weird SOLVER behavior

Discussion around the SwissMicros DM42 calculator
Post Reply
User avatar
revwillie
Posts: 133
Joined: Mon Apr 24, 2017 7:48 pm

weird SOLVER behavior

Post by revwillie »

maybe I'm just old and confused. I was verifying results from Excel for a really simple relationship and I tried to use the DM42's solver. Instead of putting the correct solution into the solve-for variable, it puts the solution in the Z register and puts zero in the variable.

I entered the exact same program in free42 android and plus42 android and my old HP42 and it worked as expected on those three platforms.

the relationship is CT=(CS*DOL)/(T-(2*DOL))
(central tension in chemically strengthened glass found with compressive stress, thickness and depth of layer of compression)

LBL "CTEN"
MVAR "CS"
MVAR "DOL"
MVAR "T"
MVAR "CT"
RCL "CS"
RCL x "DOL"
RCL "T"
2
RCL x "DOL"
-
/
RCL - "CT"
END

so when I solve for CT, I get zero every time but I can see the expected CT value in the Z register.
could the DM42 have a flag set or something? If T is used in some other program, would it interfere with the solver's use of T?
I didn't see any reference to solver function in recent changelogs for free 42.

There is probably a simple explanation that I'm missing.
DM42 SN: 00076
DM15, HP 28c, HP 12c, HP 10b, HP 20b, HP 30b, HP 17bII, HP 42s, HP 35s, WP 34s, NP-25, TI-83+, teenix multicalc
User avatar
revwillie
Posts: 133
Joined: Mon Apr 24, 2017 7:48 pm

Re: weird SOLVER behavior

Post by revwillie »

here are the Excel numbers:

t CS DOL CT


800 910 65 88
800 832 70 88
800 1025 71 110
800 790 72 87
800 1010 74 115
650 903 63 109
650 785 58 85
650 1053 59 117
650 760 65 95
650 1022 66 130
DM42 SN: 00076
DM15, HP 28c, HP 12c, HP 10b, HP 20b, HP 30b, HP 17bII, HP 42s, HP 35s, WP 34s, NP-25, TI-83+, teenix multicalc
User avatar
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: weird SOLVER behavior

Post by Walter »

EDIT: I shouldn't post being off my calcs. :?
Last edited by Walter on Fri Aug 19, 2022 11:19 pm, edited 2 times in total.
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
User avatar
revwillie
Posts: 133
Joined: Mon Apr 24, 2017 7:48 pm

Re: weird SOLVER behavior

Post by revwillie »

Thanks Walter.

My understanding is that you have to set your relationship (the function) equal to zero.

0=((CS*DOL)/(T-(2*DOL)))-CT

I verified the program works on three other types of 42. by "works" I mean it results in CT=the expected CT
DM42 SN: 00076
DM15, HP 28c, HP 12c, HP 10b, HP 20b, HP 30b, HP 17bII, HP 42s, HP 35s, WP 34s, NP-25, TI-83+, teenix multicalc
User avatar
Walter
Posts: 3070
Joined: Tue May 02, 2017 11:13 am
Location: On a mission close to DRS, Germany

Re: weird SOLVER behavior

Post by Walter »

Do you really enter SOLVE "CTEN" ?
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
Thomas Okken
Posts: 1102
Joined: Tue May 02, 2017 5:48 pm
Location: Netherlands
Contact:

Re: weird SOLVER behavior

Post by Thomas Okken »

The program is fine and solving for CT works as expected when I try it in Plus42. This should work fine on the DM42 as well.

You could try putting SOLVE "CT" in a program, and then using SST to step into it, to make sure it's executing the right function. Unless the DM42 is in a weird state, I don't see why this would fail other than the program being entered incorrectly.
User avatar
revwillie
Posts: 133
Joined: Mon Apr 24, 2017 7:48 pm

Re: weird SOLVER behavior

Post by revwillie »

Thomas Okken wrote:
Fri Aug 19, 2022 6:33 pm
The program is fine and solving for CT works as expected when I try it in Plus42. This should work fine on the DM42 as well.
yes, that's what I hoped
Thomas Okken wrote:
Fri Aug 19, 2022 6:33 pm
You could try putting SOLVE "CT" in a program, and then using SST to step into it, to make sure it's executing the right function. Unless the DM42 is in a weird state, I don't see why this would fail other than the program being entered incorrectly.
I assumed I did something wrong, which is why I tried in Plus42 and Free42.
Of course, it was nice to have an excuse to key in the program in my HP42s, so I am not at all complaining.

to answer Walter's question, you enter:
Shift 7 (solver)
select CTEN from the soft menu
enter known values for T, DOL and CS
then press the soft menu for CT.

That's how I've been using the 42S solver for .....33 years? so that's why I stared the discussion with "maybe I'm just old and confused"
DM42 SN: 00076
DM15, HP 28c, HP 12c, HP 10b, HP 20b, HP 30b, HP 17bII, HP 42s, HP 35s, WP 34s, NP-25, TI-83+, teenix multicalc
User avatar
revwillie
Posts: 133
Joined: Mon Apr 24, 2017 7:48 pm

Re: weird SOLVER behavior

Post by revwillie »

OK, something corrupted in the state, I guess. I didn't have a state file loaded but it has been sitting idle for a while since the last firmware update. I just loaded a clean state and it behaves as expected.
DM42 SN: 00076
DM15, HP 28c, HP 12c, HP 10b, HP 20b, HP 30b, HP 17bII, HP 42s, HP 35s, WP 34s, NP-25, TI-83+, teenix multicalc
Post Reply