I've never seen a POI shift on any of my rifles (everything from pencil ARs in 223 to HB bolt guns in 300 NM). Biggest limitation is difficulty mounting on auto-loading pistols if they lack a rail.
I used to think the same. Then, one day I figured I'd test it out. 50 rounds, exactly the same. 6.5CM sub 0.5MOA rifle, sub 0.5MOA ammo, and me shooting that well. Don't ask me why but all three do happen with me from time to time. So, I shot 25 with and 25 without the Magnetospeed, 5 at a time, back and forth. POI consistently shifted about 0.75" down and right with my rifle at 300m with the Magnetospeed mounted. Groups didn't change size, they just shifted a bit; I'm assuming due to the change in barrel harmonics. I have tested it out more since and found that sometimes there is absolutely no change with some rifles and a similar shift in one direction or another with other rifles. Group sizes do not appear to be affected though. So, you can definitely use a Magnetospeed to generate info for load development. Once you have settled on a load, you _MAY_ have to correct your scope POA very slightly to match the POI once the Magnetospeed is removed.
I like the Magnetospeed for it's 'always on' aspect; you never have to arm and disarm it like you have to do with the Labradar.
It is a power miser. It uses a single 9V battery and that battery will last a really long time - as in many months depending on the amount of use.
I dislike mounting the Magnetospeed on a few of my rifles. It tends to want to slide around a bit as it just doesn't 'fit' well on them.
You HAVE to check to be sure that the entire Magnetospeed bayonette is about 4mm - 1/4" away from and not intruding into the projectile path, and is very close to parallel with the bore. It has to STAY there too.
As to the Labradar, it wants to be at a certain spot with certain rifles. Anything with a muzzle brake should have the muzzle either way out in front of or better, parallel to or slightly behind the Labradar (to the side of course).
It wants to be very well aligned to get consistent readings both nearby and downrange. The alignment notch is pitiful but there are more than a few methods of improving the accuracy of its alignment on the internet, ranging from simple straw-like sighting tubes to relatively expensive machined parts. I use a cheapo/freebie pen barrel with the ink cartridge removed and a couple of elastics.
If it detects a shot but 'misses' recording the velocity of a shot, it sits and waits for you to press a button to accept that before it will continue on. Grrr.
If you take too long to fire then it 'disarms' itself and if you don't notice, it doesn't record anything until you 'arm' the damn thing again.
If you use it with a .22LR, and any kind of subsonic ammo, it pretty much requires the microphone triggering accessory. I did almost every straight forward to convoluted set-up trick I could read/find out about and nothing worked consistently with my Labradar.
The Labradar EATS regular 'AA' batteries. A couple of hours and six (or is it eight?) of them are toast. Buy a 10000-20000mA rechargeable power pack for it right off the get-go. My initial 10K lasted me about 3 or 4 x 3-4 hour long sessions before requiring recharging.
