I did a regression report on V 1.3.8 on one Linux running 

fricas --version
FriCAS 1.3.8
based on sbcl 2.2.5

And new Linux running 

>fricas --version
FriCAS 1.3.8
based on ecl 21.2.1

And noticed a timeout on this one integration problem. I build both Fricas 
on each Linux from same source tar file.

setSimplifyDenomsFlag(true)
integrate(1/(-x^3+1)/(b*x^3+a)^(1/3),x)

On the first system it finished much faster, in less than 2 minutes, on the 
second it finished in 6 minutes.

Both Linux are different Virtual boxes, running with similar Linux OS 
(Manjaro) and similar and plenty amount of RAM. All virtual boxes are on 
windows 10.

So the only difference I could think of is the lisp used?

I repeated the above more than one time, and the timings remain very 
different.  The second one is almost three times slower.

So my question: is there a preferred lisp to use? and how to tell fricas 
when building it to use it? 

I am not now sure why one used one lisp and the other used different one. I 
do not remember now setting this when configuring the build from source for 
Fricas. 

May be Linux had default lisp and it picked it up during the build?

--Nasser


-- 
You received this message because you are subscribed to the Google Groups 
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/fricas-devel/82e8f0c7-b4f6-4bbc-9540-bb7154244975n%40googlegroups.com.
  • [fricas-devel] whi... 'Nasser M. Abbasi' via FriCAS - computer algebra system

Reply via email to