Some concerns I have about KSP 2.

[lBoBl]

Hey guys! 
This is a post about all (or most of) the concerns I have about KSP 2, and what I interpret as bad signs coming from the dev team. Before starting however, I would like to apologize in advance for all the negativity that is about to follow. I am going to explain why I think KSP 2 might not be as good as a lot of us would want it to be, but I'm doing that hoping that one of you or perhaps even someone from the dev team will reply and address all of my concerns and reassure me about how great the game is actually going to be. I am not doing this with the hope that I can expose this game as a fraud, get the dev studio closed and all the personnel executed. I wish KSP 2 will be successful even if the game disappoints me personally, I know the devs are passionate and I know that none of us, no matter how fond of KSP 1 we might be, are actually entitled to a good second game. This is not a KSP 2 hate thread, please refrain from using bad words. 

 

So, let's get into it. I have been following the dev diaries and all things related to KSP 2 development very closely basically since it was announced in back 2020.

Back then, the game was being developed by an indie studio called Star Theory (Wikipedia). A certain Nate Simpson made the announcement and everyone got very excited. The game was going to allow for huge futuristic ships that could be built in orbit, and would travel to distant worlds, even as far as other stars, to establish new colonies for kerbalkind and visit those that your friends already built in multiplayer mode. And as a cherry on top, modding support would be even better than in KSP 1.  Hearing that, it's not hard to see why everyone jumped aboard the hype train. And the studio showed great engagement with the community by inviting the main KSP 1 content creators to take a look at the new game, and I remember the first impressions of these content creators being very positive. There was one thing though, that seemed a bit baffling to many, including myself : the release date of "late 2020". End of year, really? A game that big, that complex? Surely it had to be well on its way by then, right? Well, we don't know. The first trailer was not actual gameplay footage and there was none to show at the time, so that seemed a bit ambitious... But who cares, it'll probably get delayed and then what, that's not the end of the world, so fans of KSP 1 had basically no reason not to get aboard that hype train. And apparently even the game publisher, Private Division (Wikipedia) got on board big time. This is a bit of a hot topic and I don't want to discuss moral issues here, you can read more about what happened elsewhere on the internet. But from a purely technical perspective, let's just say that there was a studio and partial team change, from Star Theory to Intercept Games, with some of the initial devs transferring along with the project. And in the middle of all that, suddenly Covid decided that there was too much fun and happiness in this world, so that certainly didn't help.

The rest is basically conjecture and educated guesses because we don't really know what happens behind closed doors. Basically there are two things that I'm concerned about : the game has been delayed multiple times and we've seen very little actual gameplay footage even one week away from early access. 

I know many of you will disagree with that second statement because the devs have been doing a fair amount of communication, giving us exciting dev diaries and other videos on YouTube. But very little of that content has shown actual progress being made on the game (I'm not saying none, just very little considering it's been 3 years).  It's mostly just what I'd call wishfull thinking, devs saying we're going to have this and that and we want interstellar travel to feel long and dark and we want tutorials to be easy and intuitive and we want planets to be realistic and so on and so on... Very little "We have that, look, it's cool". There is some of that, we heard the cool sounds of rocket engines, saw some of the engine plumes and explosions being worked on, a couple colony parts and futuristic engines, but that's it. And when you think about it, these are just assets. Just bits and bobs that make the game look and sound cool but we have seen no gameplay mechanic. The only clue on what actual gameplay might be like is a glance at the new UI which admittedly does look good and functional. I'll say a little more about the gameplay footage later but first let's talk about the delays, because that ties into it. If I recall correctly the game was very rapidly delayed from "late 2020" to fiscal year 2020. Basically as a studio you can delay the game and the publisher is still happy if it's in the same fiscal year because they can still put it in their results for that year and that makes investors happy (that's of course a very crude sum up).  But then it wasn't ready, no project in the entire world at that time was on schedule mainly thanks to Covid. So it got delayed again. And again. Untill it was delayed to 2022 and then finally to February 2023. Why that looks bad to me is because this is a month before the end of fiscal year 2023 for Private Division (which is based in New York State) and therefore maybe a deadline set by the publisher. Basically they had to say 'hey guys we need to release your game before then even if it's not finished because the company needs to be profitable otherwise it will die and there will be no game at all'. Not blaming anyone here this is just the harsh reality of capitalism. Again, this is conjecture, I don't know that. And of course I'm dramatizing, they have other games and the company probably wouldn't die. So maybe this is all a coincidence, but I'm just saying, it sure looks like a profitable coincidence for Private Division.

But let's indulge in even more conjecture and try to imagine what happened when that hypothetical deadline was announced to the dev team. Remember that in game footage that we've seen? Well it had one very important characteristic that I haven't talked about yet. It looked a lot like it ran at a very low framerate and with copious amounts of stuttering. Not really something you can ship to customers. Well then, what do you do if you're a dev? You strip the game bare for early access : remove all the extra planets and star systems you were working on, remove the large scale colony and interstellar parts, scrap the actual gameplay mechanics like orbital construction and colonization, tone down post processing to a minimum and don't even think about multiplayer for now, it'll be easy to add on top of all the rest later (haha).  Reassure the player base, tell them not to worry, all the gameplay footage they've seen so far is months old and the game is in a much better state now.

But if that is truly the case, why don't they release more recent footage? We're a week away from launch, are they going to tell us that the game is not ready to be shown right now but it'll be playable in 8 days? 

So yeah that sounds kinda sus to me. But you might say hey, they never said that the game will be good or even playable in a week, it's just an early access. The devs have already addressed all of our concerns regarding what will be in the game by releasing a magnificently illustrated roadmap with happy little Kerbals on it. But no dates. How is that more than just wishful thinking intermingled with marketing bull c***? No dates, no comittment. We've gone from full game release in late 2020 to early access in 2023 with none of the features that were promised in 2020 : No interstellar travel or even big interstellar capable ships, no colonies, no resource gathering, no orbital construction, no multiplayer, and they even removed science and progression that game one had. But don't worry, the devs said on their roadmap that we will be adding all of that back into the game at one point eventually in the future probably. Except for modding support, I don't think the devs have mentioned that since 2020. Most players seem to think that this is such an obvious part of KSP's identity that they don't need to mention it but I think that if the game did have extensive modding support they would at least try to use it as a marketing point. And also, I know, they did apparently deliver on one promise: better tutorials (but I'm worried people might not properly revere He Who Knows aka Scott Manley which would be bad for humanity as a whole).  Seriously though, I think it's a very good thing if good, engaging tutorials are present, and something that was severely lacking in KSP 1. But it's mostly pre-recorded animations and nothing to do with actual gameplay so not really an encouraging sign regarding the technical state of the game.

Basically if the dev team took 3 years to bring us none of the promised features, how long will it take them to bring us all or any of these promised features? 

Anyways, I hope I'm wrong and delirious and that I just wasted an hour writing all that for nothing. I really wish KSP 2 will be better than I think it'll be and that it'll eventually have all the features we expected it to have. And regardless of whether or not it is good I hope that it will be a commercial success and that the devs get rewarded for their efforts. 

[Bej Kerman]

3 minutes ago, lBoBl said:

Anyways, I hope I'm wrong and delirious and that I just wasted an hour writing all that for nothing.

There's such thing as wishful thinking, yeah, but there's still such thing as unwishful thinking, fixating on any weird thing and drawing out-of-proportion conclusions.

Frankly, KSP 1 aged like milk. The parts list, one of my biggest peeves with the game outside broken gameplay, functions exactly the same as it did a decade ago. No mods fix it (Janitor's Closet's hacky solution of removing parts doesn't count), no-one at Squad has ever thought "hey, do you think jumbling all the parts in a category into one big pile and sorting them by whatever random digits are in their names might contribute to low player retention?". Intercept is evidently taking their time to go through the mess Squad made of KSP 1 and make sure everything is built to standards, and even without the roadmap features, it's still going to be an experience lightyears beyond what mods do to hack KSP into a bigger game. I drawn the parts list as one example to show the improvements UX is seeing in KSP 2, but you can expect it to show in the rest of the game, especially when the roadmap sees progress.

TL;DR: Even without the roadmap features, UX is still worth the game's price. Expect to see the dev's care for the UX to also reflect everywhere else, like how fleshed out the parts list is and how physics handles edge cases like orbital-speed collisions.

[Kerbart]

Imagine that. Developers who got paid by a commercial company are forced to produce something at some point in time, I never realized this is how things work in the real world. Eye opening!

There's a few things I don't understand about these "concerns"

• What's the point of sharing them?  Are you hoping for an email from Nate saying thank you for making me see my mistakes, we'll revise our plans and cancel the February 24 launch?
• Software projects overdue is coming as a surprise?
• Are you aware that a pretty solid chunk of KSP1 took place after the 1.0 launch?

So, it took longer. Likely the project was started with "we rewrite the game with a better internal engine." And then, upon further review, skeletons come out of the closet. Career mode isn;t that great. And Science is repetitive and boring. And Onboarding is horrific. You could just bolt-on solutions, like KSP21. Or you can rethink how the game works, and see how some of these things can be integrated. And as you're improving more and more, the parts that don't improve start to stand out, so you need to rethink those over as well.

If your project doesn't have precise defined milestones set out in the beginning, your timeline is never going to be met. And it looks to me that Intercept had the wisdom and courage to say, one year into the project: "we're not so sure about the project goals, let's rehash them." That's going to cost a lot of time, and surely risk shutting down the entire project prematurely. But it will improve the final product. A project that isn't finished, but has a clear vision of what parts are going to be added, with provisions in place to support that final product.

Or they're plain lying, as you say, and they're rushing to get a polished clone of KSP1 at the finish line. Time will tell. I'm confident that what's on the roadmap will be delivered. Why? Because a lot of the updates were about "this is how we imrpove gameplay elements," if what we get is only a shinier KSP1 clone we would have gotten a lot more "oooooh shiny" posts.

As for not revealing the final product right before the big reveal — who the heck does that?

 

Edited February 16, 2023 by Kerbart
Changed typo "KSP2" into "KSP1"

[The Aziz]

52 minutes ago, lBoBl said:

There was one thing though, that seemed a bit baffling to many, including myself : the release date of "late 2020". End of year, really? A game that big, that complex? Surely it had to be well on its way by then, right? Well, we don't know.

They started working on it in 2017, that's where the earliest hints point at. You could say that 3 years is enough time, but for a small studio like Star Theory/Intercept, aiming very high, it really isn't. It takes roughly 4 to 5 years to release a large game even made by large team. So 2023 mark feels appropriate.

56 minutes ago, lBoBl said:

The first trailer was not actual gameplay footage and there was none to show at the time, so that seemed a bit ambitious...

There was early footage not long after the announcement though. It didn't show much but it was there. Clearly showing that the game wasn't that far in development.

58 minutes ago, lBoBl said:

the devs have been doing a fair amount of communication, giving us exciting dev diaries and other videos on YouTube. But very little of that content has shown actual progress being made on the game (I'm not saying none, just very little considering it's been 3 years). 

Yep, that's what most small studios do. Snippets of development, some insight into how certain things are made. It's very common among indie devs.

1 hour ago, lBoBl said:

Very little "We have that, look, it's cool". There is some of that, we heard the cool sounds of rocket engines, saw some of the engine plumes and explosions being worked on, a couple colony parts and futuristic engines, but that's it. And when you think about it, these are just assets. Just bits and bobs that make the game look and sound cool but we have seen no gameplay mechanic.

Even the Creative Director said that there's no point in showing off things if they're far from playable state.

1 hour ago, lBoBl said:

Untill it was delayed to 2022 and then finally to February 2023. Why that looks bad to me is because this is a month before the end of fiscal year 2023 for Private Division (which is based in New York State) and therefore maybe a deadline set by the publisher. Basically they had to say 'hey guys we need to release your game before then even if it's not finished because the company needs to be profitable

Which is probably correct. T2 has been pouring money into the game for 5 years with no profits in sight. I have a feeling there was an agreement there, to put out whatever they can so the project can finally start earning money.

1 hour ago, lBoBl said:

Remember that in game footage that we've seen? Well it had one very important characteristic that I haven't talked about yet. It looked a lot like it ran at a very low framerate and with copious amounts of stuttering. Not really something you can ship to customers. Well then, what do you do if you're a dev? You strip the game bare for early access : remove all the extra planets and star systems you were working on, remove the large scale colony and interstellar parts, scrap the actual gameplay mechanics like orbital construction and colonization, tone down post processing to a minimum and don't even think about multiplayer for now, it'll be easy to add on top of all the rest later (haha).  Reassure the player base, tell them not to worry, all the gameplay footage they've seen so far is months old and the game is in a much better state now.

You always should optimize and polish at the end. Otherwise you'll be doing it multiple times, which is pointless and kills the productivity.

1 hour ago, lBoBl said:

But if that is truly the case, why don't they release more recent footage? We're a week away from launch, are they going to tell us that the game is not ready to be shown right now but it'll be playable in 8 days?

Games can reach "Gold" status within days from release, so yes. Besides, you wouldn't want to spoil things for yourself, would you?

1 hour ago, lBoBl said:

The devs have already addressed all of our concerns regarding what will be in the game by releasing a magnificently illustrated roadmap with happy little Kerbals on it. But no dates. How is that more than just wishful thinking intermingled with marketing bull c***? No dates, no comittment. We've gone from full game release in late 2020 to early access in 2023 with none of the features that were promised in 2020 : No interstellar travel or even big interstellar capable ships, no colonies, no resource gathering, no orbital construction, no multiplayer, and they even removed science and progression that game one had. But don't worry, the devs said on their roadmap that we will be adding all of that back into the game at one point eventually in the future probably.

No dates, because it all depends on people's feedback. You can't put timelines in software development, it leads to either rushed, buggy product or unfinished one. Things happen, unpredictable, unforeseen bugs pop out, some are harder to squash than others, and you don't want to play Cyberpunk2077 and send yellow apology cards every week.

1 hour ago, lBoBl said:

Except for modding support, I don't think the devs have mentioned that since 2020. Most players seem to think that this is such an obvious part of KSP's identity that they don't need to mention it but I think that if the game did have extensive modding support they would at least try to use it as a marketing point.

Feature Episode 6.

1 hour ago, lBoBl said:

But it's mostly pre-recorded animations and nothing to do with actual gameplay so not really an encouraging sign regarding the technical state of the game.

There are ingame tutorials, there were few seconds of footage showing them. Animated tuts are there to get a grasp on things shown in accessible way.

Cheers, I hope that helped.

[lBoBl]

16 minutes ago, The Aziz said:

Cheers, I hope that helped.

Well it sure did. I still have my doubts and I still think maybe IG isn't the best at communication, but I agree with what you said, plus they dropped an actual gameplay trailer (with some non-gameplay footage spliced in for some reason) just when I hit send on the first post.

I still think there should be (vague) dates on that roadmap, you can set deadlines in software development you should just expect them to never be met (see Hofstadter's Law), but they are necessary to keep your teams productive, it's not about yelling at them or making them feel guilty, it's about motivation

[The Aziz]

Just now, lBoBl said:

you can set deadlines in software development you should just expect them to never be met (see Hofstadter's Law), but they are necessary to keep your teams productive, it's not about yelling at them or making them feel guilty, it's about motivation

The deadlines are probably there on the internal roadmap. Only available to the team. I guess they also wouldn't like to work on the game for another decade before they hit 1.0.

But history proves that once you tell a deadline to the community, if you don't meet it, there's a load of complaining and anger and disappointment and concerns about not being true to the words, keeping "promises" or straight up calling the "lies".

But then don't tell any and you're left with complaining and anger and disappointment and concerns about not interacting with fanbase, keeping everything a secret, or hiding the fact that the game is cancelled (yes I've seen such claims).

[lBoBl]

1 hour ago, The Aziz said:

I guess they also wouldn't like to work on the game for another decade before they hit 1.0.

Yeah probably but they might! It took them three years so far to deliver none of the main promised features, the ones that were on the KSP 2 homepage from the beginning. So should I expect the very first item on that roadmap (science and tech tree) in 1 month or in 1 year? I have no idea.

What the devs have ahead of them right now is going to be pretty challenging, and I hope they planned ahead. The one big challenge I think they faced initially (just an educated guess) was to allow for interstellar scale ships with potentially thousand of parts while conserving soft joint physics and running better than the same giant interstellar ship in modded KSP1 where you would count your seconds per frame rather than the usual frames per second. So far we've seen mostly choppy footage (even the launch trailer doesn't look that smooth to me but it's a 30fps video so that might be why) and not much that would indicate that this problem is solved. Honestly if I'm limited to 50 parts per vessel before it starts lagging as all hell and going full spaghetti monster mode I'm going to be disappointed. The one thing I really hope KSP 2 will do better than KSP 1 is performance, and that is not setting the bar very high, but still, I have a feeling that I'm going to be disappointed in that domain.

If I'm wrong and it does run well (even at some point in the future) and I can play around with big ships carrying little ships in their cargo bays, I'll be happy. 100%. 

"Slay the Kraken" they said. I hope it's true, we'll see!

Edited February 16, 2023 by lBoBl

[Bej Kerman]

3 hours ago, Kerbart said:

You could just bolt-on solutions, like KSP2

Guessing you meant to type KSP 1

Either way, completely agree. KSP 1's bolt-on mentality didn't allow it to age well.

[Strawberry]

2 hours ago, The Aziz said:

or hiding the fact that the game is cancelled (yes I've seen such claims).

Oh god this has made me oddly nostalgic for these claims, these claims stopped happening basically overnight but they were shockingly common

[pss88]

18 minutes ago, Strawberry said:

Oh god this has made me oddly nostalgic for these claims, these claims stopped happening basically overnight but they were shockingly common

Well... If Early Access sales will be low, there is still a possibility they cut all roadmap features, stamp "1.0" tag on it and be done with it.

[lBoBl]

Wow, I guess you can add the fact that the devs recommend a RTX 3080  to the list of my concerns.

I guess it will not run smoothly on my PC but this time it won't be a matter of parts count

[Alexoff]

I remember that what we most expected from KSP2 was OPTIMIZATION. Looking at the system requirements, it becomes clear that we were waiting for optimization in vain. Hogwarts requires less, and there it is clear where computer resources go.

[The Aziz]

16 minutes ago, Alexoff said:

Hogwarts requires less

And was released on old gen consoles.

[Bej Kerman]

18 minutes ago, Alexoff said:

Looking at the system requirements, it becomes clear that we were waiting for optimization in vain

We weren't waiting for a definitively higher framerate, we were waiting for a game where the framerate you get with a massive vessel isn't much lower than the framerate you start with, at least compared to KSP 1.

[Alexoff]

18 minutes ago, The Aziz said:

And was released on old gen consoles.

So the developers are great, they were able to optimize such a bright and colorful game with a bunch of activities. Others simply put maximum PC hardware as a recommendation for games that definitely don't have outstanding graphics.

15 minutes ago, Bej Kerman said:

we were waiting for a game where the framerate you get with a massive vessel isn't much lower than the framerate you start with, at least compared to KSP 1.

This is exactly what I was waiting for, my stock rocket for Jupiter could not load in KSP1 RSS

[Roostergod]

If this is true, I think releasing now is the best option. Frankly they should have done it sooner. A lot of people have said that that the devs should take as long as they need before release, but after this many delays the only way to get the game where it needs to be is to put it out and get feedback from the community.

[lBoBl]

21 hours ago, Bej Kerman said:

We weren't waiting for a definitively higher framerate, we were waiting for a game where the framerate you get with a massive vessel isn't much lower than the framerate you start with, at least compared to KSP 1.

Yeah but let's face it at this point the chances that this will be the case are probably pretty slim. They said they wanted to enable big crafts three years ago when the game was being developed by a different team in a different studio and the current marketing material isn't exactly filled with massive crafts

[Bej Kerman]

43 minutes ago, lBoBl said:

the current marketing material isn't exactly filled with massive crafts

The marketing material is full to the brim with the sort of crafts you'd expect would bring KSP to a crawl. That craft flying by Duna we saw isn't exactly KSP 1 material.

[pss88]

8 hours ago, Bej Kerman said:

That craft flying by Duna we saw isn't exactly KSP 1 material.

My KSP handles crafts of that size just fine. Maybe even double that size. And I have 8 year old CPU.

[Chilkoot]

On 2/16/2023 at 12:45 PM, Kerbart said:

Imagine that. Developers who got paid by a commercial company are forced to produce something at some point in time, I never realized this is how things work in the real world. Eye opening!