Science projects vary over time and by user. Most people are getting malariacontrol (Africa@home), as it's our personal favourite. Others include Rosetta@home, Einstein@home and climateprediction.net.
Obviously, these are not the main function of Charity Engine, which is running commercial tasks and using the proceeds for charity donations and the prize draws. But, if any hardware is idle, it runs one of these projects.
We also continually donate some computing power to them even when the grid is NOT idle, as we do like the science side of things.
Your top credit contributor has won your "lottery" twice. Very interesting. I am trying to figure out the odds of this happening and to do so I have to understand the sytem. I cant quite figure out how your lottery works. The last ticket draw had a possible number range from 1 to about 27,000,000. Your site says you take the last digit from 5 stock exchanges.
That leads to question 1: How are 8 digits pulled from only 5 stock exchanges? (Only makes sense if the number draw takes at least two days)
Your site goes on to say "The prize winning number is chosen right-to-left, or smallest first". The "or smallest first." part is ambigous. Smallest what first? This could be inturperated several ways.
I understand I have taken this matter father than normal, but at this point I have turned it into a lesson of probabilities and statistics. Also, as a side note. It seems that if you had a BOINC top 500 cruncher join CE, the lottery would break down to a single highly probable winner with quicker goals met. Example: A 100 million credit per month user would have over 90% chance of winning each lottery and goals would be met every week or two at that pace.
Yes, the draw takes two days. In fact, it sometimes takes three or more, as the final (biggest denominator) digit can't be anything from 0-9 unless the total number of entries happens to begin with a nine. Which it never has, and is not likely to (see Benford's Law).
It's smallest digit first. Right-to-left. If we did it the other way, the last digit would hardly ever matter, as the winner's range would include all of them anyway. By drawing the winning number backwards, the last (biggest) digit always matters.
The double winner won it fair and square. The List of Entries is published in advance of the draw date, and nothing and nobody can predict the last digit of stock market results. He is a heavy contributor though (at least 50 PCs - I'll take your word for it that he's actually the highest, I haven't checked), so it's not really that surprising. We obviously don't kick people out of all future draws just because they've won one either, that would be completely unfair.
Interesting point with the hypothetical BOINC "superuser". I guess those guys must be running server farms or offices full of PCs or something like that. Something we'll look out for!
Okay, I now understand what you mean when you say smallest digit first. You are simply building the number from right to left with the pulled market digits.
I have to say, this is a smart project and it has high potential for the creators, etc... I am a top 0.02% BOINC user and I had never heard of your project till I found it yesterday deep in a forum that nobody reads. I need to find a way to balance the workloads in BOINC between your project and the rest of my original projects and then I will be satisfied.
Also, I know of rigs that can hit 10 million credits per month by themselves. If the aforementioned top user on your project had only 4.5 million credits (in about 2 months) and has over 50 PC's. That tells me that either your credit system is different from typical projects and or he has a bunch of slow PC's. I say this because even on my original single (fairly high-end) PC I can hit 2 million credits per month. But again, that is credit from a scope of scientific projects and not based on your credit system which may award less credits per work unit in comparison.
Lastly, I understand that the double winner was fair. Its the odds that make it interesting. Although he had about 15% of the tickets, he had a lower chance than that of winning due to the way the numbers are drawn.
Interesting discussion going here. Mark, do you have something planned to allow users to choose their favorite project? (malariacontrol, Rosetta@home, Einstein@home and climateprediction.net.)
Also, what John D said is not completely right. CE was cpu only until recently, BUT others who joined with the original BOINC client were able to use their video cards to crunch at Einstein@home.
I'd be curious to see the total number of points in the next few days/weeks. I didn't participate much, but might get back in more seriously soon.
Kevin - thanks, glad you like it...! Reason we haven't pushed CE to the BOINC community is two-fold:
1) We love BOINC and don't want to cannibalise it in any way (quite the reverse). All those PCs are already doing good, in our opinion.
2) Our model is specifically designed to appeal to everyone else, the vast majority who (sadly) aren't motivated enough by cool science to run BOINC in the first place.
In that respect, it works. Approx 85% of CE users are brand new to BOINC and some (indeed, lots) are so non-techie, they don't even know what a GPU or CPU is. No regular project would ever get these users volunteer computing.
John - yes, you're right of course. I forgot that point when replying to Kevin. Thanks!
Jo - that's not planned, for the reasons above. Our view is; if your motivation is science, then we encourage you to download BOINC (eg. from GridRepublic), pick your projects and get crunching. We do donate any computing that's not earning money for the charities (and more, in fact), but raising money is the primary purpose of Charity Engine. That said, if enough people want such an option then we'll do it.
Of course, if we throw some computing power at Charity Engine we expect it to raise money for charities, but since you also donate a portion of this compute power to various projects, I think it would not harm to let users pick their favorite project from a list when CE has no job for their computer.
This way instead of having your whole grid running CE apps in priority and malaria, users would run CE in priority and a project of their choice from a list you approve. Personally, I prefer rosetta@home over malariacontrol. So in my personal situation, I would run CE most of the time and rosetta when CE has no more work for my computer. I'm not sure if it is clear, that's why I tried to explain it a little more in detail. What I suggested is not to run a science project 100% of the time, I know BOINC allows me to do that. What I'd like to see here is just the ability to choose where the left over of my computer time would go when it's not used by income generating tasks for the charity.
Thanks, Jo. For the moment, having all users running a common set of projects is useful for benchmarking -- both individual hosts and the system as a whole. As we scale a bit more, we might be able to offer a wider range of choices.
Great discussions above...However I have to agree with Mark McA about the importantance of malaria.
Having dealt with Malaria for the last 35 years (Contracted both in West Africa and elsewhere as a teenager...Mom was MD with WHO/Loma Linda Univ.) I am a Very Enthusiastic part of BOINC.
Interesting lil bit from Malaria.com:
"A recrudescence is a renewed manifestation of symptoms due to an increase in the surviving population of erythrocytic forms; in other words, infection in the blood persists. In P. malariae a small number of parasites in the blood stream can cause a recrudescence after up to 50 years."
Mom tried Her best to get me to do the Dr thing but just tooooo much school!
I too stumbled upon CE but now am shouting the word far and wide and hope to help to continue to add to the ranks of CE workers!!
What scientific projects encompassed in CE? At this point, the only scientific project I see is malariacontrol.net.
Thanks,
-Kevin W. Willis
Hi Kevin,
Science projects vary over time and by user. Most people are getting malariacontrol (Africa@home), as it's our personal favourite. Others include Rosetta@home, Einstein@home and climateprediction.net.
Obviously, these are not the main function of Charity Engine, which is running commercial tasks and using the proceeds for charity donations and the prize draws. But, if any hardware is idle, it runs one of these projects.
We also continually donate some computing power to them even when the grid is NOT idle, as we do like the science side of things.
Cheers,
Mark
Your top credit contributor has won your "lottery" twice. Very interesting. I am trying to figure out the odds of this happening and to do so I have to understand the sytem. I cant quite figure out how your lottery works. The last ticket draw had a possible number range from 1 to about 27,000,000. Your site says you take the last digit from 5 stock exchanges.
That leads to question 1: How are 8 digits pulled from only 5 stock exchanges? (Only makes sense if the number draw takes at least two days)
Your site goes on to say "The prize winning number is chosen right-to-left, or smallest first". The "or smallest first." part is ambigous. Smallest what first? This could be inturperated several ways.
I understand I have taken this matter father than normal, but at this point I have turned it into a lesson of probabilities and statistics. Also, as a side note. It seems that if you had a BOINC top 500 cruncher join CE, the lottery would break down to a single highly probable winner with quicker goals met. Example: A 100 million credit per month user would have over 90% chance of winning each lottery and goals would be met every week or two at that pace.
Yes, the draw takes two days. In fact, it sometimes takes three or more, as the final (biggest denominator) digit can't be anything from 0-9 unless the total number of entries happens to begin with a nine. Which it never has, and is not likely to (see Benford's Law).
It's smallest digit first. Right-to-left. If we did it the other way, the last digit would hardly ever matter, as the winner's range would include all of them anyway. By drawing the winning number backwards, the last (biggest) digit always matters.
The double winner won it fair and square. The List of Entries is published in advance of the draw date, and nothing and nobody can predict the last digit of stock market results. He is a heavy contributor though (at least 50 PCs - I'll take your word for it that he's actually the highest, I haven't checked), so it's not really that surprising. We obviously don't kick people out of all future draws just because they've won one either, that would be completely unfair.
Interesting point with the hypothetical BOINC "superuser". I guess those guys must be running server farms or offices full of PCs or something like that. Something we'll look out for!
Cheers,
Mark
Okay, I now understand what you mean when you say smallest digit first. You are simply building the number from right to left with the pulled market digits.
I have to say, this is a smart project and it has high potential for the creators, etc... I am a top 0.02% BOINC user and I had never heard of your project till I found it yesterday deep in a forum that nobody reads. I need to find a way to balance the workloads in BOINC between your project and the rest of my original projects and then I will be satisfied.
Also, I know of rigs that can hit 10 million credits per month by themselves. If the aforementioned top user on your project had only 4.5 million credits (in about 2 months) and has over 50 PC's. That tells me that either your credit system is different from typical projects and or he has a bunch of slow PC's. I say this because even on my original single (fairly high-end) PC I can hit 2 million credits per month. But again, that is credit from a scope of scientific projects and not based on your credit system which may award less credits per work unit in comparison.
Lastly, I understand that the double winner was fair. Its the odds that make it interesting. Although he had about 15% of the tickets, he had a lower chance than that of winning due to the way the numbers are drawn.
Anyways, thanks for the replies.
CE was CPU only until recently. The next draw will have a much different distribution and total amount of points.
Interesting discussion going here. Mark, do you have something planned to allow users to choose their favorite project? (malariacontrol, Rosetta@home, Einstein@home and climateprediction.net.)
Also, what John D said is not completely right. CE was cpu only until recently, BUT others who joined with the original BOINC client were able to use their video cards to crunch at Einstein@home.
I'd be curious to see the total number of points in the next few days/weeks. I didn't participate much, but might get back in more seriously soon.
Jo1252
Kevin - thanks, glad you like it...! Reason we haven't pushed CE to the BOINC community is two-fold:
1) We love BOINC and don't want to cannibalise it in any way (quite the reverse). All those PCs are already doing good, in our opinion.
2) Our model is specifically designed to appeal to everyone else, the vast majority who (sadly) aren't motivated enough by cool science to run BOINC in the first place.
In that respect, it works. Approx 85% of CE users are brand new to BOINC and some (indeed, lots) are so non-techie, they don't even know what a GPU or CPU is. No regular project would ever get these users volunteer computing.
John - yes, you're right of course. I forgot that point when replying to Kevin. Thanks!
Jo - that's not planned, for the reasons above. Our view is; if your motivation is science, then we encourage you to download BOINC (eg. from GridRepublic), pick your projects and get crunching. We do donate any computing that's not earning money for the charities (and more, in fact), but raising money is the primary purpose of Charity Engine. That said, if enough people want such an option then we'll do it.
Hope that helps!
Cheers,
Mark
Of course, if we throw some computing power at Charity Engine we expect it to raise money for charities, but since you also donate a portion of this compute power to various projects, I think it would not harm to let users pick their favorite project from a list when CE has no job for their computer.
This way instead of having your whole grid running CE apps in priority and malaria, users would run CE in priority and a project of their choice from a list you approve. Personally, I prefer rosetta@home over malariacontrol. So in my personal situation, I would run CE most of the time and rosetta when CE has no more work for my computer. I'm not sure if it is clear, that's why I tried to explain it a little more in detail. What I suggested is not to run a science project 100% of the time, I know BOINC allows me to do that. What I'd like to see here is just the ability to choose where the left over of my computer time would go when it's not used by income generating tasks for the charity.
Jo1252
Thanks, Jo. For the moment, having all users running a common set of projects is useful for benchmarking -- both individual hosts and the system as a whole. As we scale a bit more, we might be able to offer a wider range of choices.
Great discussions above...However I have to agree with Mark McA about the importantance of malaria.
Having dealt with Malaria for the last 35 years (Contracted both in West Africa and elsewhere as a teenager...Mom was MD with WHO/Loma Linda Univ.) I am a Very Enthusiastic part of BOINC.
Interesting lil bit from Malaria.com:
"A recrudescence is a renewed manifestation of symptoms due to an increase in the surviving population of erythrocytic forms; in other words, infection in the blood persists. In P. malariae a small number of parasites in the blood stream can cause a recrudescence after up to 50 years."
Mom tried Her best to get me to do the Dr thing but just tooooo much school!
I too stumbled upon CE but now am shouting the word far and wide and hope to help to continue to add to the ranks of CE workers!!
Ed