MACSYMA integration

[TheFausap]

Hello all, I am using Macsyma for some times, without any issues. Today I built a KL sims to have a little bit more memory available. Anyway I stumbled upon this error (maybe it's not a bug, but at least the result should be different), doing this simple macsyma integration:

integrate(x^x,x,0,1);

The integral of x^x is defined only through series, nevertheless Macsyma should be able to perform this calculation. Instead I have a problem with "NO MORE ADDRESS SPACE". Just to say, this is a brand-new macsyma session.

Is this expected, according to your experience? I mean in a quite long script, it could happen, but after only one integration, it seems something like a bug or similar.

I tried also with KA and I have the same issue, so it seems it's not processor dependent.

[eswenson1]

Interesting. In my macsyma, I get this:

This is MACSYMA 1000

(C1) integrate(x^x,x,0,1);

DEFINT FASL DSK MACSYM being loaded
Loading done

LIMIT FASL DSK MACSYM being loaded
Loading done

RESIDU FASL DSK MACSYM being loaded
Loading done

RPART FASL DSK MACSYM being loaded
Loading done
Call to an undefined function 'WN' at Lisp level.

(C2)

You appear to be getting further than me. Not sure why anything is any different -- I should have a similar environment to yours.

[BobKerns]

In Maxima, (DOE Macsyma), I get this. Noting that x^x is undefined at 0, I poked at it a bit, doing a taylor expansion around both 1 and 0.

IIRC, INTEGRATE won't return series, and I'm pretty sure the RISCH algorithm won't either, though it's been a long time.

Comparing with the RISCH algorithm gives some insight as to why it expands as it does at zero. I'm not sure what Ramanujan might be able to do with an infinite series involving log(x) | x = 0, but me, nothing, other than observe the limit(x^x,x,0) = 1.

[image: image.png]

On Tue, Dec 22, 2020 at 7:05 AM Eric Swenson notifications@github.com wrote:

Interesting. In my macsyma, I get this:

This is MACSYMA 1000

(C1) integrate(x^x,x,0,1);

DEFINT FASL DSK MACSYM being loaded

Loading done

LIMIT FASL DSK MACSYM being loaded

Loading done

RESIDU FASL DSK MACSYM being loaded

Loading done

RPART FASL DSK MACSYM being loaded

Loading done

Call to an undefined function 'WN' at Lisp level.

(C2)

You appear to be getting farther than me. Not sure why anything is any different -- I should have a similar environment to yours.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/PDP-10/its/issues/1984#issuecomment-749585811, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAIZ6V7NC3YPZM4CNFSSHWLSWCYSLANCNFSM4VFUAU6Q .

[TheFausap]

The integral is difficult but not impossible to "solve".

https://math.stackexchange.com/questions/141347/finding-int-xxdx

Looking at the comments, you can find either the solution for the indefinite form either from 0 to 1

But the problem here is different. Why there is an allocation error?

Il ven 25 dic 2020, 03:23 Bob Kerns notifications@github.com ha scritto:

In Maxima, (DOE Macsyma), I get this. Noting that x^x is undefined at 0, I poked at it a bit, doing a taylor expansion around both 1 and 0.

IIRC, INTEGRATE won't return series, and I'm pretty sure the RISCH algorithm won't either, though it's been a long time.

Comparing with the RISCH algorithm gives some insight as to why it expands as it does at zero. I'm not sure what Ramanujan might be able to do with an infinite series involving log(x) | x = 0, but me, nothing, other than observe the limit(x^x,x,0) = 1.

[image: image.png]

On Tue, Dec 22, 2020 at 7:05 AM Eric Swenson notifications@github.com wrote:

Interesting. In my macsyma, I get this:

This is MACSYMA 1000

(C1) integrate(x^x,x,0,1);

DEFINT FASL DSK MACSYM being loaded

Loading done

LIMIT FASL DSK MACSYM being loaded

Loading done

RESIDU FASL DSK MACSYM being loaded

Loading done

RPART FASL DSK MACSYM being loaded

Loading done

Call to an undefined function 'WN' at Lisp level.

(C2)

You appear to be getting farther than me. Not sure why anything is any different -- I should have a similar environment to yours.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/PDP-10/its/issues/1984#issuecomment-749585811, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AAIZ6V7NC3YPZM4CNFSSHWLSWCYSLANCNFSM4VFUAU6Q

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/PDP-10/its/issues/1984#issuecomment-751149947, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAOCZK3EL2GENSSW6563JQDSWPZSTANCNFSM4VFUAU6Q .