Another discussion on the New Credit System

Message boards : Number crunching : Another discussion on the New Credit System

To post messages, you must log in.

Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · Next

AuthorMessage
zombie67 [MM]
Avatar

Send message
Joined: 11 Feb 06
Posts: 316
Credit: 6,347,141
RAC: 1
Message 27909 - Posted: 21 Sep 2006, 2:24:36 UTC - in response to Message 27905.  

I'd have thought that being as correct as possible is better than it being an all or nothing type thing.


It's already 100% correct, based on the rules at the time.
Reno, NV
Team: SETI.USA
ID: 27909 · Rating: 9.9920072216264E-15 · rate: Rate + / Rate - Report as offensive    Reply Quote
zombie67 [MM]
Avatar

Send message
Joined: 11 Feb 06
Posts: 316
Credit: 6,347,141
RAC: 1
Message 27917 - Posted: 21 Sep 2006, 6:33:10 UTC

Testing....

No new posts in any threads in over two hours. Just trying to see if the interest died, or if there is some other technical problem.
Reno, NV
Team: SETI.USA
ID: 27917 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
XS_Vietnam_Soldiers

Send message
Joined: 11 Jan 06
Posts: 240
Credit: 2,880,653
RAC: 0
Message 27919 - Posted: 21 Sep 2006, 7:31:14 UTC - in response to Message 27904.  
Last modified: 21 Sep 2006, 7:36:02 UTC


An interesting point. When XS got into Rosetta in January 2006 we were months behind some teams here.We had to overcome that and we did it by going from the orginal 65 members to a max of over 600 registered members.We added machines. I went from running 2 to running a max of 6. Now a lot of those didn't crunch on a regular basis, a lot just came at our attempt at a million point day and then stopped.
The point is we had to grow to catch the top teams and that option is just as open today as it was back then.


In January, the credit system was the same as it had been when those other teams started. That old credit system made it a whole lot easier for you to make up positions. Think of it this way, if the new credit system had come into effect in February, would XS have anywhere near the score they do now? No - but you would still have done the same amount of work.

Speaking just for myself, I would have no issue with backdating my credits to day one.That is me speaking strickly for myself and not for XS.
Starting from zero says to those that what you did was worthless and insults their effort.
Sort of like:" Up till now this was all a trial, now we'll do the real thing"


Starting from zero doesn't say it was worthless. The work was still done, and the sensible thing to do would be to have a snapshot of the final stats. So everybody would see that XS did more work than anyone else. Starting from zero just recognises that there's now a new credit system and counting things from that. We can't change the new credit system to fit the old one, so the other option would be to change the old one to fit the new one. And that would be backdating, which is a controversial one. But leaving things as is, and adding new credit straight on top of the old ones is a bad idea that'll have effects for years.

Tony:
A complete reset may be the ONLY way if you want to only count the work that was done from today onward but what does that say to the people that busted their backside and incurred huge expenses during the first 8 months of the year?
It says what was done is worthless and they're lot of people, myself included, who would scream to the heavens were that done.
JUST me, one little guy: My monthly electric bill was an additional $150.00-$200.00 higher during that period, all atributable to running the extra PC's for rosetta. We'll forget the $15,000.00 worth of equipment that I put solely on this project during that time.
If the developers want to say that my work was not worthy, at the very least cut me a check for the extra electric that I used just for them.
That works out to app $1400.00 for 8 months.


See above. The work was still done, as long as you got recognition for the first 8 months worth of work I don't see the problem.

You don't see the problem because you don't want to see it.
Look at the D2OL stats page. What do people see? They see the NEW stats and thats all they see unless they go looking for the OLD stats.
XS has been through this before and there were no optimised clients in D2OL, it was pure determination that got us to #1 and the science end had huge amounts of work done in that drive.
People do not "see" stored stats pages, they see what is current as that is where their interest is.
XS didn't stop crunching Rosetta when we made #1, that was in June if I remember.
We added another 20+ million points over and above what Free Dc produced after we passed them. Baker asked for more over the summer and he got it in spades from us. When it was over 110F here I kept 6 machines running 24/7 to help him get what he needed for CASP7.
One look at the stats page explains to me your interest. Your team is almost 50 million points behind XS and you are the current#2 producer since XS and Free Dc have drastically cut back in Rosetta and you see that you can never catch up as things sit today.
My answer is as stated above: You want to be #1, go add the people and add the machines and do it. If it takes you 2 years to get to #1 so be it.
We didn't get there overnight. It was 6 months of hard work.
Now you will say that the change makes that all but impossible.
My answer to that is why didn't you get on this much earlier?
You had the same playing field as anyone else this year and didn't have either the manpower or the machinery to get to #1 or you would have done so.
Points don't rule my life, but I am loyal to the guys on my team that worked their backsides off to get to a goal and I won't sit idly by while someone try's to negate that when the reality is that all that person is after is a fast way to get to #1..
Your just too transparent..


ID: 27919 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jose

Send message
Joined: 28 Mar 06
Posts: 820
Credit: 48,297
RAC: 0
Message 27920 - Posted: 21 Sep 2006, 7:50:40 UTC - in response to Message 27906.  
Last modified: 21 Sep 2006, 7:53:10 UTC

but I don't remember 5.5.0 being around before about March or April. It had the biggest effect.
5.2.13, 5.2.14, 5.3.6, are were all around at the beginning their effects were nearly the same as 5.5.0. 5.5.0 is just the newest which includes the official boinc updates. It's not much different from us regular users moving from 5.2.6 to 5.2.13 (official) to 5.4.9 to 5.4.11. They were all designed to try to increase the credit one got when they ran the optimized seti application. If the application cut your time in half, then they tried to double your credit (older opt app versions), if they cut your time to 1/3, then they claimed 3 times. It's just that some users are using them with something other than a matching app, and claiming 3x anyway. (note: actually they tried to claim 32.29 credits/wu which was the idealized value of the reference wu (one picked from a hat), unfortunately the ref WU turned out to be a long one. The average seti wu was worth 24-25, credits and the opt claims of 32 were wrong even with the opt app, but that's a whole different story).



Tony :

Can I ask you a small favor?


Please do not mention SETI work units credits or as a matter of fact no other project? I will ask the same to everyone.

If you (any of us here) brings other projects in here, you will be introducing the extraneous issue of Project Comparability in credits and I can promise you you will be asking for a not so nice discussion on Project value: questions like Why are you comparing a project like Rosetta where science is done to a Project Like SETI where the science fiction is being pursued?

Bring SETI in and I can promise you a nasty flame war: one you , nor I ,nor anyone that has seriously participate here wants or deserves.
This and no other is the root from which a Tyrant springs; when he first appears he is a protector.”
Plato
ID: 27920 · Rating: -2 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Biggles
Avatar

Send message
Joined: 22 Sep 05
Posts: 49
Credit: 102,114
RAC: 0
Message 27927 - Posted: 21 Sep 2006, 8:18:22 UTC - in response to Message 27909.  

I'd have thought that being as correct as possible is better than it being an all or nothing type thing.


It's already 100% correct, based on the rules at the time.


So was racism and discrimination. Yet thing have changed to make up for that in the past. I went to bed, that's why there was no replies.
ID: 27927 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Biggles
Avatar

Send message
Joined: 22 Sep 05
Posts: 49
Credit: 102,114
RAC: 0
Message 27929 - Posted: 21 Sep 2006, 8:30:23 UTC - in response to Message 27919.  


You don't see the problem because you don't want to see it.
Look at the D2OL stats page. What do people see? They see the NEW stats and thats all they see unless they go looking for the OLD stats.
XS has been through this before and there were no optimised clients in D2OL, it was pure determination that got us to #1 and the science end had huge amounts of work done in that drive.
People do not "see" stored stats pages, they see what is current as that is where their interest is.
XS didn't stop crunching Rosetta when we made #1, that was in June if I remember.
We added another 20+ million points over and above what Free Dc produced after we passed them. Baker asked for more over the summer and he got it in spades from us. When it was over 110F here I kept 6 machines running 24/7 to help him get what he needed for CASP7.
One look at the stats page explains to me your interest. Your team is almost 50 million points behind XS and you are the current#2 producer since XS and Free Dc have drastically cut back in Rosetta and you see that you can never catch up as things sit today.
My answer is as stated above: You want to be #1, go add the people and add the machines and do it. If it takes you 2 years to get to #1 so be it.
We didn't get there overnight. It was 6 months of hard work.
Now you will say that the change makes that all but impossible.
My answer to that is why didn't you get on this much earlier?
You had the same playing field as anyone else this year and didn't have either the manpower or the machinery to get to #1 or you would have done so.
Points don't rule my life, but I am loyal to the guys on my team that worked their backsides off to get to a goal and I won't sit idly by while someone try's to negate that when the reality is that all that person is after is a fast way to get to #1..
Your just too transparent..



So transparent you think I'm on Anandtech? You've just undermined your arguments by making a personal, unjustified attack based on a wrong premise.

I did get on this when the issue of the new credit system first appeared. I said it was only a half measure then.

If a team with as much power as XS were to run Rosetta for a year, they would get only 1/3rd of the credit that XS did in the first year of Rosetta. Why is that? Because optimised clients overclaimed by a large amount. That's not fair no matter how you swing it. They could be just as dedicated, more so even, but if they ran Rosetta after the new credit system was in place, they had no chance to overclaim.

Think about WCG for a minute. If the scoring system had changed and you could only earn 1/3rd of what you currently do, would you be able to catch Easynews?
ID: 27929 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jose

Send message
Joined: 28 Mar 06
Posts: 820
Credit: 48,297
RAC: 0
Message 27932 - Posted: 21 Sep 2006, 8:35:26 UTC - in response to Message 27929.  


You don't see the problem because you don't want to see it.
Look at the D2OL stats page. What do people see? They see the NEW stats and thats all they see unless they go looking for the OLD stats.
XS has been through this before and there were no optimised clients in D2OL, it was pure determination that got us to #1 and the science end had huge amounts of work done in that drive.
People do not "see" stored stats pages, they see what is current as that is where their interest is.
XS didn't stop crunching Rosetta when we made #1, that was in June if I remember.
We added another 20+ million points over and above what Free Dc produced after we passed them. Baker asked for more over the summer and he got it in spades from us. When it was over 110F here I kept 6 machines running 24/7 to help him get what he needed for CASP7.
One look at the stats page explains to me your interest. Your team is almost 50 million points behind XS and you are the current#2 producer since XS and Free Dc have drastically cut back in Rosetta and you see that you can never catch up as things sit today.
My answer is as stated above: You want to be #1, go add the people and add the machines and do it. If it takes you 2 years to get to #1 so be it.
We didn't get there overnight. It was 6 months of hard work.
Now you will say that the change makes that all but impossible.
My answer to that is why didn't you get on this much earlier?
You had the same playing field as anyone else this year and didn't have either the manpower or the machinery to get to #1 or you would have done so.
Points don't rule my life, but I am loyal to the guys on my team that worked their backsides off to get to a goal and I won't sit idly by while someone try's to negate that when the reality is that all that person is after is a fast way to get to #1..
Your just too transparent..



So transparent you think I'm on Anandtech? You've just undermined your arguments by making a personal, unjustified attack based on a wrong premise.

I did get on this when the issue of the new credit system first appeared. I said it was only a half measure then.

If a team with as much power as XS were to run Rosetta for a year, they would get only 1/3rd of the credit that XS did in the first year of Rosetta. Why is that? Because optimised clients overclaimed by a large amount. That's not fair no matter how you swing it. They could be just as dedicated, more so even, but if they ran Rosetta after the new credit system was in place, they had no chance to overclaim.

Think about WCG for a minute. If the scoring system had changed and you could only earn 1/3rd of what you currently do, would you be able to catch Easynews?


We are going to catch Easynews. BTW we are using standard clients in WCG and we are bound by a quorum of 3 and we dont have our full team assembled and yet we are moving past many teams in a flash.



This and no other is the root from which a Tyrant springs; when he first appears he is a protector.”
Plato
ID: 27932 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jose

Send message
Joined: 28 Mar 06
Posts: 820
Credit: 48,297
RAC: 0
Message 27933 - Posted: 21 Sep 2006, 8:43:55 UTC
Last modified: 21 Sep 2006, 8:45:49 UTC

Can someone explain to me why instead of complaining about the unfairness of catching up to XtremeSystems under the new credit system and not asking for a backdating of credits, the Dutch Power Cows are going for it?

Could it be that they really have the desire, the discipline to face a hard task. Or could it be that they have realized that winners do not whine; they work to win?


Cow Stats
This and no other is the root from which a Tyrant springs; when he first appears he is a protector.”
Plato
ID: 27933 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile carl.h
Avatar

Send message
Joined: 28 Dec 05
Posts: 555
Credit: 183,449
RAC: 0
Message 27934 - Posted: 21 Sep 2006, 8:45:42 UTC

Biggles, I respect your viewpoint and will put my view soon...as I`ve had my breakfast. This IS NOT a personal flame toward you.

The statements you have made on XS are not fact, no one knows how many were running opti, you are generalising in the same way as others. No one actually knows what effect going back over the scores with a new measure would do.

Please take this statement as just that and don`t go for my throat, you maybe surprised later what I have to say.
Not all Czech`s bounce but I`d like to try with Barbar ;-)

Make no mistake This IS the TEDDIES TEAM.
ID: 27934 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Biggles
Avatar

Send message
Joined: 22 Sep 05
Posts: 49
Credit: 102,114
RAC: 0
Message 27938 - Posted: 21 Sep 2006, 8:59:26 UTC - in response to Message 27932.  


We are going to catch Easynews. BTW we are using standard clients in WCG and we are bound by a quorum of 3 and we dont have our full team assembled and yet we are moving past many teams in a flash.


Think about it for a minute Jose. Easynews have 882 million points. Yesterday XS earned 750,000 points.

Now imagine the scoring changed and that 750,000 a day turned into 250,000 a day, but that Easynews kept 882 million points.

You would still be doing the same amount of work, but it would three times harder to catch them.

Would it be fair to change the scoring system and then make it more difficult for you to catch anyone else?
ID: 27938 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jose

Send message
Joined: 28 Mar 06
Posts: 820
Credit: 48,297
RAC: 0
Message 27940 - Posted: 21 Sep 2006, 9:10:11 UTC - in response to Message 27938.  
Last modified: 21 Sep 2006, 9:28:57 UTC


We are going to catch Easynews. BTW we are using standard clients in WCG and we are bound by a quorum of 3 and we dont have our full team assembled and yet we are moving past many teams in a flash.


Think about it for a minute Jose. Easynews have 882 million points. Yesterday XS earned 750,000 points.

Now imagine the scoring changed and that 750,000 a day turned into 250,000 a day, but that Easynews kept 882 million points.

You would still be doing the same amount of work, but it would three times harder to catch them.

Would it be fair to change the scoring system and then make it more difficult for you to catch anyone else?



Biggles:

Of course it is going to be hard!!!!! That is the beauty facing what are almost impossible odds, XS is not backing down , nor is it asking for new rules.

Hard is not bad: it shows character and desire and discipline. Only the weak like easy stuff.

That is the beauty of it all. WHEN we beat Easy news , part of the pleasure will come from not only beating a great team of most excellent crunchers but doing it without asking for special favors, belittling our opponents or calling them names (although a little bantering and the usual poking in the sides will happen)or changing ex post facto the credit granting rules under which the project was started.

The other beautiful part is that a lot and I mean a lot of work units are going to be crunched and a lot of science is going to be done as the Easynews people are not going down without a fight. Those servers at WCG may need an extra cooling soon.

Easynews is a most excellent team , a powerful team, I know that if the situation was the reverse XS on top they going after us, they wouldn't do what others have done against XS here as they are a classy team.

Aside:

Now visualize this: the DPCs are mounting a charge here. How many extra wus fully active Free DC and XS teams would have produced for Rosetta during a challenge of this magnitude ( the top three teams involved)? But we will never know as those extra wus will never happen as at least one of the teams in question swore off this project as a result of the insults and the lack of respect shown to it by the project developers. So science at Rosetta lost .

This and no other is the root from which a Tyrant springs; when he first appears he is a protector.”
Plato
ID: 27940 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile carl.h
Avatar

Send message
Joined: 28 Dec 05
Posts: 555
Credit: 183,449
RAC: 0
Message 27943 - Posted: 21 Sep 2006, 9:40:45 UTC
Last modified: 21 Sep 2006, 9:42:10 UTC

This is rather an enigmatic situation, there are valid points for each solution but is there really a totally satisfactory solution to all involved ? The answer has to be no, which is perhaps why the devs have simply kept it "as is".

We know the points system was corrupted in Boinc long before Rosetta, the obvious solution now there is a "fair" system...cancel them out and restart which seem`s logical. This negates all the previous work and dedication of people to the project, people who had given but no longer will be recognised at all. This solution was my immediate and fairly thought out, I did not see the pitfalls.Dedicated people gave $millions to those first 9-11 months are we to simply forget that ? Is re-running the race after some participants retired fair ?

Backdating a revised figure either up or down.

Since I called for others to also take up crunch3r files this would appear to me to seem very fair. Unfortunately the Devs now say for whatever reason, it technically cannot be done. There are other points against this solution but it`s a waste of time even thinking on it, apparently.

A few months ago, Jose was the loudest voice against the corrupt Boinc system, his idea of getting a credit system away from the Boinc client was taken up eventually. I think Jose would have liked to have seen Rosetta away from Boinc altogether and he put in an awful amount of work for the project away from the computer wu side. He was extremely dedicated and thought very highly of the work being done.

So I will agree on a lot of points being made by all sides but as I see it there is no solution. Boinc is still corrupt perhaps not here but in other projects, the cross project points are also a laughing stock and are not credible nor have they been since the first instance of an opti.

I do wonder how long it will take the argument to fade, when current points are 10% of the totals less, more ? I don`t know.
Not all Czech`s bounce but I`d like to try with Barbar ;-)

Make no mistake This IS the TEDDIES TEAM.
ID: 27943 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
XS_Vietnam_Soldiers

Send message
Joined: 11 Jan 06
Posts: 240
Credit: 2,880,653
RAC: 0
Message 27950 - Posted: 21 Sep 2006, 11:34:27 UTC - in response to Message 27929.  
Last modified: 21 Sep 2006, 11:53:39 UTC


You don't see the problem because you don't want to see it.
Look at the D2OL stats page. What do people see? They see the NEW stats and thats all they see unless they go looking for the OLD stats.
XS has been through this before and there were no optimised clients in D2OL, it was pure determination that got us to #1 and the science end had huge amounts of work done in that drive.
People do not "see" stored stats pages, they see what is current as that is where their interest is.
XS didn't stop crunching Rosetta when we made #1, that was in June if I remember.
We added another 20+ million points over and above what Free Dc produced after we passed them. Baker asked for more over the summer and he got it in spades from us. When it was over 110F here I kept 6 machines running 24/7 to help him get what he needed for CASP7.
One look at the stats page explains to me your interest. Your team is almost 50 million points behind XS and you are the current#2 producer since XS and Free Dc have drastically cut back in Rosetta and you see that you can never catch up as things sit today.
My answer is as stated above: You want to be #1, go add the people and add the machines and do it. If it takes you 2 years to get to #1 so be it.
We didn't get there overnight. It was 6 months of hard work.
Now you will say that the change makes that all but impossible.
My answer to that is why didn't you get on this much earlier?
You had the same playing field as anyone else this year and didn't have either the manpower or the machinery to get to #1 or you would have done so.
Points don't rule my life, but I am loyal to the guys on my team that worked their backsides off to get to a goal and I won't sit idly by while someone try's to negate that when the reality is that all that person is after is a fast way to get to #1..
Your just too transparent..



So transparent you think I'm on Anandtech? You've just undermined your arguments by making a personal, unjustified attack based on a wrong premise.

I did get on this when the issue of the new credit system first appeared. I said it was only a half measure then.

If a team with as much power as XS were to run Rosetta for a year, they would get only 1/3rd of the credit that XS did in the first year of Rosetta. Why is that? Because optimised clients overclaimed by a large amount. That's not fair no matter how you swing it. They could be just as dedicated, more so even, but if they ran Rosetta after the new credit system was in place, they had no chance to overclaim.

Think about WCG for a minute. If the scoring system had changed and you could only earn 1/3rd of what you currently do, would you be able to catch Easynews?

My point in saying "get on this much earlier" meant why didn't you get on Rosetta much earlier?Had you been on the project in force last January you'd have had the same chance and opportunity that any other team to be #1.
You evidently didn't and now you want a second chance and that second chance to come at the price of what the top teams did being rolled into the back room out of sight.
As to making it a personal attack, that was based on what I saw on the stats pages. The person calling for the zeroing of points and starting over is also the head of the currently #2 largest producing team with the most to gain from such an action. What other conclusion could I possibly draw?
As to XS only getting 1/3 of what we did get, again you are wrong.
Jan 1 to Aug 25, almost 8 full months..Optimised files used from Mid April to end of August. 3.5/8ths of the total working with the stock client..
Care to revise your comments as to what percentage of what we have now would still be there? Now matter what math you use it's way beyond the 1/3 you suggest.
As to if the scoring system changed at WCG and we got 1/3 of the amount we got now, would we still be able to catch EasyNews..The answer is yes, we'd tripple the machines or quadruple them, whatever it took.
You just don't understand goal oriented people.You do what it takes to get to what your striving for..
ID: 27950 · Rating: 9.9920072216264E-15 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Biggles
Avatar

Send message
Joined: 22 Sep 05
Posts: 49
Credit: 102,114
RAC: 0
Message 27968 - Posted: 21 Sep 2006, 14:06:51 UTC - in response to Message 27950.  

My point in saying "get on this much earlier" meant why didn't you get on Rosetta much earlier?Had you been on the project in force last January you'd have had the same chance and opportunity that any other team to be #1.
You evidently didn't and now you want a second chance and that second chance to come at the price of what the top teams did being rolled into the back room out of sight.


We have been around on Rosetta for a while. We weren't interested in diverting power from our large GIMPS, SETI or F@H teams to take number one. We still aren't. I'm bothered by things as a matter of principle. Nobody is penalising XS as they make an assault on WCG, but the new credit system penalises any team that makes an assault on Rosetta.


As to making it a personal attack, that was based on what I saw on the stats pages. The person calling for the zeroing of points and starting over is also the head of the currently #2 largest producing team with the most to gain from such an action. What other conclusion could I possibly draw?


That you can't read perhaps?

The currently 2nd largest producing team is Anandtech.


As to XS only getting 1/3 of what we did get, again you are wrong.
Jan 1 to Aug 25, almost 8 full months..Optimised files used from Mid April to end of August. 3.5/8ths of the total working with the stock client..
Care to revise your comments as to what percentage of what we have now would still be there? Now matter what math you use it's way beyond the 1/3 you suggest.


What about the math based on XS having advocated the use of optimised clients since the 31st of December 2005? This thread has talked about optimised clients and encouraged the use of them since the day after it was posted. I don't believe that it took three and a half months for XS to notice that part of the setup guide. Optimised clients and the subsequent overclaiming of credit was widespread before April.


As to if the scoring system changed at WCG and we got 1/3 of the amount we got now, would we still be able to catch EasyNews..The answer is yes, we'd tripple the machines or quadruple them, whatever it took.
You just don't understand goal oriented people.You do what it takes to get to what your striving for..


I guess EasyNews were a bad example, simply because you're only at 25% of their production. A better example would be that of IBM. You outproduce them by around 170,000 a day, going by yesterday's production. But you're also 191 million points behind them, meaning that it's going to take over 1,000 days to catch them. If the scoring changed and everybody had their production drop to 1/3rd of current levels but the same scores were kept, it would then take over 3,000 days - seven years.

I do understand goal oriented people perfectly well. Whilst saying you would triple or quadruple your production to catch them is nice fighting talk, it's also meaningless unless it's backed up, which it isn't at the moment.

It's funny that people rail against the idea of backdating credit on the basis of ex post facto, when keeping the current scores is also ex post facto - because the old scores reflect the fact that the old credit system made for far more credit being awarded than should have been.
ID: 27968 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
tralala

Send message
Joined: 8 Apr 06
Posts: 376
Credit: 581,806
RAC: 0
Message 27970 - Posted: 21 Sep 2006, 14:25:28 UTC - in response to Message 27708.  

I think the new credit system is a huge improvement. It's not perfect, but then again, nothing is. It's minimised the effect of optimised BOINC clients, which were overstating credit by 3x - 4x what it should have been. The reason it was overstated so much is that there was no corresponding optimised science application.

So the new system is a step forward. It gives credit based on how much work you do - you do four times as many structures as me and you get four times as much credit as me.

There is one remaining problem however. It's now 3x - 4x more difficult to catch other teams and users. I reckon it's going to take 2.5, perhaps 3 years before the effects of the optimised clients have become irrelevant. That's why I was in favour of a backdating (it's not a banned word is it?) which would have mostly fixed things, or freezing the stats and starting from scratch again, alá D2OL.

Let me be really clear on this - I'm not saying there was cheating. What I'm saying is that the first 9 months worth of credits were basically multiplied by 3x what they really should have been. If there was a backdating then XS would still have approx 50% more credit than Free DC and the DPC and three times that of Anandtech. But people would be catchable again.

It's like a South American country that's had hyper-inflation, but the currency has been revalued so that money is sane again. Except prices have stayed the same, so it takes longer to save up for that 50" widescreen TV.


This is exactly my view as well. However the argument of the problematic nature of an ex-post-change of the credit system has some validity as well and in fact anybody who cares a lot about credits could have used the optimized clients (I did for quite some time). Those who didn't, either didn't know (unlikely for competing teams) or didn't bother (then it should not bother them now). So although it would give a more realistic picture who did how much work with backdating it also would be a controversial act. I assume the decision not to backdate was taken to keep XS on board and avoid anger and discussions - it seems that intention did not really materialize. ;-)
ID: 27970 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
tralala

Send message
Joined: 8 Apr 06
Posts: 376
Credit: 581,806
RAC: 0
Message 27971 - Posted: 21 Sep 2006, 14:27:07 UTC - in response to Message 27950.  


As to XS only getting 1/3 of what we did get, again you are wrong.
Jan 1 to Aug 25, almost 8 full months..Optimised files used from Mid April to end of August. 3.5/8ths of the total working with the stock client..


Come on you don't believe this yourself. As biggles has pointed out there were instructions for the optimized client on your board since december 2005. Free-DC and DPC did use the optimized boinc clients all the time as well.
ID: 27971 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jose

Send message
Joined: 28 Mar 06
Posts: 820
Credit: 48,297
RAC: 0
Message 27976 - Posted: 21 Sep 2006, 14:40:41 UTC - in response to Message 27968.  
Last modified: 21 Sep 2006, 14:55:01 UTC

My point in saying "get on this much earlier" meant why didn't you get on Rosetta much earlier?Had you been on the project in force last January you'd have had the same chance and opportunity that any other team to be #1.
You evidently didn't and now you want a second chance and that second chance to come at the price of what the top teams did being rolled into the back room out of sight.


We have been around on Rosetta for a while. We weren't interested in diverting power from our large GIMPS, SETI or F@H teams to take number one. We still aren't. I'm bothered by things as a matter of principle. Nobody is penalising XS as they make an assault on WCG, but the new credit system penalises any team that makes an assault on Rosetta.


As to making it a personal attack, that was based on what I saw on the stats pages. The person calling for the zeroing of points and starting over is also the head of the currently #2 largest producing team with the most to gain from such an action. What other conclusion could I possibly draw?


That you can't read perhaps?

The currently 2nd largest producing team is Anandtech.


As to XS only getting 1/3 of what we did get, again you are wrong.
Jan 1 to Aug 25, almost 8 full months..Optimised files used from Mid April to end of August. 3.5/8ths of the total working with the stock client..
Care to revise your comments as to what percentage of what we have now would still be there? Now matter what math you use it's way beyond the 1/3 you suggest.


What about the math based on XS having advocated the use of optimised clients since the 31st of December 2005? This thread has talked about optimised clients and encouraged the use of them since the day after it was posted. I don't believe that it took three and a half months for XS to notice that part of the setup guide. Optimised clients and the subsequent overclaiming of credit was widespread before April.


As to if the scoring system changed at WCG and we got 1/3 of the amount we got now, would we still be able to catch EasyNews..The answer is yes, we'd tripple the machines or quadruple them, whatever it took.
You just don't understand goal oriented people.You do what it takes to get to what your striving for..


I guess EasyNews were a bad example, simply because you're only at 25% of their production. A better example would be that of IBM. You outproduce them by around 170,000 a day, going by yesterday's production. But you're also 191 million points behind them, meaning that it's going to take over 1,000 days to catch them. If the scoring changed and everybody had their production drop to 1/3rd of current levels but the same scores were kept, it would then take over 3,000 days - seven years.

I do understand goal oriented people perfectly well. Whilst saying you would triple or quadruple your production to catch them is nice fighting talk, it's also meaningless unless it's backed up, which it isn't at the moment.

It's funny that people rail against the idea of backdating credit on the basis of ex post facto, when keeping the current scores is also ex post facto - because the old scores reflect the fact that the old credit system made for far more credit being awarded than should have been.


Biggles : Let me see if I can make it clear. Be it one day or one thousand days, we are going to overtake every and all teams that get in front of us. That is the key to our team: determination focus even in the face of seemingly odds.

As to our strategy, pardon me ff I don't discuss it with you. We select the time and the moment to mount the big challenges, in the process we start doing some prep work as we are doing right now :)

Biggles: I think you need a crash course on the meaning and the usage of the word ex post facto. As it is used in common law and regular law it has a very specific meaning and it is not even close to the one you are giving.

Glad you mentioned Free Dc and the DPCs when we started crunching Rosetta, we were so far behind them , it was a mad man's thought that we could ever beat them, specially since they were using opti clients and in the case of the DPCs they had more people than we did. And yes we beat them, even though they kept crunching (hey the DPCs even had one of their famous stampedes)
and even after we asked almost 10% of our credits to be zeroed out when it was shown to us that two members of our team cheated and interfered with another team in another project. We did not hesitate to do it and we did it the same day we had beaten Free DC. So please, do not under estimate the crunching power and the determination behind XtremeSystems.

Ah From the description of the projects you belong, I can see why the animus against XtremeSystems. Now, I know : it has nothing to do about credits but a lot about personalities.
This and no other is the root from which a Tyrant springs; when he first appears he is a protector.”
Plato
ID: 27976 · Rating: 9.9920072216264E-15 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile carl.h
Avatar

Send message
Joined: 28 Dec 05
Posts: 555
Credit: 183,449
RAC: 0
Message 27980 - Posted: 21 Sep 2006, 14:42:49 UTC

Even as a mod Lala you cannot prove or disprove who on what team ran opti can you ? Or from when !!

The point is mute.
Not all Czech`s bounce but I`d like to try with Barbar ;-)

Make no mistake This IS the TEDDIES TEAM.
ID: 27980 · Rating: -1 · rate: Rate + / Rate - Report as offensive    Reply Quote
tralala

Send message
Joined: 8 Apr 06
Posts: 376
Credit: 581,806
RAC: 0
Message 27987 - Posted: 21 Sep 2006, 14:50:10 UTC - in response to Message 27980.  

Even as a mod Lala you cannot prove or disprove who on what team ran opti can you ? Or from when !!

The point is mute.


No I can't, but I can state which statements I'm going to believe and which not.
ID: 27987 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Nemesis
Avatar

Send message
Joined: 12 Mar 06
Posts: 149
Credit: 21,395
RAC: 0
Message 27993 - Posted: 21 Sep 2006, 14:54:52 UTC - in response to Message 27970.  
Last modified: 21 Sep 2006, 14:59:52 UTC

I think the new credit system is a huge improvement. It's not perfect, but then again, nothing is. It's minimised the effect of optimised BOINC clients, which were overstating credit by 3x - 4x what it should have been. The reason it was overstated so much is that there was no corresponding optimised science application.

So the new system is a step forward. It gives credit based on how much work you do - you do four times as many structures as me and you get four times as much credit as me.

There is one remaining problem however. It's now 3x - 4x more difficult to catch other teams and users. I reckon it's going to take 2.5, perhaps 3 years before the effects of the optimised clients have become irrelevant. That's why I was in favour of a backdating (it's not a banned word is it?) which would have mostly fixed things, or freezing the stats and starting from scratch again, alá D2OL.

Let me be really clear on this - I'm not saying there was cheating. What I'm saying is that the first 9 months worth of credits were basically multiplied by 3x what they really should have been. If there was a backdating then XS would still have approx 50% more credit than Free DC and the DPC and three times that of Anandtech. But people would be catchable again.

It's like a South American country that's had hyper-inflation, but the currency has been revalued so that money is sane again. Except prices have stayed the same, so it takes longer to save up for that 50" widescreen TV.


This is exactly my view as well. However the argument of the problematic nature of an ex-post-change of the credit system has some validity as well and in fact anybody who cares a lot about credits could have used the optimized clients (I did for quite some time). Those who didn't, either didn't know (unlikely for competing teams) or didn't bother (then it should not bother them now). So although it would give a more realistic picture who did how much work with backdating it also would be a controversial act. I assume the decision not to backdate was taken to keep XS on board and avoid anger and discussions - it seems that intention did not really materialize. ;-)


So why don't you trundle over to your hidden "Project contact thread" and get David Kim to come here and discuss whether it's at all possible to backdate. Without specific information, this is all just noise.
Nemesis n. A righteous infliction of retribution manifested by an appropriate agent.


ID: 27993 · Rating: -2 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · Next

Message boards : Number crunching : Another discussion on the New Credit System



©2022 University of Washington
https://www.bakerlab.org