Kerbal Space Program 2 to be released in 2022 [Discussion Thread]

[The Aziz]

All things considered, since they started the development around 2017, three years was a very optimistic plan, which works maybe for simple games. And ksp is not simple in any way. Five years sounds much more reasonable, considering the scope. So now from this perspective, I'm surprised they went with 2020 as release date. Sure many things happened along the way, but they shouldn't affect the schedule that much. I mean, it's like they were doing nothing since 2019 until today. No, I'm sure their work exceeded their (and our) expectations and they had to push the release a few times.

[K^2]

24 minutes ago, The Aziz said:

And ksp is not simple in any way.

KSP is actually pretty simple as far as modern games go. Even with stated ambitions for KSP2 it's not exactly AAA-scale. But they're also working with a fairly small team on (what I assume is) a modest budget. So yeah, it definitely takes time.

I'm pretty sure the original early 2020 goal was for "KSP, but with bigger tech tree, more parts, more planets, and some extras." That was reasonable for Star Theory, but then they got ambitious. Which isn't necessarily a bad thing, and some great games came out of the team deciding that they can make an even bigger game, but then you either grow the team, slip the schedule, or most realistically, both. The original 2021 goal for Intercept seems like it was in "salvage Star Theory plan" mode. The new date indicates that they rescoped, re-scheduled, and are now making the Intercept vision of KSP2. And based on some of the recent updates, it looks like a big part of it is that rather than making KSP with a fresh coat of paint on it, they're now making something that preserves the core of the original, but also feels and looks like a game made in the decade it's actually released, and not something that belongs on last gen hardware.

[dprostock]

What is the concept in which people are hopeful that in two years the promised dates will be met, when they have already had problems at the employer level and at the staff level??? 
I am amazed when I read that there are people who discuss what they are going to do about something that does not exist, where there is no certainty of compliance with dates or the final delivery of the product. 
If at least there was competition to make the proposed schedule believable, because if they do not comply, another will....
 

[K^2]

3 minutes ago, dprostock said:

What is the concept in which people are hopeful that in two years the promised dates will be met, when they have already had problems at the employer level and at the staff level???

I've worked in multiple game development studios. Currently, I am an engine lead for one of major AAA studios, and so I am involved in long-term and day-to-day planning for multi-year projects. I've seen projects that failed and I've seen projects that have suffered many delays before getting released.

I was concerned about originally stated schedule after the game was transferred from Star Theory to Intercept, in part due which reqs were unfilled. Since then, several key slots have been filled, and the new schedule matches where the game is at. Late 2022 is entirely reasonable for the team Intercept has based on progress made so far. The two unfilled reqs that concern me are graphics engineer and multiplayer engineer. There is still time to fill these, but even if these do not get filled in the near future, the game can still ship on current schedule. We might not have multiplayer at release or have fewer features, but it's not the sort of thing that would prevent the game from being shipped.

It's alright to have concerns about the delays, but jumping to, "It will never get finished, the game is doomed," is needlessly alarmist at best and bordering on shameless trolling at worst.

[dprostock]

28 minutes ago, K^2 said:

It's alright to have concerns about the delays, but jumping to, "It will never get finished, the game is doomed," is needlessly alarmist at best and bordering on shameless trolling at worst.

Ubisoft - Silent Hunter IV Wolves of the Pacific ( Aka: Bugs of the Pacific ) 

Ubisoft - Silent Hunter V Battle of the Atlantic

Far away were from the legend they created with the SHIII, even with two essential flaws that would supposedly be corrected in later versions...
And all this despite inviting modders and representatives of the Flottille to eat sausages in Wilhelmshaven and listen to the requirements of simulator enthusiasts.

 

[Incarnation of Chaos]

1 hour ago, dprostock said:

Ubisoft - Silent Hunter IV Wolves of the Pacific ( Aka: Bugs of the Pacific ) 

Ubisoft - Silent Hunter V Battle of the Atlantic

Far away were from the legend they created with the SHIII, even with two essential flaws that would supposedly be corrected in later versions...
And all this despite inviting modders and representatives of the Flottille to eat sausages in Wilhelmshaven and listen to the requirements of simulator enthusiasts

What difference does it really make at this point though? Seriously, that's the one thing i don't understand.

If our options are either KSP2 is vaporware, or KSP2 will release as a flaming pile of garbage filled with bugs 2 years from now then what does saying any of this accomplish? No money has changed hands, no game has even been available to play, literally nothing but a vague expectation has been set. If KSP2 never comes out, then no amount of posts or developer hounding would make it any less doomed. If KSP2 is in development for 2 more years, and still has multiple showstopper bugs then thankfully the internet exists and reviews will easily convince me my money is best spent elsewhere.

Development is not a nice, linear process, even in the largest of AAA studios, and if you go find a good postmortem of any of your favorite games i can almost assure you that there's a point in their development cycle that looked just as if not more bleak.  A nice steady path from point A to B in game development is the exception, not the rule I'm afraid it seems. 

 

[The Aziz]

8 hours ago, K^2 said:

KSP is actually pretty simple as far as modern games go. Even with stated ambitions for KSP2 it's not exactly AAA-scale.

Lemme rephrase it, as AAA games can also be just "go forward and shoot". It's the whole cloud of graphics, story, sometimes character progression and uh, in many cases multiplayer, around the base idea that makes them feel "big".

So, first ksp may be simple, I can agree on that, but it looks like the sequel won't be. With more places to visit, which means extension of NEW gameplay by dozens of hours (what I mean is, how many times can you go to the Mun before it starts to get boring?), with all colony mechanics - expansion, reproduction, production, resource extraction, conversion, offworld building, background resupply missions, and Nate knows what else, I see this as a big, complex game, almost a strategy, but with the ability to Lego anything you want and use it.

Sure we got a lot of this in mods, but this proves the point.

[K^2]

1 hour ago, The Aziz said:

Lemme rephrase it, as AAA games can also be just "go forward and shoot". It's the whole cloud of graphics, story, sometimes character progression and uh, in many cases multiplayer, around the base idea that makes them feel "big".

So, first ksp may be simple, I can agree on that, but it looks like the sequel won't be. With more places to visit, which means extension of NEW gameplay by dozens of hours (what I mean is, how many times can you go to the Mun before it starts to get boring?) [...]

None of these things factor into how long it takes to make a game, though. You can invest your entire life time into playing Chess, as many people have, but it doesn't mean that it will take decades to make a Chess game. I mean, you could spend that long if your goal is to write some very clever AI for it, but I can also write a simple Chess client, complete with networked multiplayer, in a couple of hours, and it will have everything that makes Chess a great game.

In terms of production, KSP is a very simple game. And I'm not saying it to diminish that game - that's part of brilliance of the original. Things were added over the years to make it a bit more complex, but I remember looking through the code of the early-ish version, around 1.0, and the physics was basically just a few lines of code for aerodynamics and gravity. Everything else was handled through Unity. Likewise, there were some clever bits to load graphics and collision around the player when you're on the surface, so that you can handle planet-sized planets without running out of memory, but the rest is Unity doing its thing. Some resource management in flight, a ship editor, some UI, and you basically have your game. Updates expanded on that. We got biomes, mining, asteroids, robotic parts - all coming with a bit more logic and a bit more UI. But there is still a very reasonable amount of work there for a very modest team.

The current vision for KSP2 is a lot bigger. I agree. But I don't think it started out that way. I think the original pitch was remake of KSP + interstellar + colonies + visual update. And depending on how you implement colonies, all the extra stuff is basically a big expansion. But then the reveal happened at 2019 E3, and everyone was excited, and I think that's when the push for larger, more ambitious KSP2 was made.

I could easily be wrong about that, but a simpler, leaner original vision for KSP2 fits well with the timeline, what we saw at E3, and how the narrative changed in the months following it, as KSP2 became a bigger project.

Even now, though, KSP2 isn't that big and complex. Things they are doing with rendering of planet surfaces and sky is pretty typical for modern games. Red Dead Redemption 2 is a good reference here. The general look of ships makes it seem like they embraced PBR, which is industry standard now. Colony building becoming an integral part of gameplay essentially means introduction of economy, but again, it's hard to come up with modern games that don't have some form of economy going. Again, KSP2 is kind of doing a lot with little here, going after bigger world-building pieces rather than getting caught in the details. And that's great, because that is what lets a mid-sized team build a game that's just as fun and long-lasting as the best of the AAA games. It's also why I think they are on reasonable schedule now despite increased ambition for the project.

[dprostock]

7 hours ago, Incarnation of Chaos said:

If our options are either KSP2 is vaporware, or KSP2 will release as a flaming pile of garbage filled with bugs 2 years from now then what does saying any of this accomplish? No money has changed hands, no game has even been available to play, literally nothing but a vague expectation has been set. If KSP2 never comes out, then no amount of posts or developer hounding would make it any less doomed. If KSP2 is in development for 2 more years, and still has multiple showstopper bugs then thankfully the internet exists and reviews will easily convince me my money is best spent elsewhere.

 

Yep. 

It would be more convenient for the same production to be more respectful of those who have bought the simulator and explain things better, not that we have to know through modders and journalists dedicated to the field of other problems not associated with the programming code itself.
In addition, the KSP is a simple simulator, lacking AI, so there is much doubt about the complexities of it. 
 

[Starhelperdude]

16 hours ago, Makrotex54 said:

How can a company estimate itself by about 2 years ? The change of the developer does not do this game any good .... I think that the new developers underestimate the game of the measures .... Probably the first pictures of this game do not show the gameplay of the actual game...

most of the developers are still here, I think it's just new features and polishes

[starcaptain]

19 hours ago, Deddly said:

I admire the work you put into this. But given the data points we have and the uncertainties regarding any projection, isn't this a little like, say.. 

 

? 

Well a polynomial forecast is better than a linear one.

Also, I wanted to put citations and graphically represent the current layout of predicted release dates in one place.

[Dr. Kerbal]

Eh. We just have to wait. No complaining. Just wait. We can blow up more Kerbals while we wait.

[K^2]

1 hour ago, starcaptain said:

Well a polynomial forecast is better than a linear one.

Which is always a good indicator that you should be careful with this, as extrapolating from a polynomial is always trouble. But either fit is missing one crucial element, the fact that release dates already have some assumptions baked into them which you don't track. There are other statistical models that are designed to track changes like this, but they don't really work with just a few points of data.

[Alexoff]

It looks like Covid-19 has infected developers' computers!

It is very interesting to know which game they talked about in the summer of 2019, which was going to be released in March 2020. Apparently Fallout 76 against its background would be a technological masterpiece. Or the developers knew that the game was far from release, but they specifically promised us the impossible so that we would not forget about the game by 2022. Although there are still two years ahead, you can postpone the game for another six months more than once.

[dprostock]

I would prefer you to recognize that the KSP2 was the same as the KSP1 with a couple of new planets, but since you realized it was an earthly primitive, considering the quality of the games and simulators that are appearing on the bussines, they decided to kick the board and start over with 21st century technologies. 
And with that said, they'd have my full support.
 

[mcwaffles2003]

17 hours ago, starcaptain said:

Well a polynomial forecast is better than a linear one.

Also, I wanted to put citations and graphically represent the current layout of predicted release dates in one place.

Based on this graph shouldn't it have released back in like 2018 then?

[Lord Aurelius]

Glad to hear the devs are taking their time to do things right. KSP1 was a brilliant game concept hobbled by a severe lack of game cohesion or polish (and is still being cleaned up to this day to address holes and shortcomings in the game), and I'm very happy to hear that the new devs have taken this to heart and are working to build a much more solid product out of the gate. Building quality software takes time, glad to hear you're committed to doing it right instead of being yet another game rushed to release (although be careful with scope creep, better to release even if it isn't perfect, but still good, than not to release at all).

[LittleBitMore]

On 11/13/2020 at 10:04 AM, starcaptain said:

Spoiler

 

Predicted release date, according to the numbers.

Polynomial trendline: y = -0.5909x² + 2389.55x - 2413583.81

The polynomial trendline predicts somewhere in mid-2022 if the actual range of "in the year 2022" is assumed to be within the first three months of 2022.

Points:

1 [1] "Spring 2020"; stated 2019-08-19

2 [2] "Fiscal Year 2021" (1 April 2020 - 31 March 2021); stated 2019-11-07

3 [3] "Fall 2021"; stated 2020-05-20

4 [4] "In 2022"; stated 2020-11-05

References:

[1] Kerbal Space Program 2 Announced, Savage, https://www.pcgamer.com/kerbal-space-program-2-announced/ , Retrieved 2020-11-13

[2] Kerbal Space Program 2 is delayed, Chalk, https://www.pcgamer.com/kerbal-space-program-2-is-delayed/ , Retrieved 2020-11-13

[3]  Kerbal Space Program 2 has been delayed to fall 2021, Lee, https://www.polygon.com/2020/5/20/21265482/kerbal-space-program-2-delayed-2021 , Retrieved 2020-11-13

[4] Kerbal Space Program 2 to be released in 2022 [Discussion Thread], UomoCapra, https://forum.kerbalspaceprogram.com/index.php?/topic/197935, Retrieved 2020-11-13

 

Spoilered for size. Also opening said spoiler causes a forum format glitch on Google Chrome, so beware.

I have no idea how you got that graph working, haha. I couldn't do error bars and I got this graph 

Edited November 15, 2020 by LittleBitMore

[Alexoff]

I think that if KSP 2 is postponed to 2030, the game will turn out completely without bugs, very modern and everyone will be happy. After all, the main thing is not to rush, but to make the game of the decade. And if you postpone the release to 2040, then it will be the game of the century!

[spock19]

On 11/5/2020 at 12:17 PM, UomoCapra said:

That involves solving problems that have never been solved before,
and that takes time.

There's soon going to be one more such problem. I'll not put a spoiler here but expect it soon (~a month).

[Alephzorg]

They could give up on a multiplayer option and save time...

[The Aziz]

31 minutes ago, Alephzorg said:

They could give up on a multiplayer option and save time...

Players (well, not me, I don't have anyone to play with) have been asking for multiplayer for over half of a decade, so they decided to provide, instead of leaving people to rely on mods again.

After all, in our world, there really are multiple players out there, working together. So I'd like to dock into my friend's station, refuel and go even further, while he expands the infrastructure around the system.

[pandaman]

On 11/16/2020 at 8:08 PM, Alexoff said:

I think that if KSP 2 is postponed to 2030, the game will turn out completely without bugs, very modern and everyone will be happy. After all, the main thing is not to rush, but to make the game of the decade. And if you postpone the release to 2040, then it will be the game of the century!

I'm no programmer, but I'm realistic, and understand enough to know that if software is 'bug free' then you just haven't found them all yet.

[Alephzorg]

3 hours ago, The Aziz said:

Players (well, not me, I don't have anyone to play with) have been asking for multiplayer for over half of a decade, so they decided to provide, instead of leaving people to rely on mods again.

After all, in our world, there really are multiple players out there, working together. So I'd like to dock into my friend's station, refuel and go even further, while he expands the infrastructure around the system.

While I understand the appeal (for some) of a multiplayer option, the never ending discussions on how to implement it let me believe that it is far from easy. And that no matter which way of doing it they choose, many will complain anyway.
And same as you, I play alone and don't have a need for it.
I wouldn't complain if they released the game without multiplayer at first and gave it to us later with an update.

[dprostock]

3 hours ago, pandaman said:

I'm no programmer, but I'm realistic, and understand enough to know that if software is 'bug free' then you just haven't found them all yet.

That's solved with a project leader, a significant number of developers, and the integration of a team of targeted beta testers with clear test goals.