Streets & Trips 2009 Routing Errors

I've been testing the trial version of Streets & Trips 2009 and am discovering that it is capable of making some gross errors when planning routes. For example, I plotted a route from Cave City, KY to Forsyth, GA with a stop at the Tennessee Valley Railroad Museum in Chattanooga, TN. The suggested route went north from the museum almost to Dunlap and then made a U-turn and came back for an additional 74 total miles. If I had Locust Grove, GA as the end instead of Forsyth, then the extraneous miles were not added. However, any point beyond Locust Grove caused the error. If I had Forsyth as the end and then added a stop along I-75 south of Chattanooga, the extraneous miles were not added. Has anyone else noticed these kinds of errors?

What steps would one need to take to reproduce the bad routing?

The route appears OK when I plot it in the free trial version of Streets and Trips 2009.

You picked Forsyth County instead of the City of Forsyth which is along I-75 south of Atlanta. I'd include a picture, but I don't know how to get it in here.

I figured it out. Notice the extra route north of Chattanooga.

Attached Thumbnails

 

Hi rmcdo,

You have indeed found a one-off routing glitch. Thanks for reporting it. I have passed this along to the team for investigation for next release.

~Larry

Larry, please have a look also at the routing issue reported by Alain. (I posted the screenshot, he described the issue on previous page). I only see reports of these types of routing problems infrequently, but they may be valuable for you guys.

I had planned a lengthy trip using Streets & Trips 2006. When I was able to install the 2009 trial version on my laptop (it won't run on my desktop), I copied these trips to the laptop and started looking at them to see how many restaurants, motels, and attractions had been updated. I found several other places where the 2009 data mis-routed my trip. The one I reported was the most severe I found, but there are others as well where the routing is illogical. These errors can usually be fixed by dragging the proposed route to the correct road, but that adds a lot of unneeded stops to the route.

Rather than dragging the route to the correct road - a better work-around would be to create an "AVOID AREA" over a portion of the illogical route. This will acheive the same result but without adding unneeded stops to the route.

Since software such as Streets & Trips no longer comes with detailed user manuals, there are probably lots of useful features that I have not stumbled across yet. I only recently discovered that I could add and label my own places of interest to make up for all the omissions in Microsoft's maps. I've often wondered if businesses are required to pay Microsoft to be included in Streets & Trips because there are so many missing motels, restaurants, gas stations, and tourist attractions.

This information is in the Help file, under Routing / Changing Your Route / Detouring around an area. Most people do not ever look at the Help file. If you do, then you can find a lot of these tricks that other people do not know about.

I believe the POIs come from Navteq too, just like the maps. You can report erroneous or missing POIs to Navteq just like you can with map errors.

The routing issue reported by Alain has already been taken in to our system and you can expect a fix for this in the next release

Let me again reiterate my suggestion to the S&T/Mappoint team to setup a public bug reporting database on Microsoft Connect. This product needs real-world testing, more than most other software.

I've noticed that, while viewing the map, the "Avoid Areas" function has a - quirk? idiosyncrasy? unexpected result - when you draw in a square area (Say, a large area from west Montana to east Minnesota, spanning both of the Dakotas north to south) and then pan the map east and west the drawn square appears to rotate sightly in place so as to remain perpendicular to the screen, and not the map (I suspect this has to do with how S&T interprets the projection of the map).

If you are expecting to avoid the areas covered by the square, you ought to place the area you want to avoid very close to the center of your screen, otherwise you may not be demarking the areas you appear to be.

[And that, believe it or not, is why this was posted in this particular thread. It took me a while to figure that out, and before I clued in, it appeared that the route S&T was sending me on went through areas I had wanted to avoid.]

Also marginally related to routing errors: Another quirk (Which has been carried through since at least S&T'05 in my experience) is that the expectation of rate of travel seems to be set much higher in the US than in Canada - resulting in S&T trying to route your trip through the states at every possible opportunity, at least when your preference is for "quickest route", anyway.
And while I've noticed that the estimates in Canada for travel time are spectacularly accurate (within 15 minutes for a 3-day trip, for example - Woo!), travelling through the US while trying to match the listed times for the various driving instructions would have been reckless in the extreme. Perhaps it could be done with the aid of a police motorcade, or in a low-flying plane...

Slightly further off topic, now, but...: A feature which would be nice to incorporate would be a warning for customs offices, which could behave just as ferry crossings do: Let you know they're there, and allow time for the stop. (Can you clear customs in less than half an hour these days? Anywhere? In either direction?)

In the overall scheme of things, S&T is very good at routing. The above bad routing examples may be a result of Microsoft's routing algorithm or Navteq's map data, or something else, which needs to be fixed, and which likely is fixable relatively easily. However, the main thing is that the underlying map data comes from a good source, Navteq. Read what happens when a popular product switches from Navteq to a less expensive data source: Google Maps: Navteq Out, Tele Atlas In

Let's hope Microsoft will not do the same fatal mistake with Streets & Trips one day.

Or like Garmin, who dropped the truly marvelous maps in their Metroguide Canada product done by DMTI Spatial in favor of Navteq's horrid Canadian coverage in City Navigator North America (which should really be called "City Navigator Continental United States with some coverage of Alaska and Canada and not including Hawaii or Mexico at all").

...ken...