Update on this; it definitely has nothing to do with DrRacket, I can reproduce 
it directly in Racket with a frame% wrapper. Also, thanks to Northwestern’s 
lengthier nightly snapshot trail, I’ve discovered that the regression was 
apparently introduced between 6.10.1.1 and 6.10.1.2, although it’s not present 
(thankfully) in the 6.11 release, suggesting that it’s related to a change that 
wasn’t picked for the release.

Anyhow, looks like a gui bug, so I submitted a GitHub issue against the gui 
repo. Hope that’s the right place… though now that I’m looking at the commit 
logs on gui, it looks like this might just belong on the racket/racket repo. 

John


> On Nov 16, 2017, at 15:55, John Clements <cleme...@brinckerhoff.org> wrote:
> 
> Since the release of 6.11, it appears there’s been a major major performance 
> regression in the display of 3d plots in the interactions window.
> 
> To see this, try running this program (from the plot docs), then try to drag 
> the image around.
> 
> On a fresh download of the snapshot, dragging is unusable; you can click and 
> drag, but all you’ll see is frantic GC. Sometimes, after releasing the 
> handle, the display updates. This problem does not appear to be present in 
> the released version of 6.11.
> 
> Also, FWIW, this is running OS X High Sierra, just upgraded about a week ago.
> 
> Also possibly related; I’ve been having serious trouble running programs that 
> display long (200-300 lines) blocks of text in the interactions window. I’ll 
> see (say) twelve and a half lines of purple output, then wait for three 
> seconds, then another few lines of output, wait a few seconds, etc. etc.
> 
> I’d be happy to submit this as a github issue…would it go into the 
> racket/drracket repo, or the racket/racket repo?
> 
> John
> 
> #lang racket
> 
> (require plot)
> 
> (require (only-in plot/utils bounds->intervals linear-seq))
> (define (norm2 x y) (exp (* -1/2 (+ (sqr (- x 5)) (sqr y)))))
> (define x-ivls (bounds->intervals (linear-seq 2 8 16)))
> (define y-ivls (bounds->intervals (linear-seq -5 5 16)))
> (define x-mids (linear-seq 2 8 15 #:start? #f #:end? #f))
> (define y-mids (linear-seq -5 5 15 #:start? #f #:end? #f))
> (plot3d (rectangles3d (append*
>                         (for/list ([y-ivl  (in-list y-ivls)]
>                                    [y  (in-list y-mids)])
>                           (for/list ([x-ivl  (in-list x-ivls)]
>                                      [x  (in-list x-mids)])
>                             (define z (norm2 x y))
>                             (vector x-ivl y-ivl (ivl 0 z)))))
>                        #:alpha 3/4
>                        #:label "Appx. 2D Normal”))
> 



-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to