in development Dynamic Campaign Discussion

[Dangerzone]

7 hours ago, Gierasimov said:

So far I understand DCE as a battle in progress where each time I join, the front would have changed based on each AI group tasking and outocmes, which I, as a player can influence further by receiving part of that tasking, for example here is a supply convoy, go kill it, and if I succeed, the convoy won't replenish another group that might loose it's next encounter with opposing force as a result. 

Given the state of AI, and all other parts of DCS dysfunctional attitude towards helping the user I doubt DCE will revolutionize DCS as such. 

For sure it will bring nice things, which will enable us to enjoy constant 'something to do' in otherwise empty DCS World. 

So we're possibly just looking at a turn-based single player/ co-op player dynamic campaign? 

I hope you're wrong, but I expect you're already right. This already exists kinda with DCS Liberation.

If I recall correctly I thought I read somewhere that Vulkan would need to be released first before the DCE for features planned? Additionally - over 5 years (to-date - let alone how many more years before it's released) to repeat what a few hobbyists have done already would be pretty disappointing - even if it has "AI" as part of the mission generation phase. 

With the time it's taking to write, I'd be hoping for a real time, ongoing mission that never ceases. To have a save / resume option for single players (or small co-op players), but also to have a continuous play for 24/7 multiplayer servers where things change and progress consistently with different missions. But I fear that you're probably closer to the ball. This was my expectation initially - I have lowered the bar significantly since, to a point where I have lost all enthusiasm for DCE. Really hoping to be wrong and pleasantly surprised. 

Quote

I guess I will stick to my missions and narrated campaigns, unless DCE is actually a revolution in AI, communication, and cooperation between player and friendly forces... THAT is almost unimaginable amount of (re)work in DCS needed.

I've been working on a proof-of-concept for DCS, where a multiplayer mission dynamically loads new objects during gameplay as objectives are completed. This allows the mission to feel continuous and dynamic, without the need to stop and start new missions. Over weeks or months, players could end up playing across the entire map. I don't know enough about AI, but could only imagine if I could incorporate AI real-time into this in the back end to control the spawning, and unit engagements, this would be incredible. 

The concept is that as objectives are met, new units are spawned further back, advancing the front line and using more of the map until the entire map is in play. If objectives are failed, the front line moves against the players, with new units spawning dynamically in a PVE environment. Players can capture or lose airfields along the way. Using some existing scripts, I can save the state of the units and progress, allowing the mission to continue if the server needs to be restarted. Maybe similar to what already exists, but being seamless - no stopping and restarting of missions - just ongoing. 

The proof-of-concept worked - and shows incredible promise, but it's been stalled due to ongoing bugs in DCS, performance issues with vehicle pathing, events not triggering, memory leaks, and occasional crashes when spawning new units in dynamically (which doesn't provide a crash report or a reliable way to reproduce the issue). I get the impression that ongoing releases will continue to have the same pattern of bugs, instabilities, etc for the fore seeable future. 

It's frustrating because there's so much potential with DCS, but the frequent bugs and unstable releases are a significant barrier. RJ (the creator of Overlordbot) had great ideas that could have taken DCS in amazing directions, but ED didn't engage and at least from my observations ignored RJ in the end, and now that opportunity is gone. This has made me rethink my own investment in DCS. I've come to realize that the instability and ongoing risk of bugs make it difficult to justify the work required, especially with frequent updates breaking things and needing constant maintenance, and the treatment shown towards RJ indicates a significant level of disinterest towards community developers. Others I've worked with, including developers and server admins, seem to also losing interest, or are burned out trying to keep up with the DCS bugs.

This isn't to have a go at DCS dev's. I can only imagine that they're stretched thin with too many things to do at present, especially with resources spread thin with MT and Vulkan rewrites. DCS is unique and I want to see it flourish.

Once these rewrites are finished, I really hope we'll see more stability and focus on the community's needs. I foresee DCS has more of a future in supporting the community dev's far more than developing it's own DCE. (Of course it's own DCE will still be nice if they nail it) But the community has already shown that it is more than capable of providing some incredible solutions, and quicker, (and in some cases better than what ED has provided), such as Olympus, Liberation, SRS, OverlordBot, VAICOM, Web Mission Editor, and MOOSE, etc. 

For now, for sanity reasons though, I've stepped back significantly from DCS to a point where I just tinker from time to time, in the attempt to avoid losing all enthusiasm for DCS completely. 

[MAXsenna]

8 minutes ago, Dangerzone said:

If I recall correctly I thought I read somewhere that Vulkan would need to be released firs

I don't believe this is true. I'm pretty sure they just said Vulkan would come first. Vulkan is graphics as you know. Multi threading on the other hand, which we now have, ST is gone, I'm sure was very much needed for the AI. My guess is that Vulkan will come first because it's easier and will give them time for the DCE. 

As far as I remember, it's been stated that it will not be like the Viper sim. Whatever that means. Warehousing woikd be a major part of it, and that is coming along. 

AI is probably the crux of everything. Transportation, path finding, battles, local mission generation etc. You name it.

16 minutes ago, Dangerzone said:

I'd be hoping for a real time, ongoing mission that never ceases. To have a save / resume option for single players (or small co-op players), but also to have a continuous play for 24/7 multiplayer servers where things change and progress consistently with different missions.

Without this. Why even bother, if you ask me. There are several other sims that have done this in the past. Not only fighter jets. In that helo som, I could just start a server and let it run. Hopping in and out of missions. But that's 25+ years ago. I had fun, while I don't remember how well it actually worked.

As for the rest of what you right. I agree. 

21 minutes ago, Dangerzone said:

Really hoping to be wrong and pleasantly surprised. 

Same! 

[Dangerzone]

20 minutes ago, MAXsenna said:

I don't believe this is true. I'm pretty sure they just said Vulkan would come first. Vulkan is graphics as you know. Multi threading on the other hand, which we now have, ST is gone, I'm sure was very much needed for the AI. My guess is that Vulkan will come first because it's easier and will give them time for the DCE. 

Thanks for the correction.  I may have misremembered that it would come first, not required. But even so, if it comes first - it would seem then we're still a way from the DCE.

As for MT - as I understand it, we don't have full MT yet. We have 'some'. Much of the processing still appears to be on a single thread - which would make sense given the CPU bottlenecking we still still see. I think MT rewrite still has quite a way to go before we see all the benefits (which is a good thing - it means that there should hopefully be more optimizations to come).  As I understand it, stand alone servers are still running mostly single thread? It's more some graphic stuff that's been moved to multi-thread for now?

22 minutes ago, MAXsenna said:

Without this. Why even bother, if you ask me. There are several other sims that have done this in the past. Not only fighter jets. In that helo som, I could just start a server and let it run. Hopping in and out of missions. But that's 25+ years ago. I had fun, while I don't remember how well it actually worked.

I agree - why bother. But this is also why I'm losing enthusiasm. I know my thoughts, foresight and expectation for DCS seems to often be at odd's with ED and the way they do things. Why implement DCS VOIP Radio's without the ability to feed audio in externally (which we've had with SRS for years)? Why implement Dynamic Spawn Slots, and leave out the much needed option to include them with dynamically created FARPS? ED does things that leave me scratching (or banging )my head - so I've learned not to take anything for granted with their plans, until it's actually released.  

It could be that ED's idea is to repeat what other sim's have done in the past, but the main difference 'never seen before' is that they're planning on using AI to generate the subsequent missions - and maybe they have no desire for a endless ongoing MP mission that never stops? Without them announcing things, we can only speculate, and/or go from experiences of how they've released things in the past. 

[MAXsenna]

19 minutes ago, Dangerzone said:

As for MT - as I understand it, we don't have full MT yet. We have 'some'. Much of the processing still appears to be on a single thread - which would make sense given the CPU bottlenecking we still still see. I think MT rewrite still has quite a way to go before we see all the benefits (which is a good thing - it means that there should hopefully be more optimizations to come).  As I understand it, stand alone servers are still running mostly single thread? It's more some graphic stuff that's been moved to multi-thread for now?

Agreed, and I believe servers are ST still. Have a feeling they might not get MT until we get the "AI enhancements" I predict is needed for the DCE actually. Pretty sure its all tied in together. 

22 minutes ago, Dangerzone said:

ED does things that leave me scratching (or banging )my head - so I've learned not to take anything for granted with their plans, until it's actually released.  

Absolutely!  

22 minutes ago, Dangerzone said:

It could be that ED's idea is to repeat what other sim's have done in the past, but the main difference 'never seen before' is that they're planning on using AI to generate the subsequent missions

Wouldn't be surprised if this is it. 

23 minutes ago, Dangerzone said:

and maybe they have no desire for a endless ongoing MP mission that never stops?

Well, then they read the room wrong.  

23 minutes ago, Dangerzone said:

Without them announcing things, we can only speculate, and/or go from experiences of how they've released things in the past. 

And we have so much fun with it, right? Speculating that is.  

[Dangerzone]

1 minute ago, MAXsenna said:

And we have so much fun with it, right? Speculating that is.  

And they probably have just as much fun laughing at our speculation. 

[MAXsenna]

And they probably have just as much fun laughing at our speculation. 

Yeah, I wrote something along those lines that summer a new module would be revealed in every upcoming newsletter.

Sent from my SM-A536B using Tapatalk

[Gierasimov]

You really flatten this @MAXsenna "Vulkan is graphics" - it certainly is related to and allows much more in terms of new shaders, visual effects and GPU scheduling, however, as we know from the newsletters already, current DirectX graphics rendering in DCS and current Vulkan implementation (internal to ED) graphics rendering produce exact the same graphics which to me means that when Vulkan comes, DCS will, hopefully, only initially look exactly the same. Unless ED decides on releasing Vulkan to public only with the new EDGE version with new visuals. 

Coming back to the DCE, for multiplayer it makes sense that battles can run long time, however in single player experience, that would mean ED had to run servers in the background. We all know this isn't going to happen, so yeah, I think it will be campaign "save game", and I expect when you load it, it will check the timestamps, calculate how much time elapsed and roll the dice on what would have happened to the battlefield, generating new positions for blue and red forces. 

I say one thing - DON'T rush DCE. Just now I placed four groups of 4 tanks each and tried to make them battle Kandahar streets... 

with all groups set on RANDOM skill level you get variable kills (who gets the first kill usually wins entire group), but man, locking into the buildings stops the progress. 

Tanks on the bridge don't see tanks in the open on the other side of it, but not only that - tanks in the open don't see tanks on the bridge until the last of the group passes it, shooting starts but it is too close and they switch to machine guns, but those are too weak to damage tanks, and they end up passing each other head on, stopping just 50 meters one group behind the other, facing back to each other and then the battle stops, no more bullets, no more action. Red and Blue tanks sitting in the open danger close, no fighting as if not knowing about each other, but without even trying to turn and check. This is just a quick example - with that kind of example, with that "logic" DCE will not work.

[MAXsenna]

3 hours ago, Gierasimov said:

You really flatten this

What do you mean? All I pointed out, was that I don't think the DCE is reliant on Vulkan. While I believe Vulkan will be released before DCE. 

 

3 hours ago, Gierasimov said:

however in single player experience, that would mean ED had to run servers in the background.

I'm positive I'm misunderstanding you, as the company, (which doesn't exist  anymore), that made that excellent dynamic campaign everyone wants for DCS, never ever ran a server to my knowledge, and there are servers still running it. I did too long time ago in my home for fun. Same goes for that heli sim I mentioned. Never accessed any company server. Ran it all at my own locally in my house. DCS has already the option to start a local server, can't see why ED needs to run any. Until the global map comes along, while that is irrelevant in this discussion. 

3 hours ago, Gierasimov said:

, I think it will be campaign "save game", and I expect when you load it, it will check the timestamps, calculate how much time elapsed and roll the dice on what would have happened to the battlefield, generating new positions for blue and red forces

I assume both. And thank for reminding of the save feature. Forgot the mention that. Until we have a save feature, we won't see DCE. 

3 hours ago, Gierasimov said:

I say one thing - DON'T rush DCE.

Agree! Don't think they will, because they do know what they're up against in regards to expectations.

3 hours ago, Gierasimov said:

with that kind of example, with that "logic" DCE will not work.

Like I said. It will be down to AI logic. So I think it is far from ready. 

[Gierasimov]

@Dangerzone

Quote

Testing of Dynamic Campaigns has resulted in a new tracking system that allows a detailed accounting of actions and behaviours. This is critical to the debugging process of such a complex task. The current DCS track file system has also been expanded with the ability to store RTS layer information. The resulting systems will allow you to review campaign events and provide a DCSDC active save system. This active save system will allow you to resume saved games and rewind time to replay an unsuccessful episode of a campaign

This update provides hints as to that the module will bring. I think it will be very similar to other strategy games where you have a 'map' at strategic level and 'missions' at tactical level. It is going to be interesting from the point of theme continuity... Although without story or narration you would normally see in paid campaigns like ones form Baltic Dragon or Reflected or others ...

[Gierasimov]

Didn't mean anything bad @MAXsenna, just you said Vulkan in graphics, that would make some of folks here think Vulkan will bring visual updates to DCS (as to take it for granted) but otherwise I agree with you. 

For now, just forget multiplayer DCE. It will be released as early access and will be a long way to get out of it, and back in 2020 ED said: 

Quote

The last stage will be the Client - Server architecture needed to support large campaigns with on-going online sessions.

So, for now DCE will come to SP, and as such, how do you expect your SP DCE to run overnight while your computer sleeps? It won't. It will save the state, load afresh and calculate battlefield to continue dynamic orders for player. 

If the order was to kill a convoy and you go sleep, next time you open DCE that mission will not be there or had changed somehow, right? 

In order to have 24/7/365 life campaign where player jumps in and out whenever player wants, that campaign cannot belong to local DCS installation.

[MAXsenna]

2 minutes ago, Gierasimov said:

Didn't mean anything bad

No worries! Didn't take it that way! I see what you meant now.

3 minutes ago, Gierasimov said:

For now, just forget multiplayer DCE. It will be released as early access and will be a long way to get out of it, and back in 2020 ED said: 

@BIGNEWYJust mentioned in another thread, that it is in testing, huge undertaking. 

5 minutes ago, Gierasimov said:

So, for now DCE will come to SP, and as such, how do you expect your SP DCE to run overnight while your computer sleeps?

You turn your computer off? Of course you are correct. So, we agree the save function needs to come first?

7 minutes ago, Gierasimov said:

If the order was to kill a convoy and you go sleep, next time you open DCE that mission will not be there or had changed somehow, right? 

Sure, if you just let the server run. That's why those other I mentioned are so immensely popular. Especially the one. You never have to play. There war will just go on. Personally I really hope CA will get some, (well, a lot), of focus. As that might bring in quite a few RTS fans.

11 minutes ago, Gierasimov said:

In order to have 24/7/365 life campaign where player jumps in and out whenever player wants, that campaign cannot belong to local DCS installation.

You better ask that other sim. It will disagree with you. 

[Gierasimov]

44 minutes ago, MAXsenna said:

You better ask that other sim. It will disagree with you. 

That other sim, assuming we are talking about the one that has got a bird in its name, does exactly that what I am expecting DCS DCE to do. 

Save campaign progress, load the save and update it with the calculated status. Roll of dice affected by player flown missions. So, let's say there were two groups about to fight each other, but you left the campaign. Next time you load it up it will not show you that battle, it will show you the result of that battle, as the time has passed. 

What I would wish for is more than that. I wish the war would go on and I could at any time log into the UI from any device to check the status and live strategic map and see how two opposing AI forces are doing, so that I could for example influence it with strategic decisions, just like you do in that other sim before it's campaign starts. 

I wish AI in that battle not only progressed to gain territory but also setup defences and reinforced positions to hold the ground for counteroffensive. I doubt it will though.

Yes, save game will definitely help set a persistent stage for DCE but do it will for single missions.

[MAXsenna]

@Gierasimov Exactly this! While I do hope, as a wrote, that down the line it will feature an RTS component, compatible with Combined Arms down the line. Doesn't have to be all in at release.

Sent from my SM-A536B using Tapatalk

[Fanki 17]

Hey Eagle Dynamics Team.

I havent played DCS since like 2 years and all the new modules and maps are cool and so on but we need an Dynamic Campaign or some sort of Mission Generator.

Right now it feels like a racing sim with bunch of cars and tracks but no actual races to compete.

Please give us something to do in the game.

Edited November 24, 2024 by Fanki 17

[Dragon1-1]

Looks like we'll get some news about it in the 2025 and beyond video, on Jan 16th.