Re: [sage-support] sage 9.0: numerical_integral of unit_step()

[Dima Pasechnik]

On Tue, Jan 21, 2020 at 1:00 PM mendes  wrote:
>
> Dear all,
>
> In previous versions of Sage I was able to do very quickly some   numerical 
> integrations  involving  unit_step(t) function .
>
> But, in  the last updates (8.9 and 9.0) ,  it takes 6 times longer to do  the 
> numerical integral of  convolution with unit_step(),  than to do the same 
> operation with a gaussian function.
>
> It does not seem to be o.k., comparing the simplicity of unit_step()  with 
> the gaussian $e^{-t^2}$.
>
> #Compare:
>
> var('x,t')
> f= e^(-(t-1)^2)
> g= sin(t)
> fg= lambda t: numerical_integral(f(t=x)*g(t=t-x),0,t,params=[0])[0]
> plot(fg,t,0,3)
>
>
> #with:
>
> var('x,t')
> f= unit_step(t-1)
> g= sin(t)
> fg= lambda t: numerical_integral(f(t=x)*g(t=t-x),0,t,params=[0])[0]
> plot(fg,t,0,3)
>
>
> Thanks for your attention.

yes, in my tests it's even worse than 6 times slower.

Open a trac ticket?

>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-support+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-support/20eca99f-2e3f-4dfe-b258-d2fc7d28b0e5%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-support/CAAWYfq3BL6aRMnj0H94UVApH8jzPu7fotgjXfv2fCMe%2BZaFq7Q%40mail.gmail.com.

[sage-support] sage 9.0: numerical_integral of unit_step()

[mendes]

Dear all,

In previous versions of Sage I was able to do very quickly some  * 
numerical* integrations  involving  unit_step(t) function . 

But, in  the last updates (8.9 and 9.0) ,  it takes 6 times longer to do  
the numerical integral of  convolution with unit_step(),  than to do the 
same operation with a gaussian function.

It does not seem to be o.k., comparing the simplicity of unit_step()  with 
the gaussian $e^{-t^2}$.

#Compare:

var('x,t')
f= e^(-(t-1)^2)
g= sin(t)
fg= lambda t: numerical_integral(f(t=x)*g(t=t-x),0,t,params=[0])[0]
plot(fg,t,0,3)


#with:

var('x,t')
f= unit_step(t-1)
g= sin(t)
fg= lambda t: numerical_integral(f(t=x)*g(t=t-x),0,t,params=[0])[0]
plot(fg,t,0,3)


Thanks for your attention.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-support/20eca99f-2e3f-4dfe-b258-d2fc7d28b0e5%40googlegroups.com.