not planned Unreal Engine 5 viable for DCS?

[WinterH]

2 hours ago, 3WA said:

Never say never. 

Never

2 hours ago, 3WA said:

Especially on a 30+ year old engine.

Render engine has, in fact, changed in 2015ish to EDGE (shorthand for Eagle Dynamics Graphics Engine, duh :p) Switching from one in house render engine to another took years. Literally years. And brought about plethora of new bugs, as such things do. Many ED and especially 3rd part products had features that worked fine got broken, and some took forever to be reintroduced. But, DCS did change engines over half a decade ago. And kept adding into/improving it, like new weather, much improved lighting, PBR, and if all goes well, multicore and Vulcan will both change things up again for better at some point in future that is nearer than what is essentially throwing everything to trash and developing a new sim.

Will it have UE5 visuals? Probably not. Does it have to? As long as it keeps being the most well developed consumer combat/flight sim with greatest flight models and a whole stable of meticulously created stable of aircraft modules, I couldn't care less that it maybe isn't as pretty as hypothetical UE5 sim. Not to mention we have a dynamic campaign coming which may or may not have engine dependent bits in it, but those things take years to develop, according to legend even dragging Spectrum Holobyte down back in the day, but are also basically the holy grail for sims.

We'll see if an UE5 sim that is a direct rival in fields like systems and flight fidelity and ability to develop multiple aircraft from all over the history and all over the world comes. If it works, I may even be among the first to get it. But with so many modules here it'd be more "get it" and less "jump ship" unless it has a lot more going on than pretty graphics, which I doubt will happen. If it happens, heck yeah, the more choices the better for us end users. ED can then decide whether it's time to jump ship for them or not then and there. But right now, or even in foreseeable future, trying to port DCS to a foreign engine is a fools errand. Both for ED and for us end users.

Also I still don't believe that UE5 means an ARMA level, even remotely so infantry experience will be possible in a flight sim. They have competing engineering choices/requirements as genres. Buildings with interiors will remain a counterproductive luxury for a flight sim for the foreseeable near future imo for example. As for the vehicles, again high fidelity vehicles have relatively little to do with game engine and more to do with limited resources who are better spent on doing work on more essential things for the flight sim.

I feel like this whole thread is "I've seen tech demos and bought into all marketing hype for the product" and not a whole lot more really.

[3WA]

13 hours ago, WinterH said:

I feel like this whole thread is "I've seen tech demos and bought into all marketing hype for the product" and not a whole lot more really.

Lol, you need to read everything about it.  This is a whole NEW LEVEL of engine.  Especially Lumen and Nanite.

It leaves UE4 in the dust.

To me, the main advantages are, it can use the raw power of new cpus and gpus.  The graphics are astounding.  We can finally get FPS and Vehicle modules into the sim.  We can make our own large maps quite easily.  It can make maps as big as Earth itself (up to 88 million km2).  It has a Google Earth plugin if you want to go the easy route.  We can make smaller maps for helicopters and vehicles / fps packed with high quality, enter-able buildings.  Personally, I wouldn't mind some homemade fantasy maps like we used to have in the old days for FPS sims.  Adding realistic Flight Sim to them would be AWESOME!  Oh, it also has destructible meshes, so tanks and such really do explode.  This is literally the Future, starting now.

They already made a hardcore flight sim for the military ( I think Heatblur was involved with it ), so we know it is well capable of that.  Exact properties of the sim are not being given out, but there is a video floating about showing its beauty.  I can't remember which fighter hud they show, but it looks amazing.

I'm not bashing DCS.  It does an amazing job for what it is.  GREAT Flight Sim.  But ED should consider whether it's time to move on, seeing as what has just come out and what the future looks to hold.

I think ED has to consider whether a loss of 5% of profits is worth getting a LOT more customers and not having to deal with ancient engine development ever again.

I hope they take their time and think about it.  It would make a Hell of a DCS 3.0 in a few years, and then you would have an actual WORLD.

Edited August 1, 2022 by 3WA

[cfrag]

5 hours ago, WinterH said:

I feel like this whole thread is "I've seen tech demos and bought into all marketing hype for the product"

 

4 hours ago, 3WA said:

Lol, you need to read everything about it.  This is a whole NEW LEVEL of engine. 

To me it reads like a somewhat Dadaesque conversation

A: "Look at Engine X. Its graphics are astounding. We need that in DCS"

B: "Wow indeed. But DCS isn't primarily about graphics"

A: "Look, it's a whole new level of graphics"

B: "DCS is about modelling systems, not graphics"

A: "I heard about a sim that's using this engine! So its possible to have these graphics"

B: "Possible doesn't mean financially attractive. DCS is about modelling systems"

A: "It can use the raw power of the CPU and GPU"

B: (blank stare, wondering what that actually means *) )

A: "And it can use really big maps"

B: (looks at the barren SA map, thinking that big isn't really the issue, but filling The Big with Interesting Things): "Uh..."

C "And the graphics are astounding! We need that in DCS!"

Ok, maybe not Dada, but definitely Beckett.

 

)* Nitpick Note:
Any app uses the CPU and GPU's raw power. A probably meant multi-threading and compute shaders to offload repetitive simple tasks to the GPU (aka "GPU computing" as used in many Crypo mining apps). Many Game Engines "can" use those. Note that actually using them is *really* advanced stuff and anything but trivial, because vectorizing an application's code is something only few people can do reliably: a single overlooked linear dependency (e.g., a shared variable) will break the entire vector - and that is *with* auto-vectorization built into your compiler. There's a big difference between marketing "can" and real-world "do".)

Edited August 1, 2022 by cfrag

[WinterH]

1 hour ago, 3WA said:

We can finally get FPS and Vehicle modules into the sim.

Sigh....  roight m8, 'ave a good one! I'm out.

[Silver_Dragon]

7 hours ago, 3WA said:

Never say never.  Especially on a 30+ year old engine.

There actually has been a massive flight simulator built on UE5 already, but so far it is only being sold to militaries.

So, ED has a chance to get in first for a public flight sim.

EDGE engine has no a 30+ engine. EDGE was release on 2015 on 1.5 Dx11 Engine and New map technology was release on 2.5, and UE5 has no a free public flight sim, has a engine when you require pay regalies and feeds to a external company with losing your hability to control and losing your product to perpetuity, force to sell them on epic store... (losing your marketplace).

DCS simulate military systems, no making funny sceneries. If your UE engine has no capable to simulate a radar sweep, a flight parameter or a weapon capablitiy, the rest has a trash. ED dont go to fire your teams and losing your develop history, to convert on a subsidiary of a game engine builder company only by a "that pretty engine has the future".

Some remember de Outerra MSFS/DCS/XP killer? And ED has building a Whole earth technology as MSFS/XP, the map sizes will no be a problem on a future.

Edited August 1, 2022 by Silver_Dragon

[MAXsenna]

39 minutes ago, Silver_Dragon said:

DCS simulate military systems, no making funny sceneries. If your UE engine has no capable to simulate a radar sweep, a flight parameter or a weapon capablitiy, the rest has a trash. 

Thank you Silver_Dragon! 

Everyone loves fantastic graphics, but in the end, this is what it it boils down to, and why we fly DCS.  

Cheers! 

[Timmy51]

It's so funny to see some guys calling others, who are trying to discuss the future of their game, "forumites waving brochures".

On the topic, I worked in a game dev company as an SWE and I know it's hard to find a good (hence expensive) one for whom you'll have to give them a lot more time to learn your in-house engine and tools compared to others. If you are not Ubisoft, Rockstar, or someone else with a big stash of money, it's hard to keep up with other engines.

So there are not only visual aspects of the new engine, it's also about tools and support.

But being realistic, there are probably stakeholders who won't be happy if they will have to pay for yet another engine upgrade if they don't have direct competitors right now.

[Silver_Dragon]

ED work with Ubisoft as a develop studio under control of a publisher on the LOMAC times, and after release your product, Ubi dont like continue de series and intent close them. ED release FC as a private inversion and repurchase the licences from Ubi, convert them on a independent study and enterprise. 

[WinterH]

3 minutes ago, Timmy51 said:

It's so funny to see some guys calling others, who are trying to discuss the future of their game, "forumites waving brochures".

Because that is precisely what they are doing.

3 minutes ago, Timmy51 said:

it's also about tools and support.

Tools you probably already have in house for your specific purposes as a FS dev, and no need for external support for the engine vs almost certainly no tools for your specific niche from vendor as well as a dependency on them for support.

Not to mention probably having to mostly/fully rewrite SDK/API ED shares with licenced 3rd party devs so they can develop their modules to interact with the underlying air combat/sensors/weapons/aerodynamics structure the sim provides. As a dev I'm pretty sure you can at least somewhat relate to "fully rewrite: oh hello things getting broken beyond recognition, new bugs breed out of nowhere etc!"

Besides, I'm fairly sure ED would have to drill in any new dev due to these proprietary foundations of the sim regardless of the engine used.

This isn't like, say, CDPR trying to (and eventually abandoning) stick to their engine vs UE5 for mainstream AAA games. Though, even in that sort of market, there have been horror stories of going to new engines that weren't specifically made for the genre negatively impacting franchises, ie people trying to hammer out RPGs from Frostbite. Now, UE5 is probably a lot more of a univerrsally supportive engine than that example, but there's still the fact that what makes DCS, DCS is a lot more than just graphics engine, and porting that leviathan to whole another engine would be an adventure to say the least imo.

[cfrag]

Just now, Timmy51 said:

So there are not only visual aspects of the new engine, it's also about tools and support.

That's indeed a very good point. At this point I assume that ED have built a complex, dedicated tool chain around the DCS ecosystem that integrates the various parts: world grid/map, textures, models, animations, audio, UX, networking, licensing and verification (modules, assets), code versioning, bug tracking, publishing pipeline with multiple target builds (at least 4: Steam Pipe and native, both beta and release trunks). Migrating a dedicated tool chain with home support to a third party is not only costly, but comes with risk. What if your provider decides to migrate their CVS to a new system that you aren't familiar with and that may introduce some subtle incompatibilities? What if the chain's physics provider suddenly deprecates your current IK system and forces you to reverse-fit all your models to a new one before December 31st? I remember when during a project I was consulting on, the Engine's provider (on contract!) scrapped the entire 'material' (properties that describe how stuff looks to the camera. Think 'rough copper' or 'dull plastic') branch of the render pipe-line and sold it to a third party. Yeah, that killed the project, because the investors didn't think it sustainable to either re-do all materials for the (extensive) asset pool or enter into a third party licensing deal with an unknown entity. 

Since ED have their own inhouse toolchain - warts and all - entering into a long-term agreement with a provider for these services can be risky on top of the incurred migration cost.   

[HoBGoBLiNzx3]

23 minutes ago, WinterH said:

Because that is precisely what they are doing.

Tools you probably already have in house for your specific purposes as a FS dev, and no need for external support for the engine vs almost certainly no tools for your specific niche from vendor as well as a dependency on them for support.

Not to mention probably having to mostly/fully rewrite SDK/API ED shares with licenced 3rd party devs so they can develop their modules to interact with the underlying air combat/sensors/weapons/aerodynamics structure the sim provides. As a dev I'm pretty sure you can at least somewhat relate to "fully rewrite: oh hello things getting broken beyond recognition, new bugs breed out of nowhere etc!"

Besides, I'm fairly sure ED would have to drill in any new dev due to these proprietary foundations of the sim regardless of the engine used.

This isn't like, say, CDPR trying to (and eventually abandoning) stick to their engine vs UE5 for mainstream AAA games. Though, even in that sort of market, there have been horror stories of going to new engines that weren't specifically made for the genre negatively impacting franchises, ie people trying to hammer out RPGs from Frostbite. Now, UE5 is probably a lot more of a univerrsally supportive engine than that example, but there's still the fact that what makes DCS, DCS is a lot more than just graphics engine, and porting that leviathan to whole another engine would be an adventure to say the least imo.

That is why he clearly said, along with others. That if you're going to create a NEW game in the future. It would likely be ideal to find an current engine such as UE5. 

People keep using the fact that DCS isn't about "visual fidelity" yet they are playing DCS and not BMS... which by a lot of accounts does things better than DCS in those realms but completely falls short in terms of content, graphics, accessibility for new users, etc. 

Graphics ARE important. Especially for VR users. That doesn't mean they have to be the MOST important thing in terms of developing their game. That might not even be the most important thing now. Not everything has to be compared to NOW because we aren't talking about now. 

This is a prime example of what I mean.. 

A MR/VR cockpit experience that right now may seem unfeasible as a standard but in 15 years this might be at the standard that VR is right now in flight sims... so, being that many features everyone is claiming we will get.. like multicore support and vulkan (wasn't that promised last year?) It's almost q32022... Never the less, MCS was a basic feature in BF3... or any other game around that time.. It's 2022 man. Now I'm really looking forward to when it's released because I enjoy the game and it's content. Be that as it may, that doesn't shield them from criticism, or discussions on future implementations. So it's not far fetched to think about what is happening now.. and throw 15 years of even bigger, more complex tech, at them and expect anything different without massive amounts of people. So most people would come to the conclusion that if you start a new game, on an engine that supports most of these features via plugin (yes all the coding, physics, etc need to be done from scratch, we know this) but it would be a NEW game. Not a port. 

Don't take this as a cheap jab at ED or their developers. It's hard enough for AAA titles and their engineers to put out a quality product with the resources and money they have at their disposal. DCS does do a fantastic job, but they aren't perfect, nor would they claim to be. This is just the reality of the speed of technological evolution. Can they keep up? Do they have the resources to keep up? I would like to think so, but I would also like to think they would have to go out of house to get some help in a lot of aspects of it because of the shear size of making a modern title let's say in 2030... Games will likely be 500gb's with 12k real time 3D textures with "world scale" technology to import and manipulate.

lol look at Falcon 4.0 when it came out. It was what? You needed a 1.8ghz CPU, 512mb or ram, 700mb of storage, and a dx7 graphics card.. DCS pushes over 20GB of VRAM in todays world in some cases... so you do the math.. 

I don't think DCS will be able to just continue on forever as the same title on the same engine. So don't berate the guy for wanting to have a logical conversation about the future of the game. 

 

[cfrag]

1 hour ago, HoBGoBLiNzx3 said:

look at Falcon 4.0 when it came out. It was what? You needed a 1.8ghz CPU

Oh, these young Whippersnappers.

In '98, a top of the line gaming pc had a Pentium II, clocked at 0.45 GHz running Windows '98 (ugh!) or a G3 Mac based on a PowerPC at 0.4 GHz running, uh, Apple's System 8 (? - I don't think Macs where much good for gaming, but F4 was available on that platform). At the time, the Internet barely existed (anyone remember IRC?), AOL was something people seemed to like, and they watched TV (mainly a series called "ER", so it seems) because you couldn't much go on-line (what with 56k modems and stuff. Yeah, that was a 0.056Mbps dial-up connection, metered by the minute). Good times. I had a life back then   

Edited August 1, 2022 by cfrag

[HoBGoBLiNzx3]

10 minutes ago, cfrag said:

Oh, these young Whippersnappers.

In '98, a top of the line gaming pc had a Pentium II, clocked at 0.45 GHz running Windows '98 (ugh!) or a G3 Mac based on a PowerPC at 0.4 GHz running, uh, Apple's System 8 (? - I don't think Macs where much good for gaming, but F4 was available on that platform). At the time, the Internet barely existed (anyone remember IRC?), AOL was something people seemed to like, and they watched TV (mainly a series called "ER", so it seems) because you couldn't much go on-line (what with 56k modems and stuff. Yeah, that was a 0.056Mbps dial-up connection, metered by the minute). Good times. I had a life back then   

 

Dude IRC existed and was still used in like 2015! lol!!! I actually really liked it during the CS and R6 Rogue Spear days... 

I hated when mom made you get off the computer cause she had to make a call. You were the coolest if you had TWO phone lines so you didn't have to get off AOL instant messenger.

Edited August 1, 2022 by HoBGoBLiNzx3

[3WA]

8 hours ago, Timmy51 said:

It's so funny to see some guys calling others, who are trying to discuss the future of their game, "forumites waving brochures".

Yeah, this is the usual here when you try to discuss the future.  You get used to it.   Carry on.

@HoBGoBLiNzx3 Awesome video!  I had not seen that one yet.  I'm glad I have not bought VR yet.  I may have to look at that goggle and consider building a pit.  This just shows how we are sailing faster and faster into the future.  Which is a good example of why there should be companies that concentrate just on Engines, and other companies concentrate on making sims for those Engines.  The workload is getting to be too much to keep up with.  And it's going to get faster and faster and faster.

Edited August 1, 2022 by 3WA

[3WA]

Lol, as I said, you get used to it.

And of course, stuff like this is just people trying to get the thread closed.  Just ignore it.

__________________

This was posted on Mudspike.  It shows a lot of what the UE5 engine is capable of.  You should pay special attention to the demo of "nanite" technology.  It is really incredible when you realize this demo is being played on a Sony Playstation 5.  I don't think my gaming computer would be anywhere near capable of this in a regular engine.  The amount of triangles being thrown around is unbelievable.

"Niagra", the new particle system, is also very interesting as well, backed by the Chaos physics engine.  Imagine what we could do with this when Mk-82 iron bombs hit the ground, throwing frag everywhere.  Destruction of buildings, infantry being taken out, tires and windows blowing out, craters being formed in the terrain, etc.

 

Now think of what we could do with this in large worlds, like flight sims.

Can you imagine DCS running on this?  If this can run on a Play Station, what can it do on a modern gaming computer?

Edited August 1, 2022 by 3WA

[cfrag]

12 hours ago, 3WA said:

"Niagra", the new particle system, is also very interesting as well, backed by the Chaos physics engine.  Imagine what we could do with this when Mk-82 iron bombs hit the ground, throwing frag everywhere. Destruction of buildings, infantry being taken out, tires and windows blowing out, craters being formed in the terrain, etc.

This is what happens when people's confidence about a subject far exceeds their expertise, a.k.a. Dunning-Kruger effect. In Graphics Engines, particle effects are purely visual (e.g., light emitters, smoke, sprites), even if they use parts of the physics engine to make the movement of these particles behave in way that looks realistic (they can bounce off the floor). For performance, a particle system does not, can not, interact with game objects (aka 'actors', they model the world), it's purely eye candy. So a particle system will not be used to achieve "buildings, infantry being taken out, tires and windows blowing out, craters being formed in the terrain, etc". That's the job of game objects. Don't "imagine what we could do". For crying out loud, this is not 4chan. Particle effects can do exactly nothing to model an Mk-82's damage. It only can make it look prettier. That's why "Niagara's" full name is "Niagara Visual Effects" in UE's docs. This is not a secret. We don't have to imagine what it can do, we frigging know.

It may be helpful if people actually invested some time into research before spouting off. 

12 hours ago, 3WA said:

of course, stuff like this is just people trying to get the thread closed

Nobody wants that, you are not being persecuted. We just hope that people learn and stop spouting nonsense and uncritically regurgitate marketing blurbs.

12 hours ago, 3WA said:

Just ignore it.

Ah. Well. Everyone seemingly has their motto.

Edited August 2, 2022 by cfrag

[WinterH]

19 hours ago, HoBGoBLiNzx3 said:

People keep using the fact that DCS isn't about "visual fidelity" yet they are playing DCS and not BMS

I'll keep this one short as it treads pretty close to no-no territory in forum rules but, for me personally, the biggest reason of not flying BMS is I don't care an iota about F-16. I didn't even have for DCS until very, very recently, and the whole reason I got it for DCS was impulse buying at a sale and FOMO  But I'll admit if we talk majority, their reason probably lies in graphics. And yet, the comparison is feel odd as DCS' graphics are already pretty strong today.

19 hours ago, HoBGoBLiNzx3 said:

Graphics ARE important.

To a degree only though.

19 hours ago, HoBGoBLiNzx3 said:

look at Falcon 4.0 when it came out. It was what? You needed a 1.8ghz CPU, 512mb or ram, 700mb of storage, and a dx7 graphics card..

Weeeelll... you kinda didn't, because 1.8 GHz CPUs didn't exist back then :)) I think the fastest were around 400 MHz maybe in 98? I had a Celeron 300A as my first PC back then, the memories...

As for the rest of your post... well... then people in this thread misunderstand each other at some level, people who appear resistive like myself, I'd say, react to the idea of "let's switch to UE5 wooohooo!!", and I frankly still think that is the idea of most of the pro-UE5 camp here. On the other hand, if we are talking about a far future, 15 years or so like you say, why even discuss it now? 15 years ago DCS didn't even exist as "DCS", and since then it got multiple major engine changes/upgrades. 15 years from now, pretty sure UE5 will be a thing of history too. I can agree yes, a lot will change by then. DCS may or may not still be around, and it may have moved to an entirely different engine even in such a long time, I see no way it'll remain as it is by then. But then, it already didn't stay what it was during the time I've been using it, which is more like 9ish years.

12 hours ago, 3WA said:

And of course, stuff like this is just people trying to get the thread closed.  Just ignore it.

So, I've just learned UE5 is apparently a religion huh?  You need to a- revere its miracles, b- ignore those who proclaim that said miracles defy logic and maybe questioned

[Mars Exulte]

3 hours ago, WinterH said:

So, I've just learned UE5 is apparently a religion huh?  You need to a- revere its miracles, b- ignore those who proclaim that said miracles defy logic and maybe questioned

  It even has holy texts!

[BIGNEWY]

thread cleaned a reminder rules can be found at the top of the forum. 

This thread has run its course now and will be closed.