Efficiency - Graphics vs text for chart annotation

[tquinn]

I like to have my Strategy NT code annotate the charts. So I end up with a lot of stuff on a chart. I'm wondering if there is a computational difference between using, say, DrawLine vs DrawText?

For example if I want a right pointing triangle I can get it 2 ways:
A) DrawLine where the line is very short, with a big arrowhead
B) DrawText using Wingdings

For the text, B), would there be any difference in using a system supplied Font, like Wingdings, vs a custom made TrueType Font?

[NinjaTrader_Joydeep]

Hello tquinn,
Thanks for your note.

Unfortunately we do not have any test parameters/profiling in this regard.

I will however leave the thread open for our forum members who can share there experience in this regard.

[sledge]

I like to have my Strategy NT code annotate the charts. So I end up with a lot of stuff on a chart. I'm wondering if there is a computational difference between using, say, DrawLine vs DrawText?

For example if I want a right pointing triangle I can get it 2 ways:
A) DrawLine where the line is very short, with a big arrowhead
B) DrawText using Wingdings

For the text, B), would there be any difference in using a system supplied Font, like Wingdings, vs a custom made TrueType Font?

more than a few complaints here with "market replay" and adding a lot of objects to the chart on backtesting... also with people who have many charts (100-1000?) and a few trades on them that cause system performance problems.

As NT_JP suggests and beyond, run a massive backtest which bogs down your system with many trades (calculate on bar close = false, simple crossover) and litter you chart with 1 draw type, then reset (go with reboot) and test with the other ...

and please report back any findings! thanks!

p.s. here is such a post with a strategy that will at some point max out everything.

http://www.ninjatrader.com/support/forum/showthread.php?t=52458&highlight=memory

[Chesterton]

Drawing line and drawing text annotation objecy within a chart image annotation library is different in some ways. Let's say drawing these two annotation objects within .net application, their annnoation parameters are different, besides the basic annotation properties, line including line startcap and endcap, while text involove multiple shadow setting.