I apologize to everyone on the forum if I kicked some kind of hornet's nest with this topic.
You didn't kick-over a thing.
We're all friends here just having something of a discussion and that's a good thing, in my opinion.
The subject of "credits" is often a hornets' nest, but it hasn't been this time.
Personally, I like the way Einstein awards credits. Remembering (as we all do) that credits are worth nothing at all, the more work units you complete successfully, the more total credits you have, and if you find that a useful measure; great!
As is pointed-out in this thread, what we used to do was count work units and equivalent CPU time. I liked that well-enough, too.
Then it started getting complicated despite the best of intentions. Can you really say that a FPU did as much work as an integer unit to multiply 100 x 1000? What about those "instruction sets" like SSE or MMX or AVX or... which computer is really doing more "work" or equivalent FLOPS?
A standard unit of measure, applied to the term "work" in this case, doesn't really translate very well between applications and among computers.
Everyone's heart was in the right place. It just wasn't so easy to do when it came to honestly applying the "idea" of equivalent "work" across projects.
The way Einstein does it, the more valid work units you complete, the more "credits" you have and if your computer does project X better than project Y, or if the credits granted are slightly skewed to one project or another, or even IF you and I were being manipulated into doing one project over another because it was a "rush-job," who cares? I'm happy to do what they want done in the order they value its completion.
It is an incentive to finish as much work as you can in your project of choice and gives you a way to measure your success or failure in doing that, both in competition against yourself and in competition with others.
And yes, you are absolutely correct -> The instruction sets change things. Since the credits are all the same, the instruction sets change the TIME it takes to earn those credits. If you can do the work more efficiently than I can and you can complete more work units in a day, I have no problem with your daily credit totals eclipsing mine even if my machine had to "work" harder at it because it lacks certain shortcuts.
The credits are not arbitrary, but they do reflect the calculation of a standard and not an actual FLOP count.
Gary, no, no, no... not the "R" word! Someone save me. I've spouted "rubbish."
Oh Gary... Oh, Gary, Gary, Gary...
You just have no idea how many times I've had to hear the statement I made spouted AT me.
You see, if I complain that a project is not able to run well, so makes a waste of the equipment I acquired specifically to run said project, I am often reminded that I was only asked for "spare" cycles on a part-time basis and that I have no reason (or right) to be disappointed by the project's inability to keep a machine busy. (Often followed by an invitation to take the farm and put it elsewhere if I don't like it.)
After all, it is argued, no project scientist ever mentioned buying hardware or running up a large electricity bill.
We all know that any sort of keeping-score would result in some competition, but I am usually brow-beaten about my own stupidity when all I was ever ASKED for were "spare cycles."
Now, thanks to you, I've been corrected for either position I took! Thanks, Gary! :-)
I think I'll go get a beer or two and cycle that! Nobody even needs to ask if they are "spare" or if they will be "wasted." The answers are "No," and "Eventually!"
in summary I disagree with the general line expressed by
Quote:
projects were intended to run on "spare cycles" via a screen saver and that it is only the most dedicated number-crunchers with their farms which have produced the situation where energy is wasted.
And, to my way of thinking, you SHOULD disagree.
Looked at an entirely different way, though... if fewer "cores" or GPUs were doing far less work by running far less often, then there would never be a time when any project "caught up" and reissued work it did not need to re-crunch for a scientific result.
So...in a way, you see that crunching a lot creates the opportunity to run out of work, which causes work to be issued and completed redundantly or at least with highly speculative parameters which would not be tried if newer data were available for distribution, which can be seen as a waste of energy.
But you really should not disagree with the "idea" that they were "intended" to run on "spare" cycles because clearly they were.
That's why the work did not run except as part of a screen saver. To understand what I mean, you really have to pay attention to the past-tense use of the word "intend," and recall that many people had a single-cylinder 486 computer with maybe 4MBs of RAM, rendering them almost incapable of crunching SETI and doing much of anything else simultaneously.
The programs were "intended" to run when you weren't doing anything else.
I apologize to everyone on the forum if I kicked some kind of hornet's nest with this topic.
If you kick a hornet's nest you get stung -- very badly. All that's happening here is that people are reminiscing about the 'good old days' in a polite and humourous way and taking the opportunity to draw attention to certain 'highlights' as they were perceived at the time. Nobody has been stung and I have every confidence it will stay that way.
I am a bit amazed with the conversation I read here...
Post count: 2 -- where have you been hiding? :-).
I'm genuinely pleased that you've decided to join in.
Quote:
When I contributed for the first time to Seti@home, around 15 years ago, the only reward, if my memory is good, was the fact that a task had been completed and validated by SETI. Then the cruncher score was no more than the number of completed tasks.
Yep, one returned WU = 1 credit. I don't think the Devs at the time understood how obsessive certain types of volunteers would become about 'climbing the ladder' at all costs. There was extensive cheating with many different variations, from deliberately selecting those tasks that would crunch (or completely fail due to RFI) very quickly to working out how to return and be credited for the same result multiple times.
Quote:
Before moving to BOINC, the SETI Team did not consider that it was his responsability to take into account the Internet connection rate of the cruncher, his computer speed or others. If one whished to make it better, he was free to move to an other city where the Internet communication was faster or he could buy a new computer...
The two factors you mention either were solved or had no relevance to the Classic -> BOINC transition. If you had very slow internet it took time to upload a result and then download the next WU. So you lost some crunch time. With caching under BOINC, this was fixed. You stored a number of tasks and didn't have to wait. If you had a slow computer you racked up your score slowly. It was just the same under BOINC.
Quote:
(note : I was proud of my 145 tasks despite the fact that some others were above 40,000. And as any cruncher when SETI moved to BOINC, I have been quite disappointed to learn that my 145 where lost for ever !)
Your classic credits were not lost by Seti. Your classic account was transferable to BOINC along with your Classic credit. I notice that the account you are currently using started at Seti in 2011. If you could access your original BOINC account from 2005 at the time of the closure of Classic, your Classic score would be showing there, I imagine. I was a volunteer that moved from Classic to BOINC and my Classic score came with me. I just had a look at my BOINC account at Seti which has been inactive for many years. I can see my BOINC credits, my Classic credits and the total number of CPU hours that were clocked up under Classic. I'm entirely satisfied with the way that Seti has preserved all that information.
Quote:
The issue is to get relevant results for a global benefit :
the project itself and the science it helps, and all of us, the crunchers who might have nothing to do if BOINC stops.
Best regards to all and thanks to Jonathan who has put the subject on the table.
No taboos here, I hope.
I completely agree that the aim of the game is the global benefit (real or potential) that will come from scientific discovery.
Genuine discussion of issues around how this is being progressed in an environment of volunteer computing will always be welcomed and never be 'taboo'. People are free to express genuinely held opinions and to debate counter opinions provided that such debates do not descend into personal attacks or insults. The rules are clearly shown each time you formulate a message.
Firstly : awkward social pauses upon mention of E@H. Been there, done that, got a wardrobe full of tee shirts on that ! So these days it may perform as a well executed conversation ending. :-)
Secondly : The Credit Issue Discussion. We have this on a regular basis here. It is our prime group-think bonding ritual. We have proposed it as a demonstration sport for the next Olympics. In the category of Arena Sports, and I think as a sub-category of the javelin. AFAIK no one has died yet on account, though much blood has been shed .... :-)
I will, per usual, put my Machiavelli suit on :
.... :-))))
To indicate, and I might add rather poncey/sage-like, that the science value is independent of your intentions. Which for me at least I flip over to give thanks to contributors absolutely regardless of how you got here, or why you stay here. So if credits float your boat then more power to you ... :-)
{ But beware a little considered piece of logic : the reason why E@H is a volunteer project ( hardware, electricity costs, internet connectivity ) is because the project doesn't have the money to do otherwise. }
Cheers, Mike.
( edit ) For myself, credits serve have valuable technical merit in measuring the performance/problems of my wee farm.
Mentioning this at parties in most of my social circles results in kind of an awkward pause ...
You're obviously going about this the wrong way ;-) :-).
The thing that works best in a social environment when people ask the awkward question about what you are doing these days, or how you spend your (obviously) copious amount of free time, is to start out slowly and ask this question in return. "Do you know that Einstein made certain predictions around 100 years ago and there is still just one that has never been directly observed?"
That tends to stop the idle chit-chat in its tracks as the others within earshot suddenly prick up their attention :-). You pause for effect and pretend to be thinking carefully and then you say, slowly and deliberately, "Well ... I'm involved in a project ... that will make that first ... direct ... observation.
Suddenly, everyone's interested, and wants to know more :-).
Of course, it's entirely up to you how you maintain that interest :-).
I do not regreat the "old good times" at all ! BOINC is much more powerful and efficient.
I just wished to state my personal point of view on the targets of BOINC, and I deeply respect the other's.
If I hurted someone (or many) I am confused because I did not wish this.
By the way, how to look at my contribution from "before" ? I don't even remember my linking codes !
That´s makes me remember an old discusion on SETI forums...
If credits means "nothing", why when we start to talk about "meaningless credits" and "spare cicles" the conversation allways goes to "agressive negotiations"?
We humans are competitives, even when we compeate for "nothing".
There are no explanation (besides madness) why we build supercrunchers or big farms, i speak for myself (before my moving, about 2 years ago i have 16 high end GPUs just for crunching), but what is the problem of that?
Science wins anyway, more processing power avaiable, more work done, etc.
The problem is not with those who do that, the problem is with the "hidden powers" who want to control all this resources to their own agenda.
RE: I apologize to everyone
)
You didn't kick-over a thing.
We're all friends here just having something of a discussion and that's a good thing, in my opinion.
The subject of "credits" is often a hornets' nest, but it hasn't been this time.
Personally, I like the way Einstein awards credits. Remembering (as we all do) that credits are worth nothing at all, the more work units you complete successfully, the more total credits you have, and if you find that a useful measure; great!
As is pointed-out in this thread, what we used to do was count work units and equivalent CPU time. I liked that well-enough, too.
Then it started getting complicated despite the best of intentions. Can you really say that a FPU did as much work as an integer unit to multiply 100 x 1000? What about those "instruction sets" like SSE or MMX or AVX or... which computer is really doing more "work" or equivalent FLOPS?
A standard unit of measure, applied to the term "work" in this case, doesn't really translate very well between applications and among computers.
Everyone's heart was in the right place. It just wasn't so easy to do when it came to honestly applying the "idea" of equivalent "work" across projects.
The way Einstein does it, the more valid work units you complete, the more "credits" you have and if your computer does project X better than project Y, or if the credits granted are slightly skewed to one project or another, or even IF you and I were being manipulated into doing one project over another because it was a "rush-job," who cares? I'm happy to do what they want done in the order they value its completion.
It is an incentive to finish as much work as you can in your project of choice and gives you a way to measure your success or failure in doing that, both in competition against yourself and in competition with others.
And yes, you are absolutely correct -> The instruction sets change things. Since the credits are all the same, the instruction sets change the TIME it takes to earn those credits. If you can do the work more efficiently than I can and you can complete more work units in a day, I have no problem with your daily credit totals eclipsing mine even if my machine had to "work" harder at it because it lacks certain shortcuts.
The credits are not arbitrary, but they do reflect the calculation of a standard and not an actual FLOP count.
RE: Rubbish - on a number
)
Gary, no, no, no... not the "R" word! Someone save me. I've spouted "rubbish."
Oh Gary... Oh, Gary, Gary, Gary...
You just have no idea how many times I've had to hear the statement I made spouted AT me.
You see, if I complain that a project is not able to run well, so makes a waste of the equipment I acquired specifically to run said project, I am often reminded that I was only asked for "spare" cycles on a part-time basis and that I have no reason (or right) to be disappointed by the project's inability to keep a machine busy. (Often followed by an invitation to take the farm and put it elsewhere if I don't like it.)
After all, it is argued, no project scientist ever mentioned buying hardware or running up a large electricity bill.
We all know that any sort of keeping-score would result in some competition, but I am usually brow-beaten about my own stupidity when all I was ever ASKED for were "spare cycles."
Now, thanks to you, I've been corrected for either position I took! Thanks, Gary! :-)
I think I'll go get a beer or two and cycle that! Nobody even needs to ask if they are "spare" or if they will be "wasted." The answers are "No," and "Eventually!"
RE: in summary I disagree
)
And, to my way of thinking, you SHOULD disagree.
Looked at an entirely different way, though... if fewer "cores" or GPUs were doing far less work by running far less often, then there would never be a time when any project "caught up" and reissued work it did not need to re-crunch for a scientific result.
So...in a way, you see that crunching a lot creates the opportunity to run out of work, which causes work to be issued and completed redundantly or at least with highly speculative parameters which would not be tried if newer data were available for distribution, which can be seen as a waste of energy.
But you really should not disagree with the "idea" that they were "intended" to run on "spare" cycles because clearly they were.
That's why the work did not run except as part of a screen saver. To understand what I mean, you really have to pay attention to the past-tense use of the word "intend," and recall that many people had a single-cylinder 486 computer with maybe 4MBs of RAM, rendering them almost incapable of crunching SETI and doing much of anything else simultaneously.
The programs were "intended" to run when you weren't doing anything else.
RE: I apologize to everyone
)
If you kick a hornet's nest you get stung -- very badly. All that's happening here is that people are reminiscing about the 'good old days' in a polite and humourous way and taking the opportunity to draw attention to certain 'highlights' as they were perceived at the time. Nobody has been stung and I have every confidence it will stay that way.
Sit back and enjoy the ride :-).
Cheers,
Gary.
RE: I am a bit amazed with
)
Post count: 2 -- where have you been hiding? :-).
I'm genuinely pleased that you've decided to join in.
Yep, one returned WU = 1 credit. I don't think the Devs at the time understood how obsessive certain types of volunteers would become about 'climbing the ladder' at all costs. There was extensive cheating with many different variations, from deliberately selecting those tasks that would crunch (or completely fail due to RFI) very quickly to working out how to return and be credited for the same result multiple times.
The two factors you mention either were solved or had no relevance to the Classic -> BOINC transition. If you had very slow internet it took time to upload a result and then download the next WU. So you lost some crunch time. With caching under BOINC, this was fixed. You stored a number of tasks and didn't have to wait. If you had a slow computer you racked up your score slowly. It was just the same under BOINC.
Your classic credits were not lost by Seti. Your classic account was transferable to BOINC along with your Classic credit. I notice that the account you are currently using started at Seti in 2011. If you could access your original BOINC account from 2005 at the time of the closure of Classic, your Classic score would be showing there, I imagine. I was a volunteer that moved from Classic to BOINC and my Classic score came with me. I just had a look at my BOINC account at Seti which has been inactive for many years. I can see my BOINC credits, my Classic credits and the total number of CPU hours that were clocked up under Classic. I'm entirely satisfied with the way that Seti has preserved all that information.
I completely agree that the aim of the game is the global benefit (real or potential) that will come from scientific discovery.
Genuine discussion of issues around how this is being progressed in an environment of volunteer computing will always be welcomed and never be 'taboo'. People are free to express genuinely held opinions and to debate counter opinions provided that such debates do not descend into personal attacks or insults. The rules are clearly shown each time you formulate a message.
Once again, thanks for joining in the discussion.
Cheers,
Gary.
Firstly : awkward social
)
Firstly : awkward social pauses upon mention of E@H. Been there, done that, got a wardrobe full of tee shirts on that ! So these days it may perform as a well executed conversation ending. :-)
Secondly : The Credit Issue Discussion. We have this on a regular basis here. It is our prime group-think bonding ritual. We have proposed it as a demonstration sport for the next Olympics. In the category of Arena Sports, and I think as a sub-category of the javelin. AFAIK no one has died yet on account, though much blood has been shed .... :-)
I will, per usual, put my Machiavelli suit on :
.... :-))))
To indicate, and I might add rather poncey/sage-like, that the science value is independent of your intentions. Which for me at least I flip over to give thanks to contributors absolutely regardless of how you got here, or why you stay here. So if credits float your boat then more power to you ... :-)
{ But beware a little considered piece of logic : the reason why E@H is a volunteer project ( hardware, electricity costs, internet connectivity ) is because the project doesn't have the money to do otherwise. }
Cheers, Mike.
( edit ) For myself, credits serve have valuable technical merit in measuring the performance/problems of my wee farm.
( edit ) I didn't expect the ....
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
RE: Mentioning this at
)
You're obviously going about this the wrong way ;-) :-).
The thing that works best in a social environment when people ask the awkward question about what you are doing these days, or how you spend your (obviously) copious amount of free time, is to start out slowly and ask this question in return. "Do you know that Einstein made certain predictions around 100 years ago and there is still just one that has never been directly observed?"
That tends to stop the idle chit-chat in its tracks as the others within earshot suddenly prick up their attention :-). You pause for effect and pretend to be thinking carefully and then you say, slowly and deliberately, "Well ... I'm involved in a project ... that will make that first ... direct ... observation.
Suddenly, everyone's interested, and wants to know more :-).
Of course, it's entirely up to you how you maintain that interest :-).
Cheers,
Gary.
RE: I've been corrected for
)
Rubbish ;-) :-).
Cheers,
Gary.
I do not regreat the "old
)
I do not regreat the "old good times" at all ! BOINC is much more powerful and efficient.
I just wished to state my personal point of view on the targets of BOINC, and I deeply respect the other's.
If I hurted someone (or many) I am confused because I did not wish this.
By the way, how to look at my contribution from "before" ? I don't even remember my linking codes !
Cheers
JFT
That´s makes me remember an
)
That´s makes me remember an old discusion on SETI forums...
If credits means "nothing", why when we start to talk about "meaningless credits" and "spare cicles" the conversation allways goes to "agressive negotiations"?
We humans are competitives, even when we compeate for "nothing".
There are no explanation (besides madness) why we build supercrunchers or big farms, i speak for myself (before my moving, about 2 years ago i have 16 high end GPUs just for crunching), but what is the problem of that?
Science wins anyway, more processing power avaiable, more work done, etc.
The problem is not with those who do that, the problem is with the "hidden powers" who want to control all this resources to their own agenda.
my 0.02 cents.