Do you think we'll ever see DCS 2?

[Tree_Beard]

Or will ED just continue to duct tape new features to the ageing original game indefinitely?

I'm just speculating but I really get the impression that the game code has become extremely convoluted over the last 20 years with new systems layered over old systems, made by different people in different countries, years apart and with different priorities... 

I think that might be the main reason why is takes so long to do anything besides add new modules... I bet there are problems so deeply hidden in the spaghetti that no one even knows where to start looking for a solution. 

Maybe a fresh start, as easy as that is to say, might be what the franchise needs in the long run. 

 

[Rudel_chw]

We already had DCS 2.0 several years ago, on 2014 or 2015   … maybe you meant DCS 3

[Tom Kazansky]

3 minutes ago, Rudel_chw said:

We already had DCS 2.0 several years ago, on 2014 or 2015   … maybe you meant DCS 3

I guess he does not mean DCS 3 neither, because we will get that as we got DCS 2. Guess he meant something like "DCS II"

Edited July 31, 2023 by Tom Kazansky

[Rudel_chw]

2 minutes ago, Tom Kazansky said:

I guess he does not mean DCS 3 neither, because we will get that as we got DCS 2. Guess he meant something like "DCS II"

 

Oh .. at least for myself, I don't want that .. I want to keep flying the modules that I've already purchased, not have to purchase them again (like it will be for the well-known civ flight sim, where its 2020 version will be superseded by a 2024 one, that everyone will have to pay again for).

I enjoy DCS as is, and the updates (better clouds, multicore support, better propeller & rotors, etc) are nice enough for me, even if they take years to develop.

[Tom Kazansky]

12 minutes ago, Rudel_chw said:

 

Oh .. at least for myself, I don't want that .. I want to keep flying the modules that I've already purchased, not have to purchase them again (like it will be for the well-known civ flight sim, where its 2020 version will be superseded by a 2024 one, that everyone will have to pay again for).

I enjoy DCS as is, and the updates (better clouds, multicore support, better propeller & rotors, etc) are nice enough for me, even if they take years to develop.

I'm torn between what you posted and the dream that a new DCS II could run close to flawlessly, almost free of bugs and with superb performance. But I'm afraid that's not realistic.

However, from time to time I'm figuring for myself, how much money I would spend for the same/DCS II modules I have bought already in DCS 2.X

Edited July 31, 2023 by Tom Kazansky
typo

[upyr1]

Due to the nature of DCS I really don't expect them to toss out the code and start from scratch. 

[Mars Exulte]

8 hours ago, Tree_Beard said:

Or will ED just continue to duct tape new features to the ageing original game indefinitely?

 How exactly do you think sequels are usually made? They do not start from scratch everytime, even if they change engines, because that's a waste of time and money. The only difference between DCS and others is the ''sequels'' are released directly as a patch, allowing all your purchases to carry over.

8 hours ago, Tree_Beard said:

I'm just speculating but I really get the impression that the game code has become extremely convoluted over the last 20 years with new systems layered over old systems, made by different people in different countries, years apart and with different priorities... 

 Probably.

8 hours ago, Tree_Beard said:

I think that might be the main reason why is takes so long to do anything besides add new modules... I bet there are problems so deeply hidden in the spaghetti that no one even knows where to start looking for a solution. 

Probably.

8 hours ago, Tree_Beard said:

Maybe a fresh start, as easy as that is to say, might be what the franchise needs in the long run. 

  That is effectively what is gradually happening. The reason multithreading support took so long is it required a complete rewrite of the executable. The reason Vulcan is taking so long is it's a similar major rewrite of the graphics side. It's like replacing the engine in a car for more power, you can't just drop a V8 into a Volkswagen without also redoing the frame, drivetrain, etc etc. So, these big changes DO involve substantial rewrites.

 Putting the game in a new box and renaming it won't fundamentally change how it all works. And throwing it in the trash and starting over would simply kill.the company, even if it didn't it would be 5-10 years before you saw anything.

 

 I'm a huge critic of the way they do their stuff around here these days. Their issue is one of priorities and almost definitely their corporate culture. It's not the engine or resources, and burning everything down won't solve anything.

[draconus]

10 hours ago, Tree_Beard said:

Or will ED just continue to duct tape new features to the ageing original game indefinitely?

There's very little left from the original and much of that stuff doesn't even need to be rewritten as they are simple lua codes and databases. You just want to ignore all the gfx engine overhauls, MT, VR, new map tech, clouds and what not?

10 hours ago, Tom Kazansky said:

I'm torn between what you posted and the dream that a new DCS II could run close to flawlessly, almost free of bugs and with superb performance. But I'm afraid that's not realistic.

Manage your expectations. DCS needs all that CPU/GPU/(V)RAM/SSD because there's so much to compute and digest. It's always compromise between performance and quality/content - that's why you can't compare it to other games, not even flight sims.

[Mr_sukebe]

Unless the comments like “spaghetti code” is from the devs themselves, I’d be careful about the assumptions.

After all, most of us have little genuine understanding of building applications and systems and the implications and typical processes.

For all we know, the MT update is a clean sheet rebuild of parts of the code and unless a dev says otherwise, it’s just speculation.

 

What I do know is that I also have other aerial combat sims and a couple of the best know Civ sims.  From my perspective, DCS:

- looks better, eg significantly more detailed than the other combat sims.  I was out in the big Civ sim yesterday and in helicopter, low to the ground, many of the buildings like like melted marshmallows.  Sure, it looked great at 10.000’, but it was awful low down.

- since the MT update, runs better than the others in VR, even more so with dynamic foveated rendering.  I’m amazed that there’s not more noise about the big Civ sim for how badly it runs.  Quite clearly it needed MT.  At least DCS players won’t have to wait until 2024 and then pay again to get it

- we know that Vulkan is coming and will hopefully further enhance things

- I assume that I’m not alone in loving how much effort it takes to fly an aircraft in DCS.  I find it quite confusing to jump into other sims where I can’t touch things in the cockpit

 

In short, I also own most of the other flight sims currently available, and for me, they look worse, run worse and feel a bit pants to be in when compared. 
Don’t get me wrong, I really want Vulkan and other things (eg AA accuracy to be resolved), but generally, I think that we’re in a great place

 

 

 

 

 

[BIGNEWY]

12 hours ago, Tree_Beard said:

Or will ED just continue to duct tape new features to the ageing original game indefinitely?

I'm just speculating but I really get the impression that the game code has become extremely convoluted over the last 20 years with new systems layered over old systems, made by different people in different countries, years apart and with different priorities... 

I think that might be the main reason why is takes so long to do anything besides add new modules... I bet there are problems so deeply hidden in the spaghetti that no one even knows where to start looking for a solution. 

Maybe a fresh start, as easy as that is to say, might be what the franchise needs in the long run. 

 

I understand what you are trying to say, however

DCS is complex, and the complexity grows as time goes on, updates are made, new features are added, this cycle will continue. 

Starting from scratch would be just stupid at this point. To get to this level of simulation takes years of work. 

The code is complex, and as Kate once said spaghetti like, but code gets rewritten as needed. 

I know your "duck tape" comment is in jest, but DCS has never been more stable than it is now, the testing being done by volunteer closed testers and our QA has much improved over the years.

So I would say as I always do, you need to be patient and try to understand the complexity the teams are working with, that complexity will always be there, it is the nature of producing simulations at this level and does take time. 

thanks

[draconus]

25 minutes ago, Mr_sukebe said:

Unless the comments like “spaghetti code” is from the devs themselves, I’d be careful about the assumptions.

This is actual word the ED's COO used:

https://stormbirds.blog/2020/02/20/behind-the-scenes-of-dcs-world-development-from-their-coo-co-founder/

[Tree_Beard]

15 hours ago, Mars Exulte said:

 Putting the game in a new box and renaming it won't fundamentally change how it all works. And throwing it in the trash and starting over would simply kill.the company, even if it didn't it would be 5-10 years before you saw anything.

I'm not suggesting throwing everything in the trash and starting from nothing. What I'm suggesting is that without some sort of a reset where you take the things that work well and you consolidate them into a new, more modern, more efficient framework, then the complexity bloat is going to keep getting worse over time and the pace of development will get slower and slower as problems compound on other, older problems. 

I really think there may be a point sometime in the (near) future where it just becomes too cost prohibitive to do anything more and the development of the core game is functionally deadlocked. We might even be reaching that point now. 

[SkateZilla]

Ummm..
DCS 2.0?

We are on 2.5.

2.0 was an Alpha of the New Terrain Engine in 2015


As for a completely new title, "DCS II", doubt it,
there's a reason developers are flocking to the Base Game as Service w/ new content and features added through patches model.

It's easier to maintain and cheaper to maintain.

DCS World in 2023 is not the same as DCS World in 2012.

Outside of standard legacy code that has no reason to change, and a few UI Elements, there's nothing left of the original DCS world 1.1.2.1 core.

We've moved from DX9 to DX11, we've changed terrain and atmosphere engines, new sounder engine, new weather engine, new physics and wake modelling, rewritten core to utilize Multi-threading, soon to have Vulkan as the new graphics API Retiring DX11 and the problems that came with it, Soon to Model the entire globe.

This isn't Madden or FIFA, ED isnt' gonna update rosters and put out DCS '23 with new cover art every year.

DCS Simulates an Environment on a Scale that No other "Game" out today does.

Edited August 1, 2023 by SkateZilla

[MiG21bisFishbedL]

What would they gain?

Given all of the content already in DCS, a DCS 2 would be a hard sell to utilize unless you just port that content immediately into DCS 2. That would be a monumental undertaking.

[draconus]

10 hours ago, Tree_Beard said:

I really think there may be a point sometime in the (near) future where it just becomes too cost prohibitive to do anything more and the development of the core game is functionally deadlocked. We might even be reaching that point now. 

This is pure conjectrue. All is assumed to be working fine unless a bug is known or reported. They keep growing and improving - which means it works for them and for us.

10 hours ago, SkateZilla said:

We are on 2.5.

I'm on 2.8.7. You should update Here's nice app:

[Hoirtel]

23 hours ago, Mr_sukebe said:

Unless the comments like “spaghetti code” is from the devs themselves, I’d be careful about the assumptions.

After all, most of us have little genuine understanding of building applications and systems and the implications and typical processes.

For all we know, the MT update is a clean sheet rebuild of parts of the code and unless a dev says otherwise, it’s just speculation.

 

What I do know is that I also have other aerial combat sims and a couple of the best know Civ sims.  From my perspective, DCS:

- looks better, eg significantly more detailed than the other combat sims.  I was out in the big Civ sim yesterday and in helicopter, low to the ground, many of the buildings like like melted marshmallows.  Sure, it looked great at 10.000’, but it was awful low down.

- since the MT update, runs better than the others in VR, even more so with dynamic foveated rendering.  I’m amazed that there’s not more noise about the big Civ sim for how badly it runs.  Quite clearly it needed MT.  At least DCS players won’t have to wait until 2024 and then pay again to get it

- we know that Vulkan is coming and will hopefully further enhance things

- I assume that I’m not alone in loving how much effort it takes to fly an aircraft in DCS.  I find it quite confusing to jump into other sims where I can’t touch things in the cockpit

 

In short, I also own most of the other flight sims currently available, and for me, they look worse, run worse and feel a bit pants to be in when compared. 
Don’t get me wrong, I really want Vulkan and other things (eg AA accuracy to be resolved), but generally, I think that we’re in a great place

 

 

 

 

 

I have to agree with this, usually disappointed with perf and appearance of the "big civ sim", among other things. I think DCS is looking and running great, not perfect but the hard work is clear and progressing well.

I do expect there will be a DCS 3.0 at some time, and I expect it will be awesome. Think of the things we do know about in one package... completed multithreading, completed vulkan, world map. Include some new ATC, maybe a few new GUI options, enhanced weather (just wind needed really). What a sim this would be...

[Raven (Elysian Angel)]

13 hours ago, Hoirtel said:

I do expect there will be a DCS 3.0 at some time

Perhaps 2.8.X to flesh out MT until ED is happy with it; 2.9.X to flesh out Vulkan until ED is happy with it, then 3.0.X as a stable version with MT + Vulkan working optimally? That would be my guess  

[wilbur81]

On 8/1/2023 at 3:22 AM, BIGNEWY said:

but DCS has never been more stable than it is now

This is 100% true... 

[Hoirtel]

12 hours ago, Raven (Elysian Angel) said:

Perhaps 2.8.X to flesh out MT until ED is happy with it; 2.9.X to flesh out Vulkan until ED is happy with it, then 3.0.X as a stable version with MT + Vulkan working optimally? That would be my guess  

Yeah, an entirely new engine in full release would surely qualify for a major version number change. I think the world map will be linked to vulkan, would make sense that this comes at a similar time. 

Edited August 3, 2023 by Hoirtel

[draconus]

13 hours ago, Raven (Elysian Angel) said:

2.9.X to flesh out Vulkan until ED is happy with it, then 3.0.X as a stable version with MT + Vulkan working optimally

It works a bit different. 3.0 will be OB first, then whenever they decide it's fine, they'll move the then-current beta to stable.

Why would users care what version it will be anyway? I'm totally fine if the next beta is 4312.3.21.0009321.

[Raven (Elysian Angel)]

11 minutes ago, draconus said:

It works a bit different

You misunderstand. Not that it matters, but with “stable” I meant working properly without many of the issues that plague both branches of current MT, such as the stuttering and memory leaks.

Aka “stable” how it is defined normally, not just “DCS lingo”…

[upyr1]

On 7/31/2023 at 11:21 PM, Mars Exulte said:

 How exactly do you think sequels are usually made? They do not start from scratch everytime, even if they change engines, because that's a waste of time and money. The only difference between DCS and others is the ''sequels'' are released directly as a patch, allowing all your purchases to carry over.

On 7/31/2023 at 2:39 PM, Tree_Beard said:

This is the first problem with anyone expecting to see DCS toss everything out and write new code. It would be prohibitively expensive to do this. The second issue is that it is wrong to assume new code will be more stable than old code. All that you can be sure is going to happen with new code is that you'll have a bunch of new stuff to debug. 

[Silver_Dragon]

On 8/2/2023 at 3:01 AM, MiG21bisFishbedL said:

What would they gain?

Given all of the content already in DCS, a DCS 2 would be a hard sell to utilize unless you just port that content immediately into DCS 2. That would be a monumental undertaking.

This already happened when the modules were changed from version 1.0 to 2.0, in fact, the module system and its construction is another building block of DCS World and has nothing to do with the graphics engine.

In fact, Vulkan would come in a 3.0 version rather than a 2.9 version, the change is as important as the jump to 2.0 with Dx11.

[SkateZilla]

3 hours ago, Silver_Dragon said:

This already happened when the modules were changed from version 1.0 to 2.0, in fact, the module system and its construction is another building block of DCS World and has nothing to do with the graphics engine.

In fact, Vulkan would come in a 3.0 version rather than a 2.9 version, the change is as important as the jump to 2.0 with Dx11.

We had DX11 in 1.5, which was out before DCS 2.0 by a whole 2 months, Both Co-Existed for 3 years until DCS 2.5.
DCS 1.5 Introduced DX11/EDGE, DCS 2.0 Alpha introduced the New T4 Terrain Engine / Structure., DCS 2.5 was the Major Release the Unified DX11/EDGE/T4

That being said, ED has already said the Alpha branch is likely to not happen again, sooooo
2.9.x would likely be something else, new feature or module wise, Vulkan would be ideal for the next major version.
Don't be surprised if 2.9.x ends up simply being MT binaries take over, no more ST Mode.

But as far as subversions and revisions go. They usually coincide with major modules (Mainly Aircraft).
We have plenty of AC in the Pipe that look close enough to be 2.x releases.
 

4 hours ago, Raven (Elysian Angel) said:

You misunderstand. Not that it matters, but with “stable” I meant working properly without many of the issues that plague both branches of current MT, such as the stuttering and memory leaks.

Aka “stable” how it is defined normally, not just “DCS lingo”…

MT also states that it's a MT PREVIEW on the main screen regardless of branch.

[Silver_Dragon]

14 minutes ago, SkateZilla said:

We had DX11 in 1.5, which was out before DCS 2.0 by a whole 2 months, Both Co-Existed for 3 years until DCS 2.5.
DCS 1.5 Introduced DX11/EDGE, DCS 2.0 Alpha introduced the New T4 Terrain Engine / Structure., DCS 2.5 was the Major Release the Unified DX11/EDGE/T4

That being said, ED has already said the Alpha branch is likely to not happen again, sooooo
2.9.x would likely be something else, new feature or module wise, Vulkan would be ideal for the next major version.
Don't be surprised if 2.9.x ends up simply being MT binaries take over, no more ST Mode.

But as far as subversions and revisions go. They usually coincide with major modules (Mainly Aircraft).
We have plenty of AC in the Pipe that look close enough to be 2.x releases.

I missing the 1.5, has correct.