Currently I use a CategoryDateTimeXAxis but in reading through some of the performance information it is recommended to use a categoryxaxis rather than the categorydatetimexaxis. The problem I am running into is that I am plotting several real time data channels that do arrive in order but the different channels may come in at different intervals and all the data sets need to be aligned by time.If I use categoryxaxis then distance of 5 seconds will look the same as 5 minutes. Also, the stepline series does not allow a numeric xaxis. Is there a suggestion on how to optimize the graphs a bit more? I need the ability to see 8 graphs containing 10 -15 series each all updating real time and at different rates. Right now, just looking at 2 puts me at 30 - 40% cpu.
Hello Mike,
Thank you for your feedback. I am glad that you resolved your issue and I believe that other community members may benefit from this as well.
Thanks again.
Yes, being able to turn off that feature might save a little more cpu for other things that the graphs need to do that are very intensive.
It is recommended to use CategoryAxis if the time intervals are not important. In your case they determine the gaps, so I beieve it is better to use CategoryDateTimeAxis. If you want I can log a Product Idea, so that you could be able to turn on and off the sorting the Axis performs.
Looking forward for your reply.
Hi Stefan,
This is not so much an issue as just help with increasing efficiency. In an article it was suggested to not use the CategoryDateTimeXAxis if the data was already in order. This is the case for me, however, the the data is time specific so a time gap of 5 minutes cannot look like a time gap of 30 minutes on the graph. It is possible that the data will come in at different rates so it is difficult to accurately pad the data points to ensure the time between them is consistent. Also, I have multiple series that all need to be synchronized by time.
My question is simply is there a way to either not use the CategoryDateTimeXAxis but still gain the ability to make time spans relative or is there a way to use the datetime axis but make it more efficient (ie do not sort collection when a point is added since it is already sorted)?
I can provide a sample but it may take a little time to rip out all the unnecessary stuff and replace all the external data inputs with local generators. Let me know if you still need a sample to answer the above question.
Thank you for your post. I have been looking into it, but it seems like that I am missing something in your scenario, so if this is still an issue for you, could you please send an isolated sample project, where the issue is reproduced, so I can investigate it further for you.