event data node time

[dennis_the_menace]

I'm using an event structure with a DAQmx event.

Inside the event structure I read the "time" of the event.

Does anyone know if this time is

a) the time the event was detected in the DAQ board

b) the time the event information was transferred to LabVIEW ?

Thank you

[johnsold]

It is the time the event structure executes.

Lynn

[dennis_the_menace]

Thanks for info.

So there is a delay between the event in the hardware, and the time the event structure executes.

Does anyone know the typical length of this delay ? I guess it is variable but might have a mean and an upper limit.

[johnsold]

Because LV is data flow driven, the delay depends on the overall program.  The event will be queued into the event structure very quickly (probably microseconds or less), but when the event case executes depends completely on how your program is written.  Without seeing the code I could not even guess.  I could write very simple code that would take a week to respond to the event and other code which would respond immediately.

Lynn

[dennis_the_menace]

Hi,

Thank you for your reply.........my dataflow does the following:

a) launches the DLL call that causes the external event - there is then an external delay of a few ms before the event happens (the DLL returns control to LabVIEW without waiting for the hardware to respond)

b) waits for the external event, using the event structure in a while loop to detect only this event

c) exits the event structure/while loop once the external event has been detected as a DAQmx event, and the event structure has executed

so there is not additional delay in my program, I designed it to avoid that

Hence, from what you say, there are a few microseconds while the event gets queued into the event structure, but before that surely there is a delay while the information gets transferred from the DAQ hardware into LabVIEW ?

So does anyone know the length of this delay ?

Thanks for support, I appreciate it

[JarodX]

Hi,

Speaking about the time node in the event structure, to say that 'time node' equals the time at the moment the thread focuses on the evt structure is not completly true.

Before my recent experience, I do thought so, but it's only true for static events and not for dynamic events.

If a dynamic event is registered, the time node contains the time of the event triggering and not the time of the event structure execution (I could provide a vi which shows that behavior).

I assume that in a dynamic event, Labview knows the time of the dynamic event triggering and transmits it to the event structure whereas in hardware event, I suppose that Labview focuses on the event structure at the time it can do it (not busy in the same event structure). So the 'time node' matches the event structure execution and the Labview event detection, but not the time the event occured physically.

I don't know if this feature is documented, but if it's not the case, it could lead to critical issues.

Regards.

[dennis_the_menace]

thanks, that is very helpful,

I can use that info this week on another project....