Jump to content

Unreal Engine 5 viable for DCS?


Bog9y

Recommended Posts

11 hours ago, DaWu said:

This thread lol. No you can’t switch engines like ur underwear. 

 

This! It would mean that DCS would need to be rebuild again from the ground up, which is pretty much impossible within a reasonable time and budget.


Edited by QuiGon

Intel i7-12700K @ 8x5GHz+4x3.8GHz + 32 GB DDR5 RAM + Nvidia Geforce RTX 2080 (8 GB VRAM) + M.2 SSD + Windows 10 64Bit

 

DCS Panavia Tornado (IDS) really needs to be a thing!

 

Tornado3 small.jpg

Link to comment
Share on other sites

17 hours ago, Xilon_x said:

gravity is not realistic those rocks so heavy they move as if they were fake polystyrene rocks.

I would like an engine that can calculate whether when you build a paper plane it is flyable or non-flyable. gravity must be realistic.

 

 

Ahem. Gravity has nothing to do with that. It's simply 9.81 m/s^2, nothing mysterious about it, and all current engines (not just Unreal, Unity and Cry) model it that way. You may be thinking about fluid / gas dynamics and how drag affects the object's trajectory. There's a flight sim that shall not be named but starts with the letter 'X' that can do lift modeling and can get pretty close to realistic flight dynamics for gliders such as planes made from paper. It can (or once was able to anyway) even model it on different planets (i.e. bodies with lower gravity (e.g. 3.72 m/s^2 on Mars) and thinner, different 'air'). And since I'm smartassing: all planes are inherently flyable, even bricks. It's just a question of how far 🙂 

 


Edited by cfrag
Link to comment
Share on other sites

e stiamo parlando di simulazione di volo, il motore grafico deve essere in grado di simulare perfettamente l'intera atmosfera terrestre dagli strati più bassi a quelli più alti.
temperatura pressione gravità venti nuvole tutti valori che variano in base all'altitudine.
 un vero motore grafico non esiste ancora.
infatti DCS ha costruito un motore grafico indipendente ma molto interessante.

 

465px-Atmosfeer_Atmosfera_-_Italiano.png


Edited by Xilon_x
Link to comment
Share on other sites

58 minutes ago, Xilon_x said:

e stiamo parlando di simulazione di volo, il motore grafico deve essere in grado di simulare perfettamente l'intera atmosfera terrestre dagli strati più bassi a quelli più alti.
temperatura pressione gravità venti nuvole tutti valori che variano in base all'altitudine.
 un vero motore grafico non esiste ancora.
infatti DCS ha costruito un motore grafico indipendente ma molto interessante.

The graphics engine is a non-factor. It has nothing to do with the simulation. There are plenty of rendering pipelines that produce sufficiently real results — Unreal being probably the top contender at the moment since it is actually being used in professional VFX pipelines. Long before that, CryEngine had managed to carve out a similar niche. But none of that has any relevance for its use in flight simulation.

 

The problem is visual scale and being fit for the purpose of delivering that scale, which is a whole different matter. Hence why, somewhat counter-intuitively, “real graphics” becomes less and less of a priority the more realistic you want the simulation to be.

 

2 hours ago, cfrag said:

Ahem. Gravity has nothing to do with that. It's simply 9.81 m/s^2, nothing mysterious about it, and all current engines (not just Unreal, Unity and Cry) model it that way. You may be thinking about fluid / gas dynamics and how drag affects the object's trajectory. There's a flight sim that shall not be named but starts with the letter 'X' that can do lift modeling and can get pretty close to realistic flight dynamics for gliders such as planes made from paper. It can (or once was able to anyway) even model it on different planets (i.e. bodies with lower gravity (e.g. 3.72 m/s^2 on Mars) and thinner, different 'air'). And since I'm smartassing: all planes are inherently flyable, even bricks. It's just a question of how far

 

That's the funny thing: the actual flight simulation is not a particularly difficult part. Even when you go beyond the utterly trivial stuff like gravity, it's largely a solved problem, and it can pretty much be considered a wholly disconnected component from the rest of the engine. You need to come up with forces that feed into the physics engine, but those calculations are more or less engine-agnostic.

 

It's more a matter of picking the right level of detail to make it run in real-time with the amount of units you want to have at any given time, and picking the aerodynamic components that are the most relevant to create that level of detail. And then it just boils down to a hideous amount of tweaking to make the resulting dynamics match real-world performance (and finding that performance data to begin with).

 

The really tricky part is… oh… everything else. Especially AI, and especially if you want the AI to be realistically scalable to the world size, much like how world size causes issues for the rendering.

  • Like 1

❧ ❧ Inside you are two wolves. One cannot land; the other shoots friendlies. You are a Goon. ❧ ❧

Link to comment
Share on other sites

artificial intelligence in common simulators is programmable and customizable but it is not intelligent.

true artificial intelligence is that which learns from its mistakes and corrects itself and develops to improve.

you can experience this intelligence in new quantum PCs but in a normal PC artificial intelligence has its logarithms and is based on these logarithms.

Link to comment
Share on other sites

3 minutes ago, Xilon_x said:

artificial intelligence in common simulators is programmable and customizable but it is not intelligent.

true artificial intelligence is that which learns from its mistakes and corrects itself and develops to improve.

you can experience this intelligence in new quantum PCs but in a normal PC artificial intelligence has its logarithms and is based on these logarithms.

I suppose that by logarithm you mean algorithm…:smartass:

A-10A, A-10C, A-10C II, AV-8B, F-5E, F-16C, F/A-18C, F-86F, Yak-52, Nevada, Persian Gulf, Syria, Supercarrier, Combined Arms, FW 190 A-8, FW 190 D-9, Spitfire LF Mk. IX, Normandy + WWII Assets Pack

 

Link to comment
Share on other sites

the best developers in the world should be asked to create a new graphics engine specifically for civilian and military flight simulations. and with an ability to recreate the entire planet earth with more detailed MESH TERRAIN below 3m with immediate computational capacity in rendering.

Link to comment
Share on other sites

22 minutes ago, Xilon_x said:

no graphic engine able to reproduce a real CRASH TEST.

That's because it's not the graphics engine's job to do so. So that's a pretty meaningless statement. It also has next to nothing to do with the topic at hand.

  • Like 1

❧ ❧ Inside you are two wolves. One cannot land; the other shoots friendlies. You are a Goon. ❧ ❧

Link to comment
Share on other sites

 

in this graphic engine of yours you see a mountain a landscape the ground the wind that moves grass and trees you see rocks that crumble into smaller rocks and you see a fall of meteorites that create cross craters in the ground and crumble the stones. smoke you can't see fire you can't see water you can't see gas you can't see it and all very clean and all unreal.

Link to comment
Share on other sites

28 minutes ago, Xilon_x said:

CRATER NOTE:

huge pit debris and fire and flame smoke.

Note about the Chelyabinsk meteor: It blew up in the sky and did not impact the ground.

https://en.wikipedia.org/wiki/Chelyabinsk_meteor

The pit of fire at the end of the second video is the Darvaza gas crater in Turkmenistan.

https://en.wikipedia.org/wiki/Darvaza_gas_crater

  • Like 1
Link to comment
Share on other sites

we are talking about a piece of stone that enters the earth's atmosphere at a speed of about 30.000Km without having light science without having considerable smoke trails without illuminating the entire area from where it flies over without any physics. and dynamic.

a graphic engine must automatically have in its DNA the thermodynamic and aerodynamic physical effects of an object that flies in the air and when it hits the ground it must know how to calculate the kinetic energy.

UNREAL ENJINE 5 is not a graphics engine for DCS.


DCS needs much more than just a graphics engine.

Link to comment
Share on other sites

26 minutes ago, Xilon_x said:

we are talking about a piece of stone that enters the earth's atmosphere at a speed of about 30.000Km without having light science without having considerable smoke trails without illuminating the entire area from where it flies over without any physics. and dynamic.

No, we're talking about real-time terrain deformation tools that can be used for map editing or in-game special effects. Nothing more.

 

26 minutes ago, Xilon_x said:

a graphic engine must automatically have in its DNA the thermodynamic and aerodynamic physical effects of an object that flies in the air and when it hits the ground it must know how to calculate the kinetic energy.

No, it must not. Those would be terribly misplaced properties to put into a graphics engine.

They're stuff that the physics engine would deal with if — and only if — they're deemed relevant, which in the vast majority of situations, they aren't. There is no “must” about it, and it has nothing to do with graphics. Hell, not even in a flight simulator would those be “must”s since they're better handled using various shortcuts and simplifications in a separate flight dynamics module that then feeds resulting forces to the physics engine.


Edited by Tippis
  • Like 1

❧ ❧ Inside you are two wolves. One cannot land; the other shoots friendlies. You are a Goon. ❧ ❧

Link to comment
Share on other sites

3 hours ago, Xilon_x said:

the best developers in the world should be asked to create a new graphics engine specifically for civilian and military flight simulations. and with an ability to recreate the entire planet earth with more detailed MESH TERRAIN below 3m with immediate computational capacity in rendering.

 

So many questions...

 

Who are the best (software?) developers in the world? I'm not even sure they agree amongst themselves who the best are (except "as long as I'm not included, the list isn't complete"). And specifically: developers for what? Massively parallel processing? Shaders? AI? Middleware? System Software? Networking? Big Data? GUI? Testing? Data Base/Retrieval? Compilers? Audio? You do realize that software development is a big and diverse field, do you?

 

And who will ask them (whoever the 'best developers' are)? Why would "they" then do what is asked? Meaning: who pays for that? How much? Who has that kind of money? How do you get that money back?

 

What is a 'graphics engine specifically for [...] flight simulators'? The challenge in FS isn't so much the graphics but engineering of the world model and the aircraft's subsystems. Bringing their visual representation on-screen is a relatively minor task.

 

The ability to re-create the entire planet earth mesh with better than 3 m^2 resolution is far easier than to actually getting the mesh data. Once you have the mesh data your work is only beginning. Now you need to fill that mesh (vertex points) with something visual (surfaces). So you need artists for the textures. Tons of them. To fill the space between the vertex points. You know, with 510067420000000 square meters to fill, you need a big team, and a really big main store. And one heck of a retrieval system to access and pipe that texture data to your GPU. One reason a certain FS from a company that shall not be named but also sells operating systems has such a crappy performance is that they can't retrieve the data fast enough. And once you have all that, you need to start working on the objects that are placed on the surfaces: you know houses and such. 

 

Let's close this subject

  • Like 2
Link to comment
Share on other sites

  • 4 months later...
  • 7 months later...

Yep.  The new UE5 can have terrains up to 88 million square kilometers.  About the surface of the Earth.

Been looking at it all of late, wondering if flight sim would be capable in it.  Because it would of course, solve the vehicle and fps problems.

Quite a step up in engine technology from what I've seen.

Also, it has a lot of free textures and objects in their library, if I recall right.  Great for making quick maps.

Still haven't seen any real proof of it's flight sim capabilities yet though.  Just a lot of talk.


Edited by 3WA
  • Like 1
Link to comment
Share on other sites

While all of that is indeed interesting and makes me happily await future development, I'm not sure if we are putting the wagon before the horse. DCS's unique selling proposition (USP) is the accuracy of it's systems modelling, followed by the accuracy of it's physics modelling. The graphics are more or less coincidental, as most of the time you aren't close to the ground where that matters. That's why there are tons of games that look better. Those games sell their looks. DCS sells it's model accuracy. Their intended target audiences don't match.

UE's USP is (stunning?) graphical fidelity, tuned mostly towards FPS. You can get jaw-dropping graphics almost out-of-the-box. You can get it to display a world-sized arena that looks good. Unreal's physics modelling? Good and fast when it comes to inverse kinematics and collisions for humanoid actors. Clothing? One of the best. Wheeled/tracked vehicles? Meh, but ok. Aerodynamics? Fluid dynamics? Nope. Unfortunately, that's DCS's bread and butter. Meaning: a migration to UE would force ED to entirely recreate their existing physics modelling in UE's environment.  

While it would be nice (especially for chopper heads like me) to have UE-level visual quality close to the ground, visuals - although welcome - are not DCS's main selling point. Switching to UE as game engine could improve visuals, and may be entirely possible in theory. It would incur a hefty development and time investment (they need to migrate DCS itself, all assets, plus every single module), plus a licensing fee. On the other hand, ED could gain access to UE's toolchain, so there may be some benefit to offset cost, but I doubt that it will come remotely close to the required investment. On the whole, I'm seeing a very difficult business case for such a migration: how many additional units will a UE-based DCS move compared to an evolution on DCS's inhouse engine? Crucially, is that enough to finance the migration and pay back investors who lend the funds? Just because something is possible to do does not mean it's financially viable. I'm sure the kind folks at EDare running the numbers regularly (and are also looking at other engines like Crytek etc.). Since they are a business, they'll go where the money is, and where they are going is answering this question. Let's see what happens.


Edited by cfrag
  • Like 4
Link to comment
Share on other sites

On 7/24/2021 at 2:20 PM, Xilon_x said:

smoke you can't see fire you can't see water you can't see gas you can't see it and all very clean and all unreal.

 

On 7/24/2021 at 2:32 PM, Xilon_x said:

REAL METEORITE REAL WORLD NOTE: GAS AND METEORITE TRAIL AND BRIGHTNESS. CRATER NOTE: huge pit debris and fire and flame smoke.

ED is obviously doing its own thing, but in terms of UR5, I think your comparing 2 things that were never meant to be compared. The UR5 demo was meant to simply show off its capacity to generate a destructible photo realistic world on the fly, not how it compares to actual video of a meteorite falling to earth.

I think the question you should be asking is whether UR5 can be used to generate smoke, fire, water and, and, and.... The short answer is yes, ...it can.


Edited by Callsign112
  • Like 1
Link to comment
Share on other sites

It's so funny reading these responses. UE5 is miles better than most other engines period. And yes they CAN do large scale rendering. If ED wanted to move into future development with DCS2.0 or whatever they want to call it.. without having an entire team try and rewrite new engines to keep up with technology they probably don't even have time to learn then they should probably have a small team start working on the next iteration... 

It's really not hard to find all of this information.. lmao.. 

https://www.unrealengine.com/en-US/blog/unreal-engine-5-offers-significant-new-potential-for-the-simulation-industry

User made clip... 

 


Edited by HoBGoBLiNzx3
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...