Hannah (Apr 16 2009)


log in

Advanced search

Message boards : Technical News : Hannah (Apr 16 2009)

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next
Author Message
1mp0£173
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 886844 - Posted: 21 Apr 2009, 1:32:56 UTC - in response to Message 886834.
Last modified: 21 Apr 2009, 1:33:09 UTC


I admit that I have pulled a few computers from S@H because they could not do AP
work within deadline. A major increase in the AP deadline would have solved the problem.


With a 1 month deadline on AP WU's i'm not sure what you're talking about when
you say you've " pulled a few computers from S@H because they could not do AP
work within deadline". The most common complaints with regard to AP work are
lame wingmen, and pending credits. Why would you even attach for AP work if your
computer isn't active long enough to complete an AP WU within the 1 month
deadline?.

John has some really, really old crunchers. I don't know if he has anything older than a Pentium II currently, but I do know that some of his active machines are as archaic as some of mine.
____________

Profile Gary CharpentierProject donor
Volunteer tester
Avatar
Send message
Joined: 25 Dec 00
Posts: 12706
Credit: 7,198,881
RAC: 15,112
United States
Message 886853 - Posted: 21 Apr 2009, 1:55:47 UTC - in response to Message 886829.

I admit that I have pulled a few computers from S@H because they could not do AP work within deadline. A major increase in the AP deadline would have solved the problem.

John, You are aware of the four separate preferences aren't you? Put those old slow machines into say "school" preferences and turn AP off for the "school" machines.


____________

Profile Mike O
Avatar
Send message
Joined: 1 Sep 07
Posts: 428
Credit: 6,670,998
RAC: 0
United States
Message 886856 - Posted: 21 Apr 2009, 2:14:21 UTC

My thoughts... if they really matter.

1. Ap's take to darn long. Its that simple. No one new to this is going to get excited watching NOTHING happen, right? I opt out of them but run them if there is no MB work. Why? hummm.. I like to see things moving along.
If I was SETI.. ID make the APs shorter so noobs 'including me in that bucket' can see things moving.

2. Yes.. there are major things going wrong in the ecomnomy ALL OVER THE WORLD.
This is and I hate to say this.. becoming a real issue with me.. Power is going up.. my income isn't.. and almost was cut off.. I guess unemployment would kick in but.. that would leave NO HEAD ROOM for hobbies like this one.

3. Sever issues for a noob are a real putoff Im sure. For me its.. 'hummm.. here we go again'

I for one am astonished by Matt and the team here. Like some one posted above.. I too have seen the servers pop backup at 2am.. Eastern time.. and on a saturday. I have and always will have the deepest respect for the team here and the work they put in to this.. AND after ten years.. they are working even harder. To say they are not.. well that just pisses me off!!

All I know is how I feel.. I love credits.. YEEHAW!! and its great to get them. BUT.. I can't buy bread with them so in reality, they are useless.. but fun. I do this for me.. and my love of space. I don't care if we find them, it.. E.T. in my life time.. but I am hoping so. Are there bigger problems in the world? Yes.. that I am sure of. Can you cure cancer with a PC? Not sure.. Maybe.. and it needs to be investegated as do any of the 'legit' projects.

Why are so many leaving SETI? You read my input.. the question is.. What can be done to solve this?

-- Mike


____________
Not Ready Reading BRAIN. Abort/Retry/Fail?

Profile Allie in Vancouver
Volunteer tester
Avatar
Send message
Joined: 16 Mar 07
Posts: 3949
Credit: 1,604,668
RAC: 0
Canada
Message 886869 - Posted: 21 Apr 2009, 2:54:36 UTC

Damn, I wish I could go to the upcoming SETI 10 year party. I bet that there will be a lot of very interesting face-to-face conversations! :O
____________
Pure mathematics is, in its way, the poetry of logical ideas.

Albert Einstein

Profile arkaynProject donor
Volunteer tester
Avatar
Send message
Joined: 14 May 99
Posts: 3688
Credit: 48,715,910
RAC: 6,764
United States
Message 886891 - Posted: 21 Apr 2009, 4:39:06 UTC - in response to Message 886838.

( Excuse me for the texte in french, but, i'm not able to translate that correctly )

En 1999, comme de nombreuses personnes, j'ai installé S@H comme "ScreenSaver" et il a fonctionné ainsi pendant plusieurs années en arriere-plan.

Aujourdhui, pour plusieurs raisons, de nombreuses personnes ont acheté ou construit des "Supers Calculateurs" uniquement dédiés à BOINC.

Un temps, ces personnes ont calculé pour SETI et SETI en a profité.
Aujourdhui, SETI a quelques problemes et ces personnes, qui ne supportent pas d'attendre, vont sur d'autres projets.

ce n'est pas un probleme !! Laissez les faire !!

Je suppose que l'equipe de SETI connait le nombre moyen d'utilisateurs sur lesquels elle peut compter.

PS: sur mes ordinateurs, SETI fonctionne toujours comme screensaver en arriere plan et les problemes de SETI ne m'ont pas dérangé. Mon RAC a un peu baissé car j'étais en vacances !!

Si quelqu'un veut bien traduire ce texte, je l'en remercie.


In 1999, as many people, I installed as S @ H "ScreenSaver" and has worked for several years and in the background.

Today, for several reasons, many people have purchased or built "supercomputer" solely dedicated to BOINC.

One time, they have calculated for SETI and SETI has benefited.
Today, SETI has some problems and these people who can not wait to go on other projects.

This is not a problem! Let them!

I guess the SETI team knows the average number of users on which it can rely.

PS: on my computers running SETI screensaver always in the background and problems of SETI I was not disturbed. My RAC has dropped a little because I was on vacation!

If someone wants to translate this text, I thank him.


____________

Profile tullioProject donor
Send message
Joined: 9 Apr 04
Posts: 3753
Credit: 387,845
RAC: 112
Italy
Message 886896 - Posted: 21 Apr 2009, 6:11:32 UTC

I think our French friend has hit the nail on the head when he says that people are building supercomputers just to run BOINC, BOINC (and SETI) has become a kind of a "mensa" club where you are admitted only if your IQ is high. Substitute RAC for IQ and you have the reason for which many leave BOINC. It has become the playground of "few, we happy few, we band of brothers" (Shakespeare).
Tullio
____________

Profile tullioProject donor
Send message
Joined: 9 Apr 04
Posts: 3753
Credit: 387,845
RAC: 112
Italy
Message 886910 - Posted: 21 Apr 2009, 8:22:25 UTC - in response to Message 886901.
Last modified: 21 Apr 2009, 8:23:21 UTC

True, SETI ,and other BOINC projects using GPUs, has become a race between the have and the have-not. This is why people are leaving it.
____________

daysteppr
Send message
Joined: 22 Mar 05
Posts: 69
Credit: 4,029,223
RAC: 6,182
United States
Message 886928 - Posted: 21 Apr 2009, 11:58:32 UTC

Well, as one of those general cruncher people. Credits? A little. Helping find ET? Sure. 200 hr wus? I have optimized apps and they still take 40+ hrs ( AMD User)

We need to keep in mind that the 'vast' majority of SETI crunchers NEVER show up on these boards. Optimized app? Never heard of it. Cuda? I saw that on an add for a video card....

IMO I think we all just need to calm down and wait it out. We've been here long enough to know that. This isnt the dam busted and we're all gonna die situation. Its somethings wrong with a sensor. Lets just work on the minor issues before we decide to rebuild a dam that isnt broke.

Thats my opinion anyway.

Sincerely,
Daysteppr
____________

Profile tullioProject donor
Send message
Joined: 9 Apr 04
Posts: 3753
Credit: 387,845
RAC: 112
Italy
Message 886941 - Posted: 21 Apr 2009, 13:29:35 UTC - in response to Message 886928.

I agree with you. I am an AMD user and have been running a climateprediction.net app for 1720 hours, with 230 more to go, Einstein@home, QMC@home, AQUA@home for 41 hours with 20 to go. I have 2 AP units in a pending credits state. I am just glad when I see a result. Credits are optional. Cheers.
Tullio
____________

Profile MikeProject donor
Volunteer tester
Avatar
Send message
Joined: 17 Feb 01
Posts: 24481
Credit: 33,796,538
RAC: 23,497
Germany
Message 886942 - Posted: 21 Apr 2009, 13:35:37 UTC - in response to Message 886869.

Damn, I wish I could go to the upcoming SETI 10 year party. I bet that there will be a lot of very interesting face-to-face conversations! :O


I second that.

____________

barbereau
Volunteer tester
Avatar
Send message
Joined: 24 May 99
Posts: 52
Credit: 94,540
RAC: 0
France
Message 887141 - Posted: 22 Apr 2009, 2:17:30 UTC - in response to Message 886891.

Thank ARKAYN for that perfect translation !!!

Profile Rhe
Avatar
Send message
Joined: 17 Dec 99
Posts: 50
Credit: 488,837
RAC: 0
United States
Message 887161 - Posted: 22 Apr 2009, 2:36:38 UTC - in response to Message 886942.

I also get the feeling that apart from a select few, we are not held in very high regard by the project administrators, probably at the level of lab rats. They appear to act in a very high handed way toward their minions, a "crunch what we give, take it or leave it" attitude.

Also all the talk about camping and hiking trips and music gigs and concentrating on incidentals instead of fixing the outage problems and preventing their ocurrence are being construed by many as a lack of caring about the project, and if the project doesn't care, then why should we?


Read these two points from posters and felt the need to add my 2 cents here.

I've lurked around the message boards for a lot longer than I have ever engaged in them and I'm a "teeny cruncher" compared to the majority here, it seems. However, never once have I felt that the project admins, project devs, scientists, etc involved with Seti@home view the lot of us as lab rat slaves who are here to crunch at the whim of Seti@home. I applaud and appreciate the consistant updates, wee-hour work fixes, and info provided to us whenever there are problems afoot...and Matt and the others have never failed to update and discuss these issues with us on the boards (sometimes in great technical detail!).

As far as the "camping and hiking trips and music gigs" comment--just as we are not Seti@home's lab rats, neither are they ours. The good folks at Seti@home that work and dedicate themselves to the benefit of this project are real people, also...with real lives to lead. I, for one, enjoy the fact that they are willing to share a small bit of their interests and lives with us here.


____________

Team SETI.USA Forums
Somewhere, something incredible is waiting to be known. ~~Carl Sagan

Profile Bob Mahoney Design
Avatar
Send message
Joined: 4 Apr 04
Posts: 178
Credit: 9,205,632
RAC: 0
United States
Message 887263 - Posted: 22 Apr 2009, 12:24:37 UTC - in response to Message 886928.

We need to keep in mind that the 'vast' majority of SETI crunchers NEVER show up on these boards. Optimized app? Never heard of it. Cuda? I saw that on an add for a video card....

IMO I think we all just need to calm down and wait it out. We've been here long enough to know that. This isnt the dam busted and we're all gonna die situation. Its somethings wrong with a sensor. Lets just work on the minor issues before we decide to rebuild a dam that isnt broke.

Thats my opinion anyway.

Sincerely,
Daysteppr

That makes a lot of sense, Daysteppr.

1mp0£173
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 887329 - Posted: 22 Apr 2009, 17:13:27 UTC - in response to Message 887161.

I also get the feeling that apart from a select few, we are not held in very high regard by the project administrators, probably at the level of lab rats. They appear to act in a very high handed way toward their minions, a "crunch what we give, take it or leave it" attitude.

Also all the talk about camping and hiking trips and music gigs and concentrating on incidentals instead of fixing the outage problems and preventing their ocurrence are being construed by many as a lack of caring about the project, and if the project doesn't care, then why should we?


Read these two points from posters and felt the need to add my 2 cents here.

I've lurked around the message boards for a lot longer than I have ever engaged in them and I'm a "teeny cruncher" compared to the majority here, it seems. However, never once have I felt that the project admins, project devs, scientists, etc involved with Seti@home view the lot of us as lab rat slaves who are here to crunch at the whim of Seti@home. I applaud and appreciate the consistant updates, wee-hour work fixes, and info provided to us whenever there are problems afoot...and Matt and the others have never failed to update and discuss these issues with us on the boards (sometimes in great technical detail!).

As far as the "camping and hiking trips and music gigs" comment--just as we are not Seti@home's lab rats, neither are they ours. The good folks at Seti@home that work and dedicate themselves to the benefit of this project are real people, also...with real lives to lead. I, for one, enjoy the fact that they are willing to share a small bit of their interests and lives with us here.


If I may... I agree with the emotion behind this, but I think the reality is a little different.

We're all taking part in two experiments.

The first one is "Can we find evidence of intelligence in the radio signals received from deep space?"

The second one: "can we harness the amazing computing power on desktops around the world?"

I do SETI because I like the idea of finding ET, but I'm fascinated by the BOINC experiment, and I am happy to be involved.

I realize that when you boil down "being part of the Grand Experiment that is BOINC" it sounds a lot like "lab rat" but there is one very important difference.

Lab rats aren't free to come and go. Their contributions to science are not optional.

I don't get the lab rat references, and I've certainly never seen anything that suggested that the project thinks of us as a disposable resource.
____________

Profile Bob Mahoney Design
Avatar
Send message
Joined: 4 Apr 04
Posts: 178
Credit: 9,205,632
RAC: 0
United States
Message 887347 - Posted: 22 Apr 2009, 17:50:13 UTC

My 2 cents: The new world has become addicted to more and more "immediate gratification." Shorter attention spans, right? Many BOINC projects prey upon this trend by giving lots of credits quickly, and they use an RAC calculation that gives you a satisfactory RAC in only a few days. Some of the projects are approaching "instant gratification", and they are retaining new users with this technique.

Perhaps the SETI@home architects might consider an adjustment to the RAC calculation policies here? I'm not saying we should give more credits for work completed. I am suggesting a much shorter time for RAC gratification.

An added benefit for the Berkeley crew: Fewer user complaints about an occasional outage. Shorter recovery times for RAC would greatly decrease user frustration after an outage. For example, maybe the RAC rampup should be no longer than the typical MTBF of the SETI@home server system.

Like it or not, this is a competitive market situation. A no-cost change of this type will enhance the user experience, resulting in improved user acquisition and retention.

Bob Mahoney

OzzFan
Volunteer tester
Avatar
Send message
Joined: 9 Apr 02
Posts: 13625
Credit: 30,981,779
RAC: 20,326
United States
Message 887366 - Posted: 22 Apr 2009, 19:08:59 UTC
Last modified: 22 Apr 2009, 19:12:12 UTC

I say we need to remove the wild guesses and personal feelings from the equation. Everyone seems to extropolate that because they feel a certain way, that everyone who leaves must be leaving for that reason, which may or may not be accurate.

My suggestion, which will only help slightly, would be to direct the BOINC uninstaller to a URL page containing an "exit questionaire" where people can leave feedback as to why they removed BOINC. This could also be done for people who detach from any particular project, but then those who detach and re-attach to solve a problem would probably become frustrated with this feature. [Edit] To work around that frustration, a separate button could be added that says "Detach and Re-attach" which would bypass going to the URL, as opposed to the "Detach" button that will direct you to the URL.

Armed with this data, projects could take the relevant info that they can actually do something about and use it to improve the BOINC infrastructure and experience.

Downside: another database that needs to be maintained, perhaps at the project level if the idea is implemented at the detach stage, for which there may not be enough manpower to maintain. This could be implemented as a third-party solution website, but I know how David Anderson feels about keeping everything under the BOINC umbrella.
____________

1mp0£173
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 887371 - Posted: 22 Apr 2009, 20:11:48 UTC - in response to Message 887347.

My 2 cents: The new world has become addicted to more and more "immediate gratification." Shorter attention spans, right? Many BOINC projects prey upon this trend by giving lots of credits quickly, and they use an RAC calculation that gives you a satisfactory RAC in only a few days. Some of the projects are approaching "instant gratification", and they are retaining new users with this technique.

Perhaps the SETI@home architects might consider an adjustment to the RAC calculation policies here? I'm not saying we should give more credits for work completed. I am suggesting a much shorter time for RAC gratification.

An added benefit for the Berkeley crew: Fewer user complaints about an occasional outage. Shorter recovery times for RAC would greatly decrease user frustration after an outage. For example, maybe the RAC rampup should be no longer than the typical MTBF of the SETI@home server system.

Like it or not, this is a competitive market situation. A no-cost change of this type will enhance the user experience, resulting in improved user acquisition and retention.

Bob Mahoney

Sorry to present the contrary argument, but if SETI raises credit (intentionally) then that does two things:

1) it encourages other projects to raise credit to "compete" for users motivated by credit.

2) it devalues the earlier work -- 1 credit today should represent the same amount of work as 1 credit in 1999.

It is unfortunate (all the way around) that calculating credit is a bit of a black art and not the simple engineering exercise that it should be.
____________

Profile Gary CharpentierProject donor
Volunteer tester
Avatar
Send message
Joined: 25 Dec 00
Posts: 12706
Credit: 7,198,881
RAC: 15,112
United States
Message 887438 - Posted: 23 Apr 2009, 0:31:30 UTC - in response to Message 887371.

My 2 cents: The new world has become addicted to more and more "immediate gratification." Shorter attention spans, right? Many BOINC projects prey upon this trend by giving lots of credits quickly, and they use an RAC calculation that gives you a satisfactory RAC in only a few days. Some of the projects are approaching "instant gratification", and they are retaining new users with this technique.

Perhaps the SETI@home architects might consider an adjustment to the RAC calculation policies here? I'm not saying we should give more credits for work completed. I am suggesting a much shorter time for RAC gratification.

An added benefit for the Berkeley crew: Fewer user complaints about an occasional outage. Shorter recovery times for RAC would greatly decrease user frustration after an outage. For example, maybe the RAC rampup should be no longer than the typical MTBF of the SETI@home server system.

Like it or not, this is a competitive market situation. A no-cost change of this type will enhance the user experience, resulting in improved user acquisition and retention.

Bob Mahoney

Sorry to present the contrary argument, but if SETI raises credit (intentionally) then that does two things:

1) it encourages other projects to raise credit to "compete" for users motivated by credit.

2) it devalues the earlier work -- 1 credit today should represent the same amount of work as 1 credit in 1999.

It is unfortunate (all the way around) that calculating credit is a bit of a black art and not the simple engineering exercise that it should be.

I also feel like you do that one credit of work today should be worth the same as it was in 1999. However David Anderson doesn't and it isn't. An old credit is actually worth more than a new credit.

Without some kind of survey no one will know for sure why people aren't continuing to send work. The first question is are they detaching or are the machines going away? Once that is known then perhaps you can begin to ask if it is because of some failing of the project they are leaving. Or has it been a while since the cron job to e-mail tickle people who aren't returning work units has run?


____________

Profile tullioProject donor
Send message
Joined: 9 Apr 04
Posts: 3753
Credit: 387,845
RAC: 112
Italy
Message 887468 - Posted: 23 Apr 2009, 2:32:00 UTC

I agree with the survey idea. In the Folding@home page on user statistics you'll find that by far the more active users are those using graphic cards. both nVidia and ATI. I suspect the same thing is happening in SETI.
Tullio
____________

Profile Bob Mahoney Design
Avatar
Send message
Joined: 4 Apr 04
Posts: 178
Credit: 9,205,632
RAC: 0
United States
Message 887482 - Posted: 23 Apr 2009, 3:33:38 UTC - in response to Message 887371.

... I'm not saying we should give more credits for work completed. I am suggesting a much shorter time for RAC gratification.
...
Shorter recovery times for RAC would greatly decrease user frustration after an outage. For example, maybe the RAC rampup should be no longer than the typical MTBF of the SETI@home server system.
...
Bob Mahoney

Ned Said:
Sorry to present the contrary argument, but if SETI raises credit (intentionally) then that does two things...

Ned and Gary,

No offense taken or meant, but this is not the old credit-whore argument.

My point: The rolling average equation with the 2-week half-life that is the basis for the Recent Average Credit (RAC) ramp-up should have a much shorter half-life. This will impact the time element of RAC only.

To summarize:

1. Don't change credit per WU on SETI@home.

2. Make the RAC reach a semi-stabilized level in 2 weeks, not 2 months. Heck, maybe make it 1 week.


Today, RAC ramp-up half-life is much more of a marketing issue than a technical issue, IMHO. RAC half-life has no impact on total credits earned or credit-per-flop. It does impact user frustration after an outage. It sells (or un-sells) the project on first impressions. Nuff said.

Bob
____________
Opinion stated as fact? Who, me?

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

Message boards : Technical News : Hannah (Apr 16 2009)

Copyright © 2014 University of California