TheTechnobear

yeah, this is what we need the highlight of the EA day, by far, was the KSP trailer... that was absolutely awesome... even though the actual EA was disappointing, and it was pretty sad also seeing its reception on YouTube. that trailer, made me want to go back and try again ... just in case

NMS's turnaround is the exception, rather than the rule .... and it still took a long time, and also with a game that also has a broader appeal anyway. put another way, Im pretty sure HG would consider their success was despite their failed release, and would not call it a good strategy! whilst NMS's story provides 'hope', its a bit clinging at straws... not what you'd want from a highly anticipated release! I guess the question is how/when will we see significant improvement (both performance and bugs). but we need to keep expectations low... a lead up to a release like this (and the EASA event), you throw everything at it, so there is not going to be huge bug fix release quickly. this is going to be a slow grind.... Id say many months! this also means, we are likely a long way from any of the new features on the roadmap. (or at least I hope they stabilise the foundation before adding more features to a wobbly rocket... :)) gaming is one of the largest markets within technology and entertainment... its great that KSP might help in STEM, but that market is much smaller..... and how many schools have a PC with a 4090

and there's the rub.... KSP2 EA will cannibalise KSP 1 sales.... even with KSP2 being labelled Early Access.... most new players, will see the price tag, and assume KSP 2 is the future, KSP 1 is dead.... then they will load it up, see its 'current state', and then likely request a refund... it is highly unlikely, they will then say.... lets go try KSP1, see if thats better. and there is a good chance they will never bother trying KSP2 again, even when it makes final release. KSP 2 taking over is a natural progression... but thats why it had to be at least as stable, and performant as KSP1

not sure this is their call, at least on Steam. Steam will give you in refund 'automatically', if you have < 2 hours playtime. thats why a quality EA is important, otherwise regardless of 'caveats', many are going to just 'give it a go', and then review/refund. (if you look at how Valve describe an EA, you can see why they don't make allowances for them) this is only going to get worst, over next few days/weeks...as there will be an increasing % of new players vs loyal KSP fans. (the latter being more forgiving, and willing to support devs for the KSP2 future etc)

indeed, as per OP, as far as Ive seen KSP2 is totally GPU bound... even on relatively humble systems, the CPU is hardly being taxed. also, I noticed that changing the graphics settings, including resolution did not seem to affect GPU usage/frame rates as much as expected. which was quite surprising given the GPU load. so, this leaves room for some hope... perhaps a GPU bottleneck, which once unblocked, may allow the processing to be better balanced over GPU/CPUs

Lots to love... Sound design and Tutorials are the stand out for me...the graphics are also great. Generally, I think the UI looks and behaves much more intuitively. definitely has potential to be great.

Unfortunately, I have to agree... this should never have been released as EA... its just way too early.. the main issue here is 'reputation risk' , look at comments here, YouTube, and steam reviews... its a dumpster fire. that reputation will take a lot of effort to now shake off... and you have also lost that initial launch day buzz. for what? the developers knew its performance was bad, it had bugs... they don't need community feedback for that. but oh well, thats their call, I don't mine that... but what does worry me, as a dev, is I don't really buy the "it'll be optimised later". sure, there is such a thing as premature optimisation, you don't overly optimise initial code, and, yes, this would be true of an EA, and can be improved later... BUT that does not mean you don't implement efficient code in the core from day 1. no dev has time to completely re-write this code continuously, its only tweaked later. if we are going to get 500+ parts craft, colonies etc...then that core design has to already be in place now. it has to be designed to work at this larger scale. so, even without the final optimisation, it should be handling 100 part craft easily.... that allows headroom for larger craft later. so unfortunately this EA signals something worrying... either the core design is no where near complete, or its not going to reach the goals that have been set. this is where the EA fails for me... I don't mind missing features, even some bugs, thats kind of expected. but I would have expected/hoped performance at this scale, would have been better than KSP1. this would have left hope KSP2 foundation was solid for its future plans. anyway, there is always hope.. but Id say we are at least a year away from KSP2 being KSP1 performance, and frankly, that is when they should have released the EA

I got it to run on the Steam Deck via Proton. Im getting around 10-15 fps. with simple rocket in flight .. (low settings) , which is better than I thought note: it won't work directly from the Steam download, as the Launcher does not work... so I had to bypass