Hi,
I want to chart large data sets of stock data. The chart should include features like
- scroll a complete page
- scroll one bar of data
- scroll a fixed number of bars (e.g. 5)
Having these requirements, I’m looking for the most efficient way (performance and memory usage!) to handle the charted dataset – which could be large.
Here are the alternatives I found:
1) Use TeeChart paging. Downside: the complete data set has to be copied to a series before displaying. Probably will not work anyway, since paging will not allow me to scroll just one bar: I only can scroll the complete page. Or did I miss a feature?
2) I have not found any way to have all data in a single OHLC series and just display say bar 901 – 1025. Unfortunately the complete series always is displays. Same downside as above: the complete data set has to be copied to the series before displaying. Am I missing something?
3) Adding and removing data points from the series on scrolling
Probably quite inefficient, I experienced a noticeable delay while clearing and filling the data series on every scroll event. This is a serious problem on charting multiple series the same time (stock data + multiple indicators).
4) I thought about implement some “smart” data manager which could be assigned as “Datasource” property to the OHLC chart series. This data source only would provide the requested data points (in the above sample point 901-1025). But I have no idea on how to build such a “smart” data source, since there is no documentation on how to subclass e.g. a OHLC series and assign it to the “Datasource” property.
Basically, I’m looking for a .NET interface to implement, which then could be used by TeeChart to retrieve the required data points. This would allow me to implement some efficient and “smart” data handling. Such an interface would be perfect if TeeChart wwould use the data directly from this interface and not (!) copy the data to internal buffers (same scenario and problsm as with (3)).
Any advice/comment on these issues is highly appreciated.
Regards
Managing large amounts of data
-
- Newbie
- Posts: 5
- Joined: Fri Nov 05, 2004 5:00 am
- Contact:
-
- Newbie
- Posts: 5
- Joined: Fri Nov 05, 2004 5:00 am
- Contact: