Re: Osmand routing calculation is very slow for long distance for many years

[Helmut Jarausch]

If that were so simple, there would the following extremely fast overall
algorithm:
Select a point in the "middle" and solve the two problems separately.
Now continue to do that recursively.

Episteme PROMENEUR  schrieb am Fr., 27. März
2020, 19:30:

>
>
>
>>
>>
> After reading all posts I was a bit puzzled and disappointed then i had an
> idea.
>
> Osmand advises cutting in several routes of 250 km our itinerary, Right ?
>
> Then I asked for Osmand to include their advice in their algorithm.
>
> See
>
> https://github.com/osmandapp/Osmand/issues/8721
>
> --
> 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 osmand+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/osmand/fbb5ad68-8570-4f28-99ed-e4733ca464f9%40googlegroups.com
> 
> .
>

-- 
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/CA%2BCgGZS%3DE6KQiA9rZKEW7Cgtp4v%2BNT_NSbFVqLo%2BxLF3gfR0pg%40mail.gmail.com.

Re: Osmand routing calculation is very slow for long distance for many years

[Episteme PROMENEUR]

 

>
>
After reading all posts I was a bit puzzled and disappointed then i had an 
idea.

Osmand advises cutting in several routes of 250 km our itinerary, Right ?

Then I asked for Osmand to include their advice in their algorithm.

See

https://github.com/osmandapp/Osmand/issues/8721

-- 
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/fbb5ad68-8570-4f28-99ed-e4733ca464f9%40googlegroups.com.

Re: Osmand routing calculation is very slow for long distance for many years

[zanny]

OK - as a very minor correction.

Actually, the mainpage of the website links to google play, amazon and 
apple app store (at the moment all with -50%).
So, as a matter of fact, the app is first put on the market to be sold, but 
also on f droid (nightly builds are available as well, for the sake of 
completeness).
F-droid itself mentions:
This app has features you may not like
NonFreeAssets  
   
   - NonFreeNet  
   - NonFreeAdd  


Il giorno venerdì 27 marzo 2020 16:30:51 UTC+1, Xavier ha scritto:
>
> On Fri, Mar 27, 2020 at 08:12:37AM -0700, zanny wrote: 
> >Last point - since Osmand is a sold product, users have the right to 
> hardly 
> >complain. "Implement your own solution" is a good answer only for 
> >free-of-charge app 
>
> Minor correction.  OsmAnd is free of charge in general, the "sale" 
> aspect on the main website is in effect more like a donation to 
> support/encourage development than a sale. 
>
> See here: https://f-droid.org/en/packages/net.osmand.plus/ 
> 
>  
>
> A free of charge, and fully functional, OsmAnd+ is available there. 
>
>

-- 
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/c42b8fef-281a-4f8a-84b7-f31d9d149f9b%40googlegroups.com.

Re: Osmand routing calculation is very slow for long distance for many years

['Xavier' via OsmAnd]

On Fri, Mar 27, 2020 at 08:12:37AM -0700, zanny wrote:

Last point - since Osmand is a sold product, users have the right to hardly
complain. "Implement your own solution" is a good answer only for
free-of-charge app


Minor correction.  OsmAnd is free of charge in general, the "sale" 
aspect on the main website is in effect more like a donation to 
support/encourage development than a sale.


See here: https://f-droid.org/en/packages/net.osmand.plus/

A free of charge, and fully functional, OsmAnd+ is available there.

--
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/20200327153042.GM14585%40d820.dp100.com.

Re: Osmand routing calculation is very slow for long distance for many years

[Poutnik Fornntp]

P.S.:  OSMand algorithm obviously can be optimized wrt speed, I and Harry 
would agree on that, but it was chosen by purpose
and will never have the speed of contraction hierarchy ( or like ) 
algorithm  speed.


And contraction hierarchy definitely is not for BRouter, as it's high 
priority is flexibility.


Those, who bought OSMand knew, or should know, what they bought.

There is enough fast car navigations, use them, if they suite your needs 
better.


Not that I am just ordinary and rather just occasional OSMAnd user, not 
speaking for developers.


I use mostly non car navigation of LocusMap+BRouter combo,
for car using Waze, MapsFactor Navigator of OSMAnd.

Dne 27. března 2020 16:11:18 Poutnik Fornntp  napsal:

..
In domain of routing, it sacrifices routing flexibility and puts on high 
load the provider resources.
And  that is why it is good to have both high and low speed routing 
applications,

wher it is on the end user, what he/she prefers.

Those blindsided with high speed of contraction hierarchy do not see
the price to achieve that speed.


--
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/1711c9d0df8.2799.a291d67f9894f806060d35c996ca15e9%40gmail.com.

Re: Osmand routing calculation is very slow for long distance for many years

[zanny]

Hello gentlemen,

I am also a long time (and hence, satisfied) user of Osmand, since back in 
2011.
Speed is (and has always been) an issue (you can check old discussion as 
well) - I believe there full consensus on this.
I had never thought it is a matter of computational power: any 100$ mobile 
phone has more computational power than a  garmin satnav, which has indeed 
always been faster.
So the bottleneck must be the algorithm, its implementation, or its 
parameters.
In any case, as a matter of fact, the speed issue has never been tackled 
properly -if it had been, an improvement would have been fond. Harry above 
cited several  offline solutions (Magic Earth for example) which are much 
faster.
I think it is a free developers choice, in which direction they want the 
app to improve. I personally would have preferred more speed (or smaller 
app dimension, just to mention another) rather than other stuff (but this 
is personal taste).
Please also notice that such an slow speed implies other issues: for 
example, Osmand does just offers one solution (THE optimal solution, 
whatever it means in real life, outside of the pure math world), whereas 
most of the "competitors" offer two-three options.
Last point - since Osmand is a sold product, users have the right to hardly 
complain. "Implement your own solution" is a good answer only for 
free-of-charge app  

-- 
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/42dc44cb-bef7-49ca-ada7-56ab87c5c86e%40googlegroups.com.

Re: Osmand routing calculation is very slow for long distance for many years

[Poutnik Fornntp]

In many areas, not limited to routing, speed can be also a weakness. Not 
necessarily the speed itself, but things leading to high speed.


It is like a half-joke about software usability:
It can be free, easy, but not powerful.
It can be free, powerful, but not easy.
It can be powerful, easy, but not free.
( Sure there are many exceptions, but the point is in trade off, not 
limited to software. Focusing on some parts and neglecting other parts ).


The focus on high speed often means sacrifying things not leading to high 
speed or blocking high speed.


In domain of WWI/WWII battle cruisers, high speed for hunting smaller 
cruisers/destroyers and escaping mighty battleships had meant sacrifying 
either firing power, either protection.

( Remember Hood vs Bismarck, where Hood weak protection was its fate.)

In domain of athletic running,  speed of sprinters sacrifices their 
endurance for long runs due high volume and mass of muscles, high oxygen 
consumption and low aerobic capacity.


In domain of routing, it sacrifices routing flexibility and puts on high 
load the provider resources.
And  that is why it is good to have both high and low speed routing 
applications,

wher it is on the end user, what he/she prefers.

Those blindsided with high speed of contraction hierarchy do not see
the price to achieve that speed.

--
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/1711c8c1638.2799.a291d67f9894f806060d35c996ca15e9%40gmail.com.

Re: Padstow & river Camel flooded?

[Dmitry Prodchenko]

Hello! Thank you for the feedback.
Could you please provide screenshots and coordinates of the place so we can 
test it.

-- 
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/3d4579b0-b7e4-4932-8af4-719bd5e17936%40googlegroups.com.

Re: Osmand routing calculation is very slow for long distance for many years

[zanny]

Il giorno giovedì 26 marzo 2020 03:32:13 UTC+1, Poutnik ha scritto:
>
> Each group is superior to the other in what it is focused to. *Speed is 
> both strength and weakness.*
>

 Could you please elaborate on this? In which cases is speed a weakness?!?!


-- 
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 osmand+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/osmand/586b-1a92-4514-a027-01b4d2e70c1b%40googlegroups.com.