NI Measurement Studio Idea Exchange

When the installer builder is called from the system shell (cmd.exe) using a command like the following:

InstallerBuilder "D:\NI Installer Builder\INSTALLERPROJECT.iip" -Build -Log "D:\NI Installer Builder\niib.log"

It spawns a background process and the application returns with a successful run BEFORE the installer is created. As the build is started in a background and a detached process; any build automation frameworks such as bamboo, jenkins, chef, etc. cannot monitor it's creation. This results in a failed operation a breaks continuous integration systems. 

As a work around I have built a light python solution that monitors for a que event; in this case a
file either being populated or updated. I am using this application to monitor the log file as a que.
The python file should be attached to this submission.

Basically:
1. InstallerBuilder is ran – begins running building process in the background
and returns successful process run before installer is constructed.

2. IndirectProcessMonitor.py is ran (max timeout and file set):
python IndirectProcessMonitor.py 300 “D:\NI_installerbuilder\niib.log”
NOTE: Probably could set it to your setup file as well

3. IndirectProcessMonitor.py returns when timeout is exceeded or que file is
created or updated.

If the NI installer builder is to be able to be dropped into any automation
frameworks this has to be fixed on NI’s end. If this bug/support is scheduled
for development I’d be interested to know as it will/should break our
workaround.

Plotting a big array of sample uses quite some time in the current version. I think this comes from Measurement Studio drawing a line explicitly between each two samples. However, in case of millions of samples, many of those lines will cover only one single pixel.

Instead of drawing all lines, it would be possible to sample the data at a rate where each sample matches just one pixel in the input axis. In order to not loose peaks, the sampling needs to do a min/max calculation and draw two lines for the minima and maxima. A shade could be used between the two curves. When zooming, the sampling needs to be repeated.

This feature would not only save much performance for big data (I think), it can also help to improve clarity of the graph, p.e. in case of noisy signals.

While it's great that Measurement Studio offers all capabilities for us to implement this functionality ourselves, such a feature could be of interest for many users.

Of course, this idea is not my intellectual property. It's a feature already implemented in other products.