The various optimisers gain their kudos without the PR.

Oh, I'm all in favor of the PR. It's always good to be around people who tell everyone else that when it comes to technology, here's the guy who really knows everything and can answer every question.

PR is good. It really makes you feel great when someone says that.

The problem is when you start to believe that you are the center of the technological universe.

So, yeah, I love it when someone says "you're mind-bogglingly clever" (or even "Sass that hoopy frood Ned, he really knows where his towel is").

PR is good, but it's a very short step from believing your own PR to dillusions of grandeur.

Never believe your own PR -- nobody is that good.

---

Oh, I'm all in favor of the PR. It's always good to be around people who tell everyone else that when it comes to technology, here's the guy who really knows everything and can answer every question.

PR is good. It really makes you feel great when someone says that.

The problem is when you start to believe that you are the center of the technological universe.

So, yeah, I love it when someone says "you're mind-bogglingly clever" (or even "Sass that hoopy frood Ned, he really knows where his towel is").

PR is good, but it's a very short step from believing your own PR to dillusions of grandeur.

Never believe your own PR -- nobody is that good.

Eloquently put as usual! :-)

Personally, I believe "Guru" status on a forum like this merely indicates you have more free time to read through the threads and contemplate that information in the context of other questions. :-D

Alinator

To put things in perspective:


Even if many crunchers are moving to other projects, SETI is still the project that cranks out the most credits.

---

Join team BOINCstats

Willy,

Any thoughts on what the story was with the big jump on the BOINC Combined Total Active Users histogram?

It didn't seem like the number of users or timing of newer projects could account for that.

Alinator

There is no big jump BOINC Combined Total Active Users histogram. There is jump in the Active Hosts histogram.

I changed the definition of an active host from granted credit in the last week to granted credit in the last month.

---

Join team BOINCstats

OOPS, my bad!! I meant Hosts.

Thanks for the info, I figured it was something simple!

Alinator

That link tells a story, project active users headed down big time. Looks like many must of lost interest in the project - hope they come back!

Since it's a lagging chart, it show that on or around April 13th the drop started. SE was released initially on May 1rst, yet there's no corresponding fall on June 1st as one might expect. But then, users probably had "pending credit" which means the drop from the release of SE would probably be seen on or about the 6-7th of June, but it's not there either. It is curious though, perhaps people had to sell puters to pay the federal tax?

It's the Electric Bill, I tell ya!

If we maybe get a penny for every 2 credits, I think that should be a break even for people.... Or a penny for every 130,000 seconds of CPU time spent. That should dissuade people from leaving...

I think there should be a SETI@Home economy with a gift shop area, money to credit - credit to money conversions, actual payouts, a credit transfer tool for people to use, actual games/contests, and so forth. The complexity would draw people's mind away from the actual computer work needed for scientific analysis and have them focus on other material goals.

Based on the last 60 days of Total number of active users' graph, the SETI@Home project is still losing active users...




Is that graph even correct? What else is going on to be causing a continueous steady decline like that?

I think the scaling on those graphs makes things look a little worse than they really are. Lets step back a bit for some perspective. Here is a graph from my stats site that spans an entire year. Be warned that I do use a different metric for "active" and my numbers are a little lower overall however they do follow the same trend.



A few things to point out. There is a steep increase in November/December of 2005. This is of course due to seti classic closing down. Immediately after that, the number of active users falls steadily for 3 months. Then there is a spike at the beginning of April. No clue why but there it is. The falloff after that is a little steeper than before but it does seem to be flattening out now.

The above graph can be seen in more context (with other graphs and time periods) here and of course for other projects just go to http://stats.kwsn.net and click on the project name.

---

A member of The Knights Who Say NI!
For rankings, history graphs and more, check out:
My BOINC stats site

Ni!

Very nicely put. It has been said before that the stats graphs can very very easily be misinterpreted. Having more than one site available to compare graphs does help -

in this case, Toby's graph seems a little more useful than the BOINCstats one. It captures the general trend better because it is compressed a bit more (includes more time).

This does help to see that the "steep drop" that you thought you saw in the BOINCstats graphs may not be as pronounced (not to mention unrelated to anything that happened on these forums).

Thanks for posting that very useful comparison Toby!

Regards,
Simon.

---

Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information

Toby, Simon

Toby Thank You and looking at some of the other projects they are also seeing a bit of decline... hmmmm could it be "summer" and in some locations school is out and people are doing silly things like vacations and travel... Where does that leave the average home computer? In most cases it is "off" or "forgotten" if the weather is nice and they are outside...

Fall as the weather starts turning to drive people "inside" we will watch users ramp up again... Right now my home office with two laptops and 5 machines has things nicely warm... and I still have Vista to get running on a couple of machines...

Regards

Pappa

Ni!

Very nicely put. It has been said before that the stats graphs can very very easily be misinterpreted. Having more than one site available to compare graphs does help -

in this case, Toby's graph seems a little more useful than the BOINCstats one. It captures the general trend better because it is compressed a bit more (includes more time).

This does help to see that the "steep drop" that you thought you saw in the BOINCstats graphs may not be as pronounced (not to mention unrelated to anything that happened on these forums).

Thanks for posting that very useful comparison Toby!

Regards,
Simon.

---

Please consider a Donation to the Seti Project.

Based on the last 60 days of Total number of active users' graph, the SETI@Home project is still losing active users...



Is that graph even correct? What else is going on to be causing a continueous steady decline like that?

Well the way I see it, since Enhanced came out - crunchers are going to projects that give out higher credit per hour ratings.
But I have not checked the graph to see if the credits have maintained a steady increase or has declined like the active users graph

Based on the last 60 days of Total number of active users' graph, the SETI@Home project is still losing active users...




Is that graph even correct? What else is going on to be causing a continueous steady decline like that?

It's just a bit off the graph now, but it peaked at 192,500...FYI. That's a lose of over 30,000 users and a lose of over 40,000 hosts.

---

Account frozen...

kevin

This is sad, IF I did not know better and the thread about Cross Project Calibration I would think that it was Funny...

In the Summer Users go outside and do things... Computers are left unattended... In the Fall and Winter they come Inside and do things on the computer...

At this point many projects are starting to get serious about Credit Wars... Obviously everyone would want to avoid that. It has hit Here and Einstien... who is next... There is nothing that you can say that will bump up credit more that Eric has acknowledged... You have really provided nothing that is "proof," there are many that have provided statistics that are "proof." I know enough about statitistic that if you give me what you want out of a set of numbers I can provide the answer that you are looking for...

Toby gave another viewpoint and I find that it probably more realistic.. But then I could really bore the Forum with Stats that I have collected to help Eric prove where "credit should be." So at this point you missed his post about the data collected and what he desired...

Overall, those that helped have given Eric a baseline... Thank You All, when 5.17 gets here it will increase credit/hour and be compatible with the new Multibeam ALFA Receiver... Or New Data to Crunch...

So the crude thing would be to say with information about "who" gives the best credit/hour then we should see a larger group move there... In the mean time those that are interested in Seti BOINC will still be here...
Otherwise as problems are fixed... It becomes a non issue, or he/she who has the fastest and largest number of machines wins... There is no inflation of "credit/hour."

Regards

Pappa

[quote]Based on the last 60 days of Total number of active users' graph, the SETI@Home project is still losing active users...

[snipped}

Well the way I see it, since Enhanced came out - crunchers are going to projects that give out higher credit per hour ratings.
But I have not checked the graph to see if the credits have maintained a steady increase or has declined like the active users graph

---

Please consider a Donation to the Seti Project.

Dogbytes

First, how people program things to represent the data is one thing... If this really concerns you, then there is nothing that you can say to convince Eric otherwise... You did not provide information to Eric when he requested it...

You now have only one choice, go to where you can get the most credits/hour... Why bore everyone else with the Cr@p?

Then people that have legitimate concerns can ask questions...

Regards

Pappa

Based on the last 60 days of Total number of active users' graph, the SETI@Home project is still losing active users...

Is that graph even correct? What else is going on to be causing a continueous steady decline like that?

It's just a bit off the graph now, but it peaked at 192,500...FYI. That's a lose of over 30,000 users and a lose of over 40,000 hosts.

---

Please consider a Donation to the Seti Project.

What concerns me is the dynamics of why the Seti project has lost so many crunchers (people/volunteers) in the recent past, and why the situation was allowed to occur.

As a matter of fact, I'm currently crunching Malaria beta (and some Rosetta) where the credit is a fraction of what I would be getting on most other projects especially those with optimized applications...so that is not the issue for me...wrong guess on your part Pappa.

Why bore everyone else with the Cr@p?

Perhaps that attitude has in some small measure something to do with the overall situation?

---

Account frozen...

Before we can start making statements like "why was this allowed to occur?" we need to find out for sure what actually happened.

It's really easy to say "well, the switch to enhance drove people off."

.... except that it appears that the change started about a month earlier.

It will be interesting to see if there is a jump in the fall -- when school starts up, vacations (or holidays) are over, etc.

... and on an unrelated subject, if you find a graphic that you like, that isn't copyrighted, put it on your own server where it uses your bandwidth. Putting a link to someone else's server is just plain wrong.

What concerns me is the dynamics of why the Seti project has lost so many crunchers (people/volunteers) in the recent past, and why the situation was allowed to occur.

As a matter of fact, I'm currently crunching Malaria beta (and some Rosetta) where the credit is a fraction of what I would be getting on most other projects especially those with optimized applications...so that is not the issue for me...wrong guess on your part Pappa.

Why bore everyone else with the Cr@p?

Perhaps that attitude has in some small measure something to do with the overall situation?

---

Dogbytes,
If you look at the graphs for BOINC and all the top projects, Seti, CPDN, Einstein etc, they all show a similar decline in hosts. So stop trying to say it is a Seti problem because they brought out enhanced and you are not getting the inflated RAC due to the optimised apps before Enhanced.

Andy

Before we can start making statements like "why was this allowed to occur?" we need to find out for sure what actually happened.

It's really easy to say "well, the switch to enhance drove people off."

.... except that it appears that the change started about a month earlier.

It will be interesting to see if there is a jump in the fall -- when school starts up, vacations (or holidays) are over, etc.

... and on an unrelated subject, if you find a graphic that you like, that isn't copyrighted, put it on your own server where it uses your bandwidth. Putting a link to someone else's server is just plain wrong.

What concerns me is the dynamics of why the Seti project has lost so many crunchers (people/volunteers) in the recent past, and why the situation was allowed to occur.

As a matter of fact, I'm currently crunching Malaria beta (and some Rosetta) where the credit is a fraction of what I would be getting on most other projects especially those with optimized applications...so that is not the issue for me...wrong guess on your part Pappa.

Why bore everyone else with the Cr@p?

Perhaps that attitude has in some small measure something to do with the overall situation?

You hit the mark here Ned. User retention issues are in no way SAH specific, this is a general BOINC problem. In fact, even a cursory examination of the data presented on any of the stat sites indicates that for the top seven projects:

1.) User retention overall is down significantly over the last 60 days, and continues to fall.

2.) For the two projects which are level or rising, the sum total of their gains is only a fraction of the loss. This would seem to indicate that project hopping due to credit "whoring", flame wars, phases of the moon, whatever is not the driving force at work for any SAH retention problems.

In addition, these trends began long before the last round of unpleasantness either here or anywhere else, and before the summer season arrived here in the northern hemisphere. I almost wish Jack Gulley would jump in here, since he was the first SAH NC "regular" I recall mentioning the trends in user retention back in January.

I will admit seasonal fluctuations makes sense, but since this is only the second summer season with the BOINC system in full operation it's hard to draw any conclusions at this point. As you say, it will be interesting to see what happens in September.

In any event, here are my working hypotheses:

1.) The "fad factor" of BOINC and DC in general is wearing off for the general public.

2.) The feeling of futility for new users (or previous users) when they jump in and discover after a while that compared to the humungous numbers posted on the leaderboards and stat sites for some individuals and/or teams their contribution is infintessimal.

3.) Given 1 and 2, with the cost of electricity increasing, and all the other factors of participation taken into account, if they have no valid reason to have a machine running all the time, what's the point to long term participation. Obviously, "It's all about the Science" isn't the answer here.

Alinator

Dogbytes,
If you look at the graphs for BOINC and all the top projects, Seti, CPDN, Einstein etc, they all show a similar decline in hosts. So stop trying to say it is a Seti problem because they brought out enhanced and you are not getting the inflated RAC due to the optimised apps before Enhanced.

Andy

Then what is the problem then? Where does it lay, I would be most interested to know. What caused this to happen then to BOINC? There are probably no simple answers but they could probably be found in a comprehensive survey. Just an idea if anyone is interested.

---

Account frozen...