Re: [Dynapi-Dev] TCanvas vs. DynLayer

[Jordi - I. - M. <jmi...@or...>]

Speed optimization can always be introduced. In fact, the latest precreation
code that caused some old bugs to reappear and some people to complain about the
API going backwards was introduced in order to speed up layer creation,
something that was not a problem until people started wanting hundreds, even
thousands of layers onscreen.

I've been tempted to suggest this many times but I didn't want to spawn another
"code split-up" argument. Some critical methods like, say, moveTo or setSize
might speed up by doing.:

if(is.ns) DynLayer.prototype.moveTo = function A
else DynLayer.prototype.moveTo = function B

I'll try myself in see what happens

Pascal wrote:

> me again :)
>
> I don't think this test is really useable.
> Your current Tcanvas code misses ALOT of things DynLayer takes care of.
>
> DynLayer sets sizes, z-index,clipping,bgimages in initialisation (and also a
> few other style properties.. note that setting a style property is what
> slows everything down)  I once did some optimization tricks to the dynlayer,
> by removing sizes/clipping etc.. this speeds up things BIG TIME, but also
> brakes useability for a large amount of widgets, and is less flexible.
>
> DynLayer has code included for fast child-creation.. even though there are
> no child layers in your test, this code is still called (function calls).
> This could be made faster in dynlayer, but for now is more readable for
> developers.  Do another test with layers containing a large amount child
> layers.. Dynlayer's precreation will probably be faster.
>
> shreded enough? ;)
>
> Pascal Bestebroer (pb...@oi...)
> Software ontwikkelaar
> Oberon Informatiesystemen b.v.
> http://www.oibv.com
>
> > -----Oorspronkelijk bericht-----
> > Van: dyn...@li...
> > [mailto:dyn...@li...]Namens Eytan
> > Heidingsfeld
> > Verzonden: vrijdag 16 februari 2001 13:19
> > Aan: Dynapi-Dev
> > Onderwerp: [Dynapi-Dev] TCanvas vs. DynLayer
> >
> >
> > I'd love to test performance one against the other. The only
> > test I did was
> > create 100 layers and check the times. In IE TCanvas was 200
> > ms faster and
> > in NS it was 1300(canvas) to 10000(dynlayer).
> >
> > I'd love you guys to start tearing my canvas to shreds.
> >
> > Included in the zip are:
> > tcanvas.js
> > browser.js
> >
> > they need to be included in the document(working on adding .include)
> >
> > 8an
> >
>
> _______________________________________________
> Dynapi-Dev mailing list
> Dyn...@li...
> http://lists.sourceforge.net/lists/listinfo/dynapi-dev