1 minute ago, fogboundturtle said:

because the issue is that we are not GPU bound but CPU bound. You wouldn't get more FPS using DLSS.

I'm GPU bound in many situations, but I don't think my AMD card will get DLSS anytime soon ;).  Hopefully by the end of 2022 we'll be able to actually buy GPUs again....

27 minutes ago, erick said:

They either don't recognize this as an issue , they are playing dumb for whatever reason or they simply don't know about it.

I think they are just completely oblivious to it. They had trouble answering the coastline glitch even. Sebastian's confused face from that segment needs to be turned to a GIF. 😁

6 minutes ago, fogboundturtle said:

because the issue is that we are not GPU bound but CPU bound. You wouldn't get more FPS using DLSS.

When the CPU optimisations along the DirectX 12 update pass through and ray-tracing is added, I think the situation could change for several high-end systems. Even ignoring that, DLSS would be of huge help to users of VR and GPUs like the RTX 2060, or the upcoming RTX 3050. At least for the next year, those models should be a lot more accessible to those looking to buy a new gaming PC.

1 minute ago, marsman2020 said:

I'm GPU bound in many situations, but I don't think my AMD card will get DLSS anytime soon ;).  Hopefully by the end of 2022 we'll be able to actually buy GPUs again....

It will get FSR though, and that's another part of the question that they totally ignored.

16 minutes ago, ChaoticBeauty said:

I think they are just completely oblivious to it. They had trouble answering the coastline glitch even. Sebastian's confused face from that segment needs to be turned to a GIF. 😁

...

It will get FSR though, and that's another part of the question that they totally ignored.

It's like being in one of those weird dreams where people respond to your questions with completely wild answers that have nothing to do with what you asked.  So frustrating.

Edited April 14, 20215 yr by bobcat999

1 minute ago, bobcat999 said:

It's like being in one of those weird dreams where people respond to you questions with completely wild answers that have nothing to do with what you asked.  So frustrating.

It's even more frustrating that Jayne asked these questions just like I had written them in the forums, with all the details and reasons about these questions getting asked, and yet both times they picked a specific part of the question and ignored the rest. I guess that asking about underrepresented features and bugs in a Q&A is not a way to learn more about them either. It's a shame, I was really enjoying the Q&A right up until that point.

If they don't fix the morphing while (slowly I guess) taking care of the coastline glitch or implementing the photogrammetry improvements in Sim Update 6, I am going to completely lose hope at that point.

2 minutes ago, ChaoticBeauty said:

It's even more frustrating that Jayne asked these questions just like I had written them in the forums, with all the details and reasons about these questions getting asked, and yet both times they picked a specific part of the question and ignored the rest. I guess that asking about underrepresented features and bugs in a Q&A is not a way to learn more about them either. It's a shame, I was really enjoying the Q&A right up until that point.

If they don't fix the morphing while (slowly I guess) taking care of the coastline glitch or implementing the photogrammetry improvements in Sim Update 6, I am going to completely lose hope at that point.

Sure.  And I would fully understand why - Immensely frustrating!

6 hours ago, fogboundturtle said:

because the issue is that we are not GPU bound but CPU bound. You wouldn't get more FPS using DLSS.

I can assure you I am well and truly gpu bound in VR with a G2

6 hours ago, fogboundturtle said:

because the issue is that we are not GPU bound but CPU bound. You wouldn't get more FPS using DLSS.

Yeah wrong... Render sclaing is very important in VR and 100% only influenced by the GPU. That is where DLSS shines

6 hours ago, ChaoticBeauty said:

It's even more frustrating that Jayne asked these questions just like I had written them in the forums, with all the details and reasons about these questions getting asked, and yet both times they picked a specific part of the question and ignored the rest.

I guess the only way to show them what is actually wrong is to actually have lots of comparison shots of the coastline glitch, because there have been many times in the previous Q&A's where they were completely oblivious to an issue or were simply not aware that what they were seeing is an issue. They finally addressed the night lighting issues only after that topic had lots of comparison picture as visual proof that they needed improvement.

7 hours ago, ChaoticBeauty said:

I think they are just completely oblivious to it. They had trouble answering the coastline glitch even. Sebastian's confused face from that segment needs to be turned to a GIF. 😁

 

It is unfortunately not the first time that they look confused by a question in a Q&A and/or answer beside the point to evade the question. They seem not get the questions beforehand to do their homework on them. That also puts in perspective their listening mode on forums. They often look like being disconnected from their user base experience of their ‘title’. 

Yeah, you would think that they would review each of the questions head of time and be prepared. Watching last night, seems they were just winging it. Not understanding a question or saying that they are not aware of an issue is to me, showing a disconnect between themselves and their customers. Jayne too should be asking full questions, in context, not cherry picking parts of them. I too was confused when she asked about 3 layers of cloud not being accurate. It was actually a question about ATIS, not clouds. Not surprised Asobo’s team sat there with blank, if not slightly irritated looking faces.

Don't want to be too hard on them as not all dev’s would take the time out to host these sessions and they are interesting, but they could be making these more effective with some homework ahead of the sessions.

All,

Also keep in mind the environment and context. Martial and Seb are great guys, and super smart engineers. I now get to talk with them a bunch. But the format is intimidating: they aren't native camera people, they're software engineers trying to understand and convey things as best they can in a second language.

Jayne is awesome too, but occasionally things get lost in translation because it might be some super deep vertical and she's a new simmer. Her skillset is community management.

I know, when you have a pet issue, it seems absolutely brain dead obvious to you what it is and it also seems completely glaring, like how could these guys not see this stuff. But even I was a bit lost on what the coastline bug was referring to, and I'm in the sim 6 hours a day or more. Now imagine if you're trying to parse that in a second language.

However @ChaoticBeauty, to answer it now, it's actually a really challenging problem. The resolution of the satellite images is generally way, way higher than the DEM data resolution. So ultimately you end up with areas where the image of the water doesn't precisely match the received DEM data, so in one area you might have 2cm/pixel visual accuracy but 10x worse DEM accuracy.  There are machine learning solutions to this that are quite complicated, but also introduce their own potential issues.

For example, this problem can be especially apparent with mountain river valleys. However, as we all know, rivers in the mountains lots of the time have a reasonable amount of elevation change to them. So if the algorithm is "find me water looking stuff and generate flat elevation data for that area" it isn't even immediately clear what elevation would be right. And then if things are running downhill, do you need to bake in some slope detection? How does it affect the surrounding terrain? Do you now, as a result, end up with areas of strange valley that have been "pulled" up or down to this new slope?

The same applies to coastlines, but maybe less in the extreme. It certainly isn't a super straightforward engineering or data problem to solve satisfactorily for all cases, as one algo that looks good for the coastline of Lake Michigan might look ridiculous for the Cliffs of Dover, and vice versa.

Once you get to these kinds of problems, I can say they can look from the outside to be conceptually simple. Of course, just adjust the elevation data so the water looking stuff doesn't ride up any coasts. But the devil is way way in the details with a lot of the issues that get forwarded in these Q/As; I could totally assign an engineer to a task like that and it would eat their entire next 2-3 months getting suitably decent.

Each on of these things is like that. ATC is an absolute time sink black hole of horrifying AI edge cases and deadlock conditions. A whole team of 10 could totally spend 6 months on it as a strike team and come up with something maybe only a little better than what exists. Many, many have tried, even commercially, in the real aviation world, with more resources. It's an incredibly difficult problem space.

I think something we could do better is communicating the depth of these engineering challenges. Lots of this is not low hanging fruit anymore that someone can spend a week or two on and fix. But it's really really hard to communicate these ideas on a good day, not in front of a camera, in my native language. So I definitely cut these guys a bit of slack for sure.

-Matt

And, to the other part of the question, about morphing. Yes, sometimes, it's obvious, for sure. But this is also one of those problems that's really hard to wave a magic wand at.

There are a few limitations: one is, the data may be streaming in, either from disk or from online, and it might be behind, due to disk or internet conditions. Then, the data shows up, and suddenly smooth hills morph into cliffs. The second case is that the distance is such that you're hitting a better LOD ring by distance, and same thing, you notice the switch. What are some ways to combat this?

On the I/O side, you could just decide that if the data hasn't gotten to you by the time you enter the high LOD ring, just don't swap it out until it's out of view. That would avoid the morphing, but might also be frustrating; some users would certainly complain that on the first pass of an area it looked like word not allowed and when they turned around suddenly it was good. That feels like a bug too, to those people.

Or, for both cases, you could smoothly move from one LOD to the other. However, now you're stretching the problem across a longer time frame, which might be more unsettling to some as the hills slowly grow into cliffs. Maybe then you think, OK, well, I'll push out the LOD transition area so that happens further away so it feels less jarring. But now what you really did is effectively push further out the high detail LOD _and_ add CPU cycles burning to figure out how to transition one set of vertices to another totally unrelated set of vertices (which, btw, is another totally enormous engineering problem that even high quality modeling software hasn't yet gotten right).

So, again, there's not really one answer that solves the issue completely for everyone with no side effects. Engineering is all about effective compromises inside resource limited systems.

I hope that helps a bit.

-Matt

I re-watched some parts of the Q&A video, and noticed a couple of things that haven't been mentioned yet in this thread.

Prop feather is coming. They've shown this in the context of turboprop improvements. (Around 20:00)

Apparently the torque/power/altitude relationship for turbo props has been fixed, and a torque limiter has been added for the Cessna 208. I'm no expert in this, so please judge for yourself... It can be found in the twitch stream at around 20 minutes.

They're also doing real-life comparisons with "test" flights in order to fine-tune the aerodynamics, so far for all C152 and C172, and all three turboprops, and coming to more planes in the future. Supposedly improves adverse yaw and the feel of twitchiness and inertia too. See around 23:10 .

For the sake of completeness: Airports EKRN Bornholm, Denmark, BIIS Ísafjörður, Iceland, ENSB Svalbard / Spitsbergen, Norway, ESSA Stockholm Arlanda, Sweden, EFVA Vaasa, Finland. Around 29:00 .

Replay functionality (minute 51:10): they want to go beyond normal replay, brain-stormed all kinds of ideas, such as recording the whole state of a plane, so that it'll be possible to enter the cockpit again at any point of the replay and re-fly a landing etc.

Aerosol density, humidity/dust parameters, visibility distance setting: There's a limitation to the atmospheric visualisation system. According to Seb, they're planning on rebooting the whole atmospheric simulation system, but this is a long process that's going to take about 6 months (no mention though when they're going to start with this). Around 58:00

As you've already pointed out, some misunderstandings about the scenery morphing.

Still I think it was a very interesting Q&A with a couple of good points. But ok, let's see when we get to play with them, and when they're in a (reasonably) bug-free state  😅

Ta Matt... your thoughtful and detailed answers are much appreciated.

The issue of a global data is indeed immense. We shall keep on, keeping on, and ask you to do likewise. If you should manage to keep this type of response to the madding crowd long term, none would be more impressed than I. It is gratifying in this brave new world.

i remain obliged- C

3 hours ago, FAZZ3 said:

I guess the only way to show them what is actually wrong is to actually have lots of comparison shots of the coastline glitch, because there have been many times in the previous Q&A's where they were completely oblivious to an issue or were simply not aware that what they were seeing is an issue.

The post I had made linked to a thread with plenty of footage, and I did post an image comparison later. I think they just had neglected to prepare their answers.

3 hours ago, Dominique_K said:

They seem not get the questions beforehand to do their homework on them.

3 hours ago, RaptyrOne said:

Yeah, you would think that they would review each of the questions head of time and be prepared. Watching last night, seems they were just winging it.

Remember the previous Q&A with the guided questions where they had read them in advance and prepared video footage to answer them? When this was posted in the Q&A forum section I thought it was going to be the same case, but it is obvious they were completely unprepared for this one.

Quote

This thread will close  April 5, 2021 10:00 PM to allow time for them to pick questions and get answers for the session.

 

40 minutes ago, MattNischan said:

Jayne is awesome too, but occasionally things get lost in translation because it might be some super deep vertical and she's a new simmer. Her skillset is community management.

I don't disagree, and she's a very quick learner too if you watch the streams with the flight lessons. I don't think she was the problem at least when it came to my questions.

40 minutes ago, MattNischan said:

But even I was a bit lost on what the coastline bug was referring to, and I'm in the sim 6 hours a day or more. Now imagine if you're trying to parse that in a second language.

As I had already explained in my question, the problem is much more visible when the Terrain Level of Detail value is set to a lower one. At 100 and above the water creeping up the coastlines is almost completely eliminated even at longer distances. Anything lower than that, and it gets really ugly.

Also it's worth mentioning that they did not really fix the water creeping up, all they did with the UK World Update is push the coastlines globally towards the mesh, so that the water does not get too close to the sawtooth pattern. While it helps make the issue less visible to an extent, it also completely botches the appearance of coastlines. You'll have buildings, trees and roads extending along the entire slope of the coastline, and the sawtooth pattern becomes even more visible.

Approaching Chambery airport or many of the Aegean islands is really, really ugly now.

40 minutes ago, MattNischan said:

However @ChaoticBeauty, to answer it now, it's actually a really challenging problem. The resolution of the satellite images is generally way, way higher than the DEM data resolution. So ultimately you end up with areas where the image of the water doesn't precisely match the received DEM data, so in one area you might have 2cm/pixel visual accuracy but 10x worse DEM accuracy.  There are machine learning solutions to this that are quite complicated, but also introduce their own potential issues.

I appreciate your post, and as I mentioned in my question thread, I know that it is not just an easy fix. But the issue does not have to do with the data source, it's rather a global problem where the shapes of the mesh are not being retained at lower levels of detail, and as CptLucky8 explained in that thread, the LOD rings are broken as of now.

Before the USA World Update, the morphing problem was much, much less noticeable, the coastline glitch did not exist at all, and the sawtooth pattern from the FSX/P3D days did not persist. And I'll stress that the experience now is much worse for those who do not play at Ultra settings. It's not like this has always been a problem, it is a post-launch regression, so I think it must be possible to somehow return the mesh quality to its previous state. It could have to do with them refactoring the terrain system to help incorporate the higher-quality mesh for the USA, but if it's breaking the mesh globally it's no good.

All I wanted was some acknowledgement of the issue, and we did not even get that. I have no problem with Asobo taking all the time they need to correct these issues, but it seems that they are not even aware which has deeply disappointed me.

Edited April 14, 20215 yr by ChaoticBeauty