Sorry, yes it does. Found it!

https://github.com/gisle/tcl.pm/blob/master/Tcl.pm lines 502-518

elsif ($ref eq 'ARRAY' && ref($arg->[0]) eq 'CODE') {
    # We have been passed something like [\&subroutine, $arg1, ...]
    # Create a proc in Tcl that invokes this subroutine with args
    my $events;
    # Look for Tcl::Ev objects as the first arg - these must be
    # passed through for Tcl to evaluate.  Used primarily for %-subs
    # This could check for any arg ref being Tcl::Ev obj, but it
    # currently doesn't.
    if ($#$arg >= 1 && ref($arg->[1]) eq 'Tcl::Ev') {
    $events = splice(@$arg, 1, 1);
    }
    $args[$argcnt] =
    $interp->create_tcl_sub(sub {
        $arg->[0]->(@_, @$arg[1..$#$arg]);
    }, $events, undef, $current_r);
    push @codes, $anon_refs{$current_r};
}
[download]

Sorry, better luck next time!

"no it doesnt create an anonymous sub it doesnt need to

If you are so sure of that then prove it, show me a link to the arrayref processing of Tkx:after calls, or tell me where to look for the code. Till then my thought that processing of a arrayref in the callback position creates a single use anon-coderef still holds as it explains why that call method works, like the the sub{...} method does, but the static coderef fails. (well at least on unix).

There is deep autoload magic going on in Tkx that i still dont understand and i have yet to find out where the parameter munging actually takes place to move across the perl<->Tk/Tcl boundary. Ive even tried to look for it.

Not going to prove anythung or fix typos. It doesnt need to create a subrkutine to be able to call a subroutine and pass arguments to it.Why would it . calling a subroutine is identical no matter a who creates it.