... the only way to resolve the problem is to reset the debts. I have had streaks where I have had to reset debts about every 3-4 days to keep enough GPU work on one system. It seems a little better with .28 and later, but I suspect that this is not to last ... for the simple reason that they have not changed code in work fetch or RR Sim ... which means that the problem is very likely not yet addressed...
Paul, you joined here about at the same time as I did. Do you remember the days 4 years ago, when there were various discussions regarding "do not use this version of boinc, it has this and that bug"?
I am wondering that after 4 years of development boinc is still nurseware :-(
Why don't the mark versions "stable" and "untested"? So users can decide themself if they want to do the nurse job? Or do they really need hundreds of thousands of testers?
I would like to download a version which I install and forget. I do not want to spend time to check it every day. And I am sure, that all the project do not prefer machines which run out of work.
Why don't the mark versions "stable" and "untested"?
Interesting point. A while ago we had specialist/specific software for our business ( medical practice ) which in it's 'second' version had manifold updates - 2.01 through to 2.93 before it finally fell over and died. Some of those had 'a', 'b' etc variants ... like 2.78c say. This was blatant laziness by the software house, in effect we were paying for the privilege of doing their testing. Fortunately the company was bought out, that 'version' was scrapped and a total re-write produced a very good stable and useful product ( of the same name ). We've had maybe 10 updates in the three years since, nearly all related to new functionality rather than firefighting old problems.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
... This was blatant laziness by the software house, in effect we were paying for the privilege of doing their testing. ...
In Germany we use the abbreviation EDV which usually means "Elektronische Datenverarbeitung", but some say, it means "Experimentelle Datenverarbeitung".
According to wikipedia, the corresponding acronym is EDP "Electronic Data Processing" which may also mean "Experimental Data Processing".
for me, it is very usual. boinc needs some time after a break or when it is newly installed, to calculate the really needed units to crunch. i´ve got several computers running and every time it switch between the projects, the workunit up- and download sucks. so i wait 2-3 or 4 weeks and then it still doing the job the way i want. plz notice, its just a computer - not a human mind, your understanding of getting the credits and new workunits is not the same, the boinc doing this. boinc experienced knowledge needs a couple of weeks to get this clear.
cheers mate
edith say´s its called "ende der vernunft" which means EDV
In Germany we use the abbreviation EDV which usually means "Elektronische Datenverarbeitung", but some say, it means "Experimentelle Datenverarbeitung".
I also like the term "Banana software" (obvious: gets ripe after being sold to the consumer). Not sure this is used in the anglo-american cultures?
... the only way to resolve the problem is to reset the debts. I have had streaks where I have had to reset debts about every 3-4 days to keep enough GPU work on one system. It seems a little better with .28 and later, but I suspect that this is not to last ... for the simple reason that they have not changed code in work fetch or RR Sim ... which means that the problem is very likely not yet addressed...
Paul, you joined here about at the same time as I did. Do you remember the days 4 years ago, when there were various discussions regarding "do not use this version of boinc, it has this and that bug"?
I am wondering that after 4 years of development boinc is still nurseware :-(
Why don't the mark versions "stable" and "untested"? So users can decide themself if they want to do the nurse job? Or do they really need hundreds of thousands of testers?
I would like to download a version which I install and forget. I do not want to spend time to check it every day. And I am sure, that all the project do not prefer machines which run out of work.
Ever since BOINC Beta it has been this way. Yes it is a sad state of affairs that we have not progressed beyond the "Don't use this version because ..." stage. But this is a natural consequence when the development types will not progress beyond the hacker mentality which resists all concepts of process...
We still have bugs in the bug base that are single or double digit and yet rated critical ... we have little or no participant input into which bugs or enhancements should be considered for various versions.
BUt the lack of discipline does give rise to this type of effect... and the randomness on the decisions of when a version is "Recommended" is even more startling...
RE: ... the only way to
)
Paul, you joined here about at the same time as I did. Do you remember the days 4 years ago, when there were various discussions regarding "do not use this version of boinc, it has this and that bug"?
I am wondering that after 4 years of development boinc is still nurseware :-(
Why don't the mark versions "stable" and "untested"? So users can decide themself if they want to do the nurse job? Or do they really need hundreds of thousands of testers?
I would like to download a version which I install and forget. I do not want to spend time to check it every day. And I am sure, that all the project do not prefer machines which run out of work.
RE: Why don't the mark
)
Interesting point. A while ago we had specialist/specific software for our business ( medical practice ) which in it's 'second' version had manifold updates - 2.01 through to 2.93 before it finally fell over and died. Some of those had 'a', 'b' etc variants ... like 2.78c say. This was blatant laziness by the software house, in effect we were paying for the privilege of doing their testing. Fortunately the company was bought out, that 'version' was scrapped and a total re-write produced a very good stable and useful product ( of the same name ). We've had maybe 10 updates in the three years since, nearly all related to new functionality rather than firefighting old problems.
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
RE: ... This was blatant
)
In Germany we use the abbreviation EDV which usually means "Elektronische Datenverarbeitung", but some say, it means "Experimentelle Datenverarbeitung".
According to wikipedia, the corresponding acronym is EDP "Electronic Data Processing" which may also mean "Experimental Data Processing".
@wurgl for me, it is very
)
@wurgl
for me, it is very usual. boinc needs some time after a break or when it is newly installed, to calculate the really needed units to crunch. i´ve got several computers running and every time it switch between the projects, the workunit up- and download sucks. so i wait 2-3 or 4 weeks and then it still doing the job the way i want. plz notice, its just a computer - not a human mind, your understanding of getting the credits and new workunits is not the same, the boinc doing this. boinc experienced knowledge needs a couple of weeks to get this clear.
cheers mate
edith say´s its called "ende der vernunft" which means EDV
RE: In Germany we use the
)
I also like the term "Banana software" (obvious: gets ripe after being sold to the consumer). Not sure this is used in the anglo-american cultures?
CU
Bikeman
RE: RE: ... the only way
)
Ever since BOINC Beta it has been this way. Yes it is a sad state of affairs that we have not progressed beyond the "Don't use this version because ..." stage. But this is a natural consequence when the development types will not progress beyond the hacker mentality which resists all concepts of process...
We still have bugs in the bug base that are single or double digit and yet rated critical ... we have little or no participant input into which bugs or enhancements should be considered for various versions.
BUt the lack of discipline does give rise to this type of effect... and the randomness on the decisions of when a version is "Recommended" is even more startling...