▼
Posts: 38
Threads: 5
Joined: Jan 1970
At my desk trying to enter a negative small number, on my 17BII in rpn mode. I can't! something like 12,12E12, the +/ (CHS) key don't let me enter a negative exponential. Strange  help?
Soon the 50g must come... Then I'll do it in "c"! Horrah!
/CFH
▼
Posts: 2,448
Threads: 90
Joined: Jul 2005
Hi,
The 17bii and the 27s are actually RPL machines internally. The CHS (actually +/) doesn't do exactly what you are used to. Rather, the minus sign is used for making negative exponents.
So you would go:
12 [+/] [E] [] 12 [ENTER]
which will display as
1.2 E11
Edited: 12 Feb 2007, 9:09 a.m.
▼
Posts: 38
Threads: 5
Joined: Jan 1970
Ah!
12,12 [+/] E [] 12 now displays 0,00000000001 "correctly"! LOL those economists (I have FIX(4)!).
Better use my 15c instead, waiting for the 50g.
/CFH
▼
Posts: 2,448
Threads: 90
Joined: Jul 2005
If your display says
"0,00000000001"
then you are not in fix4. Remember, setting fix is different in the 17bii. you go to DISPLAY, select FIX, then type in the number of digits, then pres INPUT.
I imagine you probably forgot to press INPUT, as it is not required on the voyagers.
Do you have the manual for the 17bii? All of this is in there.
The only other rather important difference as far as RPN operation is concerned is that when you switch from ALG to RPN or after a full clear (like batteries out), the stack is not 4 levels until you have actually filled it up. After that, it oeprates like any 4level stack.
▼
Posts: 38
Threads: 5
Joined: Jan 1970
Indeed, I am in fix 4! But the display sure says 0,00000000001. Pi is 3,1416, as it should be. So my 17BII is not for the scientist.
I do not have the manuals. Se linked pic for the phenomena compared to my 15c.
ftp://ftp.volvo.se/fromvolvo/cfh/hp/17BII_FIX.png
/CF
▼
Posts: 614
Threads: 66
Joined: Jul 2006
Playing around with the 17BII, RPN mode, Fix 4, some interesting results:
0.000002 INPUT gives 0.000002
0.000003 INPUT gives 0.000003
0.000004 INPUT gives 0.000004
0.000005 INPUT gives 0.00001
0.000006 INPUT gives 0.00001
0.000007 INPUT gives 0.00001
......
0.000009 INPUT Gives 0.00001
For some reason it doesn't round off to the display digits.
Bill
▼
Posts: 2,448
Threads: 90
Joined: Jul 2005
And this happens in ALG as well as RPN mode.
Of course it is worth mentioning that if you press [shift] [show] it will give the full precision correctly.
Apparently it takes "small numbers" and shows them as one significant digit, but with leading zeroes, to a maximum of 11 sigfigs east of the radix. If it is farther east than that, it goes to scinote at the sigfig that correlates to the diplacy setting ( so sigfig = displ+1 as is normal for all the other hps).
Interestingly it does not happen in the 27S.
I can't find my manual for the 17bii (I think it is in storage) but I have my 27s manual.
I wonder if this behavior of the 17bii is in the manual, or not?
Edited: 12 Feb 2007, 3:47 p.m.
▼
Posts: 614
Threads: 66
Joined: Jul 2006
Hi Bill,
Just got home and did the same test on a HP19BII. Got same wrong results as the 17BII gave  which isn't too surprising since they are very similar calculators.
Bill
▼
Posts: 2,448
Threads: 90
Joined: Jul 2005
do have either manual handy? (19 or 17) Do they discuss the handling of small numbers?
Posts: 3
Threads: 0
Joined: Jan 1970
Quote:
I can't find my manual for the 17bii
You can't find it here:
http://h10032.www1.hp.com/ctg/Manual/c00442250.pdf
Quote:
I wonder if this behavior of the 17bii is in the manual, or not?
The manual mentions nothing about this in pages 3132, only obvious information.
Gerson.
▼
Posts: 3
Threads: 0
Joined: Jan 1970
Quote:
You can't find it here:
I meant
You CAN find it...
Gerson
_{}
Sorry, I cannot edit my own post, neither I can reply to some posts using my registered account. When clicking on Respond I get back to the Forum. To reply these I had to copy the link and paste it to the address bar in a new browser window....
Posts: 1,792
Threads: 62
Joined: Jan 2005
Hi, "Bill of Smithville", and welcome back! :)
Quote:
Playing around with the 17BII, RPN mode, Fix 4, some interesting results:
0.000002 INPUT gives 0.000002
0.000003 INPUT gives 0.000003
0.000004 INPUT gives 0.000004
0.000005 INPUT gives 0.00001
0.000006 INPUT gives 0.00001
0.000007 INPUT gives 0.00001
......
0.000009 INPUT Gives 0.00001
For some reason it doesn't round off to the display digits.
Bill
Here's my answer, in a thread from a few months ago:
http://www.hpmuseum.org/cgisys/cgiwrap/hpmuseum/archv016.cgi?read=100987#100987
An edited excerpt:
"Of course, if the magnitude of a number is outside the range of the display ..., scientific notation with a base10 exponent will be utilized. For numbers of small nonzero magnitude, these (business) models will simply extend a FIX display format with more decimal digits as needed until the number can be displayed as a nonzero value."
The HP27S treats this display issue as HP scientifics do, not as the businessoriented HP17B/BII.
 KS
▼
Posts: 614
Threads: 66
Joined: Jul 2006
Hi Karl,
Thanks for the earlier thread link. I knew I remembered I had previously seen something about this, but couldn't remember where.
The HP10B also operates the same as the 17BII and 19BII.
In going over the HP19BII manual last night about display format, I found that the 19B manual is a little different from the 17B.
The 19B manual state that "Regardless of the number of decimal places displayed, the HP19B stores all numbers with 12 digits."
The 17B manual states that "Changing the number of displayed decimal places affects what you see, but does not affect the internal representation of numbers. The number inside the calculator always has 12 digits."
BUT the 19B has a footnote that the 17B does not have: "An exception is values PV, PMT, and INT used in amortization calculations (refer to amortization on page 85), which are rounded to the display setting."
Also second footnote for 19B: "During complex internal calculations, the HP19B uses 15difit numbers for intermediate results."
When I get a little time, I'll have to check this out to see if their is a difference in the way 17B and 19B handle this.
Bill
