Tim Rowledge tim at sumeru.stanford.edu
Thu Apr 3 20:25:53 UTC 2003

Aaron J Reichow <reic0024 at d.umn.edu> wrote:

> The Dynapad R0.1 image is based on Squeak 2.8, while TinySqueak is
> based on 3.0.  While not the source of all of the speed difference, a lot
> of it is likely because of the differences between 2.8 and 3.0.  There has
> been some talk on this list about the generally change in responsiveness
> between Squeak version 2.8 and 3.0.  Some people seem to think that this
> is absurd, but a lot of people have noticed that there is indeed a
> difference in how fast Squeak feels that occured somewhere in the series
> from 2.8 to 3.0.
I just ran up a 2.8 image on my Acorn and it is waaaay more responsive
than any recent image. This is, I think, largely due to it being an MVC
based image from _before_ the event system was put in. The Controllers
are rapidly polling for any input rather than waiting on a semaphore. I
can get similar response from a modern image in an MVC project _IF_ I
change the global sensor to an instance of InputSensor instead of
EventSensor. It crashes if you go to a morphic project though. :-(

Funnily enough if I do '0 tinyBenchmarks' in that 2.8 image I get.....
258 bc/sec and 496 sends/sec :-) something is a little wrong there.

Even odder, if I try to run a 2.9 or 3.0 image on my current VMs they
simply don't work; the 3.0 never opens a display, the 2.9 takes in
excess of 5 minutes to draw the screen. With the 2.9 vm instead, the 2.9
image starts up ok but takes no notice of the mouse! Funny, it used to
work just fine.

The key to improving things is Profiling. Last time we had a small blitz
on morphic performance some real improvements were made. It would be
great if somebody with a bit of time could make some good test cases for
as many UI operations as possible. I'd start with
opening a browser and setting to a particular method (Andreas provided
a recipe for that)
opening a menu (several sizes would be good, include prebuilt and JIT
scrolling a text view
entering text (various cases including into an empty view, in the middle
of a small paragraph and a large paragraph etc)
popping up/removing a flapping scrollbar
selecting an item from a list

If we could have a SUnit suite for these then at least we could run them
after any suggested UI related changes and see if they have had any bad
effects (or good ones)

Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
"E=Mc^5...nahhh...E=Mc^4...nahh...E=Mc^3...ah, the hell with it."

More information about the Squeak-dev mailing list