Helisimmer Interviews Jorg and Seb on 2024 and More

[Krakin 4,935]

https://www.helisimmer.com/interview/msfs-2024-helicopters-future-msfs-2020?fbclid=IwAR0yCiQ_QHC4v7CmKJASAXPQz0oLVJVAiO335BC6BxObxudnwkQbcQ9bymE

[Krakin 4,935]

This seems to be the best interview with the MS team so far. @lwt1971 I'm looking forward to your summary!

[Krakin 4,935]

Quote

A lot of people have been asking for improvements on the flight model. And what we've done a lot recently is all the little changes, all the little adjustments, but the big, big changes require a new architecture here again.

Yeah I really don't think 2020 is capable of pulling off that hot air balloon demo. My jaw dropped when I saw it.

[lwt1971 5,831]

2 hours ago, Krakin said:

This seems to be the best interview with the MS team so far. @lwt1971 I'm looking forward to your summary!

Hehe was just working on some quick notes 🙂  ... I'll iust focus on what I caught as new info outside of the expo and previous two interviews. MS/Asobo's quotes in italic. Understandably this interview focuses in on helicopter topics but we get a good sense of general/core sim improvements for sure.
 

• Nothing from 2020 is being taken away in 2024 ... The idea is that the planes from FlightSim 2020 are coming over. They might be better than everything else that's in FlightSim 2020 will come over, but better. So, think about it that way. The airports will be there… better. The POIs will be there, either the same or better. All the planes are going to be there. Everything that you love right now is going to be there plus, plus, plus, plus.
• All the big wishlist items, major flight model enhancements, seasons, weather etc required architectural changes ... And so at some point we said, okay, now we got to do all the long, the big, big undertakings and do the big changes, like reducing the size on the disk by changing the way the data is organized, improving drastically the flight model. A lot of people have been asking for improvements on the flight model. And what we've done a lot recently is all the little changes, all the little adjustments, but the big, big changes require a new architecture here again.
• 2020 support will continue for at least a few years after 2024 releases
• Example of what ex IL-2 tech lead Andrey has been already working on in MSFS 2024: I know that Andre, a physics engineer that's joined us recently, he worked on the governor on the turbine helicopters. He has done quite a lot of changes ... Andre joined us from Stormovik [editor’s note: IL-2 Stormovik, the flight simulator]. So that's great. Seb already had another person. So it's not just everything on Seb. Seb's brilliant, but it's a lot of work
• On CFD and physics processing:

  • You know, CFD is very important for helicopters. And so with the new system, we're very much multi-threaded, taking work off the main thread about all the simulation aerodynamics works.
    The CFD can be a lot more detailed than it could be for, without an impact on the FPS. Because basically, we stopped at the detail we had with the current implementation, because if you go further, it slows the FPS down, and we didn't want that.
    But currently, you're sort of limited in how far the ground effect goes, because that's the limit of the CFD, and how accurate the vortex effects can be. And so, for example, on a dual rotor helicopter, if you take the same CFD and you stretch it, you sort of, it is so big that with two rotors, you can't capture the, the donuts are sort of almost gone.
    And so […] for that, we can now increase the size without a performance impact. And so have a bigger CFD, more accurate, longer ground effects, all that is possible now, thanks to the architectural changes we're doing. And this with new architecture is super cheap, right? It's just offloaded, you don't see it.

• How many CPU cores can the sim use and more details on the "4x/5x and future 10x performance improvement" comment from previous interview, and what it specifically applies to. As some of us discussed on other thread, this applies to physics calculations looks like:

  • So, we don't have currently a hard limit set, it's just the limit of the memory. At some point, it's gonna drop the performance, right? What we measure right now, and we're not finished optimizing, we're already at a four to five times performance increase. And I think we're gonna maybe get 10 […] you can easily add 10 times more surfaces, right? It's very quickly, right? And so, I think if people add 20 times more surfaces than before, well, then you may be gonna offload, you know, load the other threads. And the more you load, the more you take away compute power for the terrain, or other stuff, right? Which is using thread. And at some point, it's gonna have an impact. I don't think we're gonna set a limit. We're gonna limit ourselves, we're gonna have recommendations, but just like today, a third-party aircraft can do a lot of stuff. If it wants to load up the machines, it can.

• About multi rotors and new aerodynamics framework & physics in general:

  • So I touched a little bit about this during the presentation, that the physics system has a big update and we are introducing something like, we call it templates. So instead of in 2020, where you had a, basically the plane was always, you know, fuselage, a wing, a tail, which had to be vertical and horizontal. And then the helicopter had one rotor and one rotor at the back, all you could do is spin, you know, choose the direction.

    These things still exist, but for backwards compatibility, they still exist and they are used, but they are now templates. So it's not a airplane anymore or helicopter, it's a wing, it's a fuselage.

    And there are different versions of wings and fuselage, and you can have as many as you want. So with that system, it was very easy to take the rotor and say, oh, now I just want two of them in opposite direction. And that's it. And you have a helicopter like a Chinook and it's very easy to do.

    You could have two, three, four, five rotor, 16, do a drone. It's much more flexible. Two rotors, V-Tail is now easy with a system. And it's some sort of doing a two-rotor helicopter in 2020. It would have been very difficult because it's not, it cannot do that, right?

    So, you have to hard code another rotor, implement everything, and then it's gonna eat more memory. Every plane can do everything. Can do 16 engines and then rotors and then another rotor. And it was just hitting a sort of a wall of performance and memory.

    With the new system, you only pay what you want. And you can as many as you want of different parts. And different versions of parts, more or less complex.

• AI pilot and autopilot will use same subsytems/logic

• On AI traffic:

  • There's something called the authentic air traffic. And what that means is we're going to do the right models, ideally with the right liveries as long as we can license these word not allowed things. But we are trying hard to get liveries and we're going to add things like helicopters. The data exists. We never modeled these things, but we want to. So, we're currently very busy signing licensing for for all as many helicopters as we can possibly get our hands on. As you saw, we signed with Airbus helicopters, which is great. You saw the 125.
     

Edited June 28 by lwt1971

[lwt1971 5,831]

I'm especially happy about multi-threading and CFD. Firstly, to solve the subset of Navier-Stokes equations that MSFS uses for CFD, the finer the compute granularity/resolution the better the outcome obviously.. and currently thru the debug window or flight sim model config you can increase the resolution of computation, but of course that tanks performance due to heavy CPU usage (in a non-linear fashion). Spreading out these calculations over multiple CPU cores is going to help greatly, and aircraft devs can increase the CFD computation resolution even more (i.e. 'CFD_VoxelNbVoxels' in https://docs.flightsimulator.com/html/Content_Configuration/SimObjects/Aircraft_SimO/flight_model/aerodynamics.htm, also see https://docs.flightsimulator.com/html/Developer_Mode/Aircraft_Editor/Debug/Debug_Aircraft_CFD.htm).

This, combined with the new ability to define aircraft surfaces and controls in a much more fine-grained and however complex manner is really going to make MSFS's CFD tech shine.

And as always, good to hear that Seb and Andrey (Petrovich) Solomykin are leading the charge in the core physics and aerodynamics work!
 

Edited June 28 by lwt1971

[jcomm 5,440]

As usual, great interview by Sérgio.

I had talked to him just before he left to FSExpo23, and he was already very enthusiastic about it and the perspectives on the upcoming fs2024.

It's a LOT of information, all of it great to know.

[threegreen 2,558]

I keep being startled at just how far apart this and the 'it's not a serious sim' department are. When you look at all the stuff that's going on under the hood... There just doesn't seem to be much willingness to be factual at this point.

Edited June 28 by Nixoq

[lwt1971 5,831]

1 hour ago, Nixoq said:

I keep being startled at just how far apart this and the 'it's not a serious sim' department are. When you look at all the stuff that's going on under the hood... There just doesn't seem to be much willingness to be factual at this point.

There is a small but flailing and trying-hard-to-be-vocal contingent who absolutely cannot stand how MSFS is advancing and how all this info about v2024 just breaks their pathetic narratives to pieces. Kind of fun to watch the flailing though 🙂  (their narratives were/are also foolishly clownish where v2020 is concerned). Relatedly, some of this contingent also seems to have a weirdly personal anger/bitterness towards MS and/or Jorg :s, perhaps even dating back to FSX leaving the scene.

I'll take a sim and a sim development company behind it that takes *all* aspects of flight simming seriously, and not just some. With the deep pockets, deep resources, and skills of the current MS/Asobo team, MSFS *is* the you-can-have-your-cake-and-eat-it-too sim :))
 

Edited June 28 by lwt1971

[Krakin 4,935]

Wrong thread 

Edited June 28 by Krakin
Wrong thread

[Fielder 3,318]

" What we measure right now, and we're not finished optimizing, we're already at a four to five times performance increase. And I think we're gonna maybe get 10 "

Zuckerberg thinks meta will change the world. HIs company is betting on and investing in 3D Dimensions.

Jörg thinks the cloud will change the world. His parent company MS is focused on cloud storage and Improved Performance.

 

[lwt1971 5,831]

Helisimmer's conclusions are worthy of note:
 

• MSFS 2024 is a first-class member of the MSFS franchise. No cutbacks, and no removed features. It’s MSFS (2020) on steroids. They have picked up where they were at and added features;
• Some of those added features include a lot of what helicopters do in real life. Unfortunately, I have seen community members saying the team is gamifying MSFS with the addition of those activities. Those who are saying that seem to be completely oblivious to what helicopters do in real life. Perhaps they never looked at helicopters before and realized how much more these machines do when compared with fixed-wing aircraft in real life.
  
  "If a man is in need of rescue, an airplane can come in and throw flowers on him, and that's just about all. But a direct lift aircraft could come in and save his life."
  – Igor Sikorsky
   
• Not only features were added, but the engine was also redone so we can have a lighter version that uses less space in our drives, updates quicker, and has better performance;
  • Better performance also means we will have, improved flight dynamics and more rotors for helicopters, for example;

In a nutshell, I am super excited about MSFS 2024. I think it will be an amazing new version but the current sim will not be abandoned. As you can read in this interview, there are plans for continuous support and even new features that may be added to MSFS 2024 to sneak into MSFS (2020).
 

Edited June 28 by lwt1971

[Fielder 3,318]

Imagine the poor performance of MSFS if all the files were stored on the local drive (iPad, iPhone, Xbone). But by using the cloud these underpowered devices can run MSFS right now (subscription). 

 I think the greater 2024 performance compared to FS2020 will be largely based on putting more and more in the cloud and less on our local drives.

[ca_metal 1,965]

23 minutes ago, Fielder said:

Imagine the poor performance of MSFS if all the files were stored on the local drive (iPad, iPhone, Xbone). But by using the cloud these underpowered devices can run MSFS right now (subscription). 

 I think the greater 2024 performance compared to FS2020 will be largely based on putting more and more in the cloud and less on our local drives.

Not at all. Based on all their interviews so far, putting more things in the cloud is pretty much about having a thinner client. 

The performance gains are more about improvements on the engine + focus on multi threading.

One aspect of the living world that I feel is missing so far are the trains. Not on my priority list and maybe they didn’t find live data to feed the simulator, but would be nice to have trains worldwide (again, not a priority to me, but would add to the immersion).

Edited June 28 by ca_metal

[pstrub 967]

A word of caution: Most of their comments mentioning the huge performance improvements were made in the context of the simulation of the rotor blades and aerodynamics, allowing them to improve the fidelity of the aerodynamical simulation.

Unless they mention this very specifically, I would NOT expect the performance improvements to translate into 5x better framerates. 

[Krakin 4,935]

2 minutes ago, pstrub said:

Unless they mention this very specifically, I would NOT expect the performance improvements to translate into 5x better framerates. 

Yes, the more they talk about it, the more I realize what you're saying makes sense.