S5 going backwards
2 Jul 2006 6:55:59 UTC
Topic 191500
(moderation:
On 29 Jun 2006 0:16:33 UTC Server Status was:
Total needed 388.4 days with 377.3 days remaining.
Right now, two days and almost eight hours later, Server Status is:
Total needed 398.3 days with 383.9 days remaining.
All of the above estimated of course. That is almost nine days back [6.6 days plus 2.3 days] in a little more than two days elapsed.
S5 going backwards
)
Maybe our computers can't process work units as fast as the project thinks they can. That's certainly my case. When I get a new long work unit, the estimated completion time is about 12-13 hours. It really takes me more like 17 hours to complete though, something like 5 hours longer than estimated. Add up all of these time differences between estimates and reality, and the estimated total time will have to get longer. Right?
More likely, it is people
)
More likely, it is people leaving the project because of the credit / optimization issue.
RE: Maybe our computers
)
If you ran optimised app for S4 your DCF would have been low, therefore the expected crunching time calculation now you are back to the standard application is wrong, but it should normalise fairly soon as your DCF is adjusted.
RE: More likely, it is
)
Naa, this looks more like wishfull thinking (mainly by certain people ;) )
IMO, the stats are just not completely stabilised, so early in the beginning workrun.
RE: On 29 Jun 2006 0:16:33
)
Well, if your datas are right, the S5 will be processed up in 300 days.
( 388,4 - 377,3 ) / 388,4 -> 2,858%
( 398,3 - 383,9 ) / 398,3 -> 3,615%
3,615% - 2,858% = 0,757%
100 / 0,757 * 2,3 days = 303,83 days ( total )
3,615% * 303,83 days = 10,98 days ( elapsed )
303,83 - 10,98 = 292,85 days ( remaining )
1m = 0.001%
)
1m = 0.001%
RE: 1m = 0.001% Nice
)
Nice diagram. What can we see on it?
RE: If i had done a
)
If i had done a graph like that for science class way back i'd get in trouble!
"percent per day" percent of WHAT? It doesn't say what the data of either x/y axis is. xxx m - what are they, Dates- oh yes but what year? and it doesn't indicate what point the dates are set at on the graph x axis. sun 18 could be anywhere along that first part upto tue 20. You can see that in that space the graph soares so at what point is sun 18?
Sorry. Not an attack at you Stef I just hate it when I see such a poorly done graph. Sure it looks pretty but its wrong from what was drilled into me in almost every science class. Was probly collated by some scientist as well :)
Only thing I cant remember is if its good disipline to connect all the dots of the data points. Since its a day by day thing you cant possible have accurate data for evey second/minute in between so I think it should actually just have a trend line that follows the trend of the data as close as possible and crossed indicating the real data points.
@Pepperammi: Then please make
)
@Pepperammi: Then please make a better graph, i've made this one in 5 minutes.
@Akos: It's the realtime percentage rate $now minus $now-24h which means we do right now about 0.22% per day
RE: @Akos: It's the
)
Okay! Thanks for the info! So, your graph shows the derivated values of the 'percentage function'. Really nice!
0,22% -> This tempo needs 454,5 days totally... Uh...