Data Series on Order Flow Market Depth Map

[FuriousM]

Hello,

It seems that my Data Series settings are not took into account on Chart where I have Order Flow Market Depth Map

As you can see on Screenshot I have already 153 bars loaded while my data series limit was set to 50 bars only.


It seems that my Depth Map keeps growing and growing ... Resulting in my computer to freeze as it can't handle that much this type of Depth chart...


Is this a known bug ?
Am I doing something wrong in the settings ?


Thanks for your help.

[NinjaTrader_Jim]

Hello,

Thanks for your question.

Bars To Load or Days to Load in the Data Series window determines how much historical data needs to be imported to start building the chart.

The chart then starts processing realtime data and new bars are built on the chart. As bars form, new plots are made.

The amount of data on the chart would not necessarily affect freezing, but it can increase loading times.

Adding new data points for plots is not exactly resource intensive, usually performance matters depend on what is processing realtime data, or what is preforming resource intensive calculations.

Order Flow tools in general can be very resource intensive and having multiple instances, or using Order Flow tools with other resource intensive tools/applications in use.

We recommend testing a blank chart, without anything added, in a blank workspace, to see the platform running at optimal speed. Understanding performance matters further should build off of that baseline test.

I am including some performance related information (how to understand them, what can be done about them) below.

Diagnosing freezing/performance symptoms

You may confirm by testing a basic chart without any custom barstypes, templates, or indicators added. The basic chart in the blank workspace should behave as expected because there is nothing that would be slowing NinjaTrader down.

In order to work through these issues, it is recommended to reduce what is being used in the workspace, and to configure the workspace so it is more efficient on the CPU.
If issues persist, we may then suggest isolating third party components to see if one is causing the freezing symptom.

Understanding Performance Issues and Configuring Efficient Workspaces

The impact NinjaTrader has on your PC will be dependent on how much NinjaTrader has to calculate. For example, having several charts with multiple complex indicators running can cause NinjaTrader to stutter and can cause short freezes. We typically advise the following as tips to improve platform performance.

1. Configure charts so their Days To Load is an absolute minimum, just enough for your indicators to function properly
2. Configure indicators to use Calculate.OnBarClose when possible as opposed to Calculate.OnPriceChange and Calculate.OnEachTick.
3. Use less charts and less complex indicators
4. Avoid using Tick Replay.

Various performance tips can be found here — https://ninjatrader.com/support/help...ance_tips2.htm

You may also wish to use the NinjaScript Utilization Monitor (Control Center > New > Output window > Right click > NinjaScript Utilization Monitor) This tool will tell you which components in your workspace are creating the most impact on your system.

After you leave this utility running, the items on the top will be those that are taking the most CPU time. Reducing/eliminating these items will be bring the most performance improvement. Please note that some NinjaScripts, like ChartStyles, require drawing every visible bar with every render pass. Having a ChartStyle at the top of your Utilization Monitor would be normal since this script will need to draw each and every bar that we see, each time that chart is interacted with.

I also suggest to use the Windows Task Manager to view other programs that are using your CPU, and to close resource intensive programs that are taking juice away that NinjaTrader would otherwise be using. Note that since NinjaTrader is multi-threaded, indicator work is distributed across multiple cores, but we may not see the full CPU utilized. We may also note that one single thread may never reach 100% utilization since that thread may need to wait on another to before it may resume.