RocketRockington

I think there's a more likely answer than the devs working on 30k work stations for why performance sucked despite them claiming they're having fun with it. First, the OP who stated devs work on extremely high end machines like that is just flat out wrong...I've worked for half a dozen studios and none of them give their devs machines that high end unless there's a very specific need like rendering. Why? Because systems with quaddros and HPC chips are non standard for the user base. Any special code written to make the game work on those is a waste of time. Yes, devs do get 5k machines at the top end - but it's still going to be standard commercial hardware, which would be 3/4k series Nvidia GPUs/i7 12k series CPUs. Though most devs don't get yearly hardware upgrades either - 2-3 years is more standard, so many will be lower than that. And it's not even a purely a matter of cost. I can have a $30k Playstation dev kit sitting on my desk if I need it. It's a matter of efficiency. . Is it even worth having someone resetup thier machine and lose a day for whatever marginal productivity benefit they'd get? No, the real answer to how devs manage to 'have fun' despite the plethora of bugs and optimization issues is because you can 'have fun' doing things in the office that wouldn't be entertaining if you weren't using it to goof off. Heck you can even have fun laughing at all the derpy bugs, that's certainly been the case at projects I've been on. As a dev, you don't care about maintaining a campaign or losing progress or a restart - you'll enjoy something funny happening to distract you from work, and then get back to work. You don't have any immersive investment - you used debug tools to get where you are. You know how the sausage is made. You don't care about it as a full game inasmuch as how you're playing kt. It's massively different than trying to engage immersively in a product, vs a dev having a laugh. And also because it's entirely in a PR shills interests to feed the hopium, so if Nate saw some dev enjoying building a plane, even on a slideshow, a couple times, he can spin that into the whole team loving playing the game all the time. Even if you don't think Nate shades the things past the point of even subjectively possible truth, he could just be a victim of being the boss. You think he goes to the QA pit and asks people 'How are things, you have any fun with it' and gets told by all the low-level QA staffers "It's a garbage fire, we hate what you've made'? No, they probably give him an extremely qualified yes that he can spin into 'my developers enjoy playing it'.

I put down other, because I think even with critical bugs fixed, and performance improved, I wouldn't expect full content in an EA. But also the price was too high for an EA with the amount of content it released with. So either the price should have been lower - which I don't think is a good strategy tbh for a long term even higher KSP2 price point, or it needed some more content, though not all. A progression mode of any type for instance. A free demo with even more limited content than what's out now, but still more stable and performant, would also have been a reasonable way to release something as a stopgap to keep people happy and waiting for the full product.

They don't have that kind of info. And they can't just speculate for you. The company line will literally just be 'all is well, full steam ahead' on a game unless either it's a planned sunset - like KSP1, or the axe falls from on high and then it'll be a sudden 'we regret to inform you'. And a CM wouldn't be told that sort of thing until they have to share that message.

So why didn't they release a finished, mature product? KSP1 didn't go onto Steam right out of the gate - it instead was released as a free demo. So yes, it got more time (though.. actually less time, since it was only in dev for 4 years when it got a steam release, not 6) to fix its bugs before asking the general public to pay for it. Also lots of games that release in EA get pretty good reviews. Sons of the Forest is currently sitting at 81%, just got release to EA. And the KSP community is far more positive than the Forest community. Come up with a better excuse than 'it's EA' when its clear that even EA expectations were missed with KSP2's release.

Funny you mention those two. Both are considered industry standards for 'bad launches'. But Cyberpunk's original launch date was set for April 2020 - it actually launched... October of 2020. In comparison, KSP2 got delayed by 6x as much as that. Cyberpunk's release was also considered buggy - and yet, you could play through the whole campaign, maybe people did. Performance was 'bad' because reasonable target hardware was hitting 45fps instead of 60fps. Save errors occurred very infrequently. Same with full crashes KSP2 - well, there's no campaign at all. It's just a buggy sandbox, crashes all the time. Compared to the launch of Cyberpunk, which was released in perhaps a 95% complete state, KSP2 is more like a 50% complete state. Fallout76 wasn't delayed at all, at least not publicly. It was launched in a pretty unfun state, but again, it wasn't missing nearly so much as KSP2. Its hilarious how people compare these to put KSP2 in a good light somehow - KSP2 is MILES away in worse state and after more delays than either of these historically bad industry launches. But another big difference is that the fan base for those two titles were far harsher to the poor launches than the Kerbal community is.

Sorry I'll get your name right in the future. Inadvertent misspelling. I've been a game dev for over 15 years. Not an amateur modder, actually got paid a living wage for my work. I've seen all kinds of projects succeeding and failing. I'm not saying KSP2 is unique in just how spectacularly its managed to love up, but I think it is pretty unique in how gaslit this community is after such a disastrous launch. And its pretty common to pay streamers to feature games, seemed quite fishy how several are dancing to a different tune lately. Regardless of whether its early in the game's release cycle, the time and budget involved would ordinarily have meant that such a bad first release would be met with even more critique - but that 'early access' label + a lot of hope filled fans is doing a LOT of heavy lifting for Intercept. I'm very familiar with sprints and Agile and if you read another thread I'm in, 'discussing' the matter with someone who apparently works on big IT projects but doesn't realize you can't just release bug fixes one by one, I'm with you on that. I don't even know why you brought it up, appropo of nothing I said. Though games do tend to work on non-standard schedules when they release a buggy excrements show with show stopper bugs to get specific fixes addressed asap, I'll grant you that with just how many save errors and crashes KSP2 exhibits, there's no reason to not just try and wait till they have a big batch of stuff to regress and release. Again, that doesn't matter, wasn't even what we were speaking about in this thread.

Also what happened to these YouTubers. Guys like Scott Manley, Linuxgurugsmer, and Shadowzone used to tear into KSP for relatively minor problems. I remember that scathing 'technical debt' video of Shadowzone's. Nowadays they seem to be far far far more forgiving of glaring problems, though of course not quite as biased as Nate is. Does an ESA trip really buy that much of someone's integrity? Or are there additional financial arrangements at work? I can't fault someone for making a living of course.

Yeah it was those evil businessmen that only gave them double or triple thier initial budget and let them delay the game over and over, and then forced them to ship it early...despite it being massively late. There's no way the devs could have known they actually were meant to ship it this time and should have fixed the bugs - that's on the businessmen for giving them so many extensions they just thought they'd have infinite time. Guys. Seriously. This is PR spin to make you keep hoping. Nate is a PR shill when he's doing these videos, just like Sean Murray or Chris Roberts. This isn't coming clean. This is literally the same guy who in 2020, 2021, and 2022 was telling us they're delaying so they can reach a really high quality bar. Either he was not telling the truth then, or he's not telling the truth now, can't have it both ways. I realize they did manage to implement clouds but you don't have to keep your head stuck there.

I'm surprised it took this long, they were at 4000 negative after just 2 days - Sales must be following the standard sharktooth profile, but compressed down to just a 1-2 days for the peak, vs 1-2 week like most game launches. Well, first patch might get some new blood if it's good enough.

The PR spin will tell you it's due to the 'profound nature of the technical difficulties ' or something like that - challenges a much smaller team of less experienced devs apparently managed quite well. Instead it's a vast array of poor planning/scheduling/focus on non-essentials/business screwups. Noone who knows game development or has followed this project and who isn't mainlining hopium believes things are business as usual for development of KSP2 and that it was only due to how challenging everything is that they failed to produce a game with 1. prequel-equivalent features, 2. new planned features 3. reasonably bug free/playable and 4. performing reasonably well on any hardware, even high end hardware. Even failing on 1 of those 4 points after multiple years long delays that cited trying to hit a high standard as a cause, would be reason for disappointment. Dropping the ball on all 4 of them indicates multiple major dysfunctions The people who tell you different have succumbed to the PR, or view themselves as unpaid company shills who want people to think better of the product than it deserves in the belief that eventually it'll become what they're wishing for(using your $50 to help), regardless of years of prior evidence to the contrary.

That may be an artifact of their future multiplayer implementation plans. If they intend to allow players to be on different points in time, players will have to have information about all events that are going to happen. They may be leaving the 'past' information around because they haven't written code to wipe it.

Take this with a grain of salt as it's second hand information, I didn't test this, but reports on reddit say the fairings don't shield parts inside them from aero at all. Fairings may currently be decorative items only.

From my testing pre refund, it's a simplified version of the KSP1 model (or at least with more artificial guard rails). It's still per part aero with artificial tuning around that, and no holistic occlusion, but now parts don't even have individual velocity. Or maybe I was.misunderstanding what I was seeing, aero debugging is also less robust. The game also seems to add some sort of artificial stability to craft. Stalls are even more gentle/nonexistent vs KSP1. Definitely easier for newbies to use, but also seems like it's even a notch less realistic than KSP1 aero.

Nope, was being 100% serious and not at all illustrating a point through sarcasm. How dare you not adapt to my statements. Change your mindset!

Take the OP's advice. Just change your mindset! Open up minesweeper but pretend it's KSP2. That's clearly how KSP2 wants to be played, after all. You have to adapt to it. Shame on you for expecting the game you bought to run and entertain you.

Fine but again, we're not talking about you at your.presumably much more permissive studio, and presumably much more reasonable development process, where you haven't been promising a release sometime in the next 6-12 months for the last 4 years, only to deliver an EA with terrible performance, showstopper bugs, missing both planned improvement features and parity features with your prequel, etc. We're talking about a hypothetical person slipping the beans about KSP2. Can you understand that there would be a difference, or can you only continue to relate it to your own personal status?

That's true when you're posting about someone else's game. I was talking about someone theoretically 'in the know' about what happened with KSP2. T2 is know to have hires sign pretty strict contracts about use of social media in relation to their work to be employed by them, so I do think a current employee there would be concerned about that, T2 is very litigious. Further, I think given the long delays, the state of the product, and the shennanigans with Star Theory, there's more going on with KSP2 than the standard industry turmoil you describe. Most projects are just normal workplace stuff - even ones that fail. Some are cursed though, I'm sure the Duke Nukem 3D team had a lot to say about thiers.

It's interesting how for 4 years here it was all about raising the expectations above what KSP1 - that it was going to be more performant, more stable, far less buggy, etc. And suddenly a large fraction of the community seems to be about lowering their expectations. But I get the sense it's the same portion of the community that was being unrealistic about the game before launch now being dogged about how we shouldn't expect much from it post launch.

That 60% also doesn't include costs for a lot of other things. I've often heard the # 3x pure salary costs be thrown around for game development costs.

Yeah there was a bit there with 1.4 and 1.5 where things were iffy with Squad, but I like where they headed after 1.6 and beyond. Breaking ground was a good expansion, and they reversed course on a lot of old decisions like not providing dV information.

That may also be the point? A community manager is not there to share concrete information. They're not a journalist or historian. They're there to connect with the community, to keep them happy as much as possible, get some information back to their bosses possibly, take the temperature of things - but primarily to help shape the image of the game. Sharing concrete information was a one good method of doing that with the KSP1 community, but maybe that's not the best strategy for the developing KSP2 community.

I doubt we'll get an answer for this. Noone working for the company will ever share an honest story because, as we've seen from the communications over the years, they constantly say everything was going well, and the only time they said things went badly, they blamed externalities - COVID, huge technical challenges, the need to reach quality, etc. Even with Nate's post today, claiming it was a 'bumpy' launch - it's all spin to make things sound like they're the most diligent publisher and developer ever. T2 is very very on message. Anyone who wants to publically share that information risks both lawsuits over NDAs they've signed and being blackballed by the industry for being willing to leak. And anyone who anonymously tries to speak about such things will be laughed at as a poseur. The only possibility is some sort of journalist willing to privately vet developers and then let them speak off the record, like the limited information we got about the star theory breakup. But that was a juicy story, vs this just being another dime a dozen botched launch. Gamasutra used to do a thing ages ago where developers talked about what went right and wrong with a project which was slightly more honest - but even that was spin doctored, though not as heavily as T2s public coms.

I didn't ask someone about a hardware and software project. Just software. Releasing patches for a game is not the same at all as release point fixes for a critical issue in a car. First of all, because if any car had been released in the state KSP2 was - early access or not - it would have killed half the people who tried to drive it, and the other half would be ok only because the car couldn't get over 5km/h. But again, I don't think you're arguing in good faith.

Yeah that's totally analogous to a software project. Man people really manage to stretch the limits of credibility to support thier desire to argue about things.

Based on the statement you made, that was very not obvious. Which project do you know, of any size, that releases single-bug patches?