On 2/5/2026 at 9:33, jcomm said:

@Bjoern, escrito magistralmente !!!

Concedí la versión Pro de Auto-ortho y la usé durante todo un año, hasta marzo de 2026. A partir de ahí estoy usando Xp12 sencillo con SimHeaven, y esperando tranquilamente lo que X-Plane X traerá su nueva versión.

No espero que sea tan preciso como MSFS 2024, que aunque es todo lo que es una incuraduría en la forma en que modela el vuelo, tiene una tremenda representación del mundo real, hasta el punto de que lo uso para revisar los vuelos que hago en mi IRL LS1-d.

 

https://github.com/ProgrammingDinosaur/autoortho4xplane

20 hours ago, Bjoern said:

And I will not budge from my point: Orthos are stone age tech and Laminar would be massively dumb to throw millions out of the window to license global imagery

A few weeks ago I don't think I would have agreed with you.  Autogen in XP12 is good, but unless you are flying very low there are no shadows, and that leads the autogen not looking connected to the terrain.  Ortho shadows may be fixed, but all the autogen gain the shadows making them far more believable and part of the terrain.

However, after the release of SimHeaven Pro demo, I am more convinced by the idea of no ortho.  The new farmland polygons have gone a long way to close the gap to ortho in imho.  I am eager to see what the next gen scenery has in store for us, but whatever it is, PLEASE give autogen shadows above a few hundred feet.

1 hour ago, MrBitstFlyer said:

A few weeks ago I don't think I would have agreed with you.  Autogen in XP12 is good, but unless you are flying very low there are no shadows, and that leads the autogen not looking connected to the terrain.  Ortho shadows may be fixed, but all the autogen gain the shadows making them far more believable and part of the terrain.

However, after the release of SimHeaven Pro demo, I am more convinced by the idea of no ortho.  The new farmland polygons have gone a long way to close the gap to ortho in imho.  I am eager to see what the next gen scenery has in store for us, but whatever it is, PLEASE give autogen shadows above a few hundred feet.

You really don't want too many autogen objects throwing shadows. Because shadow rendering is already massively costly on the GPU, having thousands of autogen objects in the scene is an absolute guarantee for people coming a-whining about low FPS.

On 5/2/2026 at 11:13 AM, Bjoern said:

Orthos are stone age tech and Laminar would be massively dumb to throw millions out of the window to license global imagery

Saying orthos are "stone‑age tech" without proper context is simply an attempt to downplay their importance and relevance today. Orthos aren’t mandatory, but they massively improve immersion and are more relevant now than ever before.

By your logic, cars, planes, and even transistors (invented in 1947) would also fall into the “stone‑age tech” category. This ignores the fact that the technology behind orthophotos has evolved — and continues to evolve — over the years. Orthos, just like the items mentioned above, have existed for decades, but the technology behind them has advanced dramatically. Ortho images can be color‑corrected, have clouds removed, and have shadows removed using modern techniques, including AI.

Quote

U‑Net–based shadow‑removal networks are a class of deep‑learning models that take an image with shadows and output the same image without shadows by learning to reconstruct the underlying surface. They’re one of the most widely used architectures in remote sensing, GIS cleanup, and orthoimagery enhancement.

 

18 hours ago, brinx said:

Saying orthos are "stone‑age tech" without proper context is simply an attempt to downplay their importance and relevance today. Orthos aren’t mandatory, but they massively improve immersion and are more relevant now than ever before.

By your logic, cars, planes, and even transistors (invented in 1947) would also fall into the “stone‑age tech” category. This ignores the fact that the technology behind orthophotos has evolved — and continues to evolve — over the years. Orthos, just like the items mentioned above, have existed for decades, but the technology behind them has advanced dramatically. Ortho images can be color‑corrected, have clouds removed, and have shadows removed using modern techniques, including AI.

Orthos are stone age because they're highly inefficient. The underlying global imagery takes up terabytes of disk space and requires building and vegetation footprints of unprecedented accuracy plus matching objects in order to be truly convincing - which would take up even more disk space. We need not discuss the present options being better than what XP12 offers by default, but simply throwing disk storage at an issue that should be solved a lot more efficiently is not the way of the future. Not in light of stagnant computer hardware technology, not in light of massive memory shortages of all kinds and not in light of ever more expensive supply chains.

And if orthos really are the bees knees, then why is there only MSFS with its billion dollar backing and AI and imagery access making use of it, why is there no smart startup offering licensed, processed imagery and matching overlay scenery (extracted from footprints by AI) for download or streaming for X-Plane and why is all that we have in terms ortho tile scenery generation and ortho imagery streaming for XP12 merely leeching from public sources?

Correct, because running the storage and infrastructure required for orthos is stupidly expensive precisely because of them being highly inefficient.

6 hours ago, Bjoern said:

Correct, because running the storage and infrastructure required for orthos is stupidly expensive precisely because of them being highly inefficient.

Your argument falls flat. It’s not expensive because it’s inefficient — it’s expensive because you need to launch and maintain a constellation of satellites. Only a limited number of companies do this, and Microsoft and Google must license imagery from them, which is not cheap. That’s why the cost is high. But for those companies, the process itself is not inefficient.

The real question is: inefficient for whom? If you mean someone sitting at home ripping or downloading terabytes of orthos, processing them, and trying to clean up shadows, clouds, and color issues — then yes, that’s inefficient. If this is what you mean, then you should make that clear. But even in that scenario, the inefficiency isn’t due to “stone age tech.” It is inefficient because you’re ripping tiles from map providers in a way they actively try to prevent. They want to make unauthorized tile‑ripping inefficient, which is why they ban IPs that make too many requests. This is why streaming in MSFS is such a liked feature.

For companies like Apple, Google, and Microsoft, handling orthophotos is not inefficient, and storage is not a problem for them. 2 Petabytes of high resolution orthophotos is nothing when you are talking about their storage capacities of exabytes.

8 hours ago, Bjoern said:

Orthos are stone age because they're highly inefficient.

Just completely wrong, see above.

All I can say, is we have become to reliant on the internet, nice when it works, but as real weather shows, can be a real pain.

3 hours ago, mjrhealth said:

is we have become to reliant on the internet,

The internet is not the problem, how it’s being used, abused, and sold is a problem … but without it, how would we get live weather?  How would we get live traffic to route AI aircraft?  How would we get navigation data updates?

I’m of sufficient age to remember LAN parties for Falcon 3 where we all met at a location (usually rented a hall in various event facilities), spend about 1 day with BNC getting a whopping 4-6 PC/Players connected and “hope”.  Fortunately there was enough alcohol to make all the crashes mid sortie to not matter as much.

Edited May 5May 5 by SayAgain

I will throw my two cents in. I used to run Map Enhancement and SimHeaven and Global Forests. I remember it being punishing on performance (for my laptop) so I don’t do it. I simply am running default XP12 with XASnow plugin and that’s good enough for me until LR provides a different scenery. If I want Orthos, I can always power up MSFS. The beauty of XP is that right now it runs independent of constant internet connection (besides live weather)… 

I believe LR has some ideas of how they want to do it and I don’t think it will be by streaming orthos. I will let them cook their magic and use XP12 for its wonderful flight model and environmental depiction. And when I am itching to explore places and see what they look like in real life, I can always boot up MSFS24 for that.

That’s where I stand right now.

12 hours ago, brinx said:

This is why streaming in MSFS is such a liked feature.

For companies like Apple, Google, and Microsoft, handling orthophotos is not inefficient, and storage is not a problem for them. 2 Petabytes of high resolution orthophotos is nothing when you are talking about their storage capacities of exabytes.

But for companies the size and funding of Laminar, it would be massively inefficient, so it's completely foolish to expect that future X-Plane releases feature anything close to free, streamed orthos like MSFS.

And save yourself the typing for a reply. I'm dug in on this, barbed wire, mine field, artillery kill zone, drone nets and everything else.

Edited May 5May 5 by Bjoern

38 minutes ago, Bjoern said:

But for companies the size and funding of Laminar, it would be massively inefficient, so it's completely foolish to expect that future X-Plane releases feature anything close to free, streamed orthos like MSFS.

And save yourself the typing for a reply. I'm dug in on this, barbed wire, mine field, artillery kill zone, drone nets and everything else.

You're simply confusing "inefficient" and "impossible". Orthos have nothing to do with "inefficient" or "stone age"  technology, but don't worry, all of us here understood you simply hate ortho and you're just trying to find any non-sense to discredit that kind of scenery, and there's no need to argue about that🙂

That being said, you are right when you tell that X-Plane will never include that kind of scenery per default though. It is impossible for them to acquire the necessary data and host it or ship it at once, given the price of the sim. The only "legal solution" is third party scenery products such as OrbX TrueEarth, on top of which the XPlane graphic engine would be able to alter ground colors and add snow coverage to simulate seasons. Too bad Orbx gave up on these products, though 😕

We might imagine some monthly-fee based product that could bring ortho streaming in the sim, a bit like auto-ortho but fully legal and coming with autogen and seasons. Just like the third party AI ATC products for example. However, I'm not sure many people would be interested in such a price model.

 

9 hours ago, Bjoern said:

But for companies the size and funding of Laminar, it would be massively inefficient, so it's completely foolish to expect that future X-Plane releases feature anything close to free, streamed orthos like MSFS.

It would be massively expensive, not massively inefficient. There’s a big difference. I’m not expecting X‑Plane to do any streamed ortho, nor did I advocate for that. I’m only pointing out why orthophotos are not the inefficient “stone‑age tech” you believe them to be.

I just run XP12 on CCD0 and map enhancement on CCD1. Works a dream. Zero stutters or failed tiles. Looks really good too, combined with either True Earthh GB objects or Sim Heaven. 

22 hours ago, Daube said:

You're simply confusing "inefficient" and "impossible". Orthos have nothing to do with "inefficient" or "stone age"  technology, but don't worry, all of us here understood you simply hate ortho and you're just trying to find any non-sense to discredit that kind of scenery, and there's no need to argue about that🙂

On a SSD, I have 1.786 "zOrtho4XP_" prefixed folders worth of ZL 14 imagery, which take up around 566 GB. That's 12 % or so of the planet's landmass, coasts and islands. Let's say the target for global land, coastal and island coverage is around 60%. My ortho collection thus covers 1/5 th of the planet. For simplicity's sake and with a bit of craciousness in favour of ortho, global ZL14 coverage would take up around 2.5 TB. IIRC, the factor in file size between ZL14 and ZL16 (which the masses want) is 4, so it's 10 TB for global coverage at that zoom level. And this is mesh and imagery alone, no overlays at all, plus a single season. I'm sorry but no, that's absolutely not an effcient way to store data for a global scenery for flight simulation.

Edited May 6May 6 by Bjoern

1 hour ago, Bjoern said:

that's absolutely not an effcient way to store data for a global scenery for flight simulation.

You are German, like me, and so you are maybe not aware that you are not using the correct word here - although the German "effizient" means the same thing as "efficient".

Effizienz (Efficiency) is a measure of how much effort you need to put in to get a certain result. It is not a measure of the total effort required. If you build a car that needs 100 liters to drive 100 kilometers, that is not "very efficient" by todays standards, but if other cars were to need 2000 liters, that is suddenly very efficient!

So if your goal is to store the whole world in ZL14 Orthos and it needs 10TB... you can not say anything about efficiency, it is simply the way it is. If you can do the same at 1 TB, you are ten times as efficient, but if another method needs 1000TB, then it is very efficient, compared to that.

The words you are looking for are "feasible", "practical" or "worthwhile" and "cost-effective".

If having that kind of storage requirement is being "worth it" - every user must decide for himself. It depends on your financial abilities and the importance you place on having "real world" scenery.

I think that the goal of a flight-simulator is to replicate the real world as closely as possible, and to my knowledge there is no better method for that for now than orthophotos and photogrammetry. There may be more "efficient" ways to store or stream that data, but the actual data fidelity is simply higher than other current technologies, including the current autogen method used by X-Plane.

Don´t get me wrong, I have never used any autoorthoscenery and I am totally happy with the default Global Scenery in X-Plane, but I would fully accept that I am in a minority and that realistic, true-to-earth scenery is of utmost importance to the majority of users, and there is no alternative to "real data" (i.e. Orthophoto).

What the way forward for LR will be is not decided, yet. But you can bet that they will try to get as close as possible to having the "real world" appear outside the window.

Edited May 6May 6 by Litjan