I noticed an odd behaviour in my HP11C in the HMS functions. Looking through the archives I see that this behaviour was found in the 28S and 42S. When I do 2.12 g->H 1.37 g->H (subtract) f->H.MS I get an answer of 0.3460 instead of 0.3500. As noted in the other posts the answer is "correct" but not what I was expecting. I just wanted to add this bit of information to the archives here. (I have just checked with an HP32SII, it does this also.) Cheers.
Hi;
hit the [f][PREFIX] key and you'll see:
[3459999999]As exprected, internally we have the rounding error. The problem is that 3.459999999 is a number that you can key in anytime. Tha HP11C does not "know" it's an H.MMSSssss representation. I agree with the fact the rouning error should not occur, but the rounding-up from 3.459999999 to 3.4600 is acceptable for an ordinary number.
Is that what you read about?
Cheers.
I think its just a rounding error. I think its just trying to display 0.3459999999999... but its getting rounded up to 0.3460000. In fact, you can see this is a rounding error by changing the display of the result to scientific notation by hitting f->SCI 9. You get 3.459999-01.
Spicey
Hey you beat me to it Vieira!
Spicey
Hey, Spice!
Good to know we are all in the same team... I'd rather never having you, guys, in an opposite side! And when I write you , I mean you all...
Good to know that we use our knowledge this way. I'm honored.
The best regards, Spice!
Luiz C. Vieira
Try RND ->H ->H.MS for avoiding the error.
Raul
...and using HMS- (in the 42S/48GX) you'll get the desired 0.35
Raul
Well, if I don't have my HP11C here at work, but on my 32SII if I do FIX 9 I get 0.346000000, not 0.34599999... And, if I do SCI 9 I get 3.46000000E-1 etc. I guess I am not trying to argue that it is a fully qualified bug, that is why I put "bug" in quotes in the Subject line. I was just adding to the thread in the HP Forum Archive 6 (go to that page and do a browser search for "HMS").