+1 for an FL Studio midi test 
On my setup with FL I think my midi is stable, but I have not benchmarked it thoroughly.
I recorded a 16th note click track at 120bpm for a few minutes and spot checked the intervals in Edison and all the ones I measured were 125ms. So if there’s a better stress test I’m all ears.
Unless we can convince Mzero to do it, as he’ll do it better than any of us 
Best regards,
Gino
Same here, FL Studio’s midi clock is all over the place…
With the current version I’ve installed the Digi’s even drift after just a couple bars.
With a lot of tweaking I could get them to sync up but the jitter is so bad you can hear the phase shifts in every step.
But all that is with midi playback. Recording seems ok but I use a significant lower latency for that.
I would love to see some tests on this.
have you had a chance to try any other solutions to this issue?
Recording the MIDI to Live seems to be the tightest method I’ve found. There is still some small amount of jitter I can hear that do change the groove just a tad, but it’s much better than the results I was getting in REAPER. My current kind of messy method is monitoring with everything setup and “live” in REAPER (as that’s currently where I’m used to working) and then when I’m happy with the sequencing, recording that MIDI blind into Live and import it into REAPER. It’s a bit long-winded but it does work and improve things.
The other thing is that, as someone mentioned above, when you record audio you do get the exact same thing I’m hearing from the Digitakt sequences, so I’m slowly moving over to that where I can! Because you just can’t 1 to 1 replicate that lovely tight sound I’m getting from the Digitakt any other way!
Cheers
Ben