Pre loader

Tag: Optimisation

1 vote
231 views

Hello SciCharts Team,

I have multiple data series of <date, double> Type with no:of sample points ranging in the magnitude of Millions (Not realtime). Is there a way to have a blanket model graph fitting available for the data out of the box? Right now even with default resampling provided by the Sci charts, the number of points plotted on the screen is more than what is optimal. I looking for a solution, where I can keep a constant n data points on the screen irrelevant of the zoom state.

I have attempted to use the resampling on the IRenderableSeries as provided by default by the Sci charts. Please find the screenshots for those graphs attached. The graph is not visually appealing or hard to analyze because of the high data point density. This is the major reason why I’m requesting an alternative/method to improve the visualization.

I have achieved this previously by a resampling strategy to show a maximum of n points on the screen. Let’s say the screen accommodates 2k points (n = 2k) for example, (sample total) 2M points will be aggregated to 6000k points, 2k as average, 2k as max, and 2k as min. Just an idea for thought. I’m sure there might be better ways of doing this. Please provide me with some suggestions to improve this scenario.

Thanks,
Arjun

0 votes
7k views

Hello again!
In addition to the ticket https://www.scichart.com/questions/question/memory-usage-1

I’ve modified my project to run faster. I’ve put BackgroundImageAnnotation, from which we create sprite, outside the function. But when I run the application, the image freezes and doesn’t redraw.
What is the reason? And how can I fix it?

I’ve attached modified project below.
Link: https://yadi.sk/d/TSsZuDTbipP7y

  • Egor asked 7 years ago
  • last active 7 years ago
1 vote
5k views

I am testing scichart and want to plot as much data as possible on a memory constrained system.

Our data is equi-time with no gaps in any of the series. It seems very memory wasteful to have to store an x value per data point per series when these could just be directly infered from the index of the point, this would allow memory usage to be halved.

Is it possible to achieve something similar to this in scichart.

Side question if I have a number of series that all share x values is it possible to use one set of x data for multiple series without it being duplicated in memory.

  • Hugoagogo asked 6 years ago
  • last active 6 years ago
Showing 3 results

Try SciChart Today

Start a trial and discover why we are the choice
of demanding developers worldwide

Start TrialCase Studies