Forum::BOINCstats Bug Report::Is "credit/month" The Penultimate (not Current) Month?

Guest

2006-01-18 12:14:10

Willy,
There are many users with value in Credit/Day and in Credit/Week but ZERO in Credit/Month.
This happens frequently with new users.
Does this mean that the Credit/Month is value for PENULTIMATE month?
It obviously cannot be CURRENT month to date.
Or is there a bug in the calculations?
Keith

bjango: Translator; BAM!ID: 7; Joined: 2006-01-11; Posts: 197; Credits: 598,676; World-rank: 280,557

2006-01-18 17:00:05

Month meaning the previous 30 +/- 2 days

Guest

2006-01-18 18:27:49
last modified: 2006-01-22 13:26:15

So, Credit/month, Credit/Week and Credit/Day are not in synch?
It seems strange if the Credit/Month values runs up to 2 days earlier than the other 2 values.
Does Credit/Week include the values of Credit/Day? It should do so?
Keith

P.S. As no reply is forthcoming, I must assume there is an error in the method of calculation.
Keith

Guest

2006-01-28 17:03:24

How is it possible? Nil average credit!!!
Pete B has 236,605 Total Credit, Zero Credit Day, 85,471 Credit/Week and Credit/Month.
So, how on earth can the Average Credit be Zero?
I am not saying that these stats are bug ridden, but they certainly do give the appearance of being so.

Keith

Son Goku 3SSJ: Forum moderator - Translator; BAM!ID: 8; Joined: 2006-01-11; Posts: 182; Credits: 2,572,985; World-rank: 129,190

2006-01-28 17:22:21
last modified: 2006-01-28 17:20:03

How is it possible? Nil average credit!!!
Pete B has 236,605 Total Credit, Zero Credit Day, 85,471 Credit/Week and Credit/Month.
So, how on earth can the Average Credit be Zero?
I am not saying that these stats are bug ridden, but they certainly do give the appearance of being so.

Keith

I don't know, but that is what CPDN reports also. His average credit (as reported by CPDN) is 0.43!

part of

Wenn jeder an sich denkt, wird an alle gedacht!

[BOINCstats] Willy: Forum moderator - Administrator - Developer - Tester - Translator; BAM!ID: 1; Joined: 2006-01-09; Posts: 9461; Credits: 353,172,950; World-rank: 5,071

2006-01-28 17:37:15
last modified: 2006-01-28 17:30:00

Average credit is read from the XML file, and shown in the stats without being processed. If CPDN says 0, BS also shows 0.

If that's a bug, so be it.

Ow, BTW, his recent average credit (according to BOINCstats) (as seen on the detailed stats page) is 1,424.51998. Does that suit you better?

Please do not PM, IM or email me for support (they will go unread/ignored). Use the forum for support.

Son Goku 3SSJ: Forum moderator - Translator; BAM!ID: 8; Joined: 2006-01-11; Posts: 182; Credits: 2,572,985; World-rank: 129,190

2006-01-29 18:57:41
last modified: 2006-01-29 18:58:15

Berkeley got it wrong then?
I am not an expert on the mechanics of what is shown as XML, but I accept that BOINCstats are just repeating figures fed to them by Berkeley.
I'm afraid my confidence in Berkeley leads me not to be suprised that they either got their calculations wrong or incorrectly described the results.
At least I now know that the figures I am 'trusting' were not what they appear to be. It would be extremely interesting to find out what extraordinary explanation might throw light on these anomalies, but I'll not hold my breath waiting for it from Berkeley!!!
Keith

It's not Berkeley it's CPDN. If it was Berkeley the values would be correct

part of

Wenn jeder an sich denkt, wird an alle gedacht!

Guest

2006-01-29 18:58:35
last modified: 2006-01-29 18:55:52

Berkeley got it wrong then?
I am not an expert on the mechanics of what is shown as XML, but I accept that BOINCstats are just repeating figures fed to them by Berkeley.
I'm afraid my confidence in Berkeley leads me not to be suprised that they either got their calculations wrong or incorrectly described the results.
At least I now know that the figures I am 'trusting' were not what they appear to be. It would be extremely interesting to find out what extraordinary explanation might throw light on these anomalies, but I'll not hold my breath waiting for it from Berkeley!!!
Keith

Keck_Komputers: BAM!ID: 571; Joined: 2006-05-22; Posts: 178; Credits: 23,679,569; World-rank: 31,209

2006-01-29 20:17:54

It's not Berkeley it's CPDN. If it was Berkeley the values would be correct

Correct. CPDN does not grant credit like most BOINC projects. Instead of a validator they have a script that runs every 12 hours and recalculates the credit based on trickles reported. This makes the RAC almost meaningless there. If FaH is released in it's current form it will be even worse since they don't even bother trying to create a RAC figure.

BOINC WIKI

BOINCing since 2002/12/8

Guest

2006-01-30 13:21:52

So, CPDN are the culprits. Why use them?
Why, oh why does BOINCstats repeat the inconsistent results knowing they "do not add up"?
Is there not a better, more reliable source?
It would seem that the "Active Users" figure is nearer the correct figure than looking at the descending list of Credit/Month figures that show a positive value. Maybe the former figure is obtained from a different source. If that is so, couldn't the Credit/Month figures be obtained from the same source?
Keith

[BOINCstats] Willy: Forum moderator - Administrator - Developer - Tester - Translator; BAM!ID: 1; Joined: 2006-01-09; Posts: 9461; Credits: 353,172,950; World-rank: 5,071

2006-01-30 14:27:43

Why, oh why does BOINCstats repeat the inconsistent results knowing they "do not add up"?

That's why on the detailed page you'll find 'recent average credit according to BOINCstats'.

Please do not PM, IM or email me for support (they will go unread/ignored). Use the forum for support.

Status

Shoutbox