Plausible reality - OSM hotspots

[JonRD463 146]

Glen, I don't know if it's your work or what, but I've been scanning Manitoba in OSM and have found a lot of small villages that are full of OSM buildings. There's also a lot of forest definition there. Just as a couple of examples, check out Brandon and Dauphin. There are more. I'm still looking. Gonna download the Manitoba pbf file and run it through.

[BimmerCop 332]

I recently experimented with OSM2XP and I really love it. Especially the ability to allow default XP autogen to be displayed to fill in the voids. What I really want to do is LEARN how to edit OSM data myself via the website. I tool a quick glance but it all looks overwhelming.

[JonRD463 146]

And then there's Whistler, BC...

 

I thought I'd change things up here and use a video this time. While there are buildings defined here, the main focus is on the excellent job of defining forested areas, particularly on the slopes where they match the terrain texture perfectly. The area isn't complete by any means, but what's there is done very, very well.

 

Oh, and not one word about my flying! I flew it from that POV so clipping a tree here, or skipping across the surface of a lake there was bound to happen. B)

 

[Pascal_LSGC 298]

Or with this fantastic views:

All features edited in OpenStreetMap over the past 90 days.

All features edited in OpenStreetMap over the past 7 days.

 

This is absolutely impressive to say the least !

 

While there are buildings defined here' date=' the main focus is on the excellent job of defining forested areas, particularly on the slopes where they match the terrain texture perfectly. The area isn't complete by any means, but what's there is done very, very well.[/quote']

Good finding ! pretty impressive as well !

 

:smile:

 

Have a look at that site: http://simheaven.de/, he's created a lot of orthophoto scenery and the corresponding ready to use OSM for many cities in the world. Definitely the way to go, and very handy for people who don't want to make sceneries themselves.

I downloaded OSM and ortho for Praha, it's definitely a hotspot !

 

P.S: Vienna and Bratislava also have all buildings in OSM.

[Pascal_LSGC 298]

Andras, it happens you've just pointed us to some treasury maps !

This map at the ITO map site highlights the places with buildings, and buildings with height information !!!

 

I found a big zone between Trieste and Pordenone in Italy where every single tiny building has height information !!!

 

Back to jewel places finding...

 

Pascal

[Pascal_LSGC 298]

Hotspots:

Bakersfield, CA

Hongkong (not all buildings, but many with height information).

[Muskoka 104]

Glen, I don't know if it's your work or what, but I've been scanning Manitoba in OSM and have found a lot of small villages that are full of OSM buildings. There's also a lot of forest definition there. Just as a couple of examples, check out Brandon and Dauphin. There are more. I'm still looking. Gonna download the Manitoba pbf file and run it through.

 

Jon, never done any editing myself in Manitoba. The OSM data for the most part is scarce in Canada. You can pick and choose small area's of coverage, we need mass coverage. Like I said before, Ben and I found some sites with "additional" building data, but almost all of the data would have to be edited for it to be useful. Full town / city blocks were one big building, and it looked "terrible" once converted in Xplane. The idea is great, it's implementation is going to take a ton of work and a lot of man hours. I rather do something else with my time and pay some 3rd party to offer me pre-made scenery, say for the whole of Canada for $50. My time is worth close to $50 and hour to me, I'm not going to spend hundreds more hours editing in OSM, just not going to happen.

 

We need to find a way to get outside data into OSM, then into Xplane in a manageable fashion. Like I said before, this data has to be stored somewhere, how do we get at it, and import it into OSM, or better still use it directly. To expect the individual end user of Xplane to cover the world with OSM edits is just ridiculous, and a waste of valuable time.

 

Municipal Planning Departments should have this information. This is the kind of information we need access to. Someone else has already paid "someone" to have the work done. Nobody is paying me to do all this work, and I won't. How do we get access to this "already done" information? Good luck, and I mean that sincerely.

 

Glen

[Pascal_LSGC 298]

Hotspots:

 

Reykjavik. Any Aerosoft's Keflavic owner ?

Birmingham, and all central England well covered

[greggerm 312]

What I really want to do is LEARN how to edit OSM data myself via the website. I tool a quick glance but it all looks overwhelming.

 

If you get remotely serious about OSM editing, I'd recommend using the JOSM editor. Relatively straightforward, many more features than the website, and once you get the basics, quicker to use. There's a learning curve, but it's worth it in the end.

 

I added a zillion lakes, rivers, and ponds in the New England area using JOSM back in late 2011/early 2012 when we started getting reports that the XP10 scenery data was so incomplete in the United States. Now, we just need them to re-cut the files to bring all that water in!

 

-Greg

[Muskoka 104]

Not so hotspots:

 

I'll give some examples for clarity, and this is just the tip of the iceberg. These are cities and small towns in Ontario within 2 hours of me, that basically have no OSM data available at all. There is a smattering here and there of some industrial areas, absolutely nothing residential. This is where the concept falls far too short. Where can we get the data to fill these areas without spending hundreds if not more hours manually editing OSM? As nice as it may be in some areas (some US, some Europe), it's non existent in others. To expect the end user to fill these gaps is unreasonable to say the least. When the data's good, it looks great, when it's bad, it looks really bad. Your flying over empty roads / subdivions, and it does not look good at all.

 

Again, as with all my criticisms of Xplane, I'm trying to find some positive solutions, and answers to glaring problems. The end user is not the answer, it has to come from another source, or the concept will fail. A few may care enough to do something, the majority will not. Populating the world with OSM data is not a task for the few. Sure, some areas have great coverage, good for you, most do not.

 

Where's the data going to come from?

 

Glen

 

 

 

 

 

 

[Muskoka 104]

And then we have to question "just how plausible the Xplane world is", regardless of OSM coverage?

 

Here's a screen shot (Google Earth / XPX 64bit) of part of Gravenhurst Ontario. You can see in the circled areas that the "Plausible" world fails miserably, it doesn't look anything like it should. I know the area very well, and it's just one example of "thousands". Now, it would probably look ok with accurate OSM data, but were is that going to come from? As it stands now, Xplane is sticking buildings where they don't belong, it's far too random. I suppose that could be corrected with a "Custom Scenery", but are you going to have Custom Sceneries for a large portion of the world? The concept is "great", it's implementation needs a lot of work. How are we / the developers going to correct it?

 

Notice the frame rates though (251.4), amazing (No HDR of course ).

 

Glen

 

 

 

 

 

Jon, your shots look great, and I have stated so, so don't get me wrong here. For every nice shot, anyone can come with 100's+ that don't look nice, or plausible. I'm looking for answers to improve the simulation for everyone, just not a few small areas where it does look nice. The "nice" needs to be world wide, I hope you understand.

[alpilotx 505]

The "nice" needs to be world wide, I hope you understand.

You are absolutely correct here, and no one would dispute this "dream". But you have to think about, what needs to be there to achieve that? Yes, of course, fantastic data. Because, where you have fantastic data. As soon as you have fantastic data, you don't need to come up anymore with "plausible" autogen. Autogen is just trying to "invent" data, where there is no data ... and of course NEVER can do that in a realistic way just in a plausible(!) way (and sometimes I get the feeling, that you, and many others have a very "different" understanding of what "realistiv-vs-plausible" means ... because neither of them is the other ) . AND, autogen has another very important "advantage" .... its kind of a lossy compression algorithm. It allows to represent the world with much less data ("just" a few DVDs instead of a many TBytes) in a plausible(!) but not realistic way (thats why its a "lossy" compression). .... This is also an important aspect, when someone wants the entire world "nice" ... having it that nice (like some of the above screenshots), with very good OSM based scenery and maybe even photo textures would mean a few 100 TBytes of data (or more) ... You see, reality needs quite some disc space to be represented in a ... uhm. ... realistic way :lol: . As soon as that becomes a problem (which it is today ... but maybe won't be in 10 years :rolleyes: ) you need to compromise somewhere (either don't want the entire planet, or accept some plausible, "lossy compression" style approach) ....

 

But back to the problem in the first place: you still need the "fantastic data" ... And let me tell you: its not existing! Well, at least definitely not for the entire world (which we all would like to have nice ... better today than tomorrow). There are some regions, some cities, some countries, which do already have fantastic coverage .... but this is still a far cry from the "entire world". And I am not even talking about OSM (which by the way in my eyes is still our best bet in the long - maybe very long - run). I have the "hobby" of always scouting for great geo data ... I liked this to do even before I started "cooperating" with Laminar ... and thus I have already a seen a lot of whats out there, what you can have, what quality it has. And thus I repeat: the fantastic data for the planet (or at least large parts of it), does not exist. And often, where it exists, its prohibitively expensive ... with the few notable exceptions like the - close to - fantastic New Zealand data which I used for the NZ Pro scenery. And even that, close to fantastic data has a lot of shortcomings! I had to work many weeks to make it halfways usable for X-Plane ... remember, this was with "almost fantastic data" .... now don't even imagine, how many more headaches you would have with other data out there ... Because non of them is perfect, all of them have strengths and weaknesses ... And the weaknesses are the parts which can ruin your day, or entire projects.

 

So, finally I say: even if we want that entire nice planet, its not falling from the sky, nor can you buy it (not even with lots of money ... ok, maybe with very very very very lots of money ... but I even doubt that). Thats why we have to accept compromises today, be happy what we have, and work on having more in the future .... .... Where OSM comes in the big picture. Thats one of those single, great sources (projects), which has the potential to bring us in a free (democratic) manner the "close to fantastic" data (as it will never be perfect too) ... one day, in a distant future. Even if it will take many many years, but it has the potential. Just look a what Wikipedia has become over the years ... its essentially the same as OSM is. A great, open project, which lives from the enthusiasm of a big community (we the people).

And never forget, that OSM is about much much much more than X-Plane ... we are only a tiny fraction of the users (the community) who are interested in it. So, you don't need to only count on a few enthusiastic X-Plane users to improve it, but you can expect many others to work on it (people, who don't even know that X-Plane exists) ... Look at Europe (for example Germany) .... Most of that data came from the community and see how detailed it is (still, even this data could be improved to better fit the needs of a flight sim !). So,IT IS POSSIBLE ... it just doesn't falls from sky B)!

 

(and to see, how nicely OSM grows, check my links in comment #15 of this topic).

[gheeD 62]

I also think that the key to getting more data into OSM is the opensource -part of it. If we can somehow make the public more informed about openstreetmap and get more people involved in creating the data, it will become exponentially faster. like Andras said there are quite a few other communities that take benefit of openstreetmaps data and the data is not only significant for simmers.

 

Besides... Basic building data can be done REALLY fast, just draw boxes... Or data like forests etc. Making a contribution for an hour or two might not seem quite significant, but when you have a lot of people doing the same, the end result can become quite magnificent.

[Pascal_LSGC 298]

Jon, your shots look great, and I have stated so, so don't get me wrong here. For every nice shot, anyone can come with 100's+ that don't look nice, or plausible. I'm looking for answers to improve the simulation for everyone, just not a few small areas where it does look nice. The "nice" needs to be world wide, I hope you understand.

Muskoka, are you even slightly aware that you just hijacked a potentially useful thread ?

 

Oh, you are not the only one who does that. It seems that every interesting thread around is misused by someone who needs to expose again his point of view about such and such X-plane shortcoming to the world. Except maybe the thread about aircraft roll, which stayed on topic.

 

I'm all for critics, and we must highlight the problems, and even say thousands time what's wrong in X-Plane. But please, not in every single thread. Because it draws the whole X-Plane forum down to dramatically low level of usefulness.

 

Use the existing threads for that ! Create new ones ! Many new ones !

 

For a remainder, this is what the OP had in mind:

 

What I'd like for this thread to be is a sort of repository for good OSM places to fly, particularly those areas that benefit from both OSM and orthophotos that work hand in hand.

 

This an excellent idea. Please respect what he wanted to do.

 

Pascal

[BimmerCop 332]

Muskoka, are you even slightly aware that you just hijacked a potentially useful thread ?

 

Oh, you are not the only one who does that. It seems that every interesting thread around is misused by someone who needs to expose again his point of view about such and such X-plane shortcoming to the world. Except maybe the thread about aircraft roll, which stayed on topic.

 

I'm all for critics, and we must highlight the problems, and even say thousands time what's wrong in X-Plane. But please, not in every single thread. Because it draws the whole X-Plane forum down to dramatically low level of usefulness.

 

Use the existing threads for that ! Create new ones ! Many new ones !

 

For a remainder, this is what the OP had in mind:

 

 

 

This an excellent idea. Please respect what he wanted to do.

 

Pascal

 

Amen! Couldn't have said it better. Back on topic PLEASE.

 

Sent from my Galaxy Nexus using Tapatalk 2