Hello alle,
if you build a target program for the real time target, it will typically be installed on the target and you will configure it to start up automatically after boot time. In this case, you can create a tcp listener waiting for a host client to contact it and communicate any commands. Based on the commands, you can set up a case structure doing acquisition, calibration, etc., as needed. Typically you use a timeout for the listener in case of something goes wrong, but you should not do so in this time, as the RT engine really serves as an acquisition server. Be careful, anyway, as debugging, ... this beast might be quite difficult. In case, you might set up a ftp connection to the RT engine and change the ni-rt.ini file not to start u
p the program automatically.
The other alternative (which might be much more well suited for you) is to build the RT program with the windows target set. Then, you could start the program on the host with pointing to the RT target at start-up of the host-LV-program. Setting the "close host after RT startup" at the build options and calling the host program with the -target - option enables automatic download and start of the RT program from the host side. This can be done through the host client LV program through an os exe call at its startup. The RT server program will again have a tcp listen vi and a case structure as above, now with a suitable timeout (as it is started by the host). For this way of application, you should not have a startup program on the RT engine. This second way in addition has the benefit that you can update the RT server program by delivering a new host file only, so no need to download any files on the RT system at update time and no need to have an LV dev
elopment system for this.
Hope this helps. If you have additional questions, do not hesitate to ask. Unfortunately, I will not be available in this forum until mid of July.
Regards,
acoustics