Posts by abc@home staff

11) Message boards : Problems : DOES NOT COMPUTE? (Message 6627)
Posted 2043 days ago by abc@home staff
Whatever it is, it is weird because the exact same workunit files
should be sent to clients since there is only 1 workunit and 2 results.
Maybe something went wrong with sending or on your client side
by writing other data to it somehow.
Next time please post at the appriopate place.
12) Message boards : Problems : results (Message 6602)
Posted 2044 days ago by abc@home staff
You're not being neglected Nikolaus. Verhoek has said that he is having trouble with the stats daemon.
Everyone is "affected" by this problem, just have a little patience and I'm sure it will be corrected as soon as possible.

o.k. Thanks when I know that somebody deals with this issue than i am fine.
Nikolaus


It is being fixed as I write this message, in a few hours everything should
be complete again. Sorry for the inconvenience.


Thanks , and you are entirely wellcome.
Nikolaus



Are you sure the stats daemon is fixed? The overall counter starts always with the same number, bevore the glitch i had some 38 credits per dritallen found in average since the glitch this went down to over 100 credits per dritallen found, how many dritallen found got lost during the glitch? Do we have to rekalculate these pakets to not oversee some dritallen?

sorry for bothering you but this should be of intrest for you as well.
regards Nikolaus


It's still running, I don't get your post. From 38 down to 100? Isn't that
going up? Credits/drietallen is not something generated by us btw.


It does not matter how you calculate but in average you have so many x dritallen calculated per dritallen found as a way to see if the relation between dritallen calculated and dritallen found you can take as well the crtedit got per dritallen found. As bigger the number of dritallen you count ( days or CPU hours of calculating as exacter the value. So if the distribuition of dritallen is over the whole field of numbers homogenious then the credit per dritallen found should not change. You can make this calculation as well with Cpu hours needed per dritallen found.

I am not fit eneough to claim that the ditribuition of dritallen in the field of numbers is homogenious but if it is my logic should work. So if i need more credits per dritallen found ( reverse logic) then my efficiency in finding a dritallen is going down that means i need to calculate much more packages to find one dritallen therefore i get more credit for every dritallen found.( 38 credits to over 100 credits per dritallen found)
Please correct if I am wrong.
best regards Nikolaus


Based on the assumption that it is homogeneous, you're right, but drietallen found are not homogeneous distributed over space.
Also I changed something about the number
drietallen scanned which in fact is a useless number. Also, the
stats daemon is still running and updating so it's too early to say anything
untill it is finished.
13) Message boards : Cafe ABC : LOGO Contest Here (Message 6598)
Posted 2044 days ago by abc@home staff
Hi, thanks for editing, well actually it's worse now :)

Can you remove entirely the max(a,b,c) thing,
and remove the backslash before the epsilon?
14) Message boards : Problems : results (Message 6596)
Posted 2044 days ago by abc@home staff
You're not being neglected Nikolaus. Verhoek has said that he is having trouble with the stats daemon.
Everyone is "affected" by this problem, just have a little patience and I'm sure it will be corrected as soon as possible.

o.k. Thanks when I know that somebody deals with this issue than i am fine.
Nikolaus


It is being fixed as I write this message, in a few hours everything should
be complete again. Sorry for the inconvenience.


Thanks , and you are entirely wellcome.
Nikolaus



Are you sure the stats daemon is fixed? The overall counter starts always with the same number, bevore the glitch i had some 38 credits per dritallen found in average since the glitch this went down to over 100 credits per dritallen found, how many dritallen found got lost during the glitch? Do we have to rekalculate these pakets to not oversee some dritallen?

sorry for bothering you but this should be of intrest for you as well.
regards Nikolaus


It's still running, I don't get your post. From 38 down to 100? Isn't that
going up? Credits/drietallen is not something generated by us btw.
15) Message boards : Team Recruitment Center : BOINC@AUSTRALIA (Message 6589)
Posted 2044 days ago by abc@home staff
I appreciate the support and dedication of the aussies ;)
And their AA's are fun and stimulating for projects,
here it caused a friendly third world war between Australia and France,
benefiting abc greatly.
16) Message boards : Problems : results (Message 6588)
Posted 2044 days ago by abc@home staff
You're not being neglected Nikolaus. Verhoek has said that he is having trouble with the stats daemon.
Everyone is "affected" by this problem, just have a little patience and I'm sure it will be corrected as soon as possible.

o.k. Thanks when I know that somebody deals with this issue than i am fine.
Nikolaus


It is being fixed as I write this message, in a few hours everything should
be complete again. Sorry for the inconvenience.
17) Message boards : Cafe ABC : LOGO Contest Here (Message 6587)
Posted 2044 days ago by abc@home staff
I don't think we'll hear anything from Cruncher anymore,
so I think it's over. Thanks everybody for their amazing work,
I will try to make a page with all the logo's.

The one coming on the frontpage will be one of Cori's,
the first one probably, so congratulations to Cori!

However, the mathematics in the right
corner could you replace

max (a,b,c) < C \\eps
by
c < K(\\eps)

because c is always the max, the numbers are all positive,
and C times epsilon should be C or K depending on epsilon.
18) Message boards : Number crunching : Problems sending wu! (Message 6563)
Posted 2045 days ago by abc@home staff
looks like dns problems which I can't do anything about ;(
19) Message boards : ABC@home Science : Results Listing (Message 6545)
Posted 2046 days ago by abc@home staff
Maybe you have an old list? The new one is in binary form
c b c b c b etconly, it's the most minimalistic possible afaik.

Oh yes, I definitely have an old list then (dated October 2 I think). In the form you suggest, adding the radical would increase the file size by 50% (and the usability by 200% ;).


Maybe, I didn't really think about how long it will take to factor all these numbers. How long did it take for you?
20) Message boards : Problems : results (Message 6541)
Posted 2046 days ago by abc@home staff
Probably on the Rekenmee website?

I note my own figures showed up as zero earlier today as well.


That's because I'm having trouble with the stats daemon (again).
I wasn't sure if he meant that or the client output or something else.


Next 10 posts

Return to ABC@home main page


Copyright © 2013 University of Leiden