I'm using 2009 vol 2 xamchart barchart. I'm creating a rotate3d transform that rotates the chart with the mouse. There are a mere 25 columns and 25 series (i'm plotting pixel intensities in an x,y region). The chart initializes resonably fast. The rotate transform is barely acceptable - somewhat choppy. If I increase the datapoints to say a 35 by 35 region, it takes seconds for the chart to redisplay after the rotate3d transform is changed. It's unusable. The chart is the same size, only more datapoints - why would the rotate transform perform worse? I'm not recomputing anything other than a new transform, and I have also tried enablerefresh = false. Any idea what might be going on and how to resolve? Seems like the chart is doing something behind the scenes...
Thanks in advance.
To better determine what is going on with performance, I would need to reproduce the same behavior. Is this something you could replicate with a small sample WPF project?
Thanks!
Yes. I've duplicated it in the attached zipped project.
I initialize with 25 x 25. You hold the left mouse button down and move the mouse and the chart rotates. You will see it a bit choppy, but ok. You can increase the x and y via the text boxes and click build data button, which will build a new datatable and bind it to the grid. You will see the rotation performance degrades significantly.
I've asked engineering to look into this. They suggested that I report this as a bug in order to ensure that we improve performance with an upcoming release. Until then, I am also looking into seeing if there is a suitable work-around to help you with this issue now.
Thank you!
Thanks a lot, I appreciate the help!
Robert,
I received a suggestion from our development team that modifying the transform object instead of resetting the property on the chart, made performance improvement, such as:
From:
IntensityChart.Transform3D = new System.Windows.Media.Media3D.RotateTransform3D(oQR3D);
To:
oRT3D.Rotation = oQR3D;
This has improved the performance significantly as I have tested it, and I will also submit this request in our internal issue tracking system to be reviewed for further improvement.
Thank you,
Sam
Thanks for the quick response. I agree, this modification has improved the rotate transform performance significantly. Thanks a lot!
Our developers have completed their review of the logged issue 57236, and determined that the workaround that was provided earlier is currently the best solution.
Thank you,Sam
OK, thanks a lot.