Credit, what else?


log in

Advanced search

Message boards : Number crunching : Credit, what else?

Author Message
Marek Majewski
Send message
Joined: 26 Nov 00
Posts: 14
Credit: 330,332
RAC: 0
United States
Message 27699 - Posted: 18 Sep 2004, 13:20:56 UTC

I registered 6 days ago and immediately begun crunching on one of my computers. Actually I removed it from my SETI@HOME classic farm and dedicated it to the new SETI/BOINC. I must have crunched 60-70 WUs. I have patiently waited to see the results - nothing. Is this normal? Are accounts updated really that infrequently? Are we dealing here with another half-baked, academic-doomed-to-fail initiative? Do the seti/boinc folks really swim in idle CPU cycles and don't care if people get discouraged and take they toys elswhere?

Regards,

-m(somewhat frustrated)m-

Profile Captain Avatar
Volunteer tester
Avatar
Send message
Joined: 17 May 99
Posts: 15133
Credit: 516,716
RAC: 0
United States
Message 27701 - Posted: 18 Sep 2004, 13:34:58 UTC
Last modified: 18 Sep 2004, 13:36:22 UTC

1. Take a pill
2. Have some patience
3. Make sure you update
4. How can you be discouraged after only 6 days?

"Are we dealing here with another half-baked, academic-doomed-to-fail initiative?"

If I was the seti/boinc folks I would erase your account!
This team has overcome some huge obstacles and we are enjoying the longest run yet for the new Seti and Bonic system. Part of being in this research is overcoming problems to get the answers......

Chill and if you have a problem ask nicely and maybe someone will help you.


Timmy

____________

Marek Majewski
Send message
Joined: 26 Nov 00
Posts: 14
Credit: 330,332
RAC: 0
United States
Message 27702 - Posted: 18 Sep 2004, 13:52:46 UTC

Dear Timmy:

First thing needed to "overcome problems" is admiting that there is one. The main reasons why distributed projects like SETI@HOME are a success is that people want to participate. Creating a system that encourages participation is, therefore, a key element. Seems to me that some major components of SETI/BOINC are either broken or missing. But as long as the project folks view it as a normal state of events, it won't be rectified.

I am sure you heard the phrase "The king is naked".

I take it you already took your pill this morning - perhaps you should take another one? And I am glad that you are are not the seti/boinc folks. If you were there will be no user accounts at all, since your attitude towards them is "shut-up and put-up, or I remove your account". Nice going!

BTW, I thought I asked nicely.

Regards,

-mm-

[BOINCstats] Willy
Volunteer tester
Send message
Joined: 4 Mar 01
Posts: 201
Credit: 113,951
RAC: 0
Netherlands
Message 27705 - Posted: 18 Sep 2004, 14:25:47 UTC
Last modified: 18 Sep 2004, 14:26:17 UTC

You are frustrated after six days....

Some of us are here from the start (either the alpha, the beta or the public) and we have seen times where we didn't get credit for 2~3 weeks. Sometimes we didn't get workunits either, or the site was down completely.

I never complained! It doesn't make the project run flawless, it doesn't get you credit, and it certainly doesn't earn you respect

The devs worked out most of the problems, I'm getting credit almost every day for the last 2 weeks.

I don't normaly reply to complaints but I've had it with them.

My utmost respect to the devs who have to go through all these complaints, and still manage to work on BOINC/SETI and work out the problems.

Profile Toby
Volunteer tester
Avatar
Send message
Joined: 26 Oct 00
Posts: 1005
Credit: 5,597,927
RAC: 1,977
United States
Message 27710 - Posted: 18 Sep 2004, 14:40:42 UTC
Last modified: 18 Sep 2004, 14:41:51 UTC

You did ask relatively nicely (although the 'academic-doomed-to-fail initiative' is probably the part that set Timmy off). But there have been many, MANY complainers around here in the last couple of months and some nerves are a bit frayed.

> First thing needed to "overcome problems" is admiting that there is one.

No one is denying that there are still problems with BOINC. The developers least of all. They have been working their TAILS off to try and get this thing off the ground for the past several months. We have seen people in Berkeley working at 2 AM to put the server back up and give us some work units.

> I registered 6 days ago and immediately begun crunching on one of my
> computers. Actually I removed it from my SETI@HOME classic farm and dedicated
> it to the new SETI/BOINC. I must have crunched 60-70 WUs. I have patiently
> waited to see the results - nothing. Is this normal?

Yes and no. Under normal circumstances, most credit should be granted within a day or two however since a work unit must be completed by 3 people before you get credit, it could take up to a couple of weeks to be granted credit for a particular work unit if the first person who got it stops crunching and it has to be sent out again after the 2 week timeout period. I have a team mate who has been complaining about no credits for over a week now I believe and he just got his first credit last night so I would expect for you to see some here in the very near future. Credit is only granted while the 'validator' is running. The status of the servers can be seen here: http://setiweb.ssl.berkeley.edu/sah_status.html. It was off for several days but has been up and running for the last day or two and undoubtedly has a large backlog to run through.


> Do the seti/boinc folks really swim in idle CPU cycles and don't care if
> people get discouraged and take they toys elswhere?

SETI/BOINC isn't swimming in idle CPU cycles yet since there are only 45,000 users right now but yes, seti@home classic is literally swimming in spare cycles it does not need and WANTS people to crunch for multiple projects under BOINC. this is one of the biggest reasons they developed it in the first place. Work units were/are being sent out "dozens" of times according to Dr. Anderson when they only needed 3 results for verification. Once they shut down seti-classic, they will no longer have enough work for all of us so either attach to multiple projects or expect frequent "No work from project" messages.

Well, I have to go to work but I hope that answers a few of your concernes. Welcome to the project and I hope you get some credit soon :)



-------------------------------------------
- A member of The Knights Who Say NI!
Possibly the best stats site in the universe:
http://boinc-kwsn.no-ip.info/

texasfit
Avatar
Send message
Joined: 11 May 03
Posts: 223
Credit: 500,626
RAC: 0
United States
Message 27713 - Posted: 18 Sep 2004, 14:55:26 UTC

>I have patiently waited to see the results - nothing. Is this normal?
>
>

I just converted from classic to boinc on Aug 3rd and it was almost 4 weeks before any credits started showing up for me. The system was up and down during that time. I might add that in a matter of a couple of days once the credits started I did receive over 2500 the first day. There have been quit a few problems that have been resolved over the past two months and I am sure there will be many more to come.

As was mentioned already, the answer to your question is yes and no. One variable that I don't think was mentioned is the last digit of the work unit indicates a sequence number of sorts. It will be the _0 ; _1 ; _2 ; etc. They will send out the same work unit to three computers and all three must return it for valication. If one of the computers returns it with an error then it will be sent out again to another computer. If it is a _0 wu, then it may take longer to recieve credit. If one of the computers has quit a bit of wu's to process then it could take up to 2 weeks. Generally, it should take 3-5 days to start getting credits. I would say that if you don't show any credits after 2 weeks, then there may be a problem.

In the meantime, happy crunching. :-)

----------



Join the Overclockers.com SETI Team!

Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 27721 - Posted: 18 Sep 2004, 15:35:47 UTC

Just one more side thought.

We have most of the basic issues addressed. Now we are in the dig them out stage. BOINC, which has only been a "real" proposition since about the beginniing of the year.

I did work in Beta, did work when they first went "live", back to classic, and now am back to BOINC and it looks like I will be here full time.

What we are seeing now are issues that did not show up in earlier stages because the playing field was too small. As predicted, we had, and are still having some problems.

SEI@Home is now working on both client side issues and improvements, along with changes to the server side to handle high load cconditions. Earlier, some of the beta work was involved with the issues raised with Climate Prediction (cp.net) with long work units.

LHC@Home also has joined the party and is issuing some work. It looks like we will see another client out of them next week (rumor).

Predictor@Home, was issuing work, but now is in the upgrade process to get their science applications migrated to version 4.

I think the difficulty some of us have is the apparant short, um, attention span of some people. And as explained in another post in this thread it can make people, um, testy. But, we do want your contribution, we do want you to stay, but, you have to reealize that this is not a mature environment yet.

Contrasting BOINC and any project running on it by the standard of SETI@Home Classic has to allow BOINC the same maturity level to make a valid comparison. When I started running SETI@Home classic, if I recall correctly, it was already in version 3 with nearly 2 or 3 years of operational experience behind it. We have less than one year, even less than that if you start the clock when we went "live" which (memory is dim) was about June time frame.

Two years from now, in June, then it is time to complain. But I don't think there will be much to complain about... (Paul's opinion). I now have operational experience with BOINC and with the 5 projects that are available and I have seen few, if any of the problems others have. And, though I have a signature, it is there for the documentation... I have not looked at all to see if the credit is there, not there, rising, falling ...

:)

Just some more philosophizing and practice in mispelling words ...
<p>
For BOINC Documentaion: Click Me!


Terry
Volunteer tester
Send message
Joined: 17 Sep 00
Posts: 153
Credit: 1,805,202
RAC: 0
United States
Message 27725 - Posted: 18 Sep 2004, 15:51:56 UTC

>Contrasting BOINC and any project running on it by the standard of SETI@Home >Classic has to allow BOINC the same maturity level to make a valid >comparison. When I started running SETI@Home classic, if I recall correctly, >it was already in version 3 with nearly 2 or 3 years of operational >experience behind it. We have less than one year, even less than that if you >start the clock when we went "live" which (memory is dim) was about June time >frame.

Paul, you started just before I did. Remember all the bandwidth problems? Remember when they released the version 3 client that doubled the work unit crunching time for "more science" and to help with the bandwidth problem? Personally, I think Boinc/seti 2 are ahead in the game when compared to classic.


Terry - k4vh

[/i]

Pascal, K G
Volunteer tester
Avatar
Send message
Joined: 3 Apr 99
Posts: 2343
Credit: 150,491
RAC: 0
United States
Message 27749 - Posted: 18 Sep 2004, 18:27:46 UTC - in response to Message 27699.

> I registered 6 days ago and immediately begun crunching on one of my
> computers. Actually I removed it from my SETI@HOME classic farm and dedicated
> it to the new SETI/BOINC. I must have crunched 60-70 WUs. I have patiently
> waited to see the results - nothing. Is this normal? Are accounts updated
> really that infrequently? Are we dealing here with another half-baked,
> academic-doomed-to-fail initiative? Do the seti/boinc folks really swim in
> idle CPU cycles and don't care if people get discouraged and take they toys
> elswhere?
>
> Regards,
>
> -m(somewhat frustrated)m-
>
>


6 Days hahahahahahaha OMG 6 days lololololololllll now that is funny
My name is Pascal and this message has my approval...

It is 10 oclock, do you know what your WUWUs are doing tonight...

Profile Everette Dobbins
Send message
Joined: 13 Jan 00
Posts: 263
Credit: 6,102,098
RAC: 0
United States
Message 27765 - Posted: 18 Sep 2004, 19:39:48 UTC - in response to Message 27705.

> You are frustrated after six days....
>
> Some of us are here from the start (either the alpha, the beta or the public)
> and we have seen times where we didn't get credit for 2~3 weeks. Sometimes we
> didn't get workunits either, or the site was down completely.
>
> I never complained! It doesn't make the project run flawless, it doesn't get
> you credit, and it certainly doesn't earn you respect
>
> The devs worked out most of the problems, I'm getting credit almost every day
> for the last 2 weeks.
>
> I don't normaly reply to complaints but I've had it with them.
>
> My utmost respect to the devs who have to go through all these complaints, and
> still manage to work on BOINC/SETI and work out the problems.
>
>You dont normaly reply to complaints but youve had it with them good for you. But if no one complains then how will change come about. I think the complaining helps attack issues and I think they are monitored. Some problems are easier to fix than are others. The guy has been crunching for six days and wants to know if this is how it goes. For now yes. BOINC has slowly gotten better and if you continue crunching I beleive it will get even better.

[BOINCstats] Willy
Volunteer tester
Send message
Joined: 4 Mar 01
Posts: 201
Credit: 113,951
RAC: 0
Netherlands
Message 27769 - Posted: 18 Sep 2004, 19:53:54 UTC - in response to Message 27765.

> >You dont normaly reply to complaints but youve had it with them good for
> you. But if no one complains then how will change come about. I think the
> complaining helps attack issues and I think they are monitored. Some problems
> are easier to fix than are others. The guy has been crunching for six days and
> wants to know if this is how it goes. For now yes. BOINC has slowly gotten
> better and if you continue crunching I beleive it will get even better.
>
Maybe I should clarify myself: I've had it with people trashing the devs and the project. There is nothing wrong with complaints or criticism, but some people go one step further/too far.

Please keep it civil.

Profile Robert Sullivan, MD
Avatar
Send message
Joined: 31 Oct 00
Posts: 221
Credit: 358,173
RAC: 0
United States
Message 27773 - Posted: 18 Sep 2004, 20:04:27 UTC

Hi Marek. Frustrating getting this going, isn't it. When I switched over from Classic to BOINC, I accidentally started a new account rather than activating my old one. So none of my thousands of WUs were showing up when I looked at my impressively girthful account statistics. This kind of thing happens to me all the time. It's always a good thing that feels bad. This time, it was a reminder that I am not what's important here. And these are not "my" WUs. Fortunately, not seeing my WU numbers higher than other people's reminded me that SETI is not an exercise in increasing the radius of my ego. Rather, it is a collective effort to raise humanity's gaze. Maybe find a friend out there...
Didn't intend to aim these comments directly at you, Marek. Your post just got me thinking. Keep crunching.
Clear skies.

Marek Majewski
Send message
Joined: 26 Nov 00
Posts: 14
Credit: 330,332
RAC: 0
United States
Message 27783 - Posted: 18 Sep 2004, 21:00:33 UTC

Of course we are crunching for a greater good. Of course we donate our time and money to a project which we percieve important. And, of course, we expect some gratification in return. Knowing that we are contributing to to a worth-while cause is one of them. Knowing how much we have contributed is another.

Not to mention the fact that there are certain tweaks that we could epmloy on our system which could increase throughput. But in order to get faster you need to know how fast you are going, and that's dificult when the speedometer is not functioning...

Having worked in academia, I have seen many of projects which only contribution was to the waste of budgets. They fall into "we do this because we can" or "we didn't have the time for planning" categories. I sure hope BOINC is not one of them - that's all.

But boy, some nerves are a bit on a tangled side here. You ask a question and you generate a storm of hostile reactions, my favorite of which was "SETI/BOINC should remove your account"...:)

-mm-

Profile Bartholomew T duBois
Send message
Joined: 11 Oct 03
Posts: 8
Credit: 58,846
RAC: 0
United States
Message 27795 - Posted: 18 Sep 2004, 21:29:13 UTC - in response to Message 27769.


> Maybe I should clarify myself: I've had it with people trashing the devs and
> the project. There is nothing wrong with complaints or criticism, but some
> people go one step further/too far.
>

You sound like a Republican. god forbid we should criticize and hope for change!
Bartski

[BOINCstats] Willy
Volunteer tester
Send message
Joined: 4 Mar 01
Posts: 201
Credit: 113,951
RAC: 0
Netherlands
Message 27817 - Posted: 18 Sep 2004, 22:41:15 UTC - in response to Message 27795.

> You sound like a Republican. god forbid we should criticize and hope for
> change!
> Bartski
>


I'm not voting for Bush! (Actually, not voting at all until 2006).

CyberGoyle
Avatar
Send message
Joined: 2 Jun 99
Posts: 158
Credit: 1,646,029
RAC: 544
United States
Message 27818 - Posted: 18 Sep 2004, 22:42:12 UTC - in response to Message 27773.

> Fortunately, not seeing my WU numbers higher than other people's reminded me
> that SETI is not an exercise in increasing the radius of my ego.

It isn't? Damn, I joined the wrong project again.....


[just kidding :)]


<a>

Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 27822 - Posted: 18 Sep 2004, 22:57:45 UTC - in response to Message 27783.

> Of course we are crunching for a greater good. Of course we donate our time
> and money to a project which we percieve important. And, of course, we expect
> some gratification in return. Knowing that we are contributing to to a
> worth-while cause is one of them. Knowing how much we have contributed is
> another.

Agreed. And we do have mechanisms that are doing that. The third party stat sites are doing landmark sales. :)

> Not to mention the fact that there are certain tweaks that we could epmloy on
> our system which could increase throughput. But in order to get faster you
> need to know how fast you are going, and that's dificult when the speedometer
> is not functioning...

Again, agreed. But we do have a "lead-line" so we can take soundings. The basic stats are shown on the GUI (though I am out of luck on my Macintosh for that), so, it is possible to keep track. Again, you did catch some heat. But there are constant drive-by "BOINC stinks!" that it gets hard sometimes to keep an even strain ...

Though you almost certainly did not intend it (I expect the best of people, and am usually gratified...) your start up tone was BOINC Stinks, it has already been 15 seconds and I have not been gratified yet ...

So, I was trying to explain the reaction. Perhaps I failed ...

> Having worked in academia, I have seen many of projects which only
> contribution was to the waste of budgets. They fall into "we do this because
> we can" or "we didn't have the time for planning" categories. I sure hope
> BOINC is not one of them - that's all.

I did 20 years in the U.S. Navy then 6 years in Federal Civil Service (it was a mother beautiful radar); so waste is not unfamiliar to me.

BOINC is a second generation distributed processing mechanism. The first generation proved that the concept was viable. The trouble was we spent the same dollars over and over making the same stuff over and over. BOINC is different. One of the problems right now is that we have several hairs in the soup and BOINC can get blamed for the failure of the science application, and vice versa...

> But boy, some nerves are a bit on a tangled side here. You ask a question and
> you generate a storm of hostile reactions, my favorite of which was
> "SETI/BOINC should remove your account"...:)

I don't care for that reaction either. Just like I don't like my criticisms of the system being taken out of context. And yes, I have, and do, criticise the system when it has problems. I am going to rain on you for a second, but hang in there. You raised a non-issue. And you did it in a hostile tone. You may not agree with my evaluation (your privilege) ... but read your own writing and put yourself in the shoes of those of us who have been breaking trail for you ...

This is a big, a HUGE new generation application. And it is early days. I have sympathy for those who have had problems and had their systems trashed. But that is an expense we pay for being on the "bleeding-Edge" of technology. I have been doing computers professionnally and personnaly since 1975, and electronics back into the late 60's ... and I know we don't understand our systems.

If you have a serious complaint, I am with you ... I would want it fixed. The problem with the Pinto was not the paint finish, nor that the gas guage is non-linear, the darn things blew up ...

The folks fighting the WIndows 98 problems have my sympathy and hopes for immediate resolution. The deck chairs on the Titanic are just fine thank you ... :)

So, yes, you got some negative feedback... And you may feel that I was in the posse ... I don't know ... Maybe I was. But we are not in bad shape. It works. It works reasonably well ... Heck I even got a little work from the Pirates/Einstein@Home project ...

<bap!> ball is in your court ...
<p>
For BOINC Documentaion: Click Me!


Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 27824 - Posted: 18 Sep 2004, 23:02:41 UTC - in response to Message 27769.

> Maybe I should clarify myself: I've had it with people trashing the devs and
> the project. There is nothing wrong with complaints or criticism, but some
> people go one step further/too far.

I agree ...

But for a complaint to have viability it has to be backed by some understanding. In other words, if you are just spouting I hate it, well, that is not an addressable problem.

I don't like the database schema and would have done things differently. Not that even if I had, would we have been in the clear. :)

As I stated, we probably would have run into another equally bad problem that was not easily solved. You only truly know that a schema is successful when you put real data into it. Beta was good and found problems, but we were in sandbox play ...

poo, i gotta quit ... my tyopig is goind abain
<p>
For BOINC Documentaion: Click Me!


Profile Captain Avatar
Volunteer tester
Avatar
Send message
Joined: 17 May 99
Posts: 15133
Credit: 516,716
RAC: 0
United States
Message 27832 - Posted: 18 Sep 2004, 23:37:00 UTC - in response to Message 27702.
Last modified: 18 Sep 2004, 23:38:12 UTC


>
> BTW, I thought I asked nicely.

Dear MM,

I am sorry if you felt offended to my reply to you this morning but
You just rubbed me the wrong way by your insinuations when you’re asking for help and insulting at the same time. I have a cool head and want to help and learn also. I have like the many others have been crunching for years and not days... We want to help and be proud that we are part of the team...

I have read your post several times and I get the same reaction every time I read it, Plus your reply with “BTW, I thought I asked nicely" prompted me to write the above. MM you are probably a nice person and so am I. Please bare with the problems they will work out and you'll receive your credits soon...

Good luck and best crunching



Timmy
</img>

____________

Message boards : Number crunching : Credit, what else?

Copyright © 2014 University of California