Dear masters,

I'got the error, "Result is above the max. allowed positive number error" during exponential calculation.



How I resolve this kind of errors?

Thank you!

M.S.




exponential_max_err_v0.sm (12 КиБ) скачан 40 раз(а).

Wrote

How I resolve this kind of errors?

... like this: index 'j' to de-confuse 'i' imaginary.

exponential_max_err_v0.sm (12 КиБ) скачан 32 раз(а).

Wrote

Dear masters,

I'got the error, "Result is above the max. allowed positive number error" during exponential calculation.



How I resolve this kind of errors?

Hi. Summatorie ranged have some issues. For instance, the range variable isn't local. With eval you can get punctual results, but this version can't be plotted, and I don't know why:



Whith this other version you can do both things:



Hope that this helps.

exponential_max_err_v0.sm (13 КиБ) скачан 33 раз(а).

Best regards.
Alvaro.

... more doctoring in the *.sm attached
1. local definition mode
2. index piece wise



exponential_max_err_v0 (1).sm (24 КиБ) скачан 29 раз(а).

Wrote

but this version can't be plotted, and I don't know why:

... code this to the previous doctored document.
Summation operator is very capricious, but works
and plot scalar as demonstrated. Plot scalar for
lot more involved stuff [Budapest 39, Legendere ...]



Remember Alvaro: you doctored my Barnes Hypergeometric !

Thank you your replies.
But I have a still same problem.
Here are another sample file.

exponential_max_err_v03.sm (18 КиБ) скачан 27 раз(а).

Hi. Yes, it's hard to drive those calculations. Try to avoid sum(1) too:



Best regards.
Alvaro

Wrote

Dear masters,

I'got the error, "Result is above the max. allowed positive number error" during exponential calculation.



How I resolve this kind of errors?

Thank you!

M.S.




exponential_max_err_v0.sm (12 КиБ) скачан 40 раз(а).

I am getting error "Result is above max allowed positive number",too! And l don't know how to solve it.

Hello,

I think something is wrong here in the numeric/symbolic engine of SMath which causes those errors.
Version "0.99.7109"



I can not understand this . The error message is quite confusing. The result is quite a small number close to zero. On the other hand, by mouse hoovering the vector it could be seen the fraction presentation of the numbers (symbolic engine) with quite large numbers in the denominator. Therefore the message about max. allowed positive numbers. I think something is wrong in the floating point arithmetic in SMath numerical engine or at least quite problematic. It seems that eval() must be used extensively in order to escape from these traps and to avoid the symbolic engine to make a mess here.

Regards,
Radovan

Wrote

Thank you your replies.
But I still have same problem.

Hello Novice ...
I can't see what you are attempting.
Are you trying to fit a model function to a data set ?
If so: just attach the data set, some smart cat will exercise.
A maths document that has no hint about ... Oh !

Wrote

I think something is wrong here in the numeric/symbolic engine of SMath which causes those errors.
Version "0.99.7109"

From observation in many more SS 7109, the indexing is freaked.
Doctored in pink SS 6179.

exponential_max_err_v03.sm (17 КиБ) скачан 28 раз(а).

Dear jean,


I'm just doing the calculation of an ANS standard example.



I think this problem is a kind of overflow of exponential function.
As you know if X is greater than the 709.783 then the value of exp(X) is overflown.
So I believe that SMath should do something to check the nominal boundaries of special functions (exponential, logarithmic etc.)
Thank you!


regards,

MS

Wrote

I think this problem is a kind of overflow of exponential function.
As you know if X is greater than the 709.783 then the value of exp(X) is overflown.
So I believe that SMath should do something to check the nominal boundaries of special functions (exponential, logarithmic etc.)
Thank you!

Thanks for the hint.
Functions that overflow can be managed otherwise.
About that project, just give the X,Y data you want to fit.
Smath reports overflow as "max value ..."
All those math functions aren't Smath, rather from numerical library
as implemented in Microsoft and other maths tools.

Ah! I can see XY pairs coming soon.

Cheers ... Jean

Wrote

Wrote

I think something is wrong here in the numeric/symbolic engine of SMath which causes those errors.
Version "0.99.7109"

From observation in many more SS 7109, the indexing is freaked.
Doctored in pink SS 6179.

exponential_max_err_v03.sm (17 КиБ) скачан 28 раз(а).

Still confused...



exponential_max_err_v03-corr.sm (31 КиБ) скачан 27 раз(а).

Wrote

Still confused...

Doctored a bit SS 6179 ... No red in No time.

exponential_max_err_v03-corr.sm (32 КиБ) скачан 31 раз(а).

Wrote

Doctored a bit SS 6179 ... No red in No time.

... from there => what's next step ?



exponential_max_err_v03-corr [Next].sm (39 КиБ) скачан 25 раз(а).

Wrote

... from there what's next =

... does the suite in green makes sense ?



That mystic formula has to be put in executable maths

exponential_max_err_v03-corr [Next-Suite].sm (160 КиБ) скачан 31 раз(а).

Wrote

That mystic formula has to be put in executable maths

Please, check my interpretation.



exponential_max_err_v03-corr [Next-Suite].sm (185 КиБ) скачан 23 раз(а).

Wrote

Wrote

Still confused...

Doctored a bit SS 6179 ... No red in No time.

exponential_max_err_v03-corr.sm (32 КиБ) скачан 31 раз(а).

Still red for me...

Wrote

Still red for me...

Radovan, mon ami ... you must feel Sysyphus on that boulder !

I figure the all project is as simple as in the attached SS 6179.
Novice didn't give enough ... wait and see.
On the other hand your red is for Grand Doctor Andrey.

Jean

Fission.sm (133 КиБ) скачан 24 раз(а).