Although I don't know how to solve the problem, I don't know why you would even want to use a pie chart for data that looks like this. You might have better luck presenting the information with another kind of chart?
For instance, in the "1 Large Value vs. Few Small Value" chart, maybe a bar chart with horizontal bars and a logarithmic scale. Sort by Y value descending. It will still show that Tyler Valdez has vastly more "Quantity2" than the others, but now you'll still be able to see the relative size of the others in the same chart. That's assuming your users are sophisticated enough to understand logarithmic scales.
In the chart with many items, all approximately the same, again, maybe a bar chart with horizontal bars, linear scale. Set a maximum number and allow scrolling. Sort by Y value descending. You'll very easily see the range of values for the entire chart at a glance, and with slices that small, the pie chart wasn't giving you the "part to whole" visualization that it is normally used for anyway.
Although I don't know how to solve the problem, I don't know why you would even want to use a pie chart for data that looks like this. You might have better luck presenting the information with another kind of chart?
For instance, in the "1 Large Value vs. Few Small Value" chart, maybe a bar chart with horizontal bars and a logarithmic scale. Sort by Y value descending. It will still show that Tyler Valdez has vastly more "Quantity2" than the others, but now you'll still be able to see the relative size of the others in the same chart. That's assuming your users are sophisticated enough to understand logarithmic scales.
In the chart with many items, all approximately the same, again, maybe a bar chart with horizontal bars, linear scale. Set a maximum number and allow scrolling. Sort by Y value descending. You'll very easily see the range of values for the entire chart at a glance, and with slices that small, the pie chart wasn't giving you the "part to whole" visualization that it is normally used for anyway.
Actually this is something asked by the user. From my point of view, I also think that it is not possible to fit this kind of data into a pie chart. Even if it can fit, it wont be useful as the slices will be too small to see! But I'm not sure if there is any trick that can use to 'ungroup' it. If there is no other way, then I'll have to convince my user that it is not possible to ungroup and ask them to try different chart type.
But if anyone else know how to tackle this problem, please post it! Thanks a lot!
The minimum acceptable value for the lower limit is 0.001 (1/10 of a %). I don't think anyone could differentiate numbers below that... If it's imperative to show all values in the pie, you could force all the smaller values into a minimum of 0.001:
rangemin( MyValue, 0.001)
I expect all values to be listed if you use this...
Perhaps you can use two pie charts for the second kind of data. When the first pie chart bottoms out at "Others", create a second pie chart that shows ONLY the "Others". In the attached example, I've set the first pie chart to stop at 1%, and the second chart to pick up where it left off. It would be nice if there were a way to create a nice visual reference from the small slice in the first chart to the second chart, showing how one expands into the other, but I didn't think of a good way to do it.
