Data File Writing ... cRIO vs. Driver Station

[Ziaholic]

In support of refining of our constants/gains/limits related to driving, I want to output a file of the robot's performance parameters. Then, I'll take that file and dump it into excel to plot the actual speeds vs. commanded speeds vs. accelerations vs. gyro readings etc.

I'm wondering if I should perform a local file write on the cRIO, or if I should do it at the Driver Station. Most/all of the data that we're interested in will be reported on our custom dashboard anyways, so I'm leaning towards writing it to the local disk ... either at the driver station, or a development PC on the 'bot network.

I've never done a file write in Labview before and would appreciate any tips/tricks that y'all could provide. I'm not near my development PC at the moment and don't even know where to look in the VI pallettes to find the appropriate blocks for file formatting, naming, writing, etc.

Thanks.

[Greg McKaskle]

Examples are always a good place to start. In addition to the WPI robotics examples, you'll find lots of examples on dealing with text files, binary files, XML files, etc.

Ask again if you need help narrowing it down.

Greg McKaskle

[Ziaholic]

Good advice. Saturday is an all day session w/ the team and we'll look at the examples ...

... but looking at the bigger picture ... what are the pros & cons of writing the file onto the cRIO vs. writing it onto the driver station? I've got a few ideas, but still appreciate advice from the folks that have already done this a lot.

Keep in mind that my data is simple columns of speeds, encoder values, accel values, etc. Some of my assumptions may be bogus if dealing with images or other sorts of data.

If the data is already available at both locations, then it'd probably be more "convenient" to be able to get the file copied off of the DS vs the extra file transfer from the cRIO.

I don't think resolution is a factor. Right? Putting the file-write into the teleop.vi, getting called at 50 Hz would result in the same rate as a driver station write ... giving me equal results.

The "saveImageFile.vi" example uses a target path of "c:\images" ... does the cRIO filesystem emulate a DOS-like filespace? How much filespace is there available on the cRIO to write to anyway?

I'm off to the team meeting ... looking forward to our results, and then coming back here to discuss.

[jhersh]

Originally Posted by Ziaholic

If the data is already available at both locations, then it'd probably be more "convenient" to be able to get the file copied off of the DS vs the extra file transfer from the cRIO.

Given the fact that you are already sending it to the dashboard, I tend to agree that it's just easier to do it there.

Originally Posted by Ziaholic

I don't think resolution is a factor. Right? Putting the file-write into the teleop.vi, getting called at 50 Hz would result in the same rate as a driver station write ... giving me equal results.

True, though presumably you could write more often if you did it on the cRIO assuming that was beneficial to you.

Originally Posted by Ziaholic

The "saveImageFile.vi" example uses a target path of "c:\images" ... does the cRIO filesystem emulate a DOS-like filespace? How much filespace is there available on the cRIO to write to anyway?

c:\ is considered the root of the FTP server. The cRIO has 128MB disk on chip. My controller has about 100MB free after reimaging.

-Joe