bgodette wrote:Because it's traffic dependent, you know like it's supposed to be? Additionally, that outside ring from I-25 to US-36 on the east side is all Toll. Without a LM option to disable tolls, you won't get a fastest route that doesn't use it because no one actually uses those toll roads except to go to the airport (iow they're empty about 80% of the time).
Yeah, it is kind of annoying to not have an option to disable tolls on the livemap. =/
But my point is made: 99% of the time, it's going to stick to the major highways like glue, and only if there is a SEVERE (subjective term) amount of traffic on the major highways will it consider other options.
Basically, if a non-major highway is the faster route with ZERO traffic on both that route or the major highway, it will STILL choose the major highway....because it's a major highway. It's only if the major highway is actually experiencing traffic that it doesn't use the major highway.
I mean, it's a long-distance issue (supposedly), but look at the 212 example given above and the I don't know what example given by the daknife guy. Their routes are actually faster, but they're ignored, because they aren't major highway. Can it be fixed by changing them to major highway? Sure. In fact based off FC in sketch's standard they probably are. But it just proves the point of how routes that aren't major highway get ignored or get harsh penalties for the routing system simply for not being major highway. And not EVERY route that is faster than taking major highways is going to be a primary arterial or a US route.
Can traffic cause you to go off a major highway? Yes. That's the whole point of Waze is that it uses that traffic data. But the AMOUNT of traffic it takes to get you off that major highway seems prohibitive, because that "better to stay on it in traffic rather than get off and take the back roads", not to mention detour prevention itself, gives a major penalty to not taking major highways.
bgodette wrote:Thortok2000 wrote:It still isn't perfect in my opinion because of the heavy penalties for going from a higher type to a lower type.
No such thing exists except for transiting Priv/PLot/ToD Restriction to a different type.
Experience says otherwise. In fact, so does the wiki! The entire reason we set at-grade connectors to have the same type as the roads they're connecting is because of penalties in changing type. Same with the instructions on what type to use for roundabouts depending on what roads they connect. If there wasn't a routing penalty for changing types, we wouldn't have to do that, would we?
From what I understand the penalty is always for going to a lower type, and there's no penalty for going to a higher type. This is why once you get on a major highway, it'll 'stick like glue' all the way to your destination (or the closest major highway that runs to it) unless the amount of traffic is extremely severe enough to warrant getting off of it. (And even then it'll probably get right back on again in the shortest amount of space possible to avoid the traffic it's detected.)
~
I'm flabbergasted that this extraordinarily obvious amount of 'stick like glue' that happens with major highways isn't instantly visible to everyone. I mean, even sketch says "that route was never considered until I made it a major highway"....meaning that the major highways that already existed were controlling the routing so much that even though that route was a good route, it was never considered!!!
So again, adding more major highways makes more routes considered but it makes it much, much harder for other routes to be considered instead.
That's why, with the exception of 'to fill in the gaps' for the long distance bug, I really liked AndyPoms' pure-FC based standard, particularly with principal arterials being minor highways, rather than sketch's standard. Routing was SO much better then. -sigh-
So the ONLY reason I see to go with sketch's standard is to fill in bugs on long-distance trips that error or ignore valid routes. But I can hope/wish/dream that the routing system just fixes that bug and we could go with Andy's standard instead, as it works so much better in every other way than that one issue. =/