Arjen de Korte
2008-Nov-26 12:33 UTC
[Nut-upsdev] [nut-commits] svn commit r1582 - in trunk: . data docs drivers man
Reporting current in milliAmps is not a good idea. This will lead to impractically large values, especially when we report them with two decimals resolution (10 microAmps). Even the laboratory measuring equipment that is standing right next to me now, doesn't offer this resolution in the range that is common for most UPS'es. I think you meant to write 'Ampere (A)' here and in that case, two decimals (10 milliAmps) resolution is not too unreasonable. Reporting outlet voltage with 10mV resolution is ridiculous. This would require an accuracy of better than 0.004% to have any meaning. This is at least two orders of magnitude better than the accuracy of any UPS that is on the market today. Even with *very* expensive measurement equipment, I doubt that reporting more than one decimal is justified. Basically the same goes for the power (W and VA), these shouldn't have any decimals at all. Best regards, Arjen -- Please keep list traffic on the list
Arnaud Quette
2008-Nov-28 11:54 UTC
[Nut-upsdev] [nut-commits] svn commit r1582 - in trunk: . data docs drivers man
2008/11/26 Arjen de Korte <nut+devel at de-korte.org>:> Reporting current in milliAmps is not a good idea. This will lead to > impractically large values, especially when we report them with two decimals > resolution (10 microAmps). Even the laboratory measuring equipment that is > standing right next to me now, doesn't offer this resolution in the range > that is common for most UPS'es. I think you meant to write 'Ampere (A)' here > and in that case, two decimals (10 milliAmps) resolution is not too > unreasonable. > > Reporting outlet voltage with 10mV resolution is ridiculous. This would > require an accuracy of better than 0.004% to have any meaning. This is at > least two orders of magnitude better than the accuracy of any UPS that is on > the market today. Even with *very* expensive measurement equipment, I doubt > that reporting more than one decimal is justified. Basically the same goes > for the power (W and VA), these shouldn't have any decimals at all.right, can you take over fixing these please? Arnaud