And here is the same code in unit/module form with tables commented out: The possible table lookups/optimizations are left in in case I or any one else wants to test those ;) // *** Begin of Module Code *** unit UnitStringToUint64_version_003; interface function TryStrToUint64( S : string; var Value : Uint64 ) : boolean; implementation { Skybuck's TryStrToUint64 conversion

Hello, Sometimes I google for my own code and then it sucks if I one can't find it or too it's incomplete. So one more time the complete code as it now is, all 3 versions embedded. Version 0.03 is now active... big 1600 bytes lookup tables still active but those could be commented out. Maybe somebody else could write a nice benchmark program for it ? Tomorrow maybe I write one, but

On 9/5/07, Carl Woll <carlw@wolfram.com> wrote: <snip/> I think the following explains it: In[144]:= N[ FromDigits[#, 2] & /@ RealDigits[{999.5, 99.95, 9.995, .9995}, 2], 20] Out[144]= {999.50000000000000000, 99.950000000000002842, 9.9949999999999992184, 0.99950000000000005507} ^^^ Woll FTW (again!) Thanks! (FTW is slang for, "For The Win")

René Guilbault a écrit : Tapez 0,2 dans la cellule A1, puis en dessous, tapez =A1*11-2. Bien sûr on retrouve 0,2? Tirez ensuite cette cellule vers le bas, on ne devrait trouver que des 0,2 et hélas ça diverge très vite et beaucoup. Pourquoi?? Merci! Dès a3, la quinzième décimale se prend les pieds dans le tapis. C'est la limite de précision qui ouvre la porte aux erreurs d'arrondi