+- HP Forums (https://www.hpmuseum.org/forum)
+-- Forum: HP Calculators (and very old HP Computers) (/forum-3.html)
+--- Forum: HP Prime (/forum-5.html)
+--- Thread: Prime: Fraction dependent on FIX setting (/thread-206.html)
Prime: Fraction dependent on FIX setting - Terje Vallestad - 12-22-2013 02:33 PM
Hi,
Whilst demonstrating the Fractions key to my daughter I happened to have my calculator set to FIX 2. The number on the display was 1.25 and cycling through the varieties of fractional display I was surprised to see;
[attachment=56]
(Preview does not show the picture, so to be sure the display shows 1.25 on level 4, 4/3 on level 3, 1+1/3 on level 2 and 1°15' on level 1)
Using Shift Fractions however seems to give the expected result 1°15'
Same result using the Virtual Calculator on my computer.
I could not find any reference to this behaviour in the user manual nor any previous report on the forum. I have the latest firmware installed on both the calculator and the Virtual Calculator. Whilst I can understand that this has something to do with rounding off, I do not see why the effect should not be evident for both Shifted and Un-shifted Fractions key.
Is this a bug or an undocumented feature?
Cheers, Terje
RE: Prime: Fraction dependent on FIX setting - Tim Wessman - 12-22-2013 10:30 PM
The issue is the documentation does not mention that the current digits setting impacts the way the key works. This allows you to control how you are calculating the fractional item. This is kind of similar to how changing the fix setting would make the evaluated integral precision change in older hp calcs. Change it to FIX 3 and it would returns what you expect (basically, rounds to 1 less then the digits setting).
The display of the item does not do anything to the underlying number and just flags it as "display with a fraction". Hence when you convert it to dms it works as expected. Nothing is lost and it is purely a display thing.
RE: Prime: Fraction dependent on FIX setting - Terje Vallestad - 01-01-2014 01:58 PM
(12-22-2013 10:30 PM)Tim Wessman Wrote: The issue is the documentation does not mention that the current digits setting impacts the way the key works. This allows you to control how you are calculating the fractional item. This is kind of similar to how changing the fix setting would make the evaluated integral precision change in older hp calcs. Change it to FIX 3 and it would returns what you expect (basically, rounds to 1 less then the digits setting).
The display of the item does not do anything to the underlying number and just flags it as "display with a fraction". Hence when you convert it to dms it works as expected. Nothing is lost and it is purely a display thing.
Tim,
Thank for taking time to respond
I noted this change of display when playing with various inputs. However the system does not seem to be consistent, as if you enter .75 or .25 it shows \( \frac{1}{4} \) rather than the expected \( \frac{1}{3} \) when you keep the FIX to 2 decimals.
It seems the display of the fractions is dependant on number of total digits displayed, not only digits after the decimal mark. Re-reading your response, maybe that is what you said?
If it is intended to work this way, I think it is quite confusing. For instance 100.25 (still using FIX 2) shows as 91 then 91+\( \frac{0}{1} \) for each time the \(a\) \(b/c \) key is pressed. Both displayed results are somewhat off the 100.25 start point....
So maybe this is elevated to a minor bug, or it all becomes clear in a future revision to the user's manual?
Cheers, Terje
RE: Prime: Fraction dependent on FIX setting - Michael de Estrada - 01-01-2014 02:49 PM
(01-01-2014 01:58 PM)Terje Vallestad Wrote: If it is intended to work this way, I think it is quite confusing. For instance 100.25 (still using FIX 2) shows as 91 then 91+\( \frac{0}{1} \) for each time the \(a\) \(b/c \) key is pressed. Both displayed results are somewhat off the 100.25 start point...
That is just plain ridiculous IMO. Also, I really wish the Prime would allow direct entry of whole plus fractional parts the way you can on calcs like the 35s, e.g. 3-3/8 is entered as 3.3.8 and automatically converted to 3.375. Yes, I know you can select textbook entry and use the fraction template to do this, but it's far more cumbersome.
RE: Prime: Fraction dependent on FIX setting - patrice - 01-01-2014 09:13 PM
Quote:If it is intended to work this way, I think it is quite confusing. For instance 100.25 (still using FIX 2) shows as 91 then 91+0/1 for each time the a b/c key is pressed. Both displayed results are somewhat off the 100.25 start point....Bug: Nice catch
Quote:However the system does not seem to be consistent, as if you enter .75 or .25 it shows 1/4 rather than the expected 1/3 when you keep the FIX to 2 decimals.How can you end up with expecting 1/3 when you enter 0.25 ?
RE: Prime: Fraction dependent on FIX setting - Terje Vallestad - 01-01-2014 10:36 PM
(01-01-2014 09:13 PM)patrice Wrote:Quote:If it is intended to work this way, I think it is quite confusing. For instance 100.25 (still using FIX 2) shows as 91 then 91+0/1 for each time the a b/c key is pressed. Both displayed results are somewhat off the 100.25 start point....Bug: Nice catch
Quote:However the system does not seem to be consistent, as if you enter .75 or .25 it shows 1/4 rather than the expected 1/3 when you keep the FIX to 2 decimals.How can you end up with expecting 1/3 when you enter 0.25 ?
Patrice,
It was only expected due to my understanding of Tim's explanation. I would not expect (nor wish for) such behaviour in general.
Cheers, Terje