01-23-2014, 12:07 PM
(01-23-2014 11:05 AM)Gerson W. Barbosa Wrote: [ -> ]I think I'll stick to the WP 34S and use MANT instead :-)
I share your feelings and appreciate your decision.
d;-)
(01-23-2014 11:05 AM)Gerson W. Barbosa Wrote: [ -> ]I think I'll stick to the WP 34S and use MANT instead :-)
(01-23-2014 11:05 AM)Gerson W. Barbosa Wrote: [ -> ](01-23-2014 11:02 AM)Werner Wrote: [ -> ]Hi Gerson!So does the HP-15C version for 1e-99...
Fails for 1 e-499, I'm afraid
01 ABS
02 1
03 0
04 X<Y?
05 1/x
06 *
07 ENTER
08 X≠0?
09 LOG
10 INT ' or IP on later models
11 10^x
12 /
13 1
14 X>Y?
15 10^x
16 *
(01-23-2014 03:15 PM)Dieter Wrote: [ -> ]Also, this preserves most of the stack (only T is lost). Very nice!(01-23-2014 11:05 AM)Gerson W. Barbosa Wrote: [ -> ]So does the HP-15C version for 1e-99...
Yes, that's why my previous version handles these cases separately. However, this can be done more elegantly, even shorter and without an explicit constant (-499 resp. -99), so it should work on the 42s as well as on the 15C and many other HPs.
Code:
01 ABS
02 1
03 0
04 X<Y?
05 1/x
06 *
07 ENTER
08 X≠0?
09 LOG
10 INT ' or IP on later models
11 10^x
12 /
13 1
14 X>Y?
15 10^x
16 *
All input values are shifted by one magnitude towards 10, so that over-/underflow can not occur.
Dieter
(01-22-2014 09:40 AM)Werner Wrote: [ -> ]On the 41, you can use Synthetic Programming to come up with a short and efficient routine that returns the mantissa.
(01-23-2014 09:06 PM)Werner Wrote: [ -> ]The MT routine as listed on page 339 of the PPC ROM manual, which is included in the MoHPC CDs/DVD:
(01-23-2014 09:06 PM)Werner Wrote: [ -> ]Btw your latest code is almost the same as the one I posted before ;-)
(01-24-2014 06:16 AM)Werner Wrote: [ -> ]It sounds better in English than in German ;-)