Having had the new "improved" version thrust upon me I was left with no option other than to try it.
Some of my machines (3) are slow but steady workers and would crunch a batch of work in around 3.5 hours for the fastest and 10 hours for the slowest.
This went up to 11 hours for the fastest and 46 hours for the slowest.
I do not have a problem with that however I do have a problem when more than a dozen workunits return as invalid due to noise or whatever.
This was with the standard client installed around mid May.
I consider this to be a waste of energy, energy that could be put to good use crunching work that is pre-checked for high noise levels.
A system is needed to sort out the chaff from the grain before it is issued to participants of the project enabling the project to become more environmentally friendly and cut the frustration that must be felt when you seem like you are getting more than your fair share of bad science!
I do not consider it to be too difficult to enable the system to recall work if the first two results of a workunit are returned as invalid due to noise.
This would at least spare the other two participants from crunching useless data and cut the odds of a participant's work being scored as invalid.
That by no means would solve all our problems but as I see it, it may just solve one major gripe that seems to prevail in these message boards of late.
The problem of an uneven playing field in respect of credit granted between Boinc projects will be a difficult one to solve but may I suggest that the various project leaders arrange an online meeting of all projects to level up the playing field. Maybe then some of us can go back to having some fun.

At the end of the day though is it not the science that counts?

Regards to all,
Graeme.
www.boincuk.com

---

It Might just be time to shut down the message boards... Stop the yellin and get rid of credits...

---

Do you Good Search for Seti@Home? http://www.goodsearch.com/?charityid=888957
Or Good Shop? http://www.goodshop.com/?charityid=888957

As for the drop off of active hosts/users.

Anyone want to take a stab at the number of hosts/users who are not getting no work because they have no idea that enhanced is live? I would guess a good number of machines out there are stuck with an outdated app_info.xml and optimized app because whoever put it there has changed jobs/school/etc. I do expect to see the number of active hosts/users level off shortly.

---

In actuality, the graph merely shows the tip of the real data. The visible data ranges from 177000 to 191000. The highest values occur from about 14 April thru about 25 April (approximately 188200 users), while the lowest (so far...they may get lower as time passes) are June 1 & 2 at slightly under 178400 (call it 178000 to round it off).

This works out to: (188200 - 178000) / 188200 = 0.054198 (5.4198%) loss in active users. That's 10200 active users--A lot (especially if the trend continues), but not the end of the world.

Just trying to keep things in perspective.

The graph line is pointing in only one direction - SOUTH.

Indeed it is. "Past performance is no guarantee of future results".

EDIT - removed extraneous quote keyword.

Martin, thanks for that. The current number if I calculate it correctly is indeed dropping even faster than the number of users in RandyC's users chart (about 20500 hosts from the peak to the present). And if the growth in host numbers prior to April 18th is extrapolated upwards the drop since then is something that would very greatly worry me if I were a project Admin. Thanks again, Mike

The chart was originally posted by dogbytes from the Boincstats site.

(Giving credit where it belongs).

Having had the new "improved" version thrust upon me I was left with no option other than to try it.
Some of my machines (3) are slow but steady workers and would crunch a batch of work in around 3.5 hours for the fastest and 10 hours for the slowest.
This went up to 11 hours for the fastest and 46 hours for the slowest.
I do not have a problem with that however I do have a problem when more than a dozen workunits return as invalid due to noise or whatever.
This was with the standard client installed around mid May.
I consider this to be a waste of energy, energy that could be put to good use crunching work that is pre-checked for high noise levels.

Isn;t there a basic infeasibility in that idea?
If your system is taking minutes or hours to find noisy work, then where are Berkeley going to find the same CPU time 1/2 million times over to pre-check the 1/2 million units issued daily?

On the other hand, I suppose I get noisy work too, but they must go through in seconds or a few minutes at most as I almost never see any. The CPU time 'lost' (but it's not lost as it had to be done to find it was noisy anyway), is trivial compared with the results that do process fully.

A system is needed to sort out the chaff from the grain before it is issued to participants of the project enabling the project to become more environmentally friendly and cut the frustration that must be felt when you seem like you are getting more than your fair share of bad science!
I do not consider it to be too difficult to enable the system to recall work if the first two results of a workunit are returned as invalid due to noise.
This would at least spare the other two participants from crunching useless data and cut the odds of a participant's work being scored as invalid.
That by no means would solve all our problems but as I see it, it may just solve one major gripe that seems to prevail in these message boards of late.
The problem of an uneven playing field in respect of credit granted between Boinc projects will be a difficult one to solve but may I suggest that the various project leaders arrange an online meeting of all projects to level up the playing field. Maybe then some of us can go back to having some fun.

At the end of the day though is it not the science that counts?

Regards to all,
Graeme.
www.boincuk.com

---

As for the drop off of active hosts/users.

Anyone want to take a stab at the number of hosts/users who are not getting no work because they have no idea that enhanced is live? I would guess a good number of machines out there are stuck with an outdated app_info.xml and optimized app because whoever put it there has changed jobs/school/etc. I do expect to see the number of active hosts/users level off shortly.

All the anti-optimized say "Only 1% of users were optimized."

To this I once again call BS.

---

I'm not in it for the science...only the competition.

The graph line is pointing in only one direction - SOUTH.

Reminds me of the lyrics by Status Quo:

Deeper and down
Down down deeper and down
Down down deeper and down
Down down deeper and down
Get down deeper and down

We all believe in the aims of the project and just want S@H to listen to the users and to acknowledge and even start to address the various issues with enhanced. Communication during change is key to motivating people, S@H need to provide better communication to reverse the Southerly direction.

We all believe in the aims of the project and just want S@H to listen to the users and to acknowledge and even start to address the various issues with enhanced. Communication during change is key to motivating people, S@H need to provide better communication to reverse the Southerly direction.

I It might help in future if the issue that regularly brought up is addressed and i think you touched on it there. The Seti staff could at least try to appear as thought they a doing more work. Not to say they are not I'm sure they work quite hard. It does seem though that the majoratie of what they end up doing is fixing things that it seems that they were just crossing their fingers and hoping wouldn't fall apart (again). Enhanced which was releasedd before it was really stable and ready is just the latest on a list of thinks they're having to mend on the fly.

Not meant as a discredit to the hard work they do do.

@ Halifax-Lad
I'm with you.

Cheers Nightbird, I knew someone somewhere would be!! :))

---

Join us in Chat (see the forum) Click the Sig


Join UBT

Dogbytes, the graph you posted is rather misleading.

On the surface, it seems to show a DRAMATIC dropoff in active users, with the implication that large numbers have packed-up and left because they're disgusted with how Seti is treating them or others (and that could be right for all I know/can prove).

In actuality, the graph merely shows the tip of the real data. The visible data ranges from 177000 to 191000. The highest values occur from about 14 April thru about 25 April (approximately 188200 users), while the lowest (so far...they may get lower as time passes) are June 1 & 2 at slightly under 178400 (call it 178000 to round it off).

This works out to: (188200 - 178000) / 188200 = 0.054198 (5.4198%) loss in active users. That's 10200 active users--A lot (especially if the trend continues), but not the end of the world.

Just trying to keep things in perspective.

It's not only a problem of active user.
Look at the amount of new partecipants.
They are 1/3 than two months ago.

http://www.teamocuk.com/cprojectpart.php?p=SAH

Also active teams are going down
http://www.teamocuk.com/cprojectteam.php?p=SAH

Bye
Francesco
(CH)

---

I don't know were your getting the numbers for new participants. I've been looking at the numbers daily on BoincStats (one of the sites that automatically opens when I start my browser) since a couple of months before Classic closed. The numbers then as they are now hovered between 400 and 600 daily. This started to increase in early December to between 1000 and 1500 when the hangers on and the rest of the masses finally started to move their boxes to Boinc. If memory serves this bump lasted approximately into the beginning of March when (outages aside) they slowly crept back down to current levels. The new computer attached as also remained within a factor of 2 to 3 of new users.

Despite the hysteria going on in this forum over the last month, I've seen no indication that new users have not been deterred from joining Seti at the same rate as they always did.

---

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

I don't know were your getting the numbers for new participants.
[...]

http://www.teamocuk.com/cprojectpart.php?p=SAH
They indicate now about 400 / day but 2 months ago we had 1200 / day.
I don't know before but I can look for some data.

Hysteria is caused, by my opinion by errors and bad decision taken by the management of this project.
Three version in a month and the change of calculating results. Too much.

Bye,
Franz

---

I don't know before but I can look for some data.

Ok, found!



As you can see, now for last 5 weeks, we have an average of 3200 new user / week.
Total average (7242) is double and two months ago was higher (depending on press articles, I suppose). The first two weeks of April we had an average of 10'528 and now the averge of last 5 week is 3182 (les than 1/3).
Add the decrease of active users and I think that it's better to meditate.

Bye,
Franz

---

Exactly as I said that peak is when Classic shut down and farms were forced to move over to Boinc and the downward trend is only an indication, as expected, that the most of the people that are comming over, have done so, and that fewer people are discovering that Classic is no longer working. If you spent time in the Q&A forums you would have seen a correlation with the number of questions from new users coming in from Classic.

You have to look at the entire picture not a snapshot that supports your hypothesis. If you look at data before Dec. 2005 you would see a flat line fluctuation between 400 to 600 of new users which is what we are returning to.

---

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

Exactly as I said that peak is when Classic shut down and farms were forced to move over to Boinc and the downward trend is only an indication, as expected, that the most of the people that are comming over, have done so, and that fewer people are discovering that Classic is no longer working. If you spent time in the Q&A forums you would have seen a correlation with the number of questions from new users coming in from Classic.

You have to look at the entire picture not a snapshot that supports your hypothesis. If you look at data before Dec. 2005 you would see a flat line fluctuation between 400 to 600 of new users which is what we are returning to.

So, just to fix some data.
1) I have not taken a snapshot in order to support my hypo.
I have data in my PC only starting from the first date you see.
I don't like that you suggest my imbezzlement.

2) before december the project was a "casino" (work stopped for long days or week) and very few new users because they prefer to stay in the old setiapp (not boinc).

So I think that in any case the stats before december 05 are not good to be compared with 2006.

Maybe it's better to compare today new users average with the one in 2002, 2003, 2004. Do you have data?

Bye,
Franz

---

A couple of months ago, I saw some guy who'd made all the BOINC pages XHTML get the short schrift. Although I thought the guy was treated very rudely, and I wasn't impressed, I thought it was probably a once-off, so I didn't get my knickers in a twist over it. I was/am new to this.

It seems like a waste of electricity at a time when we should be trying to reduce energy consumption.

Unlike some other projects, SETI is almost certainly pointless and worthless: we're looking for signals the like of which humans have used for less than 100 years, and won't use for much longer, representing maybe a century in the technological development of a sentient species. If I'm using valuable energy to crunch at all, I think it's better to crunch for a project that has a chance of success.

IMVHO, contrary to the accepted wisdom here, I think credit and RAC *are* a kind of status symbol. I think it's pretty cool to run up zillions of credits. I only run it on a few machines myself, so I suppose I should be one of the begrudgers, except that most, if not all, of the fun for me was in trying different optimised clients and seeing how much credit you can squeeze out of whatever boxes you have got.

I noticed that my RAC had dropped and I hadn't got any new work units for a while, so I came to have a look around the fora and found this crap going on: more people being treated rudely, significant contributors leaving, forum-postings being censored, just a pretty pervasive bad vibe.

Overall, I get the sense that the SETI tsars dismiss crunchers as serfs, rather than appreciating them as contributors. It is no surprise that the lowlier the serf, the more willing he is to kiss the tsar's ring.

It seems likely that S@H will become a salutory lesson in how not to treat the community; remember XFree86?

My $0.02 worth.

Emmet

I'm not impinging your integrity, merely pointing out why, in my opinion, you have reached the erroneous conclusion that the new users are on a decline. The chart is a good indicator of the migration since the Classic closure. Unfortunately, it is lacking two vital element to support your conclusion about new users. The first is a baseline of new user prior to beginning of migration, and second the gradual migration ramp up that occurred for a few months prior to Seti-Classic officially shutting down. Boinc had a steady flow of new to Seti users long before Classic announced its closing date.

I'm not a data collector nor statistician, I can only relate from my observation and experience, since I was here almost daily during the past two years.

Also, older Classic data would be irrelevant as it would comparing apples and oranges unless you want to bring in the entire DC community into the mix. Boinc is a totally different approach to the software and versatility by allowing other projects to be attached and syphon off users.

---

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

Overall, I get the sense that the SETI tsars dismiss crunchers as serfs, rather than appreciating them as contributors. It is no surprise that the lowlier the serf, the more willing he is to kiss the tsar's ring.
Emmet

Thus, cheerleaders are born... haha :)

---

http://www.setiusa.net

Overall, I get the sense that the SETI tsars dismiss crunchers as serfs, rather than appreciating them as contributors. It is no surprise that the lowlier the serf, the more willing he is to kiss the tsar's ring.
Emmet

Thus, cheerleaders are born... haha :)

Give me a "U"...
Give me an "H"...

*acts spacey*

Give me a clue to what you're talk about! lol :)

-Kelly
former cheerleader