My old Classic is at the shop for its annual service, and I've got a Facelift AP2 loaner for a short while ... I don't normally use AP on the back roads as it isn't really "good enough" to be allowed to drive, even with supervision, but given the opportunity I had a go. Some observations, and let me know if there are some tests I should try, and then I will repeat when the AP1 is back (although i expect it will come back with latest Maps which will prevent some re-tests ...) Maps: Delivery Driver dropped off the loaner and took off in mine. I got in the Loaner some 12 hours later and the Satnav wanted to complete his journey. That's 200M further down the road to the epicentre of my postcode ... Offhand I can't think of any Satnav I have had which has considered a journey incomplete when it has been parked up for "some time" or, at the least, offers "Continue journey?". Going out of my gate and being told to turn right (a dead end) was the first surprise! I went half a dozen miles to pick up Father-in-Law for supper and tried out SatNav. It told me to take a shortcut down a road that (even though I've lived here all my life) I had no idea existed. Sure enough, there was the requested tiny rural lane which I duly turned into ... only to then see the No Through Road sign! I had assumed that Speed Limits were done (on AP1) at least in part by database data. I'm doubting that now, or if they are it is definitely NOT the same database. AP2 location of limits was "a bit wild". This is all rural stuff, and about 15 years ago the Liberals gained majority in most councils around here, and moved all the 20 MPH limit signs a couple of hundred meters further out into the countryside ... so maybe the database is really, really old ... but AP2 was anything up to half a mile wrong (and in the wrong direction [towards the village centre]) as to where the limit was. Even so, AP2 only started to slow down as I passed the point at which the (LIMIT) symbol changed on the dash. I can understand that sign-recognition may not be able to see very far ahead, but a GPS database ought to be slowing the car down before arriving at the limit (I would think) It made me realise how important speed limit sign recognition is. Even if not for AP I do use the dashboard display as a reminder to avoid getting a ticket (and of course it can also Chime if you set it to) AP2: Driving on the rural roads with AP2 was interesting. I don't think it is significantly different to my AP1, but it was certainly "no worse". I may try a video to make some actual comparisons later. These rural roads have a centre line, and either curb or verge. On AP1 I haven't found curb / verge recognition to be "good enough, but it is "usable". I'll retest that when my car is back, but AP2 did a reasonable job. I had to take over a couple of times a mile where I thought that the car was going to actually impact the verge - I might have taken over unnecessarily of course ... but the most obvious case was where there was, say, a gateway. The car drives closer to the verge than the centre-line, and at gateways "consumed" some of the extra slack, and was then confronted with a rapidly narrowing / approaching verge at the end/edge of the gateway. Took over on pretty much every one of those occurrences, similarly on a tight-ish right-hander bend where AP2 was going "a bit straight on" (but might have corrected at last-moment). I did all this with speed set at 40, rather than risking letting the car try it at 60 - and perhaps? not slowing down for any bend ... Its pretty flat around here, but there are a couple of significant humps, some straight, some on bends, and I will be interested to see how AP1 compares ("over crest" is a weakness of AP1). Impressed with AP2 in this regard, in general. On the main road near my house (white line on verge-side, as well as centre-line) I would say that AP2 was same as AP1. "A bit drunk at times" but I didn't have to take over [unlike the back roads]. I think it might actually have succeeded in driving the whole distance (a couple of miles) so I will specifically test "can I drive without intervention" on both AP1/AP2. On my narrow back-road (no centre line) there was one short stretch, on a bend, that had a centre line and allowed me to engage AP2 , and it stayed engaged (until I disengaged for approaching traffic). That was hilarious - AP2 must have considered it a "wide lane" (its wide enough for two cars to pass easily, but not two lorries) and it ping-ponged a lot. However, I didn't have to disengage ... but I was not prepared to see what happened when traffic approached!