Hi Graham

I don't know if I have a lot of series, you tell me what you think. The software will work from 2 to 12 series, depending on users input, each series could have 30 to 3.000.000 points. A note is necessary, when the data is larger than 7.200, I don't show all at once, I bind the data but change the WindowRect area, this already improves the performance. I'm using the InfragisticsWPF4.Controls.Charts.XamDataChart.v10.3 binary.

I will try to make a sample project to emulate what is my problem, I don't guarantee that will be realist, but I'll see what can I do.

Yes, that was an idea that I had too and I tested in the release mode and didn't perceived any meaningful performance improvement. Although I didn't use any formal profiler to validate the suspect over the Clear method raised on debug session.

Tárik

Hi Graham

I'm really disappointed, with the performance. The worst is that occurs in Clear() method. I just need to say that the usability of my software is compromised by the Infragistics library. I'm already looking for an alternative vendor.

Regards

Tárik

Hi,

There is a fix for this issue that is due to be released in the next service release of the chart. If you can provide a sample that simulates the issue you are seeing I can see if I can find you any valid work arounds until the service release is available. Have you tried recreating the chart rather than calling clear on the Series collection, for example?

-Graham

which series type are you using?
I can confirm there is no appreciable delay either before or after the service release when removing the series as below for LineSeries.
But other series types may have different performance characteristics.

public partial class MainWindow : Window
    {
        public MainWindow()
        {
            InitializeComponent();

            var xAxis = theChart.Axes.OfType<CategoryXAxis>().First();
            var yAxis = theChart.Axes.OfType<NumericYAxis>().First();
            xAxis.ItemsSource = new TestData();

            for (int i = 0; i < 12; i++)
            {
                LineSeries l = new LineSeries();
                l.ItemsSource = new TestData();
                l.ValueMemberPath = "Value";
                l.XAxis = xAxis;
                l.YAxis = yAxis;

                theChart.Series.Add(l);
            }
        }

        private void remove_Click(object sender, RoutedEventArgs e)
        {
            theChart.Series.Clear();
        }


    }

    public class TestData
        : ObservableCollection<TestDataItem>
    {
        private static Random _rand = new Random();
        public TestData()
        {
            double curr = 100.0;
            for (int i = 0; i < 500000; i++)
            {
                if (_rand.NextDouble() > .5)
                {
                    curr += _rand.NextDouble() * 3.0;
                }
                else
                {
                    curr -= _rand.NextDouble() * 3.0;
                }

                Add(new TestDataItem()
                    {
                        Label = i.ToString(),
                        Value = curr
                    });
            }
        }
    }

    public class TestDataItem
    {
        public string Label { get; set; }
        public double Value { get; set; }
    }	

How would I make this sample more like your situation?

-Graham