Message boards : Number crunching : Not HAPPY people.
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · Next

AuthorMessage
Michael Karlinsky
Avatar

Send message
Joined: 18 Sep 04
Posts: 163
Credit: 1,682,370
RAC: 0
Message 13112 - Posted: 22 Mar 2006, 18:56:39 UTC - in response to Message 13111.  

I'm happy.


Me too.

Michael
Team Linux Users Everywhere
ID: 13112 · Report as offensive     Reply Quote
qaz79

Send message
Joined: 26 Nov 05
Posts: 9
Credit: 191
RAC: 0
Message 13113 - Posted: 22 Mar 2006, 19:00:04 UTC

try running other projects. i'm at least 8 or 9 of them.
ID: 13113 · Report as offensive     Reply Quote
arcturus

Send message
Joined: 27 Sep 04
Posts: 11
Credit: 116,314
RAC: 0
Message 13115 - Posted: 23 Mar 2006, 5:23:36 UTC

Let's talk about the other projects, shall we?

Climate - the linux client is HOSED, the solution uncertain, and this after months of being told it's a hardware problem.

Predictor - pathetic lack of communication, or maybe the 2004 update on the main page wasn't a good enough clue?

Rosetta - the 1% bug from hell. Everyone's blue in the face talking about it. Not a pleasant thing to discover your remote boxen have been idle for weeks.

Einstein - thumbs up! The Aksofp optimizing is impressive.

SIMAP - looks ok, low points. Wish I spoke more German.

Seti - we're alone. Live with it.
100% CPU credits.

ID: 13115 · Report as offensive     Reply Quote
qaz79

Send message
Joined: 26 Nov 05
Posts: 9
Credit: 191
RAC: 0
Message 13116 - Posted: 23 Mar 2006, 7:23:07 UTC

there's other ones like sztaki desktop grid and seti@home beta test. there are also other's in development alpha and beta stages. heres a link to some:

http://forum.setiuk.com/viewtopic.php?t=1507.
ID: 13116 · Report as offensive     Reply Quote
KWSN-GMC-Peeper of the Castle ...
Avatar

Send message
Joined: 6 Oct 05
Posts: 18
Credit: 952,091
RAC: 0
Message 13122 - Posted: 24 Mar 2006, 0:18:51 UTC

I won't run Climate. Way too many valid concerns about the quality of the data being collected.


ID: 13122 · Report as offensive     Reply Quote
Profile Ananas

Send message
Joined: 17 Jul 05
Posts: 102
Credit: 542,016
RAC: 0
Message 13131 - Posted: 26 Mar 2006, 22:50:35 UTC
Last modified: 26 Mar 2006, 22:58:43 UTC

Africa@Home (the Malaria project) seems to run reliable and QMC@Home didn't cause me trouble either.

For computers with less floating point power, SIMAP is a good thing, it needs more integer than float maths, they can catch up points there.

SZTAKI still is fighting a restart point problem but seems become more reliable too lately.

CPDN / BBC has a bug that isn't acknowledged as a bug, as long as it is possible, they will blame the hardware. It is incompatible to all kinds of software but no one accepts that the problem is in the CPDN client. I don't think that this is a good way to handle problems and although I think that it's an important project, it often makes me really mad and requires good nerves (it's one of my major projects) :-/

Einstein never caused me any trouble, it's reliable and always has enough work.

TANPAKU has an extremely(!!) short deadlines and currently wants a 4.x client - worth keeping an eye on it but no priority project for me.

SETI is currently not on my priority list - but I still like the project :-)


I cannot say much about the other projects - well, LHC has quite a high priority (the highest to be exact) on some of my boxes so they will jump on the boat when work is available.


(Still waiting for a Rothberg / TSC project under BOINC - but I guess I'm waiting in vain)
ID: 13131 · Report as offensive     Reply Quote
Maxxina

Send message
Joined: 1 Oct 04
Posts: 1
Credit: 53,202
RAC: 534
Message 13135 - Posted: 27 Mar 2006, 12:04:02 UTC

Or you should look to www.distributedcomputing.info and there you will find a very good info about running projects on internet :)
ID: 13135 · Report as offensive     Reply Quote
marshall

Send message
Joined: 28 Sep 04
Posts: 16
Credit: 2,678,563
RAC: 0
Message 13139 - Posted: 27 Mar 2006, 20:19:04 UTC - in response to Message 13122.  

I won't run Climate. Way too many valid concerns about the quality of the data being collected.


hi,
i run it as a backup when others (LHC, Predictor) dries out. could you point me to the source of the data quality concerns? just curious.
thanx


<br />
--
marshall
ID: 13139 · Report as offensive     Reply Quote
uioped1

Send message
Joined: 26 Aug 05
Posts: 18
Credit: 37,965
RAC: 0
Message 13143 - Posted: 28 Mar 2006, 1:47:17 UTC - in response to Message 13131.  

Rosetta@home is begging for more participants. The URL is http://boinc.bakerlab.org

This is one of the better projects that I have seen for communication from the project scientists and for acknoledgement of their contributors. I urge you all to check them out as they're doing great things!

-uio
ID: 13143 · Report as offensive     Reply Quote
senatoralex85

Send message
Joined: 17 Sep 05
Posts: 60
Credit: 4,221
RAC: 0
Message 13144 - Posted: 28 Mar 2006, 5:05:41 UTC - in response to Message 13143.  

Rosetta@home is begging for more participants. The URL is http://boinc.bakerlab.org

This is one of the better projects that I have seen for communication from the project scientists and for acknoledgement of their contributors. I urge you all to check them out as they're doing great things!

-uio

--------------------------------------------------------------------------

I am so disgusted with Rosetta that I will make a personal plea to everyone NOT TO JOIN that project. I clearly remember the day they moved up from beta. Everyone was soo pleased and thought it was making fast progress. Yet, the workunits are continually plagued with problems to this day. There is only so many mistakes to own up to before you just get sick of them (I did) and drop the project. What a waste of precious computer time! NO thanks!
ID: 13144 · Report as offensive     Reply Quote
Profile [B^S] Molzahn

Send message
Joined: 21 Jan 06
Posts: 46
Credit: 174,756
RAC: 0
Message 13145 - Posted: 28 Mar 2006, 5:24:32 UTC - in response to Message 13144.  
Last modified: 28 Mar 2006, 5:36:12 UTC

Rosetta@home is begging for more participants. The URL is http://boinc.bakerlab.org

This is one of the better projects that I have seen for communication from the project scientists and for acknoledgement of their contributors. I urge you all to check them out as they're doing great things!

-uio

--------------------------------------------------------------------------

I am so disgusted with Rosetta that I will make a personal plea to everyone NOT TO JOIN that project. I clearly remember the day they moved up from beta. Everyone was soo pleased and thought it was making fast progress. Yet, the workunits are continually plagued with problems to this day. There is only so many mistakes to own up to before you just get sick of them (I did) and drop the project. What a waste of precious computer time! NO thanks!

__________________
I must say, I joined Rosetta, and the first six or seven units failed. I read up and realized a reboot was a necessity. I detached, (upon a reboot), I will reattach. I have yet to think that the scientific gains could outweigh the minor problems with the project... Granted I haven't rebooted in nearly a month, I will not heed this warning. I think, after seeing a discussion on C-SPAN, this project is incredibly worth while and will only improve in the future. It seems people in this thread have been suspect of the potential for several projects, including LHC, I don't share this negativism. It's interesting no-one has mentioned malaria control.net; they are great in my opinion.

In hopes of more LCH units to crunch in the near future (if not, I will keep waiting [diligently]),
Mike Molzahn
Post Script: I may not have the experience and time that you have spent on the project, but the advances they are trying to achieve are far more important than minor (albeit consistent) errors IMHO.
I just don't see the point in a blatant call for the boycott of any project, no matter personal opinions of (possible) incompetence.

blog pictures
ID: 13145 · Report as offensive     Reply Quote
Profile anarchic teapot
Avatar

Send message
Joined: 15 Feb 06
Posts: 67
Credit: 460,896
RAC: 1
Message 13150 - Posted: 28 Mar 2006, 12:18:36 UTC - in response to Message 13139.  

I won't run Climate. Way too many valid concerns about the quality of the data being collected.


hi,
i run it as a backup when others (LHC, Predictor) dries out. could you point me to the source of the data quality concerns? just curious.
thanx


I don't know about the quality of the data, but the WUs take much too long (the 1-year expiry date is a clue), although there are regular "trickles" to feed back, and I've seen too many go into unrecoverable error.

I suspect it needs to run on much more powerful computers, or that the clients be severely optimised for every single type of chip (clients for P2, P3, P4, AMD K7, AMD64, Xeon, G3, G4, G5...). Not very practical.

It's a great shame, as this sort of study on both past and present climate variation is getting really urgent. Meanwhile, when my current WU finishes, I won't renew, although I'll keep the account info for future use.


sQuonk
Plague of Mice
Intel Core i3-9100 CPU@3.60 GHz, but it's doing its bit just the same.
ID: 13150 · Report as offensive     Reply Quote
Professor Desty Nova
Avatar

Send message
Joined: 27 Sep 04
Posts: 34
Credit: 199,100
RAC: 1
Message 13151 - Posted: 28 Mar 2006, 14:24:50 UTC

With Climate, you are runing a an "adapted" 64bit climate model, with millions of fortran code lines, that has originally been coded to run on supercomputers. Since the model started being coded in the 70s (I think I read it somewere) I guess we now all have supercomputers at home ;-)


Professor Desty Nova
Researching Karma the Hard Way
ID: 13151 · Report as offensive     Reply Quote
River~~

Send message
Joined: 13 Jul 05
Posts: 456
Credit: 75,142
RAC: 0
Message 13152 - Posted: 28 Mar 2006, 15:37:06 UTC - in response to Message 13144.  
Last modified: 28 Mar 2006, 15:43:19 UTC


I am so disgusted with Rosetta that I will make a personal plea to everyone NOT TO JOIN that project. I clearly remember the day they moved up from beta. Everyone was soo pleased and thought it was making fast progress. Yet, the workunits are continually plagued with problems to this day. There is only so many mistakes to own up to before you just get sick of them (I did) and drop the project. What a waste of precious computer time! NO thanks!


I am so disgusted with your posting that I'd counter that with a plea for everyone to join Rosetta.

1. They have the best communication with users of any BOINC project

2. They did make the mistake of moving up from Beta too soon, the project coders were all mainframe types who did not initially have the experience of DC (distributed computing), and therefore did not appreciate just how many different bugs pop out of the woodwork in a DC project.

3. They have apologised to users about this.

4. They have taken steps to make sure it does not happen again.

5. They have accepted advice from users about the standards of testing users expect.

6. There is now a separate "Ralph" (Rosetta Alpha) project where all new apps get tested before release to the mainline Rosetta project. People can choose whether to help debug the new code, or to steer clear of it simply by joining Ralph or joining Rosetta.

7. In any case, Rosetta is not a production project (in contrast say to LHC or SETI or Einstein), it is a development project. They are developing and testing new algorithms (methods) for solving the protein folding problem. In a development project it is fair to expect more errors than one where the code stays the same and only the data is varied.

OK, if they'd done 4 5 & 6 before 2 happened, then it would have been a lot better. Fair comment. Everyone agrees, including the guys from Rosetta.

But throw the project into the outer darkness where there is a wailing and a gnashing of bytes, simply because in the first three months after beta the core team were still newbies to DC, and had the honesty to say so, and the courage to try anyway, and the humility to learn from their mistakes? No, that is well over the top.

I don't know about you, I certainly make some mistakes, some of them gross errors. I don't want to live in a world where forgiveness is withheld even after saying sorry and making amends.

And if you want to live in such a world then you'd better be perfect yourself.

River~~

edit: added point 7 to list
ID: 13152 · Report as offensive     Reply Quote
Profile Yeti
Volunteer moderator
Avatar

Send message
Joined: 2 Sep 04
Posts: 453
Credit: 193,369,412
RAC: 10,065
Message 13157 - Posted: 28 Mar 2006, 17:50:39 UTC
Last modified: 28 Mar 2006, 17:50:54 UTC

HM, I think, I should post my experiences with several projects:

  • First, I miss a project, that run's smoothly: FightAids@Home. It is run via WorldCommunityGrid (WCG), with / under BOINC and, not only on linux, but also on Windows. In WCG, you can choose, wheather you want to run both projects (Rosetta and FightAids@Home) or only one of them. The WEB-Pages look different than BOINC-Users are used, but, all BOINC-Features are implemented and you can configure your client(s) as usual

  • CPDN: Meanwhile, I take it as a test-project to see, wheather my boxes are well assembled and stable ;-)

  • LHC is my favorite projekt; if it has work, I let all my clients immediatly switch over to LHC

  • Rosetta: I like Rosetta and the 1%-Bug doesn't bother me; meanwhile I have switched to RALPH

  • SIMAP: I like simap, because it is something totaly different in compare with all the other protein-predictors, ... And, the small WUs are good for older boxes

  • QMC: A german project (like SIMAP), as a german guy I think I should support german projects.

  • Malariacontrol: run's fine and smoothly




Supporting BOINC, a great concept !
ID: 13157 · Report as offensive     Reply Quote
senatoralex85

Send message
Joined: 17 Sep 05
Posts: 60
Credit: 4,221
RAC: 0
Message 13158 - Posted: 28 Mar 2006, 18:37:42 UTC - in response to Message 13152.  
Last modified: 28 Mar 2006, 18:45:26 UTC

River,

You picked a clever arguement defending Rosetta. However, When I first joined Rosetta, they were not so quick to admit their mistakes. I am pleased to hear that they have taken measures to prevent these errors, however I feel that the project was being deceptive to me and the rest of the distributed computing community. As far as I can tell, there was no pressure to move out of the Beta stage, and in fact at the time, they were touting how fast and well the transition worked. I was there when it happened and remember it clearly.

Perhaps I was being a bit harsh using the word "disgusted," I feel they have ruined their reputation by the months of problems that continually plague their project.

Your first point reads "They have the best communication with users of any BOINC project." Is that communication because the project is so generous OR is it because they have so many problems, they are forced to do so? Example: I recently bought a router. At the time, I wondered why Linksys would offer 24/7 free phone support. Well, I thought they were such a good company that cared about its customers. Once I tried to install the router, I found out why there was 24/4 tech support. Their system is plagued with so many problems, no one would buy their product if they didn't have it. My 1.5 hour hold time with tech support (during off peak times) on a daily basis for a week also backs up this arguement.

You say "In a development project it is fair to expect more errors than one where the code stays the same and only the data is varied." They should be in the beta stage then. In any case, RALPH should be used for testing the code not Rosetta, so I DO NOT EXPECT MORE ERRORS ON ROSETTA.


You also claim that my "forgiveness is withheld" however my issue is not about forgiveness. The best thing they could do for me to make amends is not to have a project out of beta plagued with the problems they are contiuously having.

How many times will an apology work before you get sick of them?? I understand projects will always have minor bugs or nusainces but the problems ROSETTA CONTINUOULSY HAS is unacceptable and I feel the BOINC community needs to hold them to higher standards than they obviously have held for themselves.
ID: 13158 · Report as offensive     Reply Quote
River~~

Send message
Joined: 13 Jul 05
Posts: 456
Credit: 75,142
RAC: 0
Message 13159 - Posted: 28 Mar 2006, 19:59:23 UTC - in response to Message 13158.  

...
Your first point reads "They have the best communication with users of any BOINC project."


Here I was thinking of their communication about the science and about the meaning of the various different WU. Irrespective of the quality of the coding, in my opinion the quality of the communication about the science is better than any other project at present.


You say "In a development project it is fair to expect more errors than one where the code stays the same and only the data is varied." They should be in the beta stage then.


Yes, hence the choice of Alpha for the testbed. But the real point is that by other projectsstandards, Rosetta will never be out of beta - they will always be testbedding new algorithms, that is the whole point.

By comparison with other projects Rosetta is a beta on a production scale - all the apps are experimental to some extent; this is not stated simply on their website (and in my opinion should be) but it is clear if you take the time to read through the detailed description of what the project is about.

For example, Einstein is only on its second app since it started, whereas Rosetta has around a dozen different apps running concurrently at present.

To complain about *that* makes no more sense than the complaints on LHC about intermittent work. Anyone is free to dislike it and leave, but in my opinion should do so courteously rahter than complaining.

You also claim that my "forgiveness is withheld" however my issue is not about forgiveness.


I really do think the issue is also partly about forgiveness.

Since they came back to work after the Xmas break they have been working flat out to address the unholy mess the got into before Xmas. There are noticeable improvements already, and it takes time to mend big problems.

How many times will an apology work before you get sick of them??


Here I disagree with you that there have been repeated problems - I see it as one *huge* problem of overconfidence last Autumn (you are right, they should not have claimed to be out of beta), one change of heart, and the fact that even after a change of heart it takes time to turn a big mistake around.


Thanks for taking the trouble to reply - I do understand more where you were coming from and I too withdraw the word "disgusted" from my posting.

River~~
ID: 13159 · Report as offensive     Reply Quote
Gaspode the UnDressed

Send message
Joined: 1 Sep 04
Posts: 506
Credit: 118,619
RAC: 0
Message 13160 - Posted: 28 Mar 2006, 20:23:39 UTC

This thread started out as a lament on the lack of work at LHC@Home.

If you're now debating the merits or otherwise of Rosetta wouldn't that be better done in the LHC Cafe, or maybe on the Rosetta boards?


Gaspode the UnDressed
http://www.littlevale.co.uk
ID: 13160 · Report as offensive     Reply Quote
uioped1

Send message
Joined: 26 Aug 05
Posts: 18
Credit: 37,965
RAC: 0
Message 13161 - Posted: 28 Mar 2006, 20:27:21 UTC - in response to Message 13159.  
Last modified: 28 Mar 2006, 20:35:32 UTC

I have only one point to add to River's last post; With the improvements that have been made in the last three months, this is all but a different project.

These improvements have come in:

  • Bug ellimination,
  • Algorithm improvement, to the point where the newest algorithms are capable of returning scientifically valuable results on certain classes of proteins, in addition to the scientific goal of the project to better understand the protein folding problem.
  • Scientific Communication, which as River stated is some of the best in DC.
  • Technical Communication, meaning acknowledging issues, explaining their plans, and asking for and listening to suggestions.
  • User Acknowledgement, in the form of credit and "result of the day" Acks.
  • Feature innovation,


And last but not least, the quality and value of the results received.

I really think that if you were to attach to their project now, you wouldn't recognize it as the one that so offended you previously.


ID: 13161 · Report as offensive     Reply Quote
Jim_

Send message
Joined: 10 Aug 05
Posts: 3
Credit: 16,473
RAC: 0
Message 13163 - Posted: 28 Mar 2006, 20:47:09 UTC - in response to Message 13160.  

This thread started out as a lament on the lack of work at LHC@Home.

If you're now debating the merits or otherwise of Rosetta wouldn't that be better done in the LHC Cafe, or maybe on the Rosetta boards?



Meanwhile more LHC work showed up!
ID: 13163 · Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · 5 · Next

Message boards : Number crunching : Not HAPPY people.


©2024 CERN