XP 10.50 South Florida Coastline

[FlyBaby 659]

The new autogen looks great for cities like New York, Chicago and Philadelphia, However, the Miami coastline looks empty; there should be many buildings along the coast. Up through Ft. Lauderdale etc., there are very few buildings along the coast.

 

Has anyone else noticed this? I am running extreme objects.

[tonywob 4,769]

Yep, I've noticed this in quite a few places. I think the reason is that there was possibly no or limited height information coded into the autogen when these areas where originally cut into the scenery. I'm sure alpilotx can correct me here. Was this the same in 10.45, if it wasn't then it might be a bug?

[strider1 575]

Yep, I've noticed this in quite a few places. I think the reason is that there was possibly no or limited height information coded into the autogen when these areas where originally cut into the scenery. I'm sure alpilotx can correct me here. Was this the same in 10.45, if it wasn't then it might be a bug?

 

Isn't this height information in OSM ?

[Colonel X 1,430]

Isn't this height information in OSM ?

It should, yes. But it's insufficent in some places. Miami being a good example. Nobody updated the OSM here since 1983...

[Lynk 4]

edited: oops wrong topic.

[tonywob 4,769]

I have no idea where the height information is in the base scenery, I guess they've used more than OSM (e.g, Building density maps, etc).

[alpilotx 505]

Well, Ben Supnik DID hint at the data sources on his own blog about two months ago, when the new autogen updates came out:

http://developer.x-plane.com/2016/06/beta-3-in-a-few-days/

 

The tall building height data that is in the global scenery DSFs now came from a mix of FAA data and user collected data from a very long time ago. Therefore tall buildings are very likely to be missing or in the wrong place outside of the US. Our future plan (which alpilotx has already made progress on; I need to examine some of the files he sent me) is to use OSM for the height data too, which should be a huge improvement.

While we do plan to use OSM data to provide DSF height data, that is not going to happen soon, and is not going to happen for X-Plane 10.50.

 

So, no, the current default scenery (and neither the HD Mesh, nor UHD Mesh) use OSM based building height data (for the default mesh, which got its OSM data mid 2011, I doubt there would have been enough such info at all!!). And this building height info has to be in the DSF, otherwise the autogen would only guess. And indeed, everywhere else, the autogen is guessing, based on some population density (as far as available - this is a topic on it own ... because this type of data is very complex to get by) data mixed into theurban landclass info.

 

In the future - as Ben already hints - X-Plane will use building height data from OSM (which can be very very good for some cities .... and also has some idiotic errors, which I corrected a few some months ago ... but will always be introduced by OSM users for sure). But this can only come with a new, re-cut DSF (which will likely only happen with XP11) ... Experiments with OSM building height data are underway (at least I have already once sent Ben such an extract).

[Colonel X 1,430]

Google launch the Google Earth API already.

[tonywob 4,769]

Thanks for the info Andras. I think as you've mentioned you'll have some fun using OSM data for heights, e.g. Buildings 1000s of meters tall, etc :-). Lots of US cities have height information now from various government sources, New York, LA and Chigaco come to mind. 

[alpilotx 505]

Thanks for the info Andras. I think as you've mentioned you'll have some fun using OSM data for heights, e.g. Buildings 1000s of meters tall, etc :-). Lots of US cities have height information now from various government sources, New York, LA and Chigaco come to mind. 

Yes, New York for example is almost perfectly "height mapped" already in OSM. I am quite confident, that as soon as Ben found a way to integrate that data (and I don't think there are too big roadblocks on that way) it will definitely improves these cities.

 

And "bad data" in OSM -- god, I have seen a few bad "building heights" indeed. Most common was the error, where people inserted the "house no." into the "height" field .... yielding some 1000m buildings in some rural, low density areas. Or inserting the elevation in the height field (thus giving us some 1700m high riser "huts" in the Alps etc.). A few months ago I did a "half manual" sweep trough the OSM height data (by previously exporting and assessing that data in QGIS) and removed a lot of bad outliers. But I am sure, we will almost never be able to really fix that (humans do make errors) ... only some plausibility checks in the scenery generator might prevent the worst cases (but also might remove some rare, legitimate cases ... if you think about it, you will realize that there is quite likely no perfect solution to this).