Welcome to the MacNN Forums.

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

You are here: MacNN Forums > Community > Team MacNN > Enhanced Optimized

Enhanced Optimized (Page 5)
Thread Tools
gulliver
Fresh-Faced Recruit
Join Date: Nov 2005
Location: Europe
Status: Offline
Reply With Quote
Jun 5, 2006, 02:27 PM
 
Originally Posted by Bad to the bone
Look at this WU!

http://setiathome.berkeley.edu/worku...?wuid=80752856

My 2.5 GHz G5 is the one with 1,952 seconds. I wish, ALL would be like this.
How did you do that? My G5 Dual 2.7 GHz (http://setiweb.ssl.berkeley.edu/resu...?hostid=633575) needs at least 5,000 seconds for that credit. My G5 Dual 2.5 GHz (http://setiweb.ssl.berkeley.edu/resu...?hostid=638749) takes even longer. Could you post the wisdom file please?
     
rhettmaxwell
Fresh-Faced Recruit
Join Date: Jun 2006
Status: Offline
Reply With Quote
Jun 5, 2006, 06:35 PM
 
Originally Posted by gulliver
How did you do that? My G5 Dual 2.7 GHz (http://setiweb.ssl.berkeley.edu/resu...?hostid=633575) needs at least 5,000 seconds for that credit. My G5 Dual 2.5 GHz (http://setiweb.ssl.berkeley.edu/resu...?hostid=638749) takes even longer. Could you post the wisdom file please?
Dunno what gave the kick. Nothing special. Went back to BOINC 4.4.4 superbench recently from 5.2.13 and another attempt to run 5.4.9 again. 5.4.9 (stock and superbench) always pukes on me with segmentation violation. Good old 4.4.4 is yielding credit very inconsistently but appears to be slightly more productive than 5.2.13. I wonder if it'll outperform 5.2.13 over time.

Check other results here. There is one Wu with 5,747 sec that harvested 31.86 credits. Not bad compared to 7.2 to 7.4 k sec worth about 30-31 credits.
http://setiathome.berkeley.edu/show_...hostid=1441575

I'm using Todd Madson's wisdom like I also did with the old worker. Gotta find some time to shut down all apps and create my own wisdom file one of these days. Will be interesting how the various combinations will perform.
     
Gecko_r7
Forum Regular
Join Date: Oct 2005
Location: Las Vegas, NV
Status: Offline
Reply With Quote
Jun 5, 2006, 10:19 PM
 
Originally Posted by Bad to the bone
Dunno what gave the kick. Nothing special. Went back to BOINC 4.4.4 superbench recently from 5.2.13 and another attempt to run 5.4.9 again. 5.4.9 (stock and superbench) always pukes on me with segmentation violation. Good old 4.4.4 is yielding credit very inconsistently but appears to be slightly more productive than 5.2.13. I wonder if it'll outperform 5.2.13 over time.

Check other results here. There is one Wu with 5,747 sec that harvested 31.86 credits. Not bad compared to 7.2 to 7.4 k sec worth about 30-31 credits.
http://setiathome.berkeley.edu/show_...hostid=1441575

I'm using Todd Madson's wisdom like I also did with the old worker. Gotta find some time to shut down all apps and create my own wisdom file one of these days. Will be interesting how the various combinations will perform.

I might be able to shed some light. Seti Enhanced uses FLOPs counting as part of the new credit scheme. Benchmarking is no longer relevant w/ the way Enhanced claims credit and does not affect credit when using the current Boinc 5.x clients. Until just a couple of weeks ago under the previous Seti project (pre-Enhanced), optimized applications that completed work quicker, did so w/ lower claimed credit vs. the standard Berkeley distributed aps. In essence, quicker computers would complete more WUs and therefore perform more work but may have claimed only similar credit to a much slower machine thaat only completed 1 WU in the same time. Optimized Boinc applications (Superbench) equalized this to a large degree so that faster rigs and those running opt. science applications (there are other reasons to run opt clients as well, esp if running mult. projects, hosts, etc.) would claim the appropriate credit due.

There has been some observation within the Seti forums that running certain Boinc 4.x clients w/ Enhanced can cause erratic operating behavior, and in the case of Opt. 4.x clients, also substantial overclaiming of credit (because of higher "multiplier" used vs. Boinc 5.x distributed clients....Gent's, please correct me if this is not correct). This WU from your computer appears to show exactly this:

http://setiathome.berkeley.edu/worku...?wuid=80785923

You're claiming 40.69 w/ client 4.44 Super Bench and computer# 2436605 is claiming 12.41 using the distributed Boinc 5.4.9 and computer #239002 is claiming the same 12.41 using 5.2.13.

The subject of "claimed credit" is an extremely passionate one right now given the new credit structure w/ Enhanced and the resultant longer WUs and disproportionate lower credit awarded for time spent. To sum it up, people used to receiving a lot of credit w/ Seti Std from running opt. clients and aps have seen their RACs fallen more sharply than those using the standard clients and aps from Berkeley. The new credit system also makes overclaiming stand-out like a sore thumb. There have been recent inquiries/suggestions/demands that Berkeley should require clients of at least 5.x to process Enhanced WUs and would do this by denying work to older clients. There have also been demands for overclaimed work to be "adjusted" or removed from the totals. There are number of issues the Berkely has to deal w/ right now w/ various priorities but given the massive consternation by the most vocal (and substantially productive group) members you will see on the Seti and team forums, the "credit issue" looms large and is perceived to be perhaps the #1 issue and source of frustration in the active Seti communities.

Depending upon your views of Seti, your participation, motivations, role in community etc. you may want to consider going back to 5.2.13....or not. There is a very good liklihood you will soon need to use a min. of 5.x anyway. It's your call, I am not passing judgement, only offering an explanation to your observations and educating on recent developments w/ the new migration to Enhanced. Regardless, this does seem to explain your observations.
Good luck!
( Last edited by Gecko_r7; Jun 6, 2006 at 12:17 AM. )
     
arkayn
Dedicated MacNNer
Join Date: Aug 2005
Location: Golden Valley, AZ
Status: Offline
Reply With Quote
Jun 5, 2006, 10:42 PM
 
Interesting read there, I have always run pretty much the newest thing that was out there. I never did get into the super bench distributions.
     
alexkan
Forum Regular
Join Date: Aug 2005
Location: Cupertino, CA
Status: Offline
Reply With Quote
Jun 6, 2006, 02:32 AM
 
Alright, I have some good news and some bad news.

First, the good news. I've finished v5, complete with Crunch3r-style stderr output. The current binary is optimized for G5 but will still run faster on G4s than any previous versions. I'll still come out with a G4-specific compile, but for now, I'll make sure that the quads take their rightful places on the Top Computers list. Give it a try, post reference WU times, etc.

http://tbp.berkeley.edu/~alexkan/set...ced-ppc-v5.zip

The source will be up shortly, although I actually played with compile flags this time, so hopefully recompilation won't be needed for G5. G4s need more than recompilation, so don't bother.

And now, the bad news. It's not going to get much faster. I estimate that at most, I can get maybe 5% out of this if I go back and schedule every single Altivec loop I've written optimally, and frankly, I don't have the time or patience to do this. Real Life calls, after all.

I'm thinking about whether or not to post this in the main SETI boards. I'd hate to rub salt in the wounds of Windows crunchers reeling from Crunch3r's departure, but Mac users need their optimized clients, too. Also, this comment by Crunch3r in this thread needs to be set straight:
Regarding the G4/G5 macs... well i know that alex kan is looking at it but i wouldn't expect much speed improvements.
Of course, I wouldn't complain if one of you decided to post the link there before I finish deciding.
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 6, 2006, 02:44 AM
 
Originally Posted by Bad to the bone
Look at this WU!

http://setiathome.berkeley.edu/worku...?wuid=80752856

My 2.5 GHz G5 is the one with 1,952 seconds. I wish, ALL would be like this.
I am amazed you got credit for it. The error it shows basically says the Client (SETI) stoped responding or communicating with BOINC for over 30 Seconds and the WU processing died, so your submittion should have been deamed invalid unless it really was finished and stalled at the very end. If that is the case it's a great time, but likely you were lucky to get points.
     
rhettmaxwell
Fresh-Faced Recruit
Join Date: Jun 2006
Status: Offline
Reply With Quote
Jun 6, 2006, 03:35 AM
 
Originally Posted by BTBlomberg
I am amazed you got credit for it. The error it shows basically says the Client (SETI) stoped responding or communicating with BOINC for over 30 Seconds and the WU processing died, so your submittion should have been deamed invalid unless it really was finished and stalled at the very end. If that is the case it's a great time, but likely you were lucky to get points.
Yeah, after I had a little more time to investigate into this behavior I reached the same conclusion like you and others. It's really interesting to see how different the 4.4.4 client behaves on various machines. I actually stuck to this old workhorse to compare performance before upgrading but on this box I accidentally launched it instead of 5.2.13 after another failed attempt to run 5.4.9 and was surprised what happened.

Thanks also to Gecko for elaborating on the various aspects involved! I've read many of the posts regarding the credit system and it's getting a little bit ridiculous at times.

But, part of the motivation to participate in BOINC projects is accumulating credits for various reasons. Otherwise, the whole line of distributed computing projects would've never got off the ground, I'm afraid...
     
applefanss
Registered User
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 6, 2006, 12:10 PM
 
Thanks, Alex, v5 goes like the proverbial clappers. I don't have the means to provide you with precise stats, but just 2 units from the quad (202253)

http://setiathome.berkeley.edu/worku...?wuid=80647824

and

http://setiathome.berkeley.edu/worku...?wuid=80774400

show a gain of some 20% over v4. That's a great contribution to the project, like all your previous efforts without which the platform, the team and all of us would look very ordinary.

Let's hope your new employers appreciate their catch.
     
arkayn
Dedicated MacNNer
Join Date: Aug 2005
Location: Golden Valley, AZ
Status: Offline
Reply With Quote
Jun 6, 2006, 12:38 PM
 
Just switched over to v5 and will see what it does on my eMac.

I am running Crunch3r's SSE2 app on both of my Dells, so far they are outstripping my eMac.
     
liebsmaschine
Registered User
Join Date: Jul 2006
Status: Offline
Reply With Quote
Jun 6, 2006, 01:29 PM
 
Originally Posted by alexkan
I've finished v5, complete with Crunch3r-style stderr output. The current binary is optimized for G5 but will still run faster on G4s than any previous versions. I'll still come out with a G4-specific compile, but for now, I'll make sure that the quads take their rightful places on the Top Computers list. Give it a try, post reference WU times, etc.
Ah, Alex, we love you. How can we ever repay you? :-)

I'll hold off until you post the G4-specific compile (to minimize the hassle of timing it right not to interfere with currently-processing WUs and then restarting BOINC, etc.). But thanks in advance for everything!

One other note (not addressed to Alex in particular): My <a href="http://setiathome.berkeley.edu/results.php?userid=8344130">results</a> show a bunch of Client Error WUs. However, I don't think it's Alex's client, because if you look at the Work Unit ID for the ones that have errored out, every computer that has attempted to process the WU has returned a Client Error. Anyone know what's up? Why is Berkeley sending out so many faulty WUs?
     
Gecko_r7
Forum Regular
Join Date: Oct 2005
Location: Las Vegas, NV
Status: Offline
Reply With Quote
Jun 6, 2006, 01:33 PM
 
Originally Posted by alexkan

Of course, I wouldn't complain if one of you decided to post the link there before I finish deciding.

Thanks Alex for V5, every bit helps. You've been so good to us and this group. I know we all wish you well in your new job At the same time, I can tell you personally that reality sometimes creeps in way to far and it's nice to have a diversion that provides a creative outlet. We hope you will continue to bless us w/ the benefit of your creativity when you feel the urge. Your efforts very much encourage and fuel the passion that many of us have for the project. You help keep it real and fun! This forum is a great gathering place of positive attitudes and "can do" spirit.

The main Seti forum is dreadful at the moment and not a very encouraging environment to spend much time.
Any newbe's checking it out could only be disuaded from joining the project.
I'll post the link as a "beacon of light" for the lost PPC crunchers who may be wandering in the current darkness of the main forum.

BTW, I PM'd you w/ an interesting read. File is dowloadable at Megaupload (you may have to click the upright link on the interim page to get to the actual download page.) If there is a problem, e-mail me at my hotmail in the PM or let me know a different way to send it to you.
Sorry, I don't have a personal host site.
     
alexkan
Forum Regular
Join Date: Aug 2005
Location: Cupertino, CA
Status: Offline
Reply With Quote
Jun 6, 2006, 02:26 PM
 
Originally Posted by jackal
I'll hold off until you post the G4-specific compile (to minimize the hassle of timing it right not to interfere with currently-processing WUs and then restarting BOINC, etc.). But thanks in advance for everything!
I wouldn't hold off on the transition if I were you. I think the performance gain over v4 is already enough to negate the inconvenience (I was of the impression that the transition is less troublesome than in the past?), and I don't have a rigid timeframe for the G4 version, anyway.
     
Andrew F
Guest
Status:
Reply With Quote
Jun 6, 2006, 04:03 PM
 
So Alex, does this new build need wisdom files? Is it 10.3 compatible? Do we just replace app and xml file?
     
alexkan
Forum Regular
Join Date: Aug 2005
Location: Cupertino, CA
Status: Offline
Reply With Quote
Jun 6, 2006, 04:09 PM
 
Originally Posted by Andrew F
So Alex, does this new build need wisdom files?
No wisdom files are needed, although you can provide pregenerated ones for a tiny performance gain.
Is it 10.3 compatible?
The build is compatible with 10.3.9, as are all its predecessors.
Do we just replace app and xml file?
Yes.
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 6, 2006, 04:43 PM
 
Originally Posted by alexkan
I wouldn't hold off on the transition if I were you. I think the performance gain over v4 is already enough to negate the inconvenience (I was of the impression that the transition is less troublesome than in the past?), and I don't have a rigid timeframe for the G4 version, anyway.
BOINC and the WU see it as the exact same app so there will be no issue. I just switched with one in the last 2 hours of Est. 9:30 hour crunch. Hopefully I will see this mess up the estimate and crank through the rest.
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 6, 2006, 05:08 PM
 
Originally Posted by alexkan
And now, the bad news. It's not going to get much faster. I estimate that at most, I can get maybe 5% out of this if I go back and schedule every single Altivec loop I've written optimally, and frankly, I don't have the time or patience to do this. Real Life calls, after all.
I wish I saw your post last night after I posted. 14 hours of crunching optimized missed. Oh well.

Thanks again Alex. As for the quote above, I think we all know there is a wall out there to how optimized you can get this. Any extra edge you can give us we appreciate.

Does anyone have a general estimate of how much faster v5 may be over the stock client? We have had a little here and a little there. It would be nice to know start to finish.
     
Gecko_r7
Forum Regular
Join Date: Oct 2005
Location: Las Vegas, NV
Status: Offline
Reply With Quote
Jun 6, 2006, 06:51 PM
 
Originally Posted by Bad to the bone
But, part of the motivation to participate in BOINC projects is accumulating credits for various reasons. Otherwise, the whole line of distributed computing projects would've never got off the ground, I'm afraid...
Yes, there is a truth to your comment. By using dist. compt. to go beyond the scope of paid professionals and the dedicated groups of core supporters, Seti had to come-up w/ a reward system to expand it's appeal to casual volunteers. Like anything, if one can create a contest and foster friendly competiton in the mix, it generally can accelerate appeal and production. With more people, come more motivations and expectations. "Credit awards" may carry very little significance vs. the fundemental intent and focus of the project as far as Seti project professionals are concerned, but it's THE prime motivator for a sizable portion of the contributors that Seti now so heavily relies. It may be considered an evil by some, but it's a necessary evil to appeal to the masses. Hopefully, they will iron-out the bugs and this whole controversy will calm down so everyone can focus on the science again.
Good crunching!
     
Todd Madson
Mac Elite
Join Date: Apr 2000
Location: Minneapolis, MN USA
Status: Offline
Reply With Quote
Jun 6, 2006, 07:26 PM
 
Some things I've noticed about V4 before upgrading to V5 - it was speeding up.
How cool is that?

It was producing work units in the 3 hours and 30 minute range and towards
the end of its short lifespan we had 2 hrs 50 mins to 3 hours 10 minutes.

There's a 3:08 block at 100% still on my screen ready to submit.

So what I'm seeing so far on my machines' screen seems VERY encouraging
with regards to work unit production. Stay tuned - I've got an errand to run
and will be back with work unit URLs to see the times,
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 6, 2006, 11:44 PM
 
Alex and I talked about he wisdom being rewritten. Maybe the wisdom fine tunes over time like the Trux Boinc, it takes a few for it to calibrate.

Now it has also been discussed that all WUs ar not created equal and some are weighted differently accordingly. I have had a lot of WUs that say they will be done in 15 hours and then recently got a few mixed in that said only 10 or 7 hours and they run in about that time frame. So keep that in mind that they may have not been fully comparible also. Enhanced is all over the board compared to the old SETI.

It would be nice if it does speed up a little though.
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 6, 2006, 11:48 PM
 
Code:
... OS X optimized [email protected] Enhanced application by Alex Kan Version info: OS X Altivec (G4/G5, G5-optimized) V5.13 by Alex Kan Work Unit Info: ............... WU true angle range is : 0.920436 OS X optimized [email protected] Enhanced application by Alex Kan Version info: OS X Altivec (G4/G5, G5-optimized) V5.13 by Alex Kan ...
That's nice. Will tell those looking at those G5 results what hit them.
     
beadman
Dedicated MacNNer
Join Date: Nov 2004
Location: Virginia
Status: Offline
Reply With Quote
Jun 7, 2006, 12:02 AM
 
Here's the first one for me on v5. http://setiathome.berkeley.edu/resul...ltid=336859060 It's on a PowerBook G4 1.67 GHz, OS X 10.4.5, 1 GB RAM. I switched from V4 about half way through. Even though it died, it still was valid!

Thanks for all your efforts, Alex!
beadman
     
Todd Madson
Mac Elite
Join Date: Apr 2000
Location: Minneapolis, MN USA
Status: Offline
Reply With Quote
Jun 7, 2006, 12:03 AM
 
See:
http://setiathome.berkeley.edu/resul...6496&offset=20
Some client errors happening - three to be specific primarily due to bad headers.

But first results are in:
http://setiathome.berkeley.edu/resul...ltid=337573544
8517.25 seconds for that work unit

And:
http://setiathome.berkeley.edu/resul...ltid=337616826
9,444.80 seconds for that work unit

Work units were generally in the 11 to 18,000 second range prior to this!
See the first link to get specific timings.
Excellent work, Alex!
     
wrhowell
Fresh-Faced Recruit
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 12:21 AM
 
Originally Posted by alexkan
Alright, I have some good news and some bad news.

First, the good news. I've finished v5, complete with Crunch3r-style stderr output. The current binary is optimized for G5 but will still run faster on G4s than any previous versions. I'll still come out with a G4-specific compile, but for now, I'll make sure that the quads take their rightful places on the Top Computers list. Give it a try, post reference WU times, etc.
Here's about 10 WU's from host 1945763 (Quad) that are completely running v5. Credit on some are pending waiting for other hosts to finish.
http://setiathome.berkeley.edu/worku...?wuid=80926771
http://setiathome.berkeley.edu/worku...?wuid=80952912
http://setiathome.berkeley.edu/worku...?wuid=80946595
http://setiathome.berkeley.edu/worku...?wuid=80939487
http://setiathome.berkeley.edu/worku...?wuid=80938304
http://setiathome.berkeley.edu/worku...?wuid=80935473
http://setiathome.berkeley.edu/worku...?wuid=80932617
http://setiathome.berkeley.edu/worku...?wuid=80929749
http://setiathome.berkeley.edu/worku...?wuid=80925998
http://setiathome.berkeley.edu/worku...?wuid=80925148

It looks like times have come down from v4.
     
fran
Baninated
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 01:03 AM
 
Hello there!

I have some questions. I have a MacBook Pro. I see that there are some Beta clients at:

http://members.dslextreme.com/~reade...boincbeta.html

1) I see that they are CLI. While I guess I could use these if someone provided instructions, I like using the GUI. Any word on if/when an intel GUI version will be available?

2) If I wanted to give these a shot, are there instructions somewhere on how to use them? The download includes only a binary.

3) What kind of performance increase should I expect if I used them?

4) Superbench vs. non-Superbench? What's Superbench? And why would I pick one over the other?


Thanks!
Erik
     
Gecko_r7
Forum Regular
Join Date: Oct 2005
Location: Las Vegas, NV
Status: Offline
Reply With Quote
Jun 7, 2006, 01:32 AM
 
Hi all,

Ran v5 vs. ref WU on G4 DP 1.33

AlexKan v5 (Test4 6.6.06)
real 276m12.600s ( =16572.600s)
user 275m38.475s
sys 0m32.771s
wu_cpu_time = ????

v5 appears 15% faster than v4 against ref WU on my G4 and 80%!!! faster than std ap.
WOW!!!

No problems in the WU. Stderr msg. is nice touch.
http://setiathome.berkeley.edu/resul...ltid=337366110

Thanks a bunch Alex!:
Excited to try the G4 compile at some point when you finish.
Lunch is on me my next trip to Bay Area!

Regards,
Ian
     
Gecko_r7
Forum Regular
Join Date: Oct 2005
Location: Las Vegas, NV
Status: Offline
Reply With Quote
Jun 7, 2006, 02:44 AM
 
Originally Posted by zombie67
Hello there!

I have some questions. I have a MacBook Pro. I see that there are some Beta clients at:

http://members.dslextreme.com/~reade...boincbeta.html

1) I see that they are CLI. While I guess I could use these if someone provided instructions, I like using the GUI. Any word on if/when an intel GUI version will be available?

2) If I wanted to give these a shot, are there instructions somewhere on how to use them? The download includes only a binary.

3) What kind of performance increase should I expect if I used them?

4) Superbench vs. non-Superbench? What's Superbench? And why would I pick one over the other?


Thanks!
Erik
If I understand your question properly and you have an Intel-based Mac Book Pro, you are pretty much limited to the distributed science application from Berkeley that you received when you connected to the project. One of our team's main developers, Rick, is working on an optimized (read: faster) science application suitable for Intel based Macs. Performance increase will depend on how how/what he tweaks, but the project developers and testers are claiming that Seti Enhanced is already substantially more optimized than the old project version that completed in May and that we should not expect the orders of magnitude speed improvements as in the past. One Seti developer has said there might be enough headroom for a doubling of speed longterm given substantial time & effort. x86 crunchers are seeing @ +20%-30% improvement w/ one of the first opt. aps recently created, but no longer available presently.

See this thread and monitor the forum for updates.
http://forums.macnn.com/showthread.php?t=292097
Rick normally includes install directions on releases, and if you have trouble, someone on the forum will help walk you through it.

Not much need for Super Bench clients any more w/ Seti Enhanced since claimed credit is no longer dependent on benchmarking. See my response to "Bad to the bone" earlier in the day in this thread. If you are running Einstein, a Super Bench client may benefit you, but you should first check-out the Einstein forum to see what the recommendations are for that project.

Hope this helps some.
Regards.
     
fran
Baninated
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 05:06 AM
 
Thanks. I think I understand the meaning of SuperBench now.

I guess I am trying to understand *how* to use these optimized clients. Are there any good instructions anywhere?

This is where I'm at now: I also have a couple of G4 laptops. So on one, I just finished off the running WU, wiped the whole thing, and then installed the CLI from Berkley. Runs fine, now that I figured out that the account key is different from the Cross-Project ID. Similar looking strings. Anyway...

Then I downloaded V5 at <http://tbp.berkeley.edu/~alexkan/seti/seti_enhanced-ppc-v5.zip>.

So my question now is, how do I use these files? Do I need to move them into a particular directory? One is a binary file, so I am guessing that I need to replace some default Berkeley file this this. But which one? And do I need to rename the new one to match the old? Or do I need to run the old binary with some argument referencing the new binary? And wat's up with the xml file? This is where I get lost, and can't seem to find instructions or rtfm that explains it.

Thanks!
     
sdubz  (op)
Junior Member
Join Date: Jun 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 06:05 AM
 
Originally Posted by Gecko_r7
I'll post the link as a "beacon of light" for the lost PPC crunchers who may be wandering in the current darkness of the main forum.
Oops, I didn't see that you were going to post this over there, and I think I beat you to it

You didn't need to erase it tho! We could have had 2 happy threads going over there!



And, Alex, your worker is awesome! I can't wait to see what a g4 version would do, because this one seems fast!
     
sdubz  (op)
Junior Member
Join Date: Jun 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 06:13 AM
 
Originally Posted by zombie67
So my question now is, how do I use these files? Do I need to move them into a particular directory? One is a binary file, so I am guessing that I need to replace some default Berkeley file this this. But which one? And do I need to rename the new one to match the old? Or do I need to run the old binary with some argument referencing the new binary? And wat's up with the xml file? This is where I get lost, and can't seem to find instructions or rtfm that explains it.

Thanks!

When you have Alex's v5 downloaded you first need to quit boinc, then you can put the files from his download into your startup disk>Library>Application Support>BOINC data>Projects>setiathome.berkeley.edu folder. If your using the command line version, you need to start in the folder where you have the boinc executeable then go into the >Projects>setiathome.berkeley.edu folder.

Then start boinc back up and hit the messages tab, and if you see something like this: "Tue Jun 6 20:59:42 2006|[email protected]|Found app_info.xml; using anonymous platform" then you did it right!
     
liebsmaschine
Registered User
Join Date: Jul 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 07:00 AM
 
Originally Posted by boog
When you have Alex's v5 downloaded you first need to quit boinc, then you can put the files from his download into your startup disk>Library>Application Support>BOINC data>Projects>setiathome.berkeley.edu folder. If your using the command line version, you need to start in the folder where you have the boinc executeable then go into the >Projects>setiathome.berkeley.edu folder.

Then start boinc back up and hit the messages tab, and if you see something like this: "Tue Jun 6 20:59:42 2006|[email protected]|Found app_info.xml; using anonymous platform" then you did it right!
Boog's right that installing Alex's packaged files in the directory will work. However, you said you have a MacBook Pro and were looking at some Intel versions, if I remember correctly. If you've found some optimized Intel app somewhere and are trying to install it, the most important files to get in your [boinc directory]/projects/setiathome.berkeley.edu folder are the worker app (seti-enhanced-ppc-v5, in our case, but if it's a Core Duo build, it'll be different) and the app_info.xml file. The rest are just extras. The app_info.xml file tells your BOINC installation to use the seti-enhanced-ppc-v5 application instead of the default setiathome_5.13_powerpc-apple-darwin worker application. (You don't even need to delete the default app--in fact, when I changed to Alex's, BOINC deleted it for me when it was finished with it.) If you're dealing with an Intel-optimized app, download and open Alex's app_info.xml file (it's in his zip package) and replace all instances of the seti-enhanced-ppc-v5 with the name of the app you have.

I have an alternate way to ensure that the optimized app is crunching (I didn't know there'd be a message in the BOINC message tab): in Terminal, type "ps -ax | grep seti". If you see the default worker listed (setiathome_5.13_powerpc-apple-darwin), it didn't work. If you see the name of the custom optimized app, you're in business.

Also, for what it's worth (and this isn't just for you, zombie), I find the best way to run BOINC is with a CLI installation, as it lets you leave it running all the time in the background, no matter who's logged in or if you close all your GUI applications. (You can still use BOINC Manager to interact with the CLI-installed client.) I use a crontab entry to make sure that BOINC starts up as soon as possible and restarts if it accidentally dies. Here's how I do that:

1. In Terminal, type "crontab -e". On my system, it opens the crontab file in the vi editor, which is extremely difficult to use (I like pico better--I know there is a way to change the default crontab editor, but I couldn't find it offhand).
2. If it opened in vi, type "i" (for insert)
3. Type (or edit, copy, and paste) something similar to the following:
0,15,30,45 * * * * cd [boinc directory]; ./boinc -redirectio
(The numbers and stars at the beginning attempt to start BOINC every 15 minutes every hour/day/week/month (you can change this as desired). I could not get the @reboot time entry to work, so this is the next best thing.)
4. Hit the Escape key
5. Type ":wq" (that's colon-w-q) to save the crontab and quit vi. Boinc should now attempt to start every 15 minutes.

Getting the BOINC Manager GUI app to talk to an existing CLI installation is a little difficult. I can't remember exactly what I had to do, especially as the BOINC Manager has its own BOINC client and tries to start that up (I think I went in the application package and deleted it). I think I also had to add a gui_rpc_auth.cfg in the BOINC directory with a custom password, but I don't know if that's required, and there may have been a second file I had to add with my IP address, though I can't find it offhand. Then, in the BOINC Manager app, you'll use the Advanced>Select computer... option, type in "localhost" and your custom password you added in the gui_rpc_auth.cfg file, and that *should* work (but my memory is fuzzy).

On second thought, if you're not mildly comfortable in Terminal and with all this stuff, the GUI-only setup may be better for ya...

And please, anyone point out some inefficiencies in the way I'm doing it! I always like to learn new tricks...
     
Gecko_r7
Forum Regular
Join Date: Oct 2005
Location: Las Vegas, NV
Status: Offline
Reply With Quote
Jun 7, 2006, 11:34 AM
 
Originally Posted by boog
Oops, I didn't see that you were going to post this over there, and I think I beat you to it

You didn't need to erase it tho! We could have had 2 happy threads going over there!
No worries Boog. I had to wait through the server outage and somehow didn't see your post when in came back up. Wouldn't you know I caught it right after I posted the link and loading instructions however. All that matters is that it's up and available. Good post!
     
fran
Baninated
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 11:36 AM
 
Thanks for the Help Jackle & Boog. But I think I'm even more confused. Just to be clear, I gave up on optimizing my intel MackBook Pro for the time being. For now, I am trying to optimize my G4 PowerBook.

Are you saying that I *can* use seti-enhanced-ppc-v5 with the GUI BOINC Manager? I thought that wasn't possible? If it can, that would be the best for me. I tried putting the files in startup disk>Library>Application Support>BOINC data>Projects>setiathome.berkeley.edu folder. BOINC Manager downloaded work, but wouldn't start the work.


As for the CLI, I don't understand what this means: "you need to start in the folder where you have the boinc executeable then go into the >Projects>setiathome.berkeley.edu folder." I'm not sure the point of changing directories after you have already started boinc. "Then start boinc back up and hit the messages tab"...huh? What tab? Is the CLI supposed to launch a GUI of some sort? I'm really confused now.

Can someone please show me the full string they use to start boinc with seti-enhanced-ppc-v5, including all the arguments?

Thanks! ...and sorry to be such a pain.

Erik
     
reader50
Administrator
Join Date: Jun 2000
Location: California
Status: Offline
Reply With Quote
Jun 7, 2006, 12:57 PM
 
zombie67, you can also read this install/optimization guide for SETI written by TiloProbst. It may answer your questions in more detail, and it has pictures.
     
fran
Baninated
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 02:10 PM
 
Originally Posted by reader50
zombie67, you can also read this install/optimization guide for SETI written by TiloProbst. It may answer your questions in more detail, and it has pictures.
Thanks for the link. I read through it all, and I think it answers some of my questions.

1) Yes, you *can* use seti_enhanced-ppc-v5 with BOINC Manager. CLI is not a requirement.

2) To install, put seti_enhanced-ppc-v5 and app_info.xml in /Library/Application Support/BOINC Data/projects/setiathome.berkeley.edu, and delete setiathome_5.13_powerpc-apple-darwin. (I tried this last night, but it didn't work. Not sure why. It seems to be working this time.)

Right so far?

Hopefully, this will be my last question for a while: How do I know it's really using seti_enhanced-ppc-v5? When I restarted BOINC Manager after the installation, I didn't see any message with "seti_enhanced-ppc-v5" in it. There is one line that says this:

"Resuming task 29mr99aa.17011.3665.267324.3.57_1 using setiathome_enhanced version 513"

...which makes me suspicious because that looks very similar to the file I deleted.

Thanks again,
Erik
     
Knightrider
Dedicated MacNNer
Join Date: Sep 2004
Location: London
Status: Offline
Reply With Quote
Jun 7, 2006, 02:25 PM
 
Originally Posted by zombie67
...which makes me suspicious because that looks very similar to the file I deleted.
If you are using the current GUI BOINC 5.4.9 and you shut it down and then put the files in the SAH project folder, then restart BOINC, it writes copies of the worker to the slots folders that it uses. The previous version you were using should still be in the slots folder as well, so that it can finish the wu it was doing before it starts the next one with the new version. It will clean up when it has finished.

K.
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 7, 2006, 03:19 PM
 
Originally Posted by zombie67
"Resuming task 29mr99aa.17011.3665.267324.3.57_1 using setiathome_enhanced version 513"

...which makes me suspicious because that looks very similar to the file I deleted.
Internally the client is called the same as the stock,only the external file name is different. Because of this you can replace the app mid stream without errors as to Boinc it is the same app. I know this is contrary to what Knightrider said, but i know from experience as do a few others in this thread that this does work (no offence to Knightrider). That was how the Previous SETI clients worked.

The reason it probably did not work for you last night is you may have forgotten to quit Boinc Manager before installing v5. This would have the Origional app still running so it could not be replaced in memory.

Also, the CLI install is like the GUI one exept not manager I would expect. You can run v5 as either, but Boinc Manager does not take too much power away and it gives you a good way to manually force results to upload as either SETI or BOINC are having issues with timely uploads. I end up ofter clicking on Update to get finished WU uploaded as I get impatient.
     
Knightrider
Dedicated MacNNer
Join Date: Sep 2004
Location: London
Status: Offline
Reply With Quote
Jun 7, 2006, 03:31 PM
 
Originally Posted by BTBlomberg
Internally the client is called the same as the stock,only the external file name is different.
I was not aware of the internals, and was speaking of the file names only, with the effects that I had observed in both slots and processing, so no offense taken, rather, knowledge extended.

K.
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 7, 2006, 03:48 PM
 
Well this is somewhat extrapolated from experince and not really on certain facts, but I know that BOINC calls the Stock and Alex's Clients the same thing from watching the messages and the task data. Replace the client and it show up the same. The WUs seam to be linked to the Client type on Download and somehow "written" to them. If you change client versions typically it will kill and dump all downloaded WUs and download new ones. I experinced this an all my PCs when I installed Crunch3r's clients. The wrong client caused an Error for the WU.

This is where I assume the internal name/version is the same and the external name does not effect things.
     
Todd Madson
Mac Elite
Join Date: Apr 2000
Location: Minneapolis, MN USA
Status: Offline
Reply With Quote
Jun 7, 2006, 04:06 PM
 
Update:
http://setiathome.berkeley.edu/resul...6496&offset=20

See this? Recent timings in the last dozen or so work units:
9417.25, 6332.11, 6267.43, 6627.30, 7972.14, 8759.14,
2718.61 2787.13, 9175.38, 9138.46, 9287.46, 9274.62,
2240.05, 5983.15, 9185.81 etc.

Three units in the six thousand seconds range, three in the
two thousand range. Very impressive! The remainder are
well below the consistent eleven thousand second average I
seemed to be getting with v4.

Alex is the bomb!
     
beadman
Dedicated MacNNer
Join Date: Nov 2004
Location: Virginia
Status: Offline
Reply With Quote
Jun 7, 2006, 04:12 PM
 
Originally Posted by zombie67
Hello there!

I have some questions. I have a MacBook Pro. I see that there are some Beta clients at:

http://members.dslextreme.com/~reade...boincbeta.html

1) I see that they are CLI. While I guess I could use these if someone provided instructions, I like using the GUI. Any word on if/when an intel GUI version will be available?

2) If I wanted to give these a shot, are there instructions somewhere on how to use them? The download includes only a binary.

3) What kind of performance increase should I expect if I used them?

4) Superbench vs. non-Superbench? What's Superbench? And why would I pick one over the other?


Thanks!
Erik
Erik:
If you take a look at this thread, about 8 or 10 posts down, http://forums.macnn.com/showthread.php?t=296948, I explaiined how to replace the stock boinc client with the superbench CLI one, and still be able to run the BOINC Manager GUI app.

beadman
     
halimedia
Dedicated MacNNer
Join Date: Oct 2005
Location: Switzerland
Status: Offline
Reply With Quote
Jun 7, 2006, 04:29 PM
 
I'd like to chime in with my thanks to Alex! Thanks to his tremendous improvement over the stock worker, my quad has been able to claim the crown in the SETI top computers list (if you discount the cheat that's holding the #1 position). Much appreciated!

NB: I'd like to point out a few naming conventions associated with SETI and BOINC that have been used incorrectly in this thread, and could therefore result in misunderstandings.

When using the term client, it is generally understood that this refers to the BOINC client software (BOINC Manager or CLI). It is the client application of the BOINC distributed computing infrastructure, and handles work unit downloads, scheduling, credit accounting, result uploads and reporting, among other things.

The application that carries out the SETI number-crunching (or the crunching of any other project, for that matter) is referred to as the worker.

There are both optimized BOINC clients (such as the CLI-clients discussed on this page) and optimized workers (such as Alex' wonderful creations). The former are only necessary if you are contributing to certain projects other than SETI and you are using an optimized worker. The latter are, of course, beneficial to the project and to your ability to accrue credits faster, as we all know.

HTH,

Ron
     
chboss
Fresh-Faced Recruit
Join Date: May 2005
Location: Switzerland
Status: Offline
Reply With Quote
Jun 7, 2006, 04:36 PM
 
Uhhh.. If Ron is running the new worker created by Alex I need to speed up my production...
I already feel his breath again in my Neck.

Thanks Alex for your hard work!
Chris Bosshard
www.bosshard-ch.net
     
fran
Baninated
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 08:33 PM
 
I think I can now answer my last question: How do I know it's really using seti_enhanced-ppc-v5?

Answer: You can't while it's running, but you can once it completes and reports back. Look at the Result ID for the WU. My first one with seti_enhanced-ppc-v5 shows this for the stderr out:

<core_client_version>5.4.9</core_client_version>
<stderr_txt>
ar=0.433832 NumCfft=71817 NumGauss= 452130524 NumPulse= 86645276032 NumTriplet= 7550392287232
OS X optimized [email protected] Enhanced application by Alex Kan
Version info: OS X Altivec (G4/G5, G5-optimized) V5.13 by Alex Kan

Work Unit Info:
...............
WU true angle range is : 0.433832
OS X optimized [email protected] Enhanced application by Alex Kan
Version info: OS X Altivec (G4/G5, G5-optimized) V5.13 by Alex Kan

Work Unit Info:
...............
WU true angle range is : 0.433832

Flopcounter: 16036587113374.705078

Spike count: 0
Pulse count: 2
Triplet count: 0
Gaussian count: 1
</stderr_txt>


DING! Winner!

Thanks everyone for helping me through this!

Erik
     
fran
Baninated
Join Date: Aug 2006
Status: Offline
Reply With Quote
Jun 7, 2006, 09:17 PM
 
Each of my G4 laptops have finally reported a WU with seti-enhanced-ppc-v5. The improvements are amazing!

1.25ghz Powerbook:
WUs with a credit of ~61 consistantly took ~41000 seconds. With seti-enhanced-ppc-v5, it took only 22342 seconds.

1.33ghz iBook:
WUs with a credit of ~61 consistantly took ~41000 seconds. With seti-enhanced-ppc-v5, it took only 29146 seconds. And this one was already ~50% completed with the stock worker. I expect it will match the Powerbook.

Wow, wow, wow!

Erik
     
BTBlomberg
Forum Regular
Join Date: Sep 2005
Location: Chicago Suburbs
Status: Offline
Reply With Quote
Jun 7, 2006, 11:31 PM
 
I had my first full WU with v5 completed and it was a 60+. On my PowerBook G4 500 it took 45,731.05 seconds (12:42 hours) and got a 60.91 Credits. June 1st was my last 60+ with v4 completed in 55,098.07 seconds (15:18 Hours) and got 63.73 Credits.

Before that on the 31st one completed in 63,546.70 seconds (17:39 Hours) and got 63.57 Credits. It was with either the stock or v3 client. I think it was the one that started stock and after 1.5 hours I switched to v3 and then to v4 1/2 way through it. So it's a mess to compare to.

Here is the link to the latest v5 WU. If you look at the "stderr_txt" it repeats the new signature 9 times???
http://setiathome.berkeley.edu/resul...ltid=336377906

The stock client originally claimed it would do a WU that would have gotten in the 60s in 24 hours +/-. If this is the case then it would be roughly a 45-50% improvement.
( Last edited by BTBlomberg; Jun 7, 2006 at 11:39 PM. )
     
alexkan
Forum Regular
Join Date: Aug 2005
Location: Cupertino, CA
Status: Offline
Reply With Quote
Jun 7, 2006, 11:36 PM
 
Originally Posted by BTBlomberg
Here is the link to the latest v5 WU. If you look at the "stderr_txt" it repeats the new signature 9 times???
http://setiathome.berkeley.edu/resul...ltid=336377906
v5 prints its identifying information to stderr every time it starts. If you pause computation (and BOINC is set up to not keep suspended applications in memory), the application will print its information again once computation resumes. Crunch3r's app used to do this as well, I think, but the problem is exacerbated by the fact that your G4 500 takes a long time to crunch in the first place and winds up being interrupted quite a few times.
     
liebsmaschine
Registered User
Join Date: Jul 2006
Status: Offline
Reply With Quote
Jun 8, 2006, 04:18 AM
 
Originally Posted by zombie67
I think I can now answer my last question: How do I know it's really using seti_enhanced-ppc-v5?

Answer: You can't while it's running, but you can once it completes and reports back. Look at the Result ID for the WU. My first one with seti_enhanced-ppc-v5 shows this for the stderr out:
Well, you've answered your own question, but for future reference for others, you *can* tell. You can't tell inside the BOINC Manager GUI app, but you can tell either by the method I posted earlier:
Originally Posted by jackal
...in Terminal, type "ps -ax | grep seti". If you see the default worker listed (setiathome_5.13_powerpc-apple-darwin), it didn't work. If you see the name of the custom optimized app, you're in business.
So, you should see "seti-enhanced-ppc-v5". I forgot that, if you're not Terminally-inclined, you can do the same thing in Activity Monitor. Make sure (just in case) that "All Processes" is selected from the popup menu, and type in "seti" in the search field. That will show any applications with the word "seti" in them that are running (you'd have to do it while SETI is crunching, since the worker isn't normally active while you're crunching other projects--look up or ask someone about the merits of running with the "Leave applications in memory while preempted?" option set in the Global preferences online).
     
Todd Madson
Mac Elite
Join Date: Apr 2000
Location: Minneapolis, MN USA
Status: Offline
Reply With Quote
Jun 8, 2006, 06:25 AM
 
Wow. Even more speedups - I'm not sure how this is happening but it is:
http://setiathome.berkeley.edu/resul...6496&offset=20

Alex, the master of black necromancy in software! Where will it end?
     
Knightrider
Dedicated MacNNer
Join Date: Sep 2004
Location: London
Status: Offline
Reply With Quote
Jun 8, 2006, 08:23 AM
 
I think the ultimate speed up would be to have (on a quad) all four processors at work on crunching one wu at a time. Slurp - Gulp, Slurp - Gulp,

K.
     
Todd Madson
Mac Elite
Join Date: Apr 2000
Location: Minneapolis, MN USA
Status: Offline
Reply With Quote
Jun 8, 2006, 08:31 AM
 
But it typically hasn't worked like that - in the present case of my machine, both
processors work on a separate work unit at the same time.

On a quad, it works on four at the same time.

I wonder how you would set it up so that 2 or 4 processors would divide the
workflow. Would it create a measureable speedup?
     
 
Thread Tools
 
Forum Links
Forum Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Top
Privacy Policy
All times are GMT -4. The time now is 08:57 AM.
All contents of these forums © 1995-2017 MacNN. All rights reserved.
Branding + Design: www.gesamtbild.com
vBulletin v.3.8.8 © 2000-2017, Jelsoft Enterprises Ltd.,