We just recorded some data from the new 64 channel flexDrive implant with two of the 32 channel open ephys headstages. This recording was done while the mouse was free to run around in an acrylic box. We're using tetrodes made from 12 micron nichrome wire, gold plated to ~300 kΩ. No matter how much the mouse moves, we see relatively little artifacts, and the signal never seems to saturate. We can leave the low-cut at 1 Hz even for freely behaving mice, where we had to go to > 5 Hz with our previous analog recording systems.
For now, we're extracting the spikes after recording the 30 kHz raw data, but we're some small software updates away from doing the same kind of extraction on the fly, which will save some post processing time.