## 602 Reputation

12 years, 121 days
Phd.
Paris, France

## Hi The code is very long....

Hi

The code is very long.

Isn't there a way to tell to maple what kind of numerical format it should use for the axis, whatever the input could be?

Thanks

S.

## Hi...

Hi januby, thanks for your suggestion. In fact what I want is the other way on: starting from an input which contains plenty of integrals of that kind, and make maple replacing all those integral with the good L(alpha,a,b). So in my final results I want the L's not the integrals.

Sorry if I wasn't that clear!

S.

## Hi...

Hi januby, thanks for your suggestion. In fact what I want is the other way on: starting from an input which contains plenty of integrals of that kind, and make maple replacing all those integral with the good L(alpha,a,b). So in my final results I want the L's not the integrals.

Sorry if I wasn't that clear!

S.

## Hi this would work, but...

Hi

this would work, but maple should every time check for the real part of theProcedure, and this, done many thousand of times in the code,  would require a lot of time. This is why I want to tell to maple from the beginning that theProcedure can give only real values.

S.

## Hi this would work, but...

Hi

this would work, but maple should every time check for the real part of theProcedure, and this, done many thousand of times in the code,  would require a lot of time. This is why I want to tell to maple from the beginning that theProcedure can give only real values.

S.

## I obtain still another...

I obtain still another ordering!!!

1+2*x*y^4+2*x^2*y^3+2*x^3*y^2+2*x^4*y+2*y^9*x+3*y^8*x^2+4*y^7*x^3

+5*y^6*x^4+6*y^5*x^5+5*x^6*y^4+4*x^7*y^3+3*x^8*y^2+2*x^9*y+x^10+y^10+2*x^5+2*y^5

## I obtain still another...

I obtain still another ordering!!!

1+2*x*y^4+2*x^2*y^3+2*x^3*y^2+2*x^4*y+2*y^9*x+3*y^8*x^2+4*y^7*x^3

+5*y^6*x^4+6*y^5*x^5+5*x^6*y^4+4*x^7*y^3+3*x^8*y^2+2*x^9*y+x^10+y^10+2*x^5+2*y^5

## Hi thanks for your post. But...

Hi

thanks for your post. But there is still something I don't understand.

My point is this: every time I launch the file, maple calculates the quantity scalar_pr(h3[5,5,5],h1[5]) in the main scope, and also calculates V31[5,5,5,5], which is exacty the same quantity, only calculated as the 5,5,5,5 element of an array. The two results are different after the 6th position (digits=10)

If that was all, I would be in agreement with your explanation: maple did a different summation on the two cases, or something like this.

But there is more: if I launch a second time the program, the first calculation (scalar ecc) gives the SAME IDENTICAL aswer, and V31 ecc gives a result different from scalar_pr and also different from the V31 of the former run.

So what: maple changes the order only inside the arrays and uses the same order and approximation in the main scope???

Doesn't this sound strange?

## Hi thanks for your post. But...

Hi

thanks for your post. But there is still something I don't understand.

My point is this: every time I launch the file, maple calculates the quantity scalar_pr(h3[5,5,5],h1[5]) in the main scope, and also calculates V31[5,5,5,5], which is exacty the same quantity, only calculated as the 5,5,5,5 element of an array. The two results are different after the 6th position (digits=10)

If that was all, I would be in agreement with your explanation: maple did a different summation on the two cases, or something like this.

But there is more: if I launch a second time the program, the first calculation (scalar ecc) gives the SAME IDENTICAL aswer, and V31 ecc gives a result different from scalar_pr and also different from the V31 of the former run.

So what: maple changes the order only inside the arrays and uses the same order and approximation in the main scope???

Doesn't this sound strange?

## Hi pagan   I'll read your...

Hi pagan

I'll read your link. But then what: if I work with typical quantity of e-50 I should set Digits:=51 to be sure?? The calculations would need weeks with that precision...

## Hi pagan   I'll read your...

Hi pagan

I'll read your link. But then what: if I work with typical quantity of e-50 I should set Digits:=51 to be sure?? The calculations would need weeks with that precision...

## Hi Georgios   but how can...

Hi Georgios

but how can this explain why in the main scope I always get the same result and in the Array scope always different??

## Hi Georgios   but how can...

Hi Georgios

but how can this explain why in the main scope I always get the same result and in the Array scope always different??

## Hi Robert thanks for you...

Hi Robert

thanks for you remark. But, as long as the number of terms if finite (like in my case and your exemple) who care about order? If I put numerical value for x and y the result shoud be the same in all the cases... will it?

S.

## Hi Robert thanks for you...

Hi Robert

thanks for you remark. But, as long as the number of terms if finite (like in my case and your exemple) who care about order? If I put numerical value for x and y the result shoud be the same in all the cases... will it?

S.

 5 6 7 8 9 10 Page 7 of 10
﻿