emacs rendering comparisson between emacs23 and emacs26.3

[rrandresf]

Hi.

On my TODO list (kind of worn hole). I have several notes.

One of them is comparing the rendering on emacs23 and emacs26.3. Today
Alan Mackenzie share an elisp Snippet:

--8<---------------cut here---------------start------------->8---
(defmacro time-it (&rest forms)
  "Time the running of a sequence of forms using `float-time'.
Call like this: \"M-: (time-it (foo ...) (bar ...) ...)\"."
  `(let ((start (float-time)))
    ,@forms
    (- (float-time) start)))

(defun time-scroll (&optional arg)
  (interactive "P")
  (message "%s"
           (time-it
            (condition-case nil
                (while t
                  (if arg (scroll-down) (scroll-up))
                  (sit-for 0))
              (error nil)))))
--8<---------------cut here---------------end--------------->8---

Alan suggested this to Andrea Corallo (BTW nice work with your branch
Andrea. and I still remember the discussion about numbering and source
code Alan)
--8<---------------cut here---------------start------------->8---
, visit .../emacs/src/xdisp.c, and do M-: (time-scroll)
--8<---------------cut here---------------end--------------->8---

When I saw this. The first thing that come to my mind was (it is
useful on one of the items of my TODO'S). So I did It on xdisp.c from emacs26-3 
source
code.

Applying it on emacs23 and on emacs26 (on an opi+2e machine with 1 Ghz
processor)
This is the result:
--8<---------------cut here---------------start------------->8---
emacs-23
162.7052059173584
emacs-26.3
281.4721345901489
--8<---------------cut here---------------end--------------->8---


The difference is very notorious. I realized it from a long time on
another ARM machine with 480 Mhz (which i still use aka my phone). But
I have not got the time and tools for letting you guys know about it. So
thanks Alan for sharing it.

Teoretically. What would be the cause of this difference on rendering
between versions?

Best Regards