Unanswered topics | Active topics


Board index » »

All times are UTC-06:00



Post new topic  Reply to topic  Page 2 of 2
  [ 20 posts ] 
  Previous topic | Next topic 
Author Message
kgardas
PostPosted: Wed Jun 29, 2011 2:24 pm 
Offline

Joined: Fri Jun 03, 2011 9:14 am
Posts: 22
Hello,
I'm really sorry, but this phoronix benchmarking is useless for i.MX53. A lot of I/O related tests which sucks with shared NFS. Anyway, here are partial results at least:
Code:
CacheBench:
pts/cachebench-1.0.0 [Test: Read]
Test 1 of 121
Estimated Time Remaining: 11 Hours, 29 Minutes
Estimated Test Run-Time: 7 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 11:39:32
Started Run 2 @ 11:41:40
Started Run 3 @ 11:43:47 [Std. Dev: 0.18%]

Test Results:
474.22025424
475.78819316
475.63455752

Average: 475.21 MB/s


CacheBench:
pts/cachebench-1.0.0 [Test: Write]
Test 2 of 121
Estimated Time Remaining: 11 Hours, 22 Minutes
Estimated Test Run-Time: 7 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 11:46:01
Started Run 2 @ 11:48:09
Started Run 3 @ 11:50:16 [Std. Dev: 0.15%]

Test Results:
830.08617308
827.9124
829.94624496

Average: 829.31 MB/s


CacheBench:
pts/cachebench-1.0.0 [Test: Read / Modify / Write]
Test 3 of 121
Estimated Time Remaining: 11 Hours, 16 Minutes
Estimated Test Run-Time: 7 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 11:52:30
Started Run 2 @ 11:54:38
Started Run 3 @ 11:56:46 [Std. Dev: 0.01%]

Test Results:
533.84464956
533.94106068
533.94294032

Average: 533.91 MB/s


7-Zip Compression:
pts/compress-7zip-1.5.0 [Compress Speed Test]
Test 4 of 121
Estimated Time Remaining: 11 Hours, 9 Minutes
Estimated Test Run-Time: 13 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 11:59:00
Started Run 2 @ 12:03:13
Started Run 3 @ 12:07:25 [Std. Dev: 0.34%]

Test Results:
337
339
337

Average: 338 MIPS


LAME MP3 Encoding:
pts/encode-mp3-1.3.1 [WAV To MP3]
Test 5 of 121
Estimated Time Remaining: 10 Hours, 57 Minutes
Estimated Test Run-Time: 45 Minutes
Expected Trial Run Count: 5
Started Run 1 @ 12:11:45
Started Run 2 @ 12:20:40
Started Run 3 @ 12:29:30
Started Run 4 @ 12:38:20
Started Run 5 @ 12:47:09 [Std. Dev: 0.29%]

Test Results:
530.95979595184
527.71042895317
527.59268903732
527.59624004364
529.70782995224

Average: 528.71 Seconds


Himeno Benchmark:
pts/himeno-1.0.0 [Poisson Pressure Solver]
Test 6 of 121
Estimated Time Remaining: 10 Hours, 13 Minutes
Estimated Test Run-Time: 4 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 12:56:08
Started Run 2 @ 12:57:14
Started Run 3 @ 12:58:18 [Std. Dev: 0.35%]

Test Results:
18.226623
18.105307
18.131521

Average: 18.15 MFLOPS


OpenSSL:
pts/openssl-1.5.0 [RSA 4096-bit Performance]
Test 7 of 121
Estimated Time Remaining: 10 Hours, 9 Minutes
Estimated Test Run-Time: 2 Minutes
Expected Trial Run Count: 4
Started Run 1 @ 12:59:30
Started Run 2 @ 12:59:52
Started Run 3 @ 13:00:15
Started Run 4 @ 13:00:38 [Std. Dev: 0.00%]

Test Results:
3.1
3.1
3.1
3.1

Average: 3.10 Signs Per Second


PyBench:
pts/pybench-1.0.0 [Total For Average Test Times]
Test 8 of 121
Estimated Time Remaining: 10 Hours, 8 Minutes
Estimated Test Run-Time: 36 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 13:01:06
Started Run 2 @ 13:12:57
Started Run 3 @ 13:24:48 [Std. Dev: 0.35%]

Test Results:
32176
32331
32397

Average: 32301 Milliseconds


AIO-Stress:
pts/aio-stress-1.1.0 [Test: Random Write]
Test 9 of 121
Estimated Time Remaining: 9 Hours, 32 Minutes
Estimated Test Run-Time: 27 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 13:36:48
Started Run 2 @ 13:47:27
Started Run 3 @ 13:59:01 [Std. Dev: 8.36%]
Started Run 4 @ 14:11:49 [Std. Dev: 7.10%]
Started Run 5 @ 14:23:52 [Std. Dev: 6.23%]
Started Run 6 @ 14:35:56 [Std. Dev: 6.73%]

Test Results:
3.3
3.07
2.79
2.95
2.97
2.75

Average: 2.97 MB/s


Apache Benchmark:
pts/apache-1.3.0 [Static Web Page Serving]
Test 10 of 121
Estimated Time Remaining: 9 Hours, 6 Minutes
Estimated Test Run-Time: 1 Hour, 1 Minute
Expected Trial Run Count: 3
Running Pre-Test Script @ 14:48:55
Started Run 1 @ 14:49:01
Started Run 2 @ 15:09:18
Started Run 3 @ 15:29:32 [Std. Dev: 0.11%]
Running Post-Test Script @ 15:49:47

Test Results:
581.55
582.73
581.72

Average: 582.00 Requests Per Second


BlogBench:
pts/blogbench-1.0.0 [Test: Read]
Test 11 of 121
Estimated Time Remaining: 8 Hours, 5 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 15:49:53
Started Run 2 @ 15:56:04
Started Run 3 @ 16:02:35 [Std. Dev: 1.64%]

Test Results:
8762
8979
9040

Average: 8927 Final Score


BlogBench:
pts/blogbench-1.0.0 [Test: Write]
Test 12 of 121
Estimated Time Remaining: 8 Hours, 5 Minutes
Expected Trial Run Count: 3
Started Run 1 @ 16:09:16
Started Run 2 @ 16:09:16
Started Run 3 @ 16:09:16 [Std. Dev: 12.39%]
Started Run 4 @ 16:09:17
The test exited with a non-zero exit status. [Std. Dev: 12.39%]
Started Run 5 @ 16:09:17
The test exited with a non-zero exit status. [Std. Dev: 12.39%]
Started Run 6 @ 16:09:18
The test exited with a non-zero exit status. [Std. Dev: 12.39%]
Started Run 7 @ 16:09:18
The test exited with a non-zero exit status. [Std. Dev: 12.39%]
Started Run 8 @ 16:09:18
The test exited with a non-zero exit status. [Std. Dev: 12.39%]
Started Run 9 @ 16:09:18
The test exited with a non-zero exit status. [Std. Dev: 12.39%]

Test Results:
22
24
28

Average: 25 Final Score


Bork File Encrypter:
pts/bork-1.0.0 [File Encryption Time]
Test 13 of 121
Estimated Time Remaining: 8 Hours, 5 Minutes
Estimated Test Run-Time: 38 Minutes
Expected Trial Run Count: 3
Running Pre-Test Script @ 16:09:25
Started Run 1 @ 16:13:11
Running Interim-Test Script @ 16:24:11
Started Run 2 @ 16:24:13
Running Interim-Test Script @ 16:35:32
Started Run 3 @ 16:35:35 [Std. Dev: 1.72%]
Running Post-Test Script @ 16:46:35

Test Results:
659.37894105911
678.43198180199
657.96553206444

Average: 665.26 Seconds


Timed Apache Compilation:
pts/build-apache-1.4.0 [Time To Compile]
Test 14 of 121
Estimated Time Remaining: 7 Hours, 27 Minutes
Estimated Test Run-Time: 55 Minutes
Expected Trial Run Count: 3
Running Pre-Test Script @ 16:46:46
Started Run 1 @ 16:56:19
Running Interim-Test Script @ 17:10:53
Started Run 2 @ 17:11:09
Running Interim-Test Script @ 17:25:40
Started Run 3 @ 17:25:59 [Std. Dev: 1.08%]
Running Post-Test Script @ 17:40:48

Test Results:
873.68577194214
870.65575695038
888.41058206558

Average: 877.58 Seconds


Timed ImageMagick Compilation:
pts/build-imagemagick-1.5.1 [Time To Compile]
Test 15 of 121
Estimated Time Remaining: 6 Hours, 32 Minutes
Estimated Test Run-Time: 3 Hours, 16 Minutes
Expected Trial Run Count: 3
Running Pre-Test Script @ 17:41:47
Started Run 1 @ 17:50:31
Running Interim-Test Script @ 18:53:09
Started Run 2 @ 18:53:30
Running Interim-Test Script @ 19:56:05
Started Run 3 @ 19:56:37 [Std. Dev: 0.04%]
Running Post-Test Script @ 20:59:13

Test Results:
3758.1075811386
3755.2809610367
3755.588835001

Average: 3756.33 Seconds


Timed Linux Kernel Compilation:
pts/build-linux-kernel-1.2.0 [Time To Compile]
Test 16 of 121
Estimated Time Remaining: 3 Hours, 16 Minutes
Expected Trial Run Count: 3
Running Pre-Test Script @ 20:59:21
Started Run 1 @ 21:26:51
The test exited with a non-zero exit status.
Running Interim-Test Script @ 21:27:23
Started Run 2 @ 21:27:40
The test exited with a non-zero exit status.
Running Interim-Test Script @ 21:28:11
Started Run 3 @ 21:28:29
The test exited with a non-zero exit status.
Running Post-Test Script @ 21:29:02

Test Results:

Average: 0 Seconds
This test failed to run properly.


Timed MPlayer Compilation:
pts/build-mplayer-1.3.0 [Time To Compile]
Test 17 of 121
Estimated Time Remaining: 3 Hours, 16 Minutes
Expected Trial Run Count: 3
Running Pre-Test Script @ 21:35:12
Started Run 1 @ 21:44:25
Running Interim-Test Script @ 22:48:22
Started Run 2 @ 22:49:01
Running Interim-Test Script @ 23:52:50
Started Run 3 @ 23:53:24 [Std. Dev: 2.34%]
Running Post-Test Script @ 00:59:54

Test Results:
3836.3186528683
3829.4304199219
3990.4527158737

Average: 3885.40 Seconds


Bullet Physics Engine:
pts/bullet-1.0.0 [Test: 3000 Fall]
Test 18 of 121
Expected Trial Run Count: 3
Started Run 1 @ 01:01:13
Started Run 2 @ 01:38:32
Started Run 3 @ 02:15:50 [Std. Dev: 0.03%]

Test Results:
504.452454
504.666504
504.404602

Average: 504.51 Seconds


Bullet Physics Engine:
pts/bullet-1.0.0 [Test: 1000 Stack]
Test 19 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:53:11
Started Run 2 @ 02:53:11
Started Run 3 @ 02:53:11 [Std. Dev: 0.21%]

Test Results:
709.17157
707.804565
710.837891

Average: 709.27 Seconds


Bullet Physics Engine:
pts/bullet-1.0.0 [Test: 136 Ragdolls]
Test 20 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:53:12
Started Run 2 @ 02:53:12
Started Run 3 @ 02:53:12 [Std. Dev: 0.03%]

Test Results:
281.619812
281.755829
281.573303

Average: 281.65 Seconds


Bullet Physics Engine:
pts/bullet-1.0.0 [Test: 1000 Convex]
Test 21 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:53:13
Started Run 2 @ 02:53:13
Started Run 3 @ 02:53:13 [Std. Dev: 0.09%]

Test Results:
390.648834
389.916443
390.290009

Average: 390.29 Seconds


Bullet Physics Engine:
pts/bullet-1.0.0 [Test: Prim Trimesh]
Test 22 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:53:14
Started Run 2 @ 02:53:14
Started Run 3 @ 02:53:14 [Std. Dev: 0.03%]

Test Results:
281.243011
281.431793
281.377777

Average: 281.35 Seconds


Bullet Physics Engine:
pts/bullet-1.0.0 [Test: Convex Trimesh]
Test 23 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:53:15
Started Run 2 @ 02:53:15
Started Run 3 @ 02:53:15 [Std. Dev: 0.03%]

Test Results:
579.605103
579.522461
579.812378

Average: 579.65 Seconds


Bullet Physics Engine:
pts/bullet-1.0.0 [Test: Raytests]
Test 24 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:53:16
Started Run 2 @ 02:53:16
Started Run 3 @ 02:53:16 [Std. Dev: 0.02%]

Test Results:
672.218872
672.408997
672.481262

Average: 672.37 Seconds


BYTE Unix Benchmark:
pts/byte-1.1.0 [Computational Test: Dhrystone 2]
Test 25 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:53:23
Started Run 2 @ 02:55:28
Started Run 3 @ 02:57:33 [Std. Dev: 0.29%]

Test Results:
2201154.1
2197872.9
2210208.9

Average: 2203078.63 LPS


BYTE Unix Benchmark:
pts/byte-1.1.0 [Computational Test: Register Arithmetic]
Test 26 of 121
Expected Trial Run Count: 3
Started Run 1 @ 02:59:44
Started Run 2 @ 03:01:50
Started Run 3 @ 03:03:58 [Std. Dev: 0.82%]

Test Results:
161721.3
160427.3
163078.3

Average: 161742.30 LPS


BYTE Unix Benchmark:
pts/byte-1.1.0 [Computational Test: Integer Arithmetic]
Test 27 of 121
Expected Trial Run Count: 3
Started Run 1 @ 03:06:11
Started Run 2 @ 03:08:17
Started Run 3 @ 03:10:23 [Std. Dev: 0.81%]

Test Results:
161974.5
164252.7
164281.3

Average: 163502.83 LPS


BYTE Unix Benchmark:
pts/byte-1.1.0 [Computational Test: Floating-Point Arithmetic]
Test 28 of 121
Expected Trial Run Count: 3
Started Run 1 @ 03:12:34
Started Run 2 @ 03:14:38
Started Run 3 @ 03:16:45 [Std. Dev: 0.07%]

Test Results:
133495.8
133489.1
133332.3

Average: 133439.07 LPS


C-Ray:
pts/c-ray-1.0.0 [Total Time]
Test 29 of 121
Expected Trial Run Count: 3
Started Run 1 @ 03:18:58
Started Run 2 @ 06:22:34
Started Run 3 @ 09:26:11 [Std. Dev: 0.01%]

Test Results:
11011.533
11013.265
11012.238

Average: 11012.35 Seconds


Compile Bench:
pts/compilebench-1.0.0 [Test: Initial Create]
Test 30 of 121
Expected Trial Run Count: 3
Started Run 1 @ 12:29:54
Started Run 2 @ 18:26:14^C

real 1966m38.118s
user 537m44.750s
sys 746m18.000s
lucid@lucid-desktop:~$
as you can see all obtained after running it for 32 hours. Also what drives me mad is that you cannot limit the number of tests (types) and also it does not push partial results to the server. :-(
Karel
Top
   
Reply with quote  
runeks
 Post subject:
PostPosted: Thu Jun 30, 2011 11:40 am 
Offline

Joined: Sun May 01, 2011 6:12 pm
Posts: 42
Location: Denmark
You are quite right that the Phoronix Test Suite seems somewhat like a half-finished application. I wish Michael would start improving on the already-existing functionality (as you mention, openbenchmarking.org's search functionality kind of sucks), instead of adding new features.
I also thought it would be smart enough to not run all the tests (that are not displayed on the said comparison page), I guess it isn't.
Just stop the tests if they're causing too much trouble, it wasn't meant to be a week-long task (I thought it'd just take a couple of hours!). I'll tinker with it myself when I get my Smartbook. I was just curious to get a preview of the results, but I guess it's best to fiddle with this myself, especially when the software (pts) doesn't seem all that well-written as I initially thought it was.
Again, thanks for enduring!

For the numerically challenged, I've made a pretty graph out of your nbench2 results. Looks interesting! A big win for PandaBoard (but not all the tests), and a win for i.MX53 over MX51:

Image
Top
   
Reply with quote  
kgardas
 Post subject:
PostPosted: Thu Jun 30, 2011 11:50 am 
Offline

Joined: Fri Jun 03, 2011 9:14 am
Posts: 22
Yes, Panda is really fast (kind of) in comparison with i.MX53, but it's kind of half baked solution! I would rather compare like: i.MX53 is very solid US made design. Everything works and is of enough quality. Panda is like Indian/Chinese solution. You need to add something to make it really working well. For example with Panda you (1) need to find SD card which works in it, also (2) you need to find suitable heatsink and put it on CPU to solve build instability issues (my panda is compiling 24 hours per day on both cores!) and finally, (3) panda's ethernet kind of sucks. Try simple ping to see it. All three issues are not presented on *rock solid* i.MX53. On the other hand when you solve those childish issues, panda is way faster (for development) than i.MX53!
Err, have I mentioned that I do have both boards on my desk now? :-)
Karel
Top
   
Reply with quote  
Neko
 Post subject:
PostPosted: Thu Jun 30, 2011 6:04 pm 
Offline
Site Admin

Joined: Fri Sep 24, 2004 1:39 am
Posts: 1589
Location: Austin, TX
I would also mention that these tests are about as real-life, real-world as a gourmet meal from McDonalds.

Phoronix's tests try to solve some of it (like the OpenSSL test, how many keysigns you can do in a second is a very useful test; it shows the performance of a real life utility a lot of people use) but do a poor job of being a benchmarking suite as a whole.

_________________
Matt Sealey
Top
   
Reply with quote  
kgardas
 Post subject:
PostPosted: Wed Jul 06, 2011 2:35 am 
Offline

Joined: Fri Jun 03, 2011 9:14 am
Posts: 22
Quote:
I would also mention that these tests are about as real-life, real-world as a gourmet meal from McDonalds.

Phoronix's tests try to solve some of it (like the OpenSSL test, how many keysigns you can do in a second is a very useful test; it shows the performance of a real life utility a lot of people use) but do a poor job of being a benchmarking suite as a whole.
I agree! For me as a developer compilation speed is nice real-world benchmark. Some numbers are here: http://dev.gentoo.org/~armin76/arm/nwbuildtimes.xml
Top
   
Reply with quote  
Display posts from previous:  Sort by  
Post new topic  Reply to topic  Page 2 of 2
  [ 20 posts ] 

Board index » »

All times are UTC-06:00

Who is online

Users browsing this forum: No registered users and 32 guests

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

Search for:
Jump to:  
PowerDeveloper.org: Copyright © 2004-2012, Genesi USA, Inc. The Power Architecture and Power.org wordmarks and the Power and Power.org logos and related marks are trademarks and service marks licensed by Power.org.
All other names and trademarks used are property of their respective owners. Privacy Policy
Powered by phpBB® Forum Software © phpBB Group