What was done

Tried running MagGyroDemo with Sprite development kit, looked at output

What was expected

When running with MagnetometerDemo magnetometer output varies with rotation as expected

When running GyroDemo keeping Sprite motionless gyroscope output is stable with bias within +/-40 as expected

Would expect same/similar outputs for magnetometer and gyroscope when running MagGyroDemo

What actually happens

If MagDemo is run first, gryroscope data will have bias much larger, around 200

If GyroDemo is run first, gyroscope data will have low bias, but be very noisy, with spikes reaching 20000(??)

Also once had it where magnetometer data was frozen and did not respond to movement. Have also had it where it worked fine(!)

Ideas

Could be some registers are not getting set to a known state? Leaving it for some time seems to make it work better.

Or maybe code simply is beyond limitations of processor. I have tried increasing delay, but seems to have no effect.

Will be experimenting more

I find the final code that was used Sprite where I can find ?

Many sketches reference SpriteRadio.h but I can't seem to find it anywhere.

Thanks a ton for putting your VMWare image. I was able to get it working. However, in gnuradio-companion I see only 2 sources (osmocom and RTL-SDR) and 1 sink (osmocom). Further, from Python interpreter, when I tried running

>>>from gnuradio import gr
>>>gr.null_source(1)

I get missing module error. I was wondering if you or anyone else had faced these issues before.

With the launch next weekend, I thought it would be a good time to check out that everything works. However, the top level ReadMe in the repository mentions GNUradio progs to decode the sprite date, there appears to be no such directory/software in your repository. Is it in another branch/fork?

Thanks,

Matt

From the wiki page on running GNURadio.