"And when travelling 700-1000 km, an extra theoretical(!) 1-2 minutes at
the start or in the end (city to city via long motorways), is negligible in
case of the above mentioned or a traffic jam."

And don't forget the extra tank stop at a busy fuel station followed by a
visit to a busy toilet, especially with small children.
:)

Op wo 28 nov. 2018 om 17:23 schreef Harry van der Wolf <[email protected]>:

> This can be an extremely long answer, so I only give the most briefest
> explanation possible with 2 examples.
>
> OsmAnd, like 95% of the navigation apps, use the A* "pathfinding"
> algorithm. When using a heuristic coefficient of 1.0, the algorithm really
> uses every path in it's calculation. Theoretically that gives the 100%
> perfect route, but is of course very slow and uses a lot of memory: hence
> the problem that only shorter routes can be caclulated.
> Increasing the heuristic coefficient to values of 1.1 ~ 1.5, makes the
> algorithm skip paths that lead to "higher costs". That makes the algorithm
> much faster and use less memory, but in theory you can miss the 100%
> perfect route in case the fastest route contains a small, slow path
> (increasing cost), but possibly followed by very fast motorways.
> However, a value of 1.2 ~ 1.4 is really good for car navigation.
> The OsmAnd development team on the other hand has chosen to always
> calculate the 100% (theoretical) perfect path by using the 1.0 value.
> Like I mentioned: your perfect path can contain a red traffic light or a
> very busy intersection, and then a less perfect path can be (much) faster.
> And when travelling 700-1000 km, an extra theoretical(!) 1-2 minutes at
> the start or in the end (city to city via long motorways), is negligible in
> case of the above mentioned or a traffic jam.
>
> Please read this article at Wikipedia:
> https://en.wikipedia.org/wiki/A*_search_algorithm
> You can find more articles once you know to search for A* algorithm (or
> Dijkstra algorithm: the original)
>
> Harry
>
> Op wo 28 nov. 2018 om 13:56 schreef Stefan Monnier <
> [email protected]>:
>
>> > I also use a heuristicCoefficient="1.4" since several years following
>> the
>> > example of Harry van der Wolf, which makes the calculation of routing
>> much
>> > faster. I like it!
>>
>> BTW, what does this magic button do?
>>
>>
>>         Stefan
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Osmand" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> For more options, visit https://groups.google.com/d/optout.
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Osmand" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to