Poul-Henning, On 2021-03-30 10:15, Poul-Henning Kamp wrote: > -------- > Magnus Danielson writes: > >> A number of different approaches to new SI definition of Hz (was second) >> was put forward, by among others me, and that range from a single >> species to actually use a ensemble of species. The later would have a >> benefit in using the advancement of several species and allow a larger >> range of clock realizations. > In one of his lectures Feynmann mentions that the energy levels of > hydrogen have been calculated to 15 or 16 digits "from first > principles". As I recall it, for very long there was a huge mismatch between measured and calculated frequency. It seems only recently that things improved in that sense, 12 digits if my memory serves me right. It would be fun investigating. > As far as I can figure out, that must have happened in the 1970'ies, > which means it was probably done on one of Seymour Cray's designs. Mmm, yes, with that approximate multiplication, lovely. a*b =/= b*a in general. But it was fast and it was documented. > > Hydrogen is obviously "the easy one", but even if the calculation > scales to the power of four on the number of particles in the atom, > we should still be well inside current super-computing capacities. > > Do you know if anybody has considered that as way to nail the frequencies ?
To the best of what I've seen, experimentalists have been better crunching out numbers thant the teorists. Then again, I would not say I have researched that field as exhaustively as it should have been done. It becomes increasingly difficult as the core and shells becomes more complex. Only recently it was shown definitively that some cores have clearly non-spheric shape, and had a pear-like disformation. Now any such deviation will shift the electron shell. A whole range of theory for corrections have been added for this and that. Doing credible models would be a nice PhD topic. I've tried to read up on it several times. First principles does not help all that much. Cheers, Magnus _______________________________________________ time-nuts mailing list -- time-nuts@lists.febo.com -- To unsubscribe send an email to time-nuts-le...@lists.febo.com To unsubscribe, go to and follow the instructions there.
