Add barns/farms etc to autogen buildings please

All over the UK there are office block type buildings where there should be farms/barns. I’m not asking for miracles but surely the AI could learn that in a countryside area there is a higher percentage chance that it is a farm rather than an office block and more often than not, generate a more appropriate building.

I thought the “I” in “AI” stood for intelligence! If it was intelligent it would know that office buildings aren’t appropriate structures in those places!

And I also have yet to see a single church anywhere in the UK!

I do hope the AI is learning, as I was lead to believe it does, in one of the trailers.

If you saw the Japan world update notes, they talked about Blackshark.ai adding a new Japanese building type, so that leads me to believe that their AI must fit whatever it sees to a type of pre-defined building and take it from there. So it sounds like someone needs to do some manual work to make that happen.

1 Like

Yes I live in Cheshire and round here we suddenly have lots of jobs in offices I will have to come out of retirement and find a job. They must be crying out for people. :slight_smile::slight_smile:

1 Like

I’d caution you against taking a job at one of those fancy new office buildings. In the next release, your office could turn into a grain silo!

3 Likes

I’ve noticed this all over the vast farming communities of Canada. Thousands of square kilometers of cattle, corn, and various other staples, and suddenly…an apartment building!

For anyone who doesn’t know what’s involved in making this change and is curious, I have added some details below. Keep in mind that I am not an expert in this area; I’ve only worked with neural nets on a small scale. Please don’t do your dissertation using my explanation:

The way these neural nets work is that first blackshark, (the company that provides the AI building generation capabilities) will need to train the neural network how to spot farms and rural structures.

Training a neural net involves building a set of training data (satellite pictures of farms, rural roads, crops in various conditions, etc.) to learn from. Next, people will individually annotate that enormous collection of pictures with labels (this is a farm, that is a silo, that is an irrigation ditch, and so on). It’s a lot like teaching someone using flash cards. Do this work a hundred thousand times, and the AI should pick up fairly quickly on the pattern. Then you send it off on its own without any help, and see what it comes up with, review for mistakes, and try to correct them.

The challenge can come when the AI does something really crazy and unexpected, like turning a grain silo, which seems so obvious to us, into some sort of garrish satantic monolith or maybe an apartment block. The AI has no way to communicate WHY it made the choice it did (unlike teaching a person), so the trainer is left to use their engineering intuition for determine why the neural net is acting crazy in one particular edge case. This can be a time consuming process, as there are many reasons why it can break; not enough training data, too much training data, bad labelling, etc.

You’ve also got some requirements that make this extra interesting: The AI system will need to stick a building in exactly the right spot so it matches up with the map. It will need to generate, within a set of parameters or “limits”, the right building, out of a big set of choices it will have. It needs to be able to work with various qualities of satellite images, deal with weather, shadows, and so on.

As far as I know, every time they make a change, the ENTIRE dataset for the WORLD needs be completely re-generated - this involves spinning up a large fleet of compute instances (servers specifically configured for doing this type of resource-intensive work), often hundreds or thousands of machines, and having them run at full-tilt for many hours, possibly even days.

Now granted, they’ve streamlined the process quite a lot, but this is still an involved and often tedious process.

The takeaway is that Blackshark.ai seems to have a pretty good system for world generation - I have no doubt this will continue to improve. Every time they train the system, it will get better and better. All hope is not lost! :slight_smile:

2 Likes

Thank you for that. From what limited knowledge I have of AI from when I got my computer science degrees (cough, 20 years ago, cough), that is my understanding, as well. Except for the part about regenerating the world every time they make a change. That is not something I even considered! Wow!

Thank you for that interesting and informative reply. I am a little naïve I suppose - I didn’t realise it was as complicated as that. Pretty amazing stuff, and I hope that it is learning and that they will add more building types as they go along. Cheers!

Hi all
Could the generic 5/6 story white/cream office blocks/appartment buildings be removed from the rural areas of the sim.
They have their place in the towns & cities but in fields, villages, farms and on top of mountains they reduce the reality that has been created by the rest of the sims amazing visuals and stand out like “sore thumbs”.

I’ve flown mainly over the UK and they are all over the place and have been since day one.

Thanks

1 Like

Agreed! I zen desked this almost on day 1. Where are the farm buildings and barns etc!?

Hi all, there is already a topic for this:


I’m hoping for this, too!

1 Like

Definitely agree, especially with the really large chicken houses that are scattered all over the United States. Here is a post I made about them recently with comparison pictures:

https://forums.flightsimulator.com/t/interesting-read-concerning-our-new-3d-world/275940/3

When the AI turns rural chicken houses into massive apartment complexes, it drastically changes the look & feel of the area.

1 Like

Hey, turkey farms matter, too! :stuck_out_tongue:
:turkey: :small_airplane: :smile_cat:

Yeah, they do! Hogs too. The houses all look pretty much the same though. :slight_smile:

1 Like

Apparently it is in the rural part that this issue exists. The same is true for Denmark. Lots of highrise barns and mansion type farmhouses. Hope AI will learn.

2 Likes

In the same idea : Hay bales piles are also identified as buildings by AI.

Big let-down for me is flying over some of the excellent immersive UK countryside but then you come across a 5-story modern office block where an old farmhouse of barn should be. It seems in rural areas nearly every isolated building gets the block-of-concrete-apartments look which ruins the othrewise excellent and immersive landscape. I’d love to get this fixed with low farm buildings & barns in the isolted areas of the countryside.

1 Like

Hey I rather enjoyed the thought of skyscrapers in Jerome, Arizona, an old mining community build on the side of a mountain with winding narrow streets and turn of the century buildings. Laughs. In other words it’s not just the UK.

1 Like

Having now navigated to some other countries - I concur! I do hope they are working on this…

1 Like

From the Fens in rural Cambridgeshire. There are other apartment blocks in the background. It really is poor and completely ruins flying around these areas. These are supposed to be farm sheds. Does anyone from Microsoft ever check these forums? The Land’s End area in Cornwall is similar, though the apartment blocks are harder to see due to the forest that has overgrown the heathland. Disappointed.

1 Like