. I had originally wanted to run it with Turbo mode turned on, but the Core Clock down a bit so that when it upped the multiplier it would bring it up to approx 4 GHz, but I couldn't get it to stay absolutely rock-solid with those settings (it would error out in Prime95 after 6-8 hours). I was making headway by bumping the VCore and VCC voltages, but it got to where I didn't want to push it much higher.

Once I disabled Turbo and just set it to 4.0 GHz full-time, I was able to drop my voltages down substantially and it was easily 30 hours+ stable under Prime95.

glynor · « **Reply #80 on:** December 07, 2009, 12:24:29 pm »

Here's a screenie:

Click to enlarge

p7389 · « **Reply #81 on:** December 07, 2009, 12:42:52 pm »

Yeah, I haven't really put in that much work into my OC efforts (not saying that you have either - it's pretty simple and you don't really need to know much), basically just turned off a few things (like turbo), bumped the bus a bit, and set the multiplier to 21.

It's ridiculously easy. When I have the time and will I will try for something closer to yours.

HiFiTubes · « **Reply #82 on:** December 07, 2009, 02:28:15 pm »

I downgraded the HTPC from a Q6600 clocked at 3gHz (on air) to a 45w AMD dual-core and a mini-ITX board running 120watts psile psu; has onboard HD3200:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 9.132 seconds
Single-threaded floating point math... 3.738 seconds
Multi-threaded integer math... 9.544 seconds
Multi-threaded mixed math... 3.923 seconds
Score: 721

Running 'Image' benchmark...
Image creation / destruction... 1.887 seconds
Flood filling... 4.320 seconds
Direct copying... 3.732 seconds
Small renders... 6.116 seconds
Bilinear rendering... 5.663 seconds
Bicubic rendering... 4.636 seconds
Score: 835

Running 'Database' benchmark...
Create database... 0.981 seconds
Populate database... 3.575 seconds
Save database... 0.783 seconds
Reload database... 0.146 seconds
Search database... 5.411 seconds
Sort database... 4.995 seconds
Group database... 3.893 seconds
Score: 1087

JRMark (version 14.0.106): 881

morrison · « **Reply #83 on:** April 02, 2010, 06:44:30 pm »

few MC15 results:

server i5@4 8ram raid

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3,329 seconds
    Single-threaded floating point math... 1,887 seconds
    Multi-threaded integer math... 1,729 seconds
    Multi-threaded mixed math... 1,065 seconds
Score: 2372

Running 'Image' benchmark...
    Image creation / destruction... 0,731 seconds
    Flood filling... 0,458 seconds
    Direct copying... 0,691 seconds
    Small renders... 2,092 seconds
    Bilinear rendering... 0,991 seconds
    Bicubic rendering... 0,967 seconds
Score: 3711

Running 'Database' benchmark...
    Create database... 0,412 seconds
    Populate database... 1,499 seconds
    Save database... 0,124 seconds
    Reload database... 0,068 seconds
    Search database... 1,987 seconds
    Sort database... 2,601 seconds
    Group database... 2,075 seconds
Score: 2452

JRMark (version 15.0.16): 2845

client - atom@1,6 dualcore, ion

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 22,282 seconds
    Single-threaded floating point math... 13,454 seconds
    Multi-threaded integer math... 12,605 seconds
    Multi-threaded mixed math... 7,564 seconds
Score: 340

Running 'Image' benchmark...
    Image creation / destruction... 9,269 seconds
    Flood filling... 5,641 seconds
    Direct copying... 4,776 seconds
    Small renders... 15,835 seconds
    Bilinear rendering... 9,474 seconds
    Bicubic rendering... 9,522 seconds
Score: 404

Running 'Database' benchmark...
    Create database... 4,198 seconds
    Populate database... 13,029 seconds
    Save database... 0,813 seconds
    Reload database... 0,332 seconds
    Search database... 16,654 seconds
    Sort database... 15,308 seconds
    Group database... 14,151 seconds
Score: 333

JRMark (version 15.0.16): 359

client with MC14

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 22,222 seconds
    Single-threaded floating point math... 13,456 seconds
    Multi-threaded integer math... 12,610 seconds
    Multi-threaded mixed math... 7,553 seconds
Score: 340

Running 'Image' benchmark...
    Image creation / destruction... 7,778 seconds
    Flood filling... 5,518 seconds
    Direct copying... 4,533 seconds
    Small renders... 15,567 seconds
    Bilinear rendering... 9,052 seconds
    Bicubic rendering... 8,908 seconds
Score: 428

Running 'Database' benchmark...
    Create database... 4,124 seconds
    Populate database... 12,729 seconds
    Save database... 0,777 seconds
    Reload database... 0,292 seconds
    Search database... 16,193 seconds
    Sort database... 15,367 seconds
    Group database... 12,913 seconds
Score: 345

JRMark (version 14.0.159): 371

and server with MC14

Code: [Select]

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
    Single-threaded integer math... 3,335 seconds
    Single-threaded floating point math... 1,881 seconds
    Multi-threaded integer math... 1,701 seconds
    Multi-threaded mixed math... 1,044 seconds
Score: 2387

Running 'Image' benchmark...
    Image creation / destruction... 0,732 seconds
    Flood filling... 0,466 seconds
    Direct copying... 0,695 seconds
    Small renders... 2,094 seconds
    Bilinear rendering... 0,992 seconds
    Bicubic rendering... 0,975 seconds
Score: 3695

Running 'Database' benchmark...
    Create database... 0,432 seconds
    Populate database... 1,659 seconds
    Save database... 0,118 seconds
    Reload database... 0,049 seconds
    Search database... 2,040 seconds
    Sort database... 2,605 seconds
    Group database... 2,024 seconds
Score: 2408

JRMark (version 14.0.159): 2830

HiFiTubes · « **Reply #84 on:** May 22, 2010, 08:40:33 pm »

New impromptu file server after adding a Vertex I SSD

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 9.979 seconds
   Single-threaded floating point math... 5.278 seconds
   Multi-threaded integer math... 10.190 seconds
   Multi-threaded mixed math... 5.329 seconds
Score: 617

Running 'Image' benchmark...
   Image creation / destruction... 1.302 seconds
   Flood filling... 1.664 seconds
   Direct copying... 1.855 seconds
   Small renders... 5.301 seconds
   Bilinear rendering... 4.782 seconds
   Bicubic rendering... 4.150 seconds
Score: 1155

Running 'Database' benchmark...
   Create database... 0.691 seconds
   Populate database... 2.753 seconds
   Save database... 0.262 seconds
   Reload database... 0.086 seconds
   Search database... 3.953 seconds
   Sort database... 4.231 seconds
   Group database... 3.196 seconds
Score: 1417

JRMark (version 15.0.44): 1063

And with Nod32 disabled:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 9.919 seconds
Single-threaded floating point math... 5.261 seconds
Multi-threaded integer math... 10.181 seconds
Multi-threaded mixed math... 5.308 seconds
Score: 620

Running 'Image' benchmark...
Image creation / destruction... 1.446 seconds
Flood filling... 1.665 seconds
Direct copying... 1.839 seconds
Small renders... 5.275 seconds
Bilinear rendering... 4.758 seconds
Bicubic rendering... 4.121 seconds
Score: 1152

Running 'Database' benchmark...
Create database... 0.681 seconds
Populate database... 2.714 seconds
Save database... 0.248 seconds
Reload database... 0.080 seconds
Search database... 3.901 seconds
Sort database... 4.273 seconds
Group database... 2.946 seconds
Score: 1448

JRMark (version 15.0.48): 1073

steveklein · « **Reply #85 on:** May 28, 2010, 01:35:07 am »

I have what I consider a pretty strong system (Quad core, 8GB ram, 7200rpm drives, ATI 4xxx video card, etc) and I got benchmark scores in the 1700s and 1800s. However, for the last score, the database score, it was like 1200something. Could this be due to a large database? I guess what I'm asking is would that bench score be higher if my library had 10,000 files instead of 80,000?

JimH · « **Reply #86 on:** May 28, 2010, 07:36:07 am »

Quote from: steveklein on May 28, 2010, 01:35:07 am

I have what I consider a pretty strong system (Quad core, 8GB ram, 7200rpm drives, ATI 4xxx video card, etc) and I got benchmark scores in the 1700s and 1800s. However, for the last score, the database score, it was like 1200something. Could this be due to a large database?

No. It's a test. It isn't reading or writing MC's database.

Try turning off your virus checker, etc. Check your Windows power settings. Choose "performance".

It's the overall score at the bottom that really matters. What was it?

steveklein · « **Reply #87 on:** May 28, 2010, 07:55:13 am »

On my old system

Intel Dual Core 2.16 GHz, 2GB Ram, Win 7 64 w/ aero turned on, onboard video

Running 'Math' benchmark...
Single-threaded integer math... 15.644 seconds
Single-threaded floating point math... 9.751 seconds
Multi-threaded integer math... 16.614 seconds
Multi-threaded mixed math... 10.350 seconds
Score: 363

Running 'Image' benchmark...
Image creation / destruction... 8.439 seconds
Flood filling... 6.328 seconds
Direct copying... 5.902 seconds
Small renders... 16.547 seconds
Bilinear rendering... 12.284 seconds
Bicubic rendering... 13.491 seconds
Score: 349

Running 'Database' benchmark...
Create database... 3.523 seconds
Populate database... 11.493 seconds
Save database... 0.815 seconds
Reload database... 0.251 seconds
Search database... 14.057 seconds
Sort database... 14.170 seconds
Group database... 12.195 seconds
Score: 381

JRMark (version 14.0.165): 364

I'll run my good computer when I have time.

marko · « **Reply #88 on:** June 01, 2010, 02:38:44 pm »

Intel i7 @ 3.6GHz, 6Gb RAM, Windows 7 x64:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 3.690 seconds
Single-threaded floating point math... 2.081 seconds
Multi-threaded integer math... 1.116 seconds
Multi-threaded mixed math... 0.709 seconds
Score: 2501

Running 'Image' benchmark...
Image creation / destruction... 0.901 seconds
Flood filling... 0.456 seconds
Direct copying... 0.644 seconds
Small renders... 2.296 seconds
Bilinear rendering... 1.132 seconds
Bicubic rendering... 1.067 seconds
Score: 3387

Running 'Database' benchmark...
Create database... 0.492 seconds
Populate database... 1.722 seconds
Save database... 0.133 seconds
Reload database... 0.039 seconds
Search database... 2.019 seconds
Sort database... 2.549 seconds
Group database... 2.073 seconds
Score: 2382

JRMark (version 15.0.51): 2757

That'll do for me

tjobbins · « **Reply #89 on:** September 17, 2010, 03:20:55 pm »

Quote from: JimH on May 28, 2010, 07:36:07 am

No. It's a test. It isn't reading or writing MC's database.

But does it do any disk IO at all? Or are the Database tests in RAM and CPU bound only? Based on the test scores I'm thinking it probably doesn't do IO right now? Especially as one guy was testing with a RAM disk but only got 1200.

If not, I'd suggest that'd be a really useful thing to add - because in real life, MC does do a lot of IO operations and it'd be great to have some representative ones benchmarked. You could create a temp DB in a separate file and run the same DB operations against that but with data actually written/read. Ideally I'd say this would be a whole new section, i.e. keep Database as it is but then add also Database-IO, so that the impact of real IO can be clearly seen.

Also because I'd like to see what my SATA3 128GB SSD can achieve

It benchmarks outside MC at 350MB/s reads, 140MB/s writes. And I'm considering buying a second one and RAID-0ing them..

For the record, here are my scores, on both 101 and 113:

JRMark (version 15.0.101): 2660
JRMark (version 15.0.113): 2764

System is:
i7-930 OC'd to 3.3ghz (3.5 ghz turbo)
Corsair 12GB DDR3-1600 RAM
ATI 5850 1GB GFX
Crucial 128GB SATA3 SSD.

My OC is pretty conservative, especially considering I have a watercooled CPU (albeit a basic system.) I might try for higher later. Actually the OC was done for me by the company I bought the PC from; they tried for 3.8 but apparently had stability issues with the 12GB RAM, but I'm fairly sure higher should be possible with more time/testing done. Only reason I've not done it yet is that the current OC comes with a warranty. But I probably won't let that stand in the way of ultimate performance

Full results for 101 are below (more representative against the rest of this thread than the new improved 113 scores):

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 3.789 seconds
Single-threaded floating point math... 2.133 seconds
Multi-threaded integer math... 1.163 seconds
Multi-threaded mixed math... 1.017 seconds
Score: 2345

Running 'Image' benchmark...
Image creation / destruction... 0.884 seconds
Flood filling... 0.459 seconds
Direct copying... 0.547 seconds
Small renders... 2.352 seconds
Bilinear rendering... 1.203 seconds
Bicubic rendering... 1.192 seconds
Score: 3315

Running 'Database' benchmark...
Create database... 0.495 seconds
Populate database... 1.749 seconds
Save database... 0.149 seconds
Reload database... 0.064 seconds
Search database... 2.111 seconds
Sort database... 2.665 seconds
Group database... 2.032 seconds
Score: 2321

JRMark (version 15.0.101): 2660

glynor · « **Reply #90 on:** September 17, 2010, 04:16:05 pm »

Wow. Still no one has toppled my Core i5 750 server system's record of 2867 (though I haven't re-run it, that was way back on version 14.0.83). Marko, tjobbins, and morrison each got pretty darn close up there though. The 4GHz OC is dead-ass simple to hit with the modern Core i5 and i7 chips though, so someone could if they put a little time into it.

If I remember, I'll give the server a good old re-run this weekend and see how it has changed with the newer versions of stuff.

glynor · « **Reply #91 on:** September 17, 2010, 05:45:19 pm »

So, I re-ran the Benchmark. Same machine, same config. New MC, and more modern software compliment (newer drivers and whatnot). Big difference:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 3.328 seconds
   Single-threaded floating point math... 1.877 seconds
   Multi-threaded integer math... 1.710 seconds
   Multi-threaded mixed math... 0.989 seconds
Score: 2404

Running 'Image' benchmark...
   Image creation / destruction... 0.760 seconds
   Flood filling... 0.462 seconds
   Direct copying... 0.674 seconds
   Small renders... 2.106 seconds
   Bilinear rendering... 0.999 seconds
   Bicubic rendering... 0.977 seconds
Score: 3681

Running 'Database' benchmark...
   Create database... 0.361 seconds
   Populate database... 1.193 seconds
   Save database... 0.127 seconds
   Reload database... 0.049 seconds
   Search database... 1.259 seconds
   Sort database... 2.469 seconds
   Group database... 0.956 seconds
Score: 3353

JRMark (version 15.0.114): 3146

Now I'm curious if my Mac Pro at work could beat it. That's an 8-core Xeon monster with a metric ton of RAM stuffed in it.

Matt · « **Reply #92 on:** September 17, 2010, 05:59:39 pm »

The database in Media Center has gotten much faster recently. It has pushed benchmark scores up.

We may re-level the benchmark so it's more comparable with older versions at some point. But it's also fun to see how Media Center has gotten faster at doing the same work.

rick.ca · « **Reply #93 on:** September 17, 2010, 06:12:53 pm »

Quote

It has pushed benchmark scores up.

Rats. I thought you had found a way to make my computer faster...

I'll have to stay tuned for glynor's OC'ing tutorial.

JustinChase · « **Reply #94 on:** September 17, 2010, 08:06:55 pm »

I figured I'd give it a go and see how my $400 laptop does (my low expectations were met

)

However, I was surprised to see that multi-threaded math was slower than single-threaded math. Any thoughts as to why?

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 5.736 seconds
Single-threaded floating point math... 3.409 seconds
Multi-threaded integer math... 6.483 seconds
Multi-threaded mixed math... 3.472 seconds
Score: 995

Running 'Image' benchmark...
Image creation / destruction... 2.340 seconds
Flood filling... 3.429 seconds
Direct copying... 3.185 seconds
Small renders... 4.504 seconds
Bilinear rendering... 4.067 seconds
Bicubic rendering... 4.198 seconds
Score: 1013

Running 'Database' benchmark...
Create database... 1.059 seconds
Populate database... 3.591 seconds
Save database... 0.290 seconds
Reload database... 0.088 seconds
Search database... 2.917 seconds
Sort database... 5.964 seconds
Group database... 4.540 seconds
Score: 1165

JRMark (version 15.0.113): 1058

Media Center 15.0.113 Registered -- C:\Program Files (x86)\J River\Media Center 15\

Workstation 6.1 (Build 7600)
Intel Pentium III 2094 MHz MMX / Memory: Total - 2025 MB, Free - 506 MB

Internet Explorer: 8.0.7600.16385 / ComCtl32.dll: 5.82.7600 / Shlwapi.dll: 6.1.7600 / Shell32.dll: 6.1.7600 / wnaspi32.dll: N/A
Ripping / Drive D: SlimtypeDVD A DS8A4SH Mode:ModeSecure Type:Auto Speed:Max
Digital playback: Yes / Get cover art: Yes / Calc replay gain: Yes / Copy volume: 100
Eject after ripping: Yes / Play sound after ripping: No

Burning / Drive D: Slimtype DVD A DS8A4SH Addr: 1:0:0 Speed:40 MaxSpeed:40 BurnProof:Yes
Test mode: No / Eject after writing: Yes / Direct decoding: Yes / Write CD-Text: Yes
Use playback settings: No /

Portable Device Info
Removed devices: E:\

Interface Plugins:
TiVo Server (Active)
Playing Now

=== From Windows ===

More details about my computer

Component Details Subscore Base score
Processor Pentium(R) Dual-Core CPU T4300 @ 2.10GHz 5.4 3.3
Determined by lowest subscore

Memory (RAM) 2.00 GB 5.4
Graphics Mobile Intel(R) 4 Series Express Chipset Family 3.6
Gaming graphics 796 MB Total available graphics memory 3.3
Primary hard disk 85GB Free (137GB Total) 5.3
Windows 7 Home Premium

System
--------------------------------------------------------------------------------

Manufacturer Acer
Model Aspire 5732Z
Total amount of system memory 2.00 GB RAM
System type 64-bit operating system
Number of processor cores 2

Storage
--------------------------------------------------------------------------------

Total size of hard disk(s) 137 GB
Disk partition (C:) 85 GB Free (137 GB Total)
Media drive (D:) CD/DVD

Graphics
--------------------------------------------------------------------------------

Display adapter type Mobile Intel(R) 4 Series Express Chipset Family
Total available graphics memory 796 MB
Dedicated graphics memory 64 MB
Dedicated system memory 0 MB
Shared system memory 732 MB
Display adapter driver version 8.15.10.1892
Primary monitor resolution 1366x768
DirectX version DirectX 10

Network
--------------------------------------------------------------------------------

Network Adapter Atheros AR5B93 Wireless Network Adapter
Network Adapter Atheros AR8132 PCI-E Fast Ethernet Controller
Network Adapter Microsoft Virtual WiFi Miniport Adapter
Network Adapter Juniper Network Connect Virtual Adapter

marko · « **Reply #95 on:** September 18, 2010, 05:08:22 am »

Quote from: marko on June 01, 2010, 02:38:44 pm

Intel i7 @ 3.6GHz, 6Gb RAM, Windows 7 x64:

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 3.690 seconds
Single-threaded floating point math... 2.081 seconds
Multi-threaded integer math... 1.116 seconds
Multi-threaded mixed math... 0.709 seconds
Score: 2501

Running 'Image' benchmark...
Image creation / destruction... 0.901 seconds
Flood filling... 0.456 seconds
Direct copying... 0.644 seconds
Small renders... 2.296 seconds
Bilinear rendering... 1.132 seconds
Bicubic rendering... 1.067 seconds
Score: 3387

Running 'Database' benchmark...
Create database... 0.492 seconds
Populate database... 1.722 seconds
Save database... 0.133 seconds
Reload database... 0.039 seconds
Search database... 2.019 seconds
Sort database... 2.549 seconds
Group database... 2.073 seconds
Score: 2382

JRMark (version 15.0.51): 2757

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 3.692 seconds
Single-threaded floating point math... 2.081 seconds
Multi-threaded integer math... 1.155 seconds
Multi-threaded mixed math... 0.710 seconds
Score: 2488

Running 'Image' benchmark...
Image creation / destruction... 0.789 seconds
Flood filling... 0.459 seconds
Direct copying... 0.635 seconds
Small renders... 2.297 seconds
Bilinear rendering... 1.173 seconds
Bicubic rendering... 1.105 seconds
Score: 3407

Running 'Database' benchmark...
Create database... 0.379 seconds
Populate database... 1.257 seconds
Save database... 0.127 seconds
Reload database... 0.036 seconds
Search database... 1.279 seconds
Sort database... 2.637 seconds
Group database... 1.092 seconds
Score: 3159

JRMark (version 15.0.114): 3018

So, from 0.51 to 0.114, on the same basic hardware (new sound card now) there's quite an improvement. My maths score has gone down slightly (story of my life!!) but the database score shows a dramatic increase, and the benefits of the recent optimisations are very real in day to day MC usage.

Respect to the team with their work on this.

Glynor, I passed on the 4 GHz, thinking that the cooling would be too noisy for me, and bought an already OC'd i7 system running at 3.6. It's the first time I've bought a pre-built system having built all my previous boxes at home. I got it from overclockers.co.uk and have to say I'm quite pleased with how my new lazy approach to system choice worked out!!

-marko

glynor · « **Reply #96 on:** September 18, 2010, 11:27:39 am »

This server is noisy, but it is in the basement so I don't care. The noise is mostly from the hard drives though (there are currently 9 drives in the thing, including the four in the external RAID box). For the CPU, I have a Noctua NH-U12P SE2 which does an awesome job cooling and is quite a bit quieter than the fan on the video card and the 120mm case fan I have cooling the hard drives. Also, the Lynnfield chips overclock so easily that I'm running at effectively stock voltage (I did bump it up a bit just to ensure stability, but not enough to make any substantial heat difference).

If you ever need a good cooler, check out Noctua. They make great stuff. The U12P actually outperforms a bunch of liquid cooling setups in independent tests.

overclockers.co.uk is an awesome store. They have a great set of enthusiast forums over there.

BushPilot · « **Reply #97 on:** September 18, 2010, 11:32:34 am »

When I ran JRM 15.0.95 benchmark test with and without anti-virus, I was happily suprised of the small and insignificant difference. HP Laptop DV5000 series, 2 GB mem, Windows 7 - Hi-performance power setting.

Anti-virus ON 822
Anti-virus OFF 827

glynor · « **Reply #98 on:** September 20, 2010, 08:43:29 pm »

Excellent work, guys.

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
   Single-threaded integer math... 3.333 seconds
   Single-threaded floating point math... 1.881 seconds
   Multi-threaded integer math... 1.726 seconds
   Multi-threaded mixed math... 0.990 seconds
Score: 2396

Running 'Image' benchmark...
   Image creation / destruction... 0.856 seconds
   Flood filling... 0.492 seconds
   Direct copying... 0.698 seconds
   Small renders... 2.080 seconds
   Bilinear rendering... 1.112 seconds
   Bicubic rendering... 0.987 seconds
Score: 3534

Running 'Database' benchmark...
   Create database... 0.359 seconds
   Populate database... 1.204 seconds
   Save database... 0.134 seconds
   Reload database... 0.050 seconds
   Search database... 1.283 seconds
   Sort database... 1.253 seconds
   Group database... 0.870 seconds
Score: 4172

JRMark (version 15.0.115): 3367

marko · « **Reply #99 on:** September 21, 2010, 02:17:07 am »

Something not right here. My maths and Image scores have taken a pounding, dropping my overall score way down below even my build 51 score. I've rebooted, score is the same lower than expected figure. What's changed in the past 3 days?

I've updated to the latest Catalyst 10.9 driver release and reinstalled my sound drivers.
My system also developed PSU coil whine over the weekend.

Would you expect any of those three events to affect the benchmark tool quite so dramatically?

=== Running Benchmarks (please do not interrupt) ===

Running 'Math' benchmark...
Single-threaded integer math... 4.491 seconds
Single-threaded floating point math... 2.535 seconds
Multi-threaded integer math... 1.359 seconds
Multi-threaded mixed math... 0.911 seconds
Score: 2044

Running 'Image' benchmark...
Image creation / destruction... 0.980 seconds
Flood filling... 0.670 seconds
Direct copying... 0.738 seconds
Small renders... 2.754 seconds
Bilinear rendering... 1.558 seconds
Bicubic rendering... 1.325 seconds
Score: 2742

Running 'Database' benchmark...
Create database... 0.503 seconds
Populate database... 1.589 seconds
Save database... 0.154 seconds
Reload database... 0.045 seconds
Search database... 1.527 seconds
Sort database... 1.587 seconds
Group database... 1.093 seconds
Score: 3309

JRMark (version 15.0.115): 2698

INTERACT FORUM

Author Topic: JRMark -- JRiver's Benchmark Tool (Read 219785 times)