Hi Graham,

Thanks for the quick response.  A bit of detail - we're displaying the data acquired from a scientific instrument that collects signal from a "plate" of 396 wells over time.  Each series represent the amount of fluorescence detected in each well on the plate.  Once the data is captured, the scientists need to see all signal data at once to pick out what they are looking for.  But once the data is acquired - it will never change.

I had already deduced a couple of your suggestions (the markers and legend) and we don't have much else in the charts airspace.  (We have to add a dozen or so widgets to mark some peaks on one of the curves (some issues here which I will post about later if we stay with Infragistics) but that doesn't seem to cause any issues.)

I did not know about the bitmap caching mode and turned that on.  Can't see too much of a difference with that tweak.  When I get some more time I will play with the mouse wheel and panning suggestions.

I noticed in your sample code you are using a LineSeries.  We're using a ScatterLineSeries.  Would that make much of a difference performance wise?  We could certainly use LineSeries if the X data is evenly spaced (which it will be most of the time) and when it isn't throw in the odd ScatterLineSeries to handle it....  thoughts?

Thanks for the advice! :)

Chris

Please look for an email from me.

Thanks.

Chris

Chris,

I experimented with using LineSeries rather than ScatterLineSeries in your scenario and the experience is much smoother. This, of course, would only be appropriate if the sampling interval was the same for all your instruments, or could be aligned to be the same, does that seem like it would work?

Profiling the application, it definitely seems like you are losing some performance to ScatterLineSeries and it boils down to us being less able to simplify the visuals efficiently, without losing graphical fidelity. The crux with ScatterLineSeries is that we can't assume that the data you provide proceeds in an orderly "line-like" fashion, since this constraint does not actually exist with that series type, you could technically pick an entirely different X-Y location for every data point, and this would generate an absolutely worst case scenario for the renderer.

But your dat is, in reality, much more orderly and constrained. You can indicate these constraints to the chart by using LineSeries rather than ScatterLineSeries. However, it is possible that those constraints are too restrictive, in your scenario. Here's what Category series expect:

• Every category series (like line series) on the same category axis should have the same number of data points. If the do not, they will be aligned by INDEX. This will result in series with fewer data points being smooshed (technical term) toward the left edge of the plot area. It's on you to align the data values. Note, you can have "missing" data, and you can tell the series to either interpolate over the gaps, or create discontinuities in the line shape.
• A category series will treat each data point as being equidistant from its neighbors. This is fine if you are sampling at a regular interval, but may not be appropriate if you are sampling at a more random interval. This is one of the constraints that give us a lot of power to simplify the shapes being rendered efficiently without losing graphical fidelity.
• The labels along the x axis may hold dates, or times, but they are treated essentially as strings, the chart doesn't have any special processing for displaying interesting date related info in this mode (though you can play tricks with DataTemplates in the labels, of course)

We could also try to tune the ScatterLineSeries more to your scenario, but that would be a feature request. For example, we could add an option whereby you indicate that your data has additional constraints (monotonically increases on x values, for example) which could help performance.

Let me also dig a bit. I'm remembering creating a sample once that created a custom scatter line series with some additional constraints on the data provided which may perform better in your scenario. Still, I'd recommend looking into whether LineSeries, is right for you.

-Graham

Hi Graham.

Thanks again for the valuable insight.  I'm going to be speaking with our domain expert soon to get more information about the data and how we will expect it to arrive.  This will provide me with better knowledge on whether or not we can switch to a LineSeries or not.

Please let me know if you have any more information or changes to suggest.  Feel free to send back the code I sent if you can so I can see some comparisons to the original.

If you don't have anything more, go ahead and mark this question as "answered".

Thanks again for the support.  Infragistics has a good product.  Perhaps this area is something that can be tweaked to make graphs with lots of series perform even better.  (Maybe if there were more parameters we could set to provide hints about the data...hmmm?)

Chris

Chris,

I've got something neat for you here: https://gist.github.com/gmurray81/52776872c9cdac3a6efb

Please note this also requires a generic.xaml (included in the gist) that you'll need to set up in the project that you add OrderedScatterLineSeries to. The easiest thing to do to get this set up is to do an Add New Item in VS and select WPF => Custom Control and add one to your project called OrderedScatterLineSeries. This will add the Generic.Xaml to your project and make sure that is all set up correctly. You should then be able to use the stuff from the gist.

The gist is an old sample I dug up that I wrote quote a while ago which implements a custom scatter line series which assumes more constraints on the data provided to it. I also tweaked it a bit more for your scenario, and fixed a few bugs. You should be able to drop it in as a replacement for ScatterLineSeries in your sample and comment out the lines that relate to the markers (this series doesn't support those as of yet). You should see its a lot smoother than ScatterLineSeries, and maybe even a bit smoother than LineSeries.

Now, you can also try setting UseHairline=true on that series (added this just for you). This will lower the fidelity a tad, but will make things even smoother. Normally this custom series, and the series built into the chart, are trying to simplify the line somewhat, but they want to make sure to represent the data faithfully, so what they actually end up drawing is a polygon that spans the min/max y values. Specifying UseHairline=true on this custom series will make sure that it only every uses a Polyline to render the series, and if the data points are especially dense in a region, it will only actually represent the max values, rather than spanning both the min and max values over that range. Let me know if I should elaborate further there.

Anyhow, this decreases the complexity of the visuals we are passing to WPF quite a bit so can boost things more for you, at the sacrifice of a bit of fidelity. The line will still make sure to encompass your max values in that mode, but doesn't necessarily encompass the min peaks. Anyhow, for greater fidelity you can leave that off. And in either case case, if you zoom in, you'll see all the data.

To decrease fidelity even further, to boost perf, you can start playing with the resolution setting on this series, or the built in chart series. Raising that to higher values will decrease fidelity further, but simplify the visuals pushed to WPF.

Try all the above with BitmapCache on or off, I'm not sure if it will help or hinder you with your scenario. Basically turn it on if the scenario you have set up seems to be slowing down the crosshair/tooltip/etc in addition to the zoom performance.

Hope this helps!

-Graham

Hi Graham.

This looks cool.  I'm deep in the bowels of coding up a module to read NetCDF scientific data files right now so I can't get to this right away.  Hopefully this afternoon (I'm in the PST timezone) I'll be able to get to it.  If not, Monday.  I'll let you know how it goes.

Chris

Hi Graham,

Thanks for further suggestions.  We're good with peaks and such - we've got our own library to do all the "heavy lifting" math on the data - we'll just be adding a dozen or so annotations to the chart at the proper places (I will be raising an issue with this later though...).

Chris

Chris,

Awesome, glad I could help! Do let me know if you need any more assistance.

One note on the hairline mode, if you guys are also interested in negative peaks, you may want to alter the behavior whereby rather than it picking the min or max as the representative points for a range that it could first compute the median value for the series data, and select either min or max at any given point in time based on whether you are greater or less than the median, and maybe that will ensure that the maxima and minima are well represented.

As for praise, totally not required, just glad I could help! I'll make sure to send my manager this forum thread, though. I believe there is an opportunity to share feedback when we close a case, whether it positive or negative, but I'm not super familiar with the process, myself.

-Graham

Hi Graham,

I've gotten a chance to look at what you posted WRT the OrderedScatterLineSeries.  First off, thanks again for working with me.  Let me know who I should email to give you some public praise! :)

So with the new series type.  Seems to be working. :) I can see/feel an improvement in responsiveness.  Especially WRT drag zooming.  Of course this is all subjective but it does feel faster.  We like the hairline so that is a plus.   The BitmapCache on/off makes a big difference in performance so we'll stay with it on.  In my (limited) playing with the Resolution setting, I found that the loss of resolution was too much for any gains in performance - even going from 1 to 2 I saw smooth peaks going to sharp peaks. And since we're interested in peaks.... that wasn't what we needed so we'll probably stay away from changing that.

So to wrap up, I think we'll stick with your enhancement.  At some point soon, I'll try to get to your suggestions page to add this as an enhancement.

Chris

Neat, let me know how it goes!

-Graham