Heatmap changed with latest version

The latest download (OxyPlot-2013.2.140.1) shows me another heatmap then the versions before.! Is there a change, it looks so different...

Please note: If I do a CTRL-ALT-C on the graph screen, the data is identical but the graph differs.

Are any of your axes reversed? The new code takes axis reversal into account, but I think there may be a bug if X0>X1 or Y0>Y1.
I did not see any differences in the examples included in the example library.

Thanks for your quick respons.

The Y-Axis is reversed because these are the rotors of a cold mill. The bottom is rotor 21 and at the top of the Y-axis is rotor 5. The view is from the operator room of the cold-mill. Till version 139 it worked well. If you need some pictures I can send it to you.
Thanks in advance.

I have corrected the bug for X0>X1 and Y0>Y1.
I guess you can swap Y0 and Y1 in your HeatMapSeries to correct the problem.
Y0 should define the y-coordinate of the Data[*,0] elements. If this is on the top or the bottom depends on the vertical axis.
Let me know if there are better ways to define this to make it less confusing!

Thank you objo, now it's working the same as before. (with the latest v 2013.2.141.1)

Sorry there's still a problem if the lowest number of the Y-axis is at the bottom (say Cartesisch) and the highest above, the graph is not the same as before.
Reason to do this: Sometimes they want to swap the rotors of the roll(steel); if the roll is handled in another mill.

I only use:

            if (cbSwapRotors)
            {
                linearAxisRotors.EndPosition = 1;
                linearAxisRotors.StartPosition = 0;
            }
            else
            {
                 linearAxisRotors.EndPosition = 0;
                 linearAxisRotors.StartPosition = 1;
            }

and of course the data is refilled from the database in a two dimensional array. That's all I do.

Can you create a small example that fails? I think all the current examples in the example library are working (correct me if I am wrong)

When I was creating a test model for you I saw that using start- and endposition was enough for the right result (in the latest version), but in version 139 and before I had to load the peak data the other way around. (much more code) With the latest version the following routine is doing the whole fill:

        for (int iRow = 0; iRow <= iMaxArrayIndex; iRow++)
                {
                    this.listScheefstand.Add(new DataPoint((int)(dataTableVormData.Rows[iRow]["length"]),
                        Math.Abs((int)(dataTableVormData.Rows[iRow]["flatn_error_steer"]) - MiddelsteBedekteRotor)));
                    meters[iRow] = (int)dataTableVormData.Rows[iRow]["length"];

                    for (int iMeting = iLeftsensorPLUSsideOffset; iMeting <= iRightsensorMINsideOffset; iMeting++) // sensors
                    {
                        if ((string)(dataTableVormData.Rows[iRow]["SS" + iMeting.ToString().PadLeft(2, '0')]) != "U")
                        {
                            // Covered
                            peaksData[iRow, iRightsensorMINsideOffset - (iMeting - 1)] =
                                (int)dataTableVormData.Rows[iRow]["SD" + iMeting.ToString().PadLeft(2, '0')];
                        }
                    }
            }

Thanks for your help objo.