Hello,
I am looking for a way to reduce the CPU usage of the pc by writing an optimized code of an indicator.
I know that PRT runs an indicator for every tick arriving on the instrument and, within this, it executes the indicator for every bar on the worktable.
That is, if I have 1000 bars in the work plan, for every incoming tick PRT executes the indicator 1000 times.
My indicatorit already uses calculateonlastbars = 20 so for every tick it run 20 times, one for last 20 bars, and draws objects with DRAW commands only if islastbarupdate.
But the use of CPU is very high due to complex calculation.
because object drawing are very static (starting from a certain date in the past), it would be sufficient to activate drawing not for each tick,
but:
– at each new opening bar or every 10 seconds
– in any case when the market is closed (because no new tick arriving to force draw).
in other words, there is a way to recognize if we are at a closed market period, and therefore if ticks do not arrive?
I can always draw, or I am at an open market and therefore I can count on the fact that the ticks arrive and then use
currentsecond to make the indicator work every 10 seconds?
Or its’ possible to know the my pc Time and date at indicator execution? I found only reference to closing bar time/date/timestamp.
thank you.
.
