Mac OS Sierra

Message boards : Number crunching : Mac OS Sierra
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 5 · 6 · 7 · 8 · 9 · Next

AuthorMessage
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1844832 - Posted: 27 Jan 2017, 23:17:10 UTC - in response to Message 1844823.  

I receive the Same results in Yosemite, KWSN-OSX-bench-MB\ 2.01.08 Fails while KWSN-OSX-bench-MB_v2.1.07 Works.
KWSN-OSX-bench-MB_v2.1.07 also works in Mountain Lion & Mavericks.

Darwin 14.5
Last login: Fri Jan 27 17:51:57 on console
TomsMacPro:~ Tom$ cd /Users/Tom/KWSN-OSX-bench-MB\ 2.01.08 
TomsMacPro:KWSN-OSX-bench-MB 2.01.08 Tom$ ./benchmark
KWSN-Darwin-MBbench v2.1.08
Running on TomsMacPro.local at Fri Jan 27 22:59:51 2017
---------------------------------------------------
Starting benchmark run...
---------------------------------------------------
Listing wu-file(s) in /testWUs :
reference_work_unit_r3215.wu

Listing executable(s) in /APPS :
setiathome_x41p_zi+_x86_64-apple-darwin_cuda75

Listing executable in /REF_APPs :
MBv8_8.05r3344_sse41_x86_64-apple-darwin
---------------------------------------------------
usage: cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file target_file
       cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file ... target_directory
Current WU: reference_work_unit_r3215.wu
---------------------------------------------------
usage: cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file target_file
       cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file ... target_directory
Running default app with command : MBv8_8.05r3344_sse41_x86_64-apple-darwin
./MBv8_8.05r3344_sse41_x86_64-apple-darwin: No such file or directory
        0.00 real         0.00 user         0.00 sys
cp: stderr.txt: No such file or directory
cp: result.sah: No such file or directory
Elapsed Time: ………………………………… 0 seconds
---------------------------------------------------
usage: cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file target_file
       cp [-R [-H | -L | -P]] [-fi | -n] [-apvX] source_file ... target_directory
Running app with command : setiathome_x41p_zi+_x86_64-apple-darwin_cuda75 -bs -unroll 6
./setiathome_x41p_zi+_x86_64-apple-darwin_cuda75: No such file or directory
        0.00 real         0.00 user         0.00 sys
setiathome_x41p_zi+_x86_64-apple-darwin_cuda75 failed to run on your system. Continuing with next (if any)...
---------------------------------------------------
Done with reference_work_unit_r3215.wu.
2017-01-27 17:59:52.161 system_profiler[739:7787] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0

Done with Benchmark run! Removing temporary files!

######################

TomsMacPro:KWSN-OSX-bench-MB 2.01.08 Tom$ cd /Users/Tom/KWSN-OSX-bench-MB_v2.1.07 
TomsMacPro:KWSN-OSX-bench-MB_v2.1.07 Tom$ ./benchmark
KWSN-Darwin-MBbench v2.1.07
Running on TomsMacPro.local at Fri Jan 27 23:02:30 2017
---------------------------------------------------
Starting benchmark run...
---------------------------------------------------
Listing wu-file(s) in /testWUs :
reference_work_unit_r3215.wu

Listing executable(s) in /APPS :
setiathome_x41p_zi+_x86_64-apple-darwin_cuda75 setiathome_x41p_zi3n_x86_64-apple-darwin_cuda65

Listing executable in /REF_APPs :
MBv8_8.05r3344_sse41_x86_64-apple-darwin
---------------------------------------------------
Current WU: reference_work_unit_r3215.wu
---------------------------------------------------
Skipping default app MBv8_8.05r3344_sse41_x86_64-apple-darwin, displaying saved result(s)
Elapsed Time: ………………………………… 2198 seconds
---------------------------------------------------
Running app with command : setiathome_x41p_zi+_x86_64-apple-darwin_cuda75
      114.31 real        96.39 user        13.32 sys
Elapsed Time : ……………………………… 115 seconds
Speed compared to default : 1911 %
-----------------
Comparing results
Result      : Strongly similar,  Q= 99.81%
---------------------------------------------------
Running app with command : setiathome_x41p_zi3n_x86_64-apple-darwin_cuda65 -bs -unroll 5 -device 0
      107.83 real        13.55 user         8.35 sys
Elapsed Time : ……………………………… 108 seconds
Speed compared to default : 2035 %
-----------------
Comparing results
Result      : Strongly similar,  Q= 99.81%
---------------------------------------------------
Done with reference_work_unit_r3215.wu.
2017-01-27 18:06:13.573 system_profiler[851:9455] platformPluginDictionary: Can't get X86PlatformPlugin, return value 0

Done with Benchmark run! Removing temporary files!
TomsMacPro:KWSN-OSX-bench-MB_v2.1.07 Tom$

I suppose I better hold on to my KWSN-OSX-bench-MB_v2.1.07.zip file.
ID: 1844832 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1844835 - Posted: 27 Jan 2017, 23:45:07 UTC - in response to Message 1844804.  

check that init_data.xml file present in binary and data directories (to be sure app can find it).
Copy that file from BOINC directories.
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1844835 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1844840 - Posted: 28 Jan 2017, 0:20:35 UTC - in response to Message 1844835.  

TomsMacPro:KWSN-OSX-bench-MB 2.01.08 Tom$ cd /Users/Tom/KWSN-OSX-bench-MB_v2.1.07
TomsMacPro:KWSN-OSX-bench-MB_v2.1.07 Tom$ ./benchmark
Do you see it????
The NAME.
KWSN-OSX-bench-MB_v2.1.07 = Works
KWSN-OSX-bench-MB 2.01.08 = Has a Blank Space.
Last login: Fri Jan 27 18:53:53 on console
TomsMacPro:~ Tom$ cd /Users/Tom/KWSN-OSX-bench-MB_2.01.08 
TomsMacPro:KWSN-OSX-bench-MB_2.01.08 Tom$ ./benchmark
KWSN-Darwin-MBbench v2.1.08
Running on TomsMacPro.local at Sat Jan 28 00:11:00 2017
---------------------------------------------------
Starting benchmark run...
---------------------------------------------------
Listing wu-file(s) in /testWUs :
reftiny.wu

Listing executable(s) in /APPS :
MBv8_8.22r3603_ati5_ssse3_x86_64-apple-darwin

Listing executable in /REF_APPs :
MBv8_8.05r3344_sse41_x86_64-apple-darwin
---------------------------------------------------
Current WU: reftiny.wu
---------------------------------------------------
Running default app with command : MBv8_8.05r3344_sse41_x86_64-apple-darwin
       82.72 real        79.52 user         0.67 sys
Elapsed Time: ………………………………… 82 seconds
ID: 1844840 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34846
Credit: 261,360,520
RAC: 489
Australia
Message 1844894 - Posted: 28 Jan 2017, 4:57:00 UTC
Last modified: 28 Jan 2017, 4:57:25 UTC



That Core i7 has no AVX2 so result is as expected.

I asked before, so I'll ask again,

TBar why are you, "playing around with apps that your hardware doesn't support?".

I know that I may seem to be an idiot times, but looks can be deceiving. ;-)

I admire the direction that you took after all our debates about the lack of AP's for you and I admire your current work with MB's with your selective OS's, but I'm puzzled by you doing this.

Cheers.
ID: 1844894 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1844900 - Posted: 28 Jan 2017, 5:23:16 UTC - in response to Message 1844894.  



That Core i7 has no AVX2 so result is as expected.

I asked before, so I'll ask again,

TBar why are you, "playing around with apps that your hardware doesn't support?".

I know that I may seem to be an idiot times, but looks can be deceiving. ;-)

I admire the direction that you took after all our debates about the lack of AP's for you and I admire your current work with MB's with your selective OS's, but I'm puzzled by you doing this.

Cheers.


I would step in to defend the pokings, on the basis that what's being seen, with respect to changes happening, hasn't been seen in my time here since circa 2007. With the operating systems, compilers, tools, and hardware all changing at a faster rate than ever before, there is a massive number of unresolved questions/holes on what the next generation of applications will look like.

While I am definitely taking a back seat and watching a lot for the time being, there are valuable insights to be gleaned from what appears to be abject chaos. In that sense i'd argue that any number of good, bad and ugly approaches can pay off in the long run. i.e. it saves me work ;)
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1844900 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1844949 - Posted: 28 Jan 2017, 10:35:28 UTC - in response to Message 1844894.  
Last modified: 28 Jan 2017, 10:40:32 UTC



That Core i7 has no AVX2 so result is as expected.

I asked before, so I'll ask again,

TBar why are you, "playing around with apps that your hardware doesn't support?".

I know that I may seem to be an idiot times, but looks can be deceiving. ;-)

I admire the direction that you took after all our debates about the lack of AP's for you and I admire your current work with MB's with your selective OS's, but I'm puzzled by you doing this.

Cheers.

My question would be why are you concerned?
I don't have an OSX Intel iGPU, but neither does Urs or Raistmer. You don't seem to be bothered by Raistmer or Urs "playing around with apps that your hardware doesn't support".
I'm curious why you've singled out me for this question.
ID: 1844949 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1844954 - Posted: 28 Jan 2017, 12:04:37 UTC - in response to Message 1844949.  

My question would be why are you concerned?
I don't have an OSX Intel iGPU, but neither does Urs or Raistmer. You don't seem to be bothered by Raistmer or Urs "playing around with apps that your hardware doesn't support".
I'm curious why you've singled out me for this question.


No It's a logical question. A common perception is that to develop for a given architecture, then you need access to it. In some respects that's reasonable, and in others not. FWIW my first operational Cuda Fermi builds existed well before I had such hardware. Naturally without access to the hardware they were imperfect. So to me it just speaks to the argument that all avenues need to be explored. So to all sides my response is 'Suck it up buttercup', then strap yourselves in and prepare to feel the cheese,
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 1844954 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1844956 - Posted: 28 Jan 2017, 12:51:33 UTC - in response to Message 1844954.  

My question would be why are you concerned?
I don't have an OSX Intel iGPU, but neither does Urs or Raistmer. You don't seem to be bothered by Raistmer or Urs "playing around with apps that your hardware doesn't support".
I'm curious why you've singled out me for this question.


No It's a logical question. A common perception is that to develop for a given architecture, then you need access to it. In some respects that's reasonable, and in others not. FWIW my first operational Cuda Fermi builds existed well before I had such hardware. Naturally without access to the hardware they were imperfect. So to me it just speaks to the argument that all avenues need to be explored. So to all sides my response is 'Suck it up buttercup', then strap yourselves in and prepare to feel the cheese,

OK, I'll add you to the list. Jason, Raistmer, Urs, and myself have all compiled Apps while not having access to the particular hardware. So, why am I being singled out?
Yes, things have changed quite a bit since the Only GPU work available to OSX were APs. The choice was simple back then, you ran APs on your Mac GPUs or you ran NOTHING. It was a very simple choice if you had a Mac. Now things are quite different, and I'm proud to say Four of the Current Mac Apps on SETI Main were complied by myself. There would be more if SETI wasn't so concerned about a few people demanding a Screen Saver on the CPU App. For the want of a Screen Saver the Science runs at half Speed. It's extremely simple to change the configure line from SSE41 to AVX2, as long as the code is correct the compiled App will run just as well. Since the current machine testing the Intel App also has AVX2, why Not provide it with a suitable App?
ID: 1844956 · Report as offensive
Urs Echternacht
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 692
Credit: 135,197,781
RAC: 211
Germany
Message 1844973 - Posted: 28 Jan 2017, 17:51:10 UTC - in response to Message 1844956.  

TBar,
thanks for pointing out what your problem was with the v2.01.08 benchmark script. I've changed the naming and reuploaded the package at lunatics. Nice job compiling all these apps.

By the way : Testing the AVX2 app on a host that has no AVX2 with a "new" version of the benchmark script is a routine test for the benchmark, checking that it behaves as it should (outcome as expected!). Nothing more nothing less.
_\|/_
U r s
ID: 1844973 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1845216 - Posted: 29 Jan 2017, 19:26:28 UTC - in response to Message 1844973.  
Last modified: 29 Jan 2017, 19:27:37 UTC

Replacing the Blank Space with an underscore fixed the problem. The New download with the shortened name is working fine too. The last builds have the ATI Non-SoG builds just as fast as the SoG builds while reducing the Idle Wake Ups by half. The last non-SoG build compiled in Mountain Lion has almost normal Idle Wake Ups when running the Non-VLAR BLCs. The Arecibo tasks are running around 200 to 400, about half the SoG numbers on this machine. I'll post the new files so Laurent can test them on his machine which is showing much higher Idle Wake Ups than mine.
MBv8_8.20r3556_ati5-SoG_ssse3_x86_64-apple-darwin is the Current Stock App on Main.
TomsMacPro:~ Tom$ cd /Users/Tom/KWSN-OSX-bench-MB
TomsMacPro:KWSN-OSX-bench-MB Tom$ ./benchmark
KWSN-Darwin-MBbench v2.1.08
Running on TomsMacPro.local at Sun Jan 29 10:25:36 2017
---------------------------------------------------
Starting benchmark run...
---------------------------------------------------
Listing wu-file(s) in /testWUs :
reference_work_unit_r3215.wu

Listing executable(s) in /APPS :
MBv8_8.20r3556_ati5-SoG_ssse3_x86_64-apple-darwin MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin MBv8_8.22r3611_ati5-SoG_ssse3_x86_64-apple-darwin

Listing executable in /REF_APPs :
MBv8_8.05r3344_sse41_x86_64-apple-darwin
---------------------------------------------------
Current WU: reference_work_unit_r3215.wu
---------------------------------------------------
Skipping default app MBv8_8.05r3344_sse41_x86_64-apple-darwin, displaying saved result(s)
Elapsed Time: ………………………………… 2521 seconds
---------------------------------------------------
Running app with command : MBv8_8.20r3556_ati5-SoG_ssse3_x86_64-apple-darwin -device 0
      342.95 real        40.89 user        32.08 sys
Elapsed Time : ……………………………… 343 seconds
Speed compared to default : 734 %
-----------------
Comparing results
Result      : Strongly similar,  Q= 99.52%
---------------------------------------------------
Running app with command : MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin -device 0
      330.20 real        37.45 user        35.00 sys
Elapsed Time : ……………………………… 330 seconds
Speed compared to default : 763 %
-----------------
Comparing results
Result      : Strongly similar,  Q= 99.78%
---------------------------------------------------
Running app with command : MBv8_8.22r3611_ati5-SoG_ssse3_x86_64-apple-darwin -device 0
      335.45 real        35.24 user        31.66 sys
Elapsed Time : ……………………………… 336 seconds
Speed compared to default : 750 %
-----------------
Comparing results
Result      : Strongly similar,  Q= 99.95%
---------------------------------------------------
Done with reference_work_unit_r3215.wu.
ID: 1845216 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1845237 - Posted: 29 Jan 2017, 20:34:46 UTC

I replaced the Old ATi5r3552 with the new package ATi5r3610&CPU-AVX.zip here, http://www.arkayn.us/forum/index.php?topic=191.msg4368#msg4368
To replace the ATi5r3587 by itself, pull the two r3610 files out of the package and place them in your setiathome.berkeley.edu folder. Then change your app_info.xml from the old r3587 names to the new r3610 names. Do Not change the <version_num>xxx</version_num> or <plan_class>xxxx</plan_class> in your current app_info.xml or you will Ghost All your existing GPU tasks. Or, you could finish your existing ATI tasks then replace the whole package. Run the BOINC Installer to set the file permissions after changing or editing the files.
ID: 1845237 · Report as offensive
W3Perl Project Donor
Volunteer tester

Send message
Joined: 29 Apr 99
Posts: 251
Credit: 3,696,783,867
RAC: 12,606
France
Message 1845404 - Posted: 30 Jan 2017, 10:24:00 UTC - in response to Message 1845237.  

I replaced the Old ATi5r3552 with the new package ATi5r3610&CPU-AVX.zip here, http://www.arkayn.us/forum/index.php?topic=191.msg4368#msg4368
To replace the ATi5r3587 by itself, pull the two r3610 files out of the package and place them in your setiathome.berkeley.edu folder. Then change your app_info.xml from the old r3587 names to the new r3610 names. Do Not change the <version_num>xxx</version_num> or <plan_class>xxxx</plan_class> in your current app_info.xml or you will Ghost All your existing GPU tasks. Or, you could finish your existing ATI tasks then replace the whole package. Run the BOINC Installer to set the file permissions after changing or editing the files.


Sorry but last post is from January 12th.
About the error I get running 3587 outside BOINC (using the benchmark tool), stderr-MBv8_8.22r3587_ati5-SoG.....txt :

OpenCL-kernels filename : Multibeam_Kernels_r3587.cl
INFO : can't open binary kernel file : .//MultiBeam_Kernels_r3587.clHD5_IntelRHDGraphics530.bin_V7_SoG_16.1.0_12Nov42016182, continue with recompile...
ERROR : can't open CL file, exiting....
ID: 1845404 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1845410 - Posted: 30 Jan 2017, 11:10:35 UTC - in response to Message 1845404.  
Last modified: 30 Jan 2017, 11:43:46 UTC

INFO : can't open binary kernel file : .//MultiBeam_Kernels_r3587.clHD5_IntelRHDGraphics530.bin_V7_SoG_16.1.0_12Nov42016182, continue with recompile...
ERROR : can't open CL file, exiting....

That shows the Machine is trying to run the ATI App on the Intel iGPU. That is one of the Bugs listed in the 10.12.3 Update;
- Improves automatic graphics switching on MacBook Pro (15-inch, October 2016)
- Resolves graphics issues while encoding Adobe Premiere Pro projects on MacBook Pro with Touch Bar (13- and 15-inch, October 2016)
Download macOS Sierra 10.12.3 Combo Update

List of suggestions;
1) Install the 10.12.3 Update
2) Download New Benchmark Tool http://lunatics.kwsn.info/index.php?action=downloads;sa=view;down=500
3) Download New ATI App http://www.arkayn.us/forum/index.php?action=dlattach;topic=191.0;attach=898
4) Make sure the MultiBeam_Kernels_r3610.cl file is in the root level of KWSN-OSX-bench-MB, and the App MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin is in KWSN-OSX-bench-MB/APPS
The new App MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin should give much lower Idle Wake Ups than the older r3587 App.
5) In the file KWSN-OSX-bench-MB/comlineoptions.txt add the device number to the cmdline; MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin -device 0
6) Go to your BOINC Data folder and copy the file coproc_info.xml, then paste coproc_info.xml into the KWSN-OSX-bench-MB folder

Besides fixing the Known Bugs with the New OS, the 10.12.3 Update will install a new ATI Driver. Note My ATI Driver is newer than Yours;
My Driver version: 1.2 (Jan 10 2017 00:23:26) http://setiathome.berkeley.edu/result.php?resultid=5472273713
Your Driver version: 1.2 (Oct 19 2016 21:05:08) https://setiathome.berkeley.edu/result.php?resultid=5470005676

see how all that works... ;-)
ID: 1845410 · Report as offensive
W3Perl Project Donor
Volunteer tester

Send message
Joined: 29 Apr 99
Posts: 251
Credit: 3,696,783,867
RAC: 12,606
France
Message 1845652 - Posted: 31 Jan 2017, 10:44:05 UTC - in response to Message 1845410.  


1) Install the 10.12.3 Update
2) Download New Benchmark Tool http://lunatics.kwsn.info/index.php?action=downloads;sa=view;down=500


OS have been upgraded to 10.12.3.
New benchmark tool have been uploaded (btw, the space character should be changed to '_' in the directory name)
No real speed improvement on Sierra ....


KWSN-Darwin-MBbench v2.1.08
Running on domisses-MacBook-Pro.local at Tue Jan 31 09:54:10 2017
---------------------------------------------------
Starting benchmark run...
---------------------------------------------------
Listing wu-file(s) in /testWUs :
18dc09ah.26284.16432.6.33.125.wu reference_work_unit_r3215.wu

Listing executable(s) in /APPS :
MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin

Listing executable in /REF_APPs :
MBv8_8.22r3587_ati5-SoG_ssse3_x86_64-apple-darwin
---------------------------------------------------
Current WU: 18dc09ah.26284.16432.6.33.125.wu
---------------------------------------------------
Running default app with command : MBv8_8.22r3587_ati5-SoG_ssse3_x86_64-apple-darwin -device 0
837.48 real 49.94 user 81.08 sys
Elapsed Time: ………………………………… 838 seconds
---------------------------------------------------
Running app with command : MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin -device 0
795.22 real 49.04 user 91.28 sys
Elapsed Time : ……………………………… 795 seconds
Speed compared to default : 105 %
-----------------
Comparing results
Result : Strongly similar, Q= 99.47%
---------------------------------------------------
Done with 18dc09ah.26284.16432.6.33.125.wu.
Current WU: reference_work_unit_r3215.wu
---------------------------------------------------
Running default app with command : MBv8_8.22r3587_ati5-SoG_ssse3_x86_64-apple-darwin -device 0
486.74 real 68.42 user 52.22 sys
Elapsed Time: ………………………………… 487 seconds
---------------------------------------------------
Running app with command : MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin -device 0
490.98 real 75.97 user 59.31 sys
Elapsed Time : ……………………………… 491 seconds
Speed compared to default : 99 %
-----------------
Comparing results
Result : Strongly similar, Q= 99.53%
---------------------------------------------------
Done with reference_work_unit_r3215.wu.
ID: 1845652 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1845658 - Posted: 31 Jan 2017, 12:10:18 UTC - in response to Message 1845652.  
Last modified: 31 Jan 2017, 12:17:50 UTC

New benchmark tool have been uploaded (btw, the space character should be changed to '_' in the directory name)
No real speed improvement on Sierra ....
The Benchmark Tool was changed, the current download uses the folder name KWSN-OSX-bench-MB, which doesn't have any spaces. Just download it again and you should get the newer version.
So, using the -device 0 tag runs the ATI App on the ATI GPU? I have the same problem on my machine. If I don't use the -device 0 tag the ATI App runs on the nVidia GPUs. It appears hopeless with the Intel App though, it always runs on the nVidia GPUs unless I run it in BOINC, then I can get it to run on the ATI card.

The difference between r3587 & r3610 is how the App runs on the System. I think you said r3587 was giving up to 4000 Idle Wake Ups per sec? That is Much too high, the limit is 150. High Idle Wake Ups usually happen just before the App Crashes. Go to Applications/Utilities and run the Console App, look for entries such as;
1/31/17 5:00:40.000 AM kernel[0]: process MBv8_8.22r3610_a[11143] caught causing excessive wakeups. Observed wakeups rate (per sec): 215; Maximum permitted wakeups rate (per sec): 150; Observation period: 300 seconds; Task lifetime number of wakeups: 45023
There have been complaints about the SoG App causing System reboots, probably due to Driver Crashes. Are the Idle Wake Ups any better with the New ATI Driver? Can you post the difference between the Idle Wake Ups with r3587 and r3610? Even though r3610 is much better on my system the system is Still complaining about it. To see the logs go to;
1/31/17 5:00:42.147 AM spindump[438]: Saved wakeups_resource.diag report for MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin version ??? (???) to /Library/Logs/DiagnosticReports/MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin_2017-01-31-050042_TomsMacPro.wakeups_resource.diag
You will see reports similar to;
Date/Time: 2017-01-31 06:20:45 -0500
OS Version: Mac OS X 10.11.6 (Build 15G1217)
Architecture: x86_64
Report Version: 19
Command: MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin
Path: */MBv8_8.22r3610_ati5_ssse3_x86_64-apple-darwin
Version: ??? (???)
Parent: boinc [564]
Responsible: BOINCManager [561]
PID: 11938
Event: wakeups (microstackshots only)
Wakeups: 277 wakeups per second for 163 seconds
Duration: 163.00s
Steps: 167
Hardware model: MacPro3,1
....
Compare that to the SoG App r3611;
Date/Time: 2017-01-29 05:18:43 -0500
OS Version: Mac OS X 10.11.6 (Build 15G1217)
Architecture: x86_64
Report Version: 19
Command: MBv8_8.22r3611_ati5-SoG_ssse3_x86_64-apple-darwin
Path: MBv8_8.22r3611_ati5-SoG_ssse3_x86_64-apple-darwin
Version: ??? (???)
Parent: time [850]
Responsible: Terminal [649]
PID: 851
Event: wakeups (microstackshots only)
Wakeups: 643 wakeups per second for 70 seconds
Duration: 70.00s
Steps: 70
Hardware model: MacPro3,1
.....
If r3610 gives Half the Wake Ups as the SoG Apps I'll recommend it to Beta.
ID: 1845658 · Report as offensive
W3Perl Project Donor
Volunteer tester

Send message
Joined: 29 Apr 99
Posts: 251
Credit: 3,696,783,867
RAC: 12,606
France
Message 1845661 - Posted: 31 Jan 2017, 12:23:05 UTC - in response to Message 1845658.  

Another test using your latest Intel GPU :

KWSN-Darwin-MBbench v2.1.08
Running on domisses-MacBook-Pro.local at Tue Jan 31 10:48:57 2017
---------------------------------------------------
Starting benchmark run...
---------------------------------------------------
Listing wu-file(s) in /testWUs :
18dc09ah.26284.16432.6.33.125.wu reference_work_unit_r3215.wu

Listing executable(s) in /APPS :
MBv8_8.22r3606_Intel-mod_ssse3_x86_64-apple-darwin

Listing executable in /REF_APPs :
MBv8_8.22r3587_ati5-SoG_ssse3_x86_64-apple-darwin
---------------------------------------------------
Current WU: 18dc09ah.26284.16432.6.33.125.wu
---------------------------------------------------
Running default app with command : MBv8_8.22r3587_ati5-SoG_ssse3_x86_64-apple-darwin -device 0
836.72 real 49.76 user 81.04 sys
Elapsed Time: ………………………………… 837 seconds
---------------------------------------------------
Running app with command : MBv8_8.22r3606_Intel-mod_ssse3_x86_64-apple-darwin -device 1
1191.32 real 59.24 user 96.06 sys
Elapsed Time : ……………………………… 1191 seconds
Speed compared to default : 70 %
-----------------
Comparing results
Result : Strongly similar, Q= 100.0%
---------------------------------------------------
Done with 18dc09ah.26284.16432.6.33.125.wu.
Current WU: reference_work_unit_r3215.wu
---------------------------------------------------
Running default app with command : MBv8_8.22r3587_ati5-SoG_ssse3_x86_64-apple-darwin -device 0
488.29 real 68.72 user 51.39 sys
Elapsed Time: ………………………………… 489 seconds
---------------------------------------------------
Running app with command : MBv8_8.22r3606_Intel-mod_ssse3_x86_64-apple-darwin -device 1
809.59 real 211.43 user 67.09 sys
Elapsed Time : ……………………………… 809 seconds
Speed compared to default : 60 %
-----------------
Comparing results
Result : Strongly similar, Q= 99.99%
---------------------------------------------------
Done with reference_work_unit_r3215.wu.
ID: 1845661 · Report as offensive
W3Perl Project Donor
Volunteer tester

Send message
Joined: 29 Apr 99
Posts: 251
Credit: 3,696,783,867
RAC: 12,606
France
Message 1845663 - Posted: 31 Jan 2017, 12:48:57 UTC - in response to Message 1845658.  


So, using the -device 0 tag runs the ATI App on the ATI GPU? I have the same problem on my machine. If I don't use the -device 0 tag the ATI App runs on the nVidia GPUs. It appears hopeless with the Intel App though, it always runs on the nVidia GPUs unless I run it in BOINC, then I can get it to run on the ATI card.


Only got ATI cards (it's a laptop). I guess ATI is device 0 and Intel device 1 ? How canI know for sure ?


The difference between r3587 & r3610 is how the App runs on the System. I think you said r3587 was giving up to 4000 Idle Wake Ups per sec? That is Much too high, the limit is 150. High Idle Wake Ups usually happen just before the App Crashes. Go to Applications/Utilities and run the Console App, look for entries such as;


4000 was the max value...
MBv8_8.22r3587_ATI5-SOG_sse3 : between 1000 and 1500
MBv8_8.22r3604-ati5_ssse3 : between 300 and 400
MBv8_8.22r3610-ati5_ssse3 : between 500 and 750
MBv8.8.22r3606_Intel-mod_sse3 : between 200 and 600

If r3610 gives Half the Wake Ups as the SoG Apps I'll recommend it to Beta.


go for beta ! :)
ID: 1845663 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1845665 - Posted: 31 Jan 2017, 13:04:10 UTC - in response to Message 1845661.  
Last modified: 31 Jan 2017, 13:38:37 UTC

Current WU: 18dc09ah.26284.16432.6.33.125.wu
---------------------------------------------------
Running default app with command : MBv8_8.22r3587_ati5-SoG_ssse3_x86_64-apple-darwin -device 0
836.72 real 49.76 user 81.04 sys
Elapsed Time: ………………………………… 837 seconds
---------------------------------------------------
Running app with command : MBv8_8.22r3606_Intel-mod_ssse3_x86_64-apple-darwin -device 1
1191.32 real 59.24 user 96.06 sys
Elapsed Time : ……………………………… 1191 seconds
Speed compared to default : 70 %
-----------------
Comparing results
Result : Strongly similar, Q= 100.0%
Interesting.
Are you sure it's running the Intel App on the iGPU? I think to run it on the Intel iGPU you shouldn't use the device tag, just leave it blank.
To tell which device it's using look at the .bin files while the App is running. It will list which device as such;
MultiBeam_Kernels_r3587.clHD5_IntelRHDGraphics530.bin_V7_SoG_16.1.0_12Nov42016182
MultiBeam_Kernels_r3587.clHD5_AMD450...bin_V7_SoG_16.1.0_12Nov42016182
The .bin files will appear at the root level of the Benchmark folder.

BTW, did you test the last version of the CPU AVX2 App?
ID: 1845665 · Report as offensive
W3Perl Project Donor
Volunteer tester

Send message
Joined: 29 Apr 99
Posts: 251
Credit: 3,696,783,867
RAC: 12,606
France
Message 1845693 - Posted: 31 Jan 2017, 15:45:59 UTC - in response to Message 1845665.  



BTW, did you test the last version of the CPU AVX2 App?


KWSN-Darwin-MBbench v2.1.08
Running on domisses-MacBook-Pro.local at Tue Jan 31 14:23:50 2017
---------------------------------------------------
Starting benchmark run...
---------------------------------------------------
Listing wu-file(s) in /testWUs :
18dc09ah.26284.16432.6.33.125.wu reference_work_unit_r3215.wu

Listing executable(s) in /APPS :
MBv8_8.22r3588_avx2_x86_64-apple-darwin

Listing executable in /REF_APPs :
MBv8_8.06r3366_avx_x86_64-apple-darwin
---------------------------------------------------
Current WU: 18dc09ah.26284.16432.6.33.125.wu
---------------------------------------------------
Running default app with command : MBv8_8.06r3366_avx_x86_64-apple-darwin
1443.24 real 1438.85 user 2.17 sys
1560.16 real 1555.76 user 2.24 sys
Elapsed Time: ………………………………… 1561 seconds
---------------------------------------------------
Running app with command : MBv8_8.22r3588_avx2_x86_64-apple-darwin
1161.59 real 1158.33 user 1.25 sys
Elapsed Time : ……………………………… 1161 seconds
Speed compared to default : 134 %
-----------------
Comparing results
------------- R1:R2 ------------ ------------- R2:R1 ------------
Exact Super Tight Good Bad Exact Super Tight Good Bad
Spike 0 0 0 0 0 0 0 0 0 0
Autocorr 0 0 0 0 0 0 0 0 0 0
Gaussian 0 0 0 0 0 0 0 0 0 0
Pulse 0 0 0 0 1 0 0 0 0 0
Triplet 0 1 1 1 2 0 1 1 1 6
Best Spike 0 0 0 0 1 0 0 0 0 1
Best Autocorr 0 0 0 0 1 0 0 0 0 1
Best Gaussian 1 1 1 1 0 1 1 1 1 0
Best Pulse 0 0 0 0 1 0 0 0 0 1
Best Triplet 0 0 0 0 1 0 0 0 0 1
---- ---- ---- ---- ---- ---- ---- ---- ---- ----
1 2 2 2 7 1 2 2 2 10

Unmatched signal(s) in R1 at line(s) 359 376 393 419 435 473 498
Unmatched signal(s) in R2 at line(s) 359 376 393 410 427 444 461 477 515 539
For R1:R2 matched signals only, Q= 100.0%
Result : Different.
---------------------------------------------------
Done with 18dc09ah.26284.16432.6.33.125.wu.
Current WU: reference_work_unit_r3215.wu
---------------------------------------------------
Running default app with command : MBv8_8.06r3366_avx_x86_64-apple-darwin
983.03 real 979.93 user 1.00 sys
Elapsed Time: ………………………………… 983 seconds
---------------------------------------------------
Running app with command : MBv8_8.22r3588_avx2_x86_64-apple-darwin
934.18 real 931.26 user 0.88 sys
Elapsed Time : ……………………………… 935 seconds
Speed compared to default : 105 %
-----------------
Comparing results
------------- R1:R2 ------------ ------------- R2:R1 ------------
Exact Super Tight Good Bad Exact Super Tight Good Bad
Spike 0 0 0 0 13 0 0 0 0 0
Autocorr 0 0 0 0 1 0 0 0 0 0
Gaussian 0 0 0 0 6 0 0 0 0 0
Pulse 0 0 0 0 0 0 0 0 0 0
Triplet 0 0 0 0 2 0 0 0 0 0
Best Spike 0 0 0 0 1 0 0 0 0 1
Best Autocorr 0 0 0 0 1 0 0 0 0 1
Best Gaussian 0 0 0 0 1 0 0 0 0 1
Best Pulse 0 0 0 0 1 0 0 0 0 1
Best Triplet 0 0 0 0 1 0 0 0 0 0
---- ---- ---- ---- ---- ---- ---- ---- ---- ----
0 0 0 0 27 0 0 0 0 4

Unmatched signal(s) in R1 at line(s) 240 257 273 289 305 321 337 353 369 385 401 417 433 449 466 482 499 526 553 580 607 634 661 677 694 720 745
Unmatched signal(s) in R2 at line(s) 240 256 273 299
Result : Different.
---------------------------------------------------
Done with reference_work_unit_r3215.wu.
ID: 1845693 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1845694 - Posted: 31 Jan 2017, 16:01:42 UTC - in response to Message 1845693.  

BTW, did you test the last version of the CPU AVX2 App?
Running app with command : MBv8_8.22r3588_avx2_x86_64-apple-darwin
r3588 was the old version. The new version was in the Intel-mod-1-24.zip download and named MBv8_8.22r3605_avx2_x86_64-apple-darwin
ID: 1845694 · Report as offensive
Previous · 1 . . . 5 · 6 · 7 · 8 · 9 · Next

Message boards : Number crunching : Mac OS Sierra


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.