-
Posts
104 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Everything posted by BmB
-
It has taken a year to release the first substantial update, this is obviously not okay. If it takes a year for each item on the roadmap it might literally not be done by the year 2030. But the update is pretty good, the tech tree is well designed, the missions are fun enough, and it's good to have some sensible structure to the gameplay for the first time since 2013. However the game is just too buggy, it has been buggy for a year with little to no sign of improvement, urgency, or caring. So I'll make a list of the biggest gripes I have right now which is keeping the game from being fun or playable. The parachutes, why do they not deploy? I had to do a descent 5 times just now simply to get the parachutes to deploy. The stages, why do they not respond? I often press space and nothing happens. I had a stage of sepratrons which I was going to use in case of parachute failure, of course you have to fire them at just the right moment, and if the stage does not go off, then you can imagine what happens. Even a backup using a totally different game system is non-functional. This most critical mechanic being unresponsive and unpredictable makes the game simply awful to play. Staging needs to happen consistently and immediately, without any uncertainty or doubt. The VAB, after using it for some time it will just start to bug out, maybe you can recover it by reloading or maybe you'll have to start a whole new campaign (which with this update just became unacceptable.) you just don't know. You can sometimes not rearrange or edit stages, you can sometimes not move parts, you can sometimes not set the launch assembly. The delta-v calculator is all over the place, it's nice to have a built in dV reference, but it is totally useless if you cannot see what your dV actually is. Sometimes it is stuck at 0 and will not update until you rebuild the entire rocket. The stage dV disagrees with the assembly dV, which is correct? Sometimes removing and adding the same part in the same place will give different dV readings, at that point not having it would be less confusing. The save system, just why? I don't understand the point of having a different name for the "workspace" and the craft, that is just confusing, as far as I can tell they should always be the same name. And your list gets polluted with random autosaves that delete the description. There was nothing wrong with how saving craft worked in KSP1. Building a rocket and pressing "launch" DOES NOT SAVE THE ROCKET. If you do not remember to manually save before launching, YOUR BUILD IS DELETED. What the actual love? In KSP1 you would always get your last build back whenever you went into the VAB unless you specifically pressed "load" or "new". This is just a blatant disregard for the users critical data. If you're lucky an autosave might have happened in the last 10 minutes but it does not save the final state of the craft, so good luck with that. Why is there no interstage fairing? You can't make apollo. Switching the launch assembly loves up your staging, each assembly should really have it's own set of stages. Autostaging in general is terrible. In KSP1 pressing the reset button you could generally expect a reasonable result as long as you weren't being too clever in your build. Here you HAVE to adjust staging every time you make a change because the default staging is so bad and obviously wrong, main engines at the bottom of the rocket will appear at the top of the staging list, endless empty stages get created. All in all building is a horrible nightmare due to all these issues. The "landed" bug after taking off with a lander and being unable to see your trajectory in the map has been there from the start. Between the light bug where you can't turn on lights and the lack of helmet lamps on kerbals means night time or even landing in a crater in shadow is a no-go. Everything being "default-name-2734" and having the capsule icon means I have no idea what anything is. As soon as you detach or dock something like a lander it loses its name, and everything has the same icon. It's easy to get lost. Again this was not a problem in KSP1, with auto-naming and icons being mostly sensible and only rarely you had to manually fix it. You can't even change the icon in KSP2. I can only describe this as a state that makes you feel that"nothing works". I fear I'll have to wait another year or more for things to improve. And releasing a major early access update just before the holidays knowing full well it may have issues is such a genius move, making any kind of rapid response to a buggy update impossible.
-
If the game was stable enough, there's loads of things you could give feedback on, like the new UI, the new parts, the music, the tutorials etc. But you can't, because none of it works. KSP2 has no modding support, which is worrying. The mods you see now are done using hacks originally developed for japanese porn games. KSP1 had asset loading, plugins and data driven config files almost from the start.
-
I will say I'm fine with it being just sandbox mode with a limited selection of parts right now, it's the game breaking bugs that are the issue.
-
Once again, Steam Early Access means a finished product.
-
By releasing in Steam Early Access for 50€ it did make that claim, it's implicit to the platform. [snip]
-
Early Access is, once again, NO EXCUSE. Early Access games must still be working games. If this was a closed alpha I'd say whatever. But this is a finished product that costs 50€. Stop using that excrementsty excuse. So it was recalled. On Steam, Batman Arkham Knight was recalled, and that was way more functional than this is.
-
Yet good companies do push out hotfixes for critical issues within sometimes a few hours or days.
-
The VAB does not work, it often breaks. Many times your flights will be destroyed by bugs, sometimes you can't launch, sometimes you can't save or load. And yes, there are crashes. The game is basically non-functional.
-
Exactly, I didn't pre-purchase fixes in 2 weeks, I purchased the game now, the game must work now. Fixes HAVE to be released now. [snip]
-
They can only improve the game, it's that bad. If they have the fixes listed in the OP, then they can be released. As we've already discussed over and over, EA is not an excuse, there are guidelines and rules for the quality that an EA product should meet, it is still a finished release.
-
We didn't pay for it years ago, we paid for it now. That leaves an obligation on the part of the developer. Cyberpunk 2077 was recalled, this is honestly no better. They need to release fixes ASAP.
-
You keep acting as if it's guaranteed to blow up the game rather than a small risk, a risk that is insignificant next to how broken the game is right now. The worst that can happen is it's broken in a slightly different way. If they don't release anything the worst that can happen is it stays this broken, which I might remind you is very bad. Not my job to care, it's his job to fix his product. They're not your friends, just because he makes a puppy face on camera doesn't earn him anything. Maybe if he isn't capable of getting this under control he should be axed.
-
Sure it has.
-
Leaving us with a broken product for extended periods is what will ruin their reputation. If they release fixes promptly that is the only thing that will save that. If, and that's only IF they accidentally make something else broken, they apply the same principle and fix it promptly or roll back a version. You should do this until the game is stable enough to justify waiting for QA. And I'm talking about the showstoppers here, if they have a fix for KSC appearing in front of you and destroying your ship, then they need to release it now, not in two weeks. If they have a fix for rockets bending out of shape and destroying themselves, they need to release it now. The launch button randomly stops working? Fix now. Ships loading in all broken and messed up? Now. There's plenty of game breaking bugs like this which can't wait. These are the exact type of bug that you are more worried about hypothetical versions of that may or may not get introduced, than the actual bugs that are in the game right now, to justify not fixing what is there that makes it unplayable. You don't need to defend such excrementsty development practices, you really don't. You get nothing out of it.
-
Releasing patches isn't going to make the game worse [snip] Yes, it may introduce some new bugs, emphasis on MAY, that is still better than leaving bugs this bad unfixed for weeks, and those new bugs can themselves be fixed. It's something you only worry about when the game is already stable and you don't want to accidentally make it unstable. But that's not the case right now, the case is they have fixes and they are withholding them for no reason.
-
No, not when the game is in this poor of a state. Once it gets more stable only then it will make sense to test patches more thoroughly
-
I am complaining that they released too early, I already said I'm empathetic to the fact that such release dates are set by publishers although the devs do have a hand in not meeting whatever internal deadlines were set. But they did release too early, so unless they're going to do a recall they need to release the fixes they have.
-
So those aren't what they're planning, but what is already completed? That is even worse, that is how much better the game could be RIGHT NOW, but which we are forced to wait weeks for for no reason.
-
It means solar system sized simulations are not supported in any engine and you'll need to make modifications to any engine you choose, unity is if anything more friendly to this kind of modification than most engines. Another major problem is game engine physics are not meant for large stacks and complex constraint systems like KSP rockets, they are optimized for performance over stability, things like a few props you can pick up and throw around in HL2. I would say many of the problems with KSP (the "kraken") are due to the instability of the physics engine, If I had a say I would make it a goal to create a custom internal solver for rocket stacks, and let physx worry about only the entire assembly and loose debris. Looking at the most recent unity docs I can see they have actually added a constraint system like that already, for robotics, very handy.
-
KSP is a game that demands custom systems that aren't well supported by any engine, except perhaps I think unigine had a version for large scale simulations
-
They ought to have not released it like this. I understand that it is ultimately the publisher that sets release dates and prices, but that does not absolve the developers, as we know of multiple public delays, which probably cover several more internal delays. So what we can get from that is intercept failing multiple times to meet milestones and take 2 forcing them to release to recoup some costs, at the end of the day consumers are the ones getting shafted with an overpriced and underdeveloped release, and you get nothing good out of defending this or making excuses. The least they can do is not keep us waiting any longer than we have to by releasing critical fixes for the current version as soon as possible. "Weeks" is not an acceptable answer here, never was and never will be.
-
As we've already discussed, Early Access is not an excuse, it's still a release that is not exempt from quality standards.
-
And I'm saying the current bugs are more critical than whatever you're afraid of introducing by accident. QA for a patch makes sense when you're trying to maintain some kind of baseline stability, but this live version does not have that. So any delay is for no reason. Hotfixes that circumvent the normal patch development process to get critical fixes out fast are not some foreign alien concept I just invented. Let's look at how an actually good company handles this stuff, I bought Half-Life 2 Episode 2 on release, it had a game breaking bug in the antlion caves, later that afternoon a fix was released and I could continue playing. I didn't have to read excuses on a forum about how QA takes so much time and they need to test everything properly so it'll probably maybe come somewhere down the roadmap in some weeks, they just fixed it, that's what we should expect with issues like this.
-
I'm going to tell them that delaying fixes is not a good idea.