XY graphs under-perform on large data sets

[Mads]

If for example you have 3 signals with 8 million points each and you plot these on a regular waveform graph, the user interface is able to display the data smoothly. All graph palette operations (zoom, scroll etc.) respond in "real-time".

Put the same 3x8 million points on an XY graph, and you have one sluuuuggish user interface. Scrolling is for example no longer possible in any practical fashion.

I'm sure a lot of it has to do with the overhead of having all those X-values (often unnecessarily many - as discussed in this idea), but the performance degradation compared to a regular waveform graph (even if the latter is fed twice the amount of Y values for example) is severe.

Are there ways around this performance issue? Sure. We can e.g. write code that decimates the data we send to the indicator, and refills it when the user zooms or scrolls and therefore needs additional data points. But this requires lots of code, and can never become as transparent/integrated and smooth as an implementation within the indicator itself. 

And competing products are already there, that's what bugs me right now. I've got colleagues that get such functionality "for free" with the graphing tools they have.

So, we're about to develop an XControl that makes it possible to present such large non-continuous data sets in a smooth manner. (Ironically one solution is to add data points so that I have continuous data - and then use the regular graph...) But has anyone already done this? Andhowfaroffisa nativeXYgraphindicatorthatmakes such code obsolete?