JavaFX performance for complex visualisations

[Pedro Duque Vieira]

If you want to have a look or try it, the site where you can download the
app is back up: http://modellus.co/

Just wanted to point out that the biggest issue I find on this topic with
javaFX is smoothness, as someone already pointed out before.

Cheers,


On Tue, Dec 18, 2012 at 12:18 AM, Pedro Duque Vieira <
pedro.duquevieira at gmail.com> wrote:

> Hi,
>
> Just wanted to leave my positive experience with javafx on this subject
> (performance).
> I've been porting a previous app that I've also made called Modellus from
> swing to javafx, it's a free app you can get it here (the official site is
> down temporarily): http://code.google.com/p/modellus5/downloads/list(current version is 0.2)
>
> The main component - the animator part, is all finished porting to javafx.
> In general users (with no concrete stats I'd say it has about 2000 - 5000
> users) are happy with the performance, only some corner cases where people
> are using old computers or are creating extremely complex simulations is
> performance slower than required and slower than the swing version. You can
> get and test some simulations from here:
> http://fisicanalixa.blogspot.pt/search/label/Modellus (I'm sorry that the
> site is in Portuguese but you can use google translate). The videos on that
> site are from the older version written in swing, but you can download and
> open the examples on the newer javafx Modellus.
>
> There are some simulations there like the "onda longitudinal"
> (longitudinal wave) posted on 25 November 2010, that have about 100 objects
> on screen + a chart + a table with hundreds of lines and performance is
> still acceptable. All this running on a custom made calculation engine that
> can respond to user input in real-time allowing the simulations to be
> interactive.
>
> Note that this is an hobby project of mine that is getting done on my
> spare time, so only a few hours a week, and also that I haven't done any
> optimization or been worried about performance. And finally that I know
> that here and there I can do some simple optimizations and bring
> it theoretically to a 10x performance improvement.
>
> On a final note I think that IMHBO (In My Humble Biased Opinion) this app
> is significantly more complex than any of the apps mentioned so far (on
> this topic) including the port of tower defense that is being done, and
> still it performs acceptably.
>
> Thanks, cheers.
>
> --
> Pedro Duque Vieira
>



-- 
Pedro Duque Vieira