@Joe Riel 

Maplev output buffer:

(**) reset;
                                     reset
(**) plotsetup(maplet);
(**) plotsetup();
  preplot = [], postplot = [], plotdevice = maplet, plotoutput = terminal,

        plotoptions =
(**) plot(sin(x));
(**) plotsetup(x11);
(**) plotsetup();
  preplot = [], postplot = [], plotdevice = x11, plotoutput = terminal,

        plotoptions =
(**) plot(sin(x));
Error, plot driver not found

@Thomas Richard 

This happens with Emacs/maplev-mode but not with command line in an xterm

@Thomas Richard 

sudo apt-get install t1-xfree86-nonfree ttf-xfree86-nonfree ttf-xfree86-nonfree-syriac xfonts-75dpi xfonts-100dpi
cd /usr/share/fonts/X11/100dpi/
sudo mkfontdir
xset fp+ /usr/share/fonts/X11/100dpi

@Thomas Richard 

I installed the missing fonts (and maybe a lot more...). mkfontdir,...

Starting commandline maple over ssh, plot(sin(x) now works with plotsetup(x11).

emacs maplev-mode still presents the plot problem.

@Thomas Richard 

I went to the '2nd' computer and started command line maple 2020.

> plotsetup(x11);
> plot(sin(x));
> Maple X driver failure:BadName (named color or font does not exist)Warning: Cannot convert string "-adobe-helvetica-bold-r-normal--14-*" to type FontStruct
Warning: Cannot convert string "-adobe-helvetica-medium-r-normal--14-*" to type FontStruct

So, it has a missing set of fonts...

@Thomas Richard 

I am running command line maple withinh emacs, on a remote computer via ssh. The local computer displays X things like emacs OK.

The lib that is missing is libimf.so, included in the maple2022 distriibution.

> sudo lddconfig /usr/local/Maple2022/bin.X86_64_LINUX ## resolved the missing lib

After that, I have a different message:

> plotsetup(x11);

> plot(sin(x));

plotting was not implemented by the application

I tried ssh to a different computer.  plotsetup(11); plot(sin(x)); resulted in a font problem, most likely a missing font. So, 'x11 should work on that system.

Maplet works on this system.

I am sorry, I should know better. I did not include the context:

I am using command line maple in emacs.

> interface(version);

`Command-line Interface, Maple 2022, X86 64 LINUX, Mar 8 2022, Build ID 1599809`

> interface(prettyprint=3)

fixed it.

I really did not twiddle with anything - new computer - new install. Sorry for the noise.

have been here before.

plotdevice=maplet, plotoutput=terminal

@Carl Love 

In my case the data was collected with 10 runs of similar sized inputs. So, LengthSplit was applicable.

Your solution of using table(sparse) makes sequential operations on lists much faster than operating on lists.

I did not know to use a table in this type application. Now, I do.

Thanks.

@Carl Love 

Thanks for all the input.

Of all the methods, using ListTools is the fastest.

See indexed_sum.mw

@acer 

f := proc(a,b)
a+=b;
end proc

print(`f`)

print(`+=`)

@Carl Love 

I ran the factor 10 times for each of 30,40,50,60,70. The mean of each group is strictly increasing.

See cado-noise.mw

How do I add groups of 10 values in a list?

@Carl Love 

The times for N>100 is measured in days...

I can do an average 10 tries for N<80. But, since this is a parallel operation, I suspect the time for smaller N is masked by other issues.

I will try this in a couple days. The process is still running, projected to complete Wednesday 4/6/2022

@Carl Love 

I should have included  the source of the measurements. The measured times were for factoring integers with N digits using parallel NFS calculations.

Tcpu is the total number of CPU seconds, the sum of all threads. Telapsed is real time.

But, a more general question, not specifically related to Maple.

I first assumed polynomial time. I looked at at some variations of log time, but, polynomial time seemed to match the later time values more closely. But, fitting a poly to data is closer as the degree approached the number of points...

Is there a way to decide this other than trial and error?

@Joe Riel

I believe button-lock is from emacs package-install.