Since September 2004 i am crunching for Seti, but i have to admit that i am getting upset about it quite often.

Since a couple of days my machines are running almost dry and i barely get new work on the regular bases.

After studing the forum, i believe that this is a well known issue that is persident in that project for ages. but no one really seemt to even try to fix it.

This is not a senseless complaint, but i have deceidet to let my machines finish all workunits in queue and then leave Seti for the following reasons:

We are all volunteers here, but regarding the "no work avaiable" issue those in charge here seem to be a bit ignorant.

I also believe that maintaining a steady workflow is a responsibility of those in charge here, but they even fail doing that. When people complain about it, they will be either ignored or insulted by others here in the community. in so far i believe the best way to show how much i do disagree with such kind of operation, is to leave that project, and join other projects with a more matured infrastructure and better conditions as have i been seen in here in the last 2 yrs.


Ohh by the way, i am not a native english speaker, so if u will find wrong grammar or spellings, you are authorized to keep them.

;-)

---

Wie wäre es, wenn Du einfach den WU-Speicher erhöhst? ;-D

---

I dunno which Seti project you're connected to, but I have over 100 Seti workunits in my cache at the moment. Are you sure you're setting your WU buffer high enough?

---

You will be assimilated...bunghole!

I know we get a little animated here in the forums, but you should not let that color your judgement of the project.


One thing to remember is that the project never promised work 24/7/365. If you run dry, this is bound to happen. This isn't ignorance, nor carelessness, nor necessarily a "problem", and it is definitely not the project admin's "responsibility". Their responsibility as far as we're concerned is to send out work, if any, and accept the results and scientifically analyze them.

Other projects have an easier time sending out data because they do not have the massive user base that SETI has, so this in no way means that they have a more "mature infrastructure", but I could see how it could be perceived as "better conditions" due to fewer users, which means less stress on the servers, which means far less demand on them to provide work.


At any rate, its unfortunate for SETI that you feel they must provide work all the time. I do not know why this expectation is there, but more than a few people have this expectation. Most users do not have issues, or are willing to work right through them when there's problems.

If you feel you must go, then that is certainly your choice. Maybe if you decide that you still believe in the project's goals, perhaps you'll try again in the future - but please keep in mind if you come back that the project has never guaranteed there will always be work.

---

Hi, what version of BOINC are you using and for how long? How about optimized ({S}SSE3{x} applications), AK_V8_WINxx_SSxE3/4.1* and/or AP 5 (and r103*;[Don't Panic]). *(Raistmer,Segur,Whale, et al.)
You can always UP your work-cache, in your account page, to, what ever, between 1 and 10 day's :)

And seti@home has never said/garanteed, that work, will be always available, that's why they advice you, to choose a Back-Up Project, in case you don't want to run out of work.
I would just wait, in the meantime you can choose another (back-up) project or play a game or what ever.

_ _
|-._|. \/.
==========

---

Since September 2004 i am crunching for Seti, but i have to admit that i am getting upset about it quite often.

Since a couple of days my machines are running almost dry and i barely get new work on the regular bases.

After studing the forum, i believe that this is a well known issue that is persident in that project for ages. but no one really seemt to even try to fix it.

This is not a senseless complaint, but i have deceidet to let my machines finish all workunits in queue and then leave Seti for the following reasons:

We are all volunteers here, but regarding the "no work avaiable" issue those in charge here seem to be a bit ignorant.

I also believe that maintaining a steady workflow is a responsibility of those in charge here, but they even fail doing that. When people complain about it, they will be either ignored or insulted by others here in the community. in so far i believe the best way to show how much i do disagree with such kind of operation, is to leave that project, and join other projects with a more matured infrastructure and better conditions as have i been seen in here in the last 2 yrs.


Ohh by the way, i am not a native english speaker, so if u will find wrong grammar or spellings, you are authorized to keep them.

;-)

I feel your pain, Daniel, as one disgraced former President is famous for saying. But don't take it out on the project. Your observations are not new; for example I've been fighting the good fight against the nay-sayers for a week or so over in this thread, more or less about the topic you bring up here. It is a tough fight and I feel a bit like Davey Crockett at the Alamo. (One guy more or less insinuated that I'm a Nazi!) But the project and improving the project are worth the bloodshed, and suffering the obstructionists. So please consider the advice below and stay with us. We need your RAC!

And then there is this guy 4611299! He has 65 tasks due tomorrow, 13 Feb, and only returns a few tasks per day. He obviously caches many more units than he needs and puts the hurt on those with pending results, as well as those willing crunchers that cannot get new work.

Don't be so quick to condemn Daniel Ahlborn and his opinion.

Something is going on that does not add up.

The AP splitters are splitting work out but the numbers on the server status do not bear this out. "Results in the field" are decreasing. Result creation rate does not show any being created.

After an outage such as the last few days the bandwidth graph should be near maximum, especially if the new 5.03 AP app and work units are being sent out. The graph has always maxed out after a outage or a new app release, such as it did with CUDA. This time it has not done so.

The lack of used bandwidth indicates to me that other problems are happening to restrict the work being issued.

And again I will ask....has anybody actually downloaded the AP 5.03 app and a work unit for AP 5.03???

---

Boinc....Boinc....Boinc....Boinc....

I feel your pain, Daniel, as one disgraced former President is famous for saying. But don't take it out on the project. Your observations are not new; for example I've been fighting the good fight against the nay-sayers for a week or so over in this thread, more or less about the topic you bring up here. It is a tough fight and I feel a bit like Davey Crockett at the Alamo. (One guy more or less insinuated that I'm a Nazi!) But the project and improving the project are worth the bloodshed, and suffering the obstructionists. So please consider the advice below and stay with us. We need your RAC!

The problem is we're still applying E-Commerce standards (where the customer and his credit card can take their business elsewhere, in a heartbeat) to BOINC (where the only thing inconvenienced by delays are the computers).

The standard should be simple. If Daniel has at least 1 work unit per core per computer, he's crunching along fine.

I'm not arguing that BOINC can't work better, that there is no incentive for improvement. If they can find ways to reduce the impact BOINC clients have on the BOINC servers, they can do more with less capacity.

But slowing down the client to ease the load on the servers will not make the Daniels' of the world happier. It will do the opposite.

I'm not calling you a naysayer. I did point out that it is a short step between labeling those who disagree with you "naysayers" and calling them Nazis. Big difference.

Your argument is that the project does not expose all of the metrics, and if you can't see them, they obviously do not exist. You claim that if you can't measure "quality" then no one else is doing it, formally or informally.

Daniel is upset because he's not carrying as much in his cache as he'd like, and he's upset because he sees requests that go unfilled.

My recommendation to Daniel: if you are running an optimized client, make sure you're on the latest version. Otherwise, let BOINC worry about it.

---

Surely this only matters if you're more bothered about the credits than the search? I downloaded Seti@home 'cause I wanted to help (hopefully) find aliens, not 'cause I wanted to play king-of-the-mountain one-upmanship games about who did the most work, and I'm sure the majority of users feel the same. If there's no work available then there's no work available; why worry unless your scared someone else is going to 'beat' you while you're waiting?

---

Surely this only matters if you're more bothered about the credits than the search? I downloaded Seti@home 'cause I wanted to help (hopefully) find aliens, not 'cause I wanted to play king-of-the-mountain one-upmanship games about who did the most work, and I'm sure the majority of users feel the same. If there's no work available then there's no work available; why worry unless your scared someone else is going to 'beat' you while you're waiting?

Agreed.

---

Surely this only matters if you're more bothered about the credits than the search? I downloaded Seti@home 'cause I wanted to help (hopefully) find aliens, not 'cause I wanted to play king-of-the-mountain one-upmanship games about who did the most work, and I'm sure the majority of users feel the same. If there's no work available then there's no work available; why worry unless your scared someone else is going to 'beat' you while you're waiting?

I can see another reason why someone would be bothered about no work, that has nothing to do with credits, and more to do with wasting electricity when you thought your pc was running WU's, just to come back to it and find it's not. It's not always about credits.

---

Good point. We each need to feel appreciated and that our efforts/resources are not being squandered by screw-ups and poor planning, which are out of our control.

Look that, copied from the home page. Is for science or for credits ?


" February 3, 2009
Congratulations to SETIBR, the Brazilian team which, across several BOINC-based projects, has achieved the levels of 100K average credit and 50M total credit. "

Daniel

I volunteer. I coach hockey. Sometimes the games are spread too closely for my liking. Sometimes the practices are scheduled too far apart.

I didn't volunteer to coach hockey for my greater glory. I coach because I like hockey, and I like helping out the local league give the kids a great experience.

I also volunteer to crunch. By my id you can see I am one of the old timers - November 2000. When old SETI ended and BOINC came along I didn't get around to it. Recently I remembered the old account and thought "Time to volunteer some cycles".

It *is not* about you. It is about the science of the search. If they don't have need of you today - so be it - you can be pleased with what you did to help. When you are again needed (and WUs come spilling down to your PC) you can be pleased with yourself that you were there when the call came.

For anything more than that, you cannot impune upon the good graces of those who strive mightily for our collective benefit. I think you are a better person than that. This is not about what SETI can do for you, but what you can do for SETI (I know, galactically cheesy but I couldn't resist).

By the way - I've only once run out of units since starting up on BOINC and then only on one or two machines - manage your cache and it will solve your problem. If you don't know how then do a little reading and you will find the solution to your concerns.

Regards,

---

I can see another reason why someone would be bothered about no work, that has nothing to do with credits, and more to do with wasting electricity when you thought your pc was running WU's, just to come back to it and find it's not. It's not always about credits.

... which is a good argument for carrying a good sized cache.

... and a good argument for crunching more than one project.

... and for not running 24/7 if you are nearly out of work.

This page says:

Finite work supply

Each work unit is now processed a limited number of times (typically two). When we have no work for your computer, you'll get a 'no work available' message. We encourage you to participate in other BOINC-based projects; then, when SETI@home has no work, your computer can stay busy doing other scientific research.

---

If a few pence worth of wasted electricity is the price of me helping to find aliens, something thats interested me since my first Andre Norton novel, 'several' years ago, I can cope with that.

---

The simple fact is that demanding continuous work from a single project is not realistic. The Seti people know this and have always said that there would be dry spells. They have no control over when they can collect data from the telescope and don't have the money nor the staff to keep the system operating and feeding work 24/7/365. They have done a good job innovating new software to keep up with the even increasing capability of the volunteers computer.

The Seti originators know that Seti alone can't keep up with demand for work. That was one of the primary reason they developed Boinc, so that they could share their wealth of volunteers computing power with other needy projects that would probably never exist if they had to set up networks of their own from scratch.

If you want uninterrupted work for your computer there only one way, and that is to connect to several projects. Even the well financed Boinc projects (and there are only a couple of those) have period of no connection or no work. Most of the 50+ Boinc project have extended down time.

I've personally have never had an idle CPU, but then I use Boinc as it was intended and crunch for several worthy projects. I hardly notice when 4 or 5 of them are not supplying work for long periods of time.

---

Boinc V7.2.42
Win7 i5 3.33G 4GB, GTX470

I can see another reason why someone would be bothered about no work, that has nothing to do with credits, and more to do with wasting electricity when you thought your pc was running WU's, just to come back to it and find it's not. It's not always about credits.

... which is a good argument for carrying a good sized cache.

... and a good argument for crunching more than one project.

... and for not running 24/7 if you are nearly out of work.

This page says:

Finite work supply

Each work unit is now processed a limited number of times (typically two). When we have no work for your computer, you'll get a 'no work available' message. We encourage you to participate in other BOINC-based projects; then, when SETI@home has no work, your computer can stay busy doing other scientific research.

All valid arguements to keep the pc crunching, but wasn't really what I was addressing. There are several ways as you pointed out to keep our pc's busy, but I was addressing why other than credits someone might get upset about not having work when they probably haven't done the things you've suggested.

As far as what impact the price of electricity means to a person when there is no work on a running computer, I think can be idividual and dependent on other factors than our interest in the project, especially in an economy like we find ourselves in today.

I keep mine busy on two projects, but when I don't feel like crunching Einstein on a pc for whatever reason, and there's no work at SETI for it, I just turn it off till there is.

---

Surely this only matters if you're more bothered about the credits than the search? I downloaded Seti@home 'cause I wanted to help (hopefully) find aliens, not 'cause I wanted to play king-of-the-mountain one-upmanship games about who did the most work, and I'm sure the majority of users feel the same. If there's no work available then there's no work available; why worry unless your scared someone else is going to 'beat' you while you're waiting?

I can see another reason why someone would be bothered about no work, that has nothing to do with credits, and more to do with wasting electricity when you thought your pc was running WU's, just to come back to it and find it's not. It's not always about credits.

Wouldn't that be their own fault for wasting electricity by not having their computer go into suspend or standby when the CPU is idle? Its great that volunteers leave their computers on to crunch SETI, but SETI never asked that. They only want what you aren't using while its powered on, such as during word processing or other non-CPU intensive tasks.

If you choose to leave your computer on, knowing that SETI may run out of work from time to time, and you don't properly configure your computer, I'd say that's your fault more than the project's.

---