steveman0
-
Posts
181 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Bug Reports
Posts posted by steveman0
-
-
21 hours ago, chefsbrian said:
Its not KSP2 or Unity specific, its physics specific. You can't really do parallel physics calculations, since you need the results of the last physics frame to start computing the future physics frames. In theory, you can parallelize calculations for groups of entities that are guaranteed to not interact in a frame, but in practice doing that without a bajillion bugs is near impossible. Best you get is KSP's attempts to on-rails stuff with predictable outcomes.
So while its inaccurate to say KSP2 is single-threaded, KSP2 is only able to effectively utilize a single thread for what is probably the overwhelmingly most expensive part of its CPU work.
Unity is not really central to the situation. It is possible to solve these issues either through multithreading, as difficult as this would be, or through batching parts to reduce simulation load. The choice of engine is irrelevant to this though as any good solution here isn't going to rely on the simplified toolset of any off-the-shelf engine.
-
3 hours ago, calabus2 said:
Unity strikes again
Is this based on actual knowledge of KSP2 specifically? Unity games are perfectly capable of leveraging multiple threads just as nearly any other engine when designed as such. Can you elaborate on the KSP2 design of their threading approach as to why it's an issue?
-
59 minutes ago, MARL_Mk1 said:
That works for a game already past 1.0. Not for an Early Access borderline unplayable preview build of what's yet to become a videogame (current KSP2)
It's this kind of exageration that makes it understandable why they wouldn't want to engage with the community right now. You hurt your credibility by making claims like this. 0.2+ has been quite playable. Not bug free, but easily playable for 100+ hours as a solid game with the addition of the science and mission objectives.
-
2 hours ago, chefsbrian said:
more towards being a long term activity
But... how? We already have science experiments that require time in orbit, or orbits, to capture the science for a biome. Just take that to the extreme? I suppose if they simulated it so that it could work in the background when you are controlling other craft, but how does this change the gameplay outside of encouraging time warp to finish it? This is a notable problem that hasn't been answered. What exactly would be different in this ideal new science scenario?
I already find science gathering to be quite an engaging, long-term activity to the extent that it requires me planning many missions to explore new places. Going places is a large part of exploring the game after all. Pressing the button to register I've done it is a nice conclusion to the efforts of the mission for me.
-
1 hour ago, Presto200 said:
Here's to hoping that them specifically saying "Primary" means that there will be expansion of side missions
They asked for feedback on missions for ideas before. I recall mention that part of the hold back on more missions was the logic to evaluate the objectives are met. As they implement these, I'm sure we will see new side mission types. For example, I could imagine a case where we get an "Observe an eclipse on a body other than Kerbin". They just haven't written the code to detect this event to make this a mission.
2 hours ago, Presto200 said:I think every roadmap milestone (besides multiplayer) will bring more missions
That'd be a shame. Imagine, multiplayer specific mission objectives!
1 hour ago, The Aziz said:Sadly, one that teaches docking ain't there yet
This was one I submitted to the mission/tutorial feedback threads. I wasn't the only one either, so I would expect it's high up on the priority list for addition.
-
5 hours ago, chefsbrian said:
it was also considered 'obvious' that they'd overhaul science which didn't come to pass
I think it was sufficiently overhauled. The one-click use is so much easier to mamage and the new parts and use methods are far more interesting to plan for. I've seen several people express disappointment but no one seems to have a better idea for what it should have been. Were people expecting quick-time events or puzzle minigames for completing science?
The science systems does a very effective job of couraging particular mission builds and to explore the system for new places to run the science. That's the core purpose it was intended to accomplish and it worked substantially better for me than KSP1. I've collected roughly 250k science in KSP2. Never even finished science mode in 1 as it and the career structure were too tedious by comparison.
-
I'm so glad the repetitive missions are gone. I hated them in KSP1. The boredom of them was a big factor in me never leaving Kerbin's SOI. KSP2's missions actually encouraged me to go and try new things. This is what missions are supposed to do, not make you repeat the same thing over and over.
I haven't even finished all the missions FS introduced and I've already clocked at least 130 hours in it. If they continue to add side missions and extend the story missions with the milestones, the tailored missions can easily add up to hundreds of hours by the time they reach 1.0. If you managed to finish all of them, I suspect by then you will either be satisfied, have your own list of things of your own to do, or have a list of mods that will fill in the gap.
It doesn't strike me as an important focus for core development to invest in repetitive missions when they can instead provide more incentives to explore.
-
16 hours ago, RileyHef said:
Correct! Here at around 15:40. "Significantly shorter turnaround time," with, of course, an large asterisk including everything Nate said in the minute prior to the timestamp about not wanting to make concreate guarantees.
In my eyes, a 6 month period between 0.2 and 0.3 would definitely be a significant difference. We will see come June.
I always like to highlight that the "significantly" here was notably understated compared to all of the caution about the timeline. I think it's a bit extreme to put too much stock in that considering there were doubts that it could even be faster. These kinds of interpretations are the kinds of things that set people up for disappointment. I think a fairer assessment is that it will be marginally faster.
-
1 hour ago, Scarecrow71 said:
The biggest difference is in the orbital mechanics involved in KSP2; Kithack doesn't have to worry about that...yet
And this is the elephant in the room. KSP2 needs to use moving, splitting, and merging frames of reference for compatibility of spatial scales of interstellar distances on temporal scales varying by many orders of magnitude for the simulation while also supporting multiplayer synchronization. These differences heap a ton of requirements above and beyond what Kithack requires as well as the original KSP where multiplayer wasn't in the mix of all of this. Maybe your experience is just different enough more than mine not to see this in the same light.
My background includes simulation including some real-time work that has taught me a few of the pitfalls of simulation resolution, sim time scales, boundary conditions, and the nuances of network sync that can uterly wreck havoc on the expected performamce of the sim when these all combine. It's a whole lot simpler when you can work in a simple single-res, single frame of reference, simulation space.
-
4 hours ago, Scarecrow71 said:
My point about bringing it up is to show that a dedicated person with a small team and almost no external funding is doing better than a professional team with the backing of a large corporation. Why can't TT/PD/IG do what Felipe is doing?
Am I mistaken in recalling that you were a software developer of some sort? Surely a software developer would recognize the substantial difference in the technical complexity of these two games. From the looks of it, it looks to be a fairly simplistic simulation compared to everything that KSP2 has to account for. I'm not seeing how the scope of these projects can be remotely compared.
-
6 hours ago, RileyHef said:
Putting on my skeptic hat here, but given that this message is not a confirmation that project cancellations and/of layoffs did not impact KSP2, could it be possible that this news impacts overall development after 0.3? Could we see the project be cancelled midway through EA?
Did you stop to think how silly that would be? The game is at what is likely the inflection point of the development cycle where it can start to turn the investment into profit. Cancelling would not only burn a huge amount of investment effort already committed, but also likely sink the value of the IP for the foreseeable future all while the game still has substantial financial value if it can get through to 1.0. Things would have to be very, very bad to come to that. FS! showed that they can still do it. There's still millions on the table in future sales if they finish it.
-
5 hours ago, Fizzlebop Smith said:
I think construction times could be reduced (engineer) and hydroponic growth increased (science) as two .. probably already listed..
That makes sense when considered for colonies of a few Kerbals. From the looks of the scales in mind, colonies might have hundreds of Kerbals late game once they become self sufficient. At that point, an individual Kerbal's specialty won't matter as much. Are we then going to be balancing whether we want 60 engineers and 40 scientists, or 40-60, 55-45? It strikes me as the kind of tedium found in some strategy games that doesn't fit KSP. If specialties return, I think the focus on applications directly to missions would be more meaningful.
With colonies taking the responsibilities of orbital science and assembly, individual kerbal specialties no longer have a central role. I'm open to ideas on how it could be done. Haven't seen any good suggestions that feel like they would fit the new model.
-
7 hours ago, cocoscacao said:
@steveman0 I mean... I can suspend my belief that much, but I like to picture it as a single purpose, highly specialized part.
Yeah, that would be cool eventually. I would guess they thought of it and it's somewhere low on the priority list if I were to guess. Good to put it out there as something you'd like to see. Just wanted to make sure you knew that there's at least a gameplay solution in the interim.
-
12 minutes ago, cocoscacao said:
I'm specific about surface samples. You need Kerbals for that. Storing those in capsules... yeah ok... but in probe cores? Not as much. It's supposed to be just a piece of electronic, without any sort of storage space.
I like to picture a little pneumatic mail tube that shoots the sample up the side of the craft into a slot on the probe core. It's a box, it could have room for a tiny bit of dirt.
-
1 minute ago, cocoscacao said:
Really? Probe cores transfer samples???
Yep, any control module can store any science collected. Whichever returns home first will contribute the science points to your total.
-
I got the impression with the past discussions and the shear size of some of the passenger modules that colonies may require many, many kerbals to be fully functional for late game such that it doesn't make as much sense to think of them on an individual basis as you might need a whole team to keep some colony modules operational.
That said, the careers aspect did add a nice angle to play for the ones who would be regular kerbolnauts. Needing to bring the right skill set for the mission added a small bit to mission planning. Overall though, I still felt it to be a pretty minor element that I don't feel it'd be a loss if they did away with it. If it was to return, I'd like to see much more dome with it to justify it as a feature to invest in.
-
On 4/16/2024 at 4:09 AM, cocoscacao said:
Sample containers
It would be nice to have a way to send back surface samples, without shipping Kerbals back with it. A storage unit for remote guided small rockets.You can use any probe core for this.
+1 for radial heatshields. I know several asked for this for spaceplanes. Good to have for general use.
-
9 hours ago, PDCWolf said:
Since the FS! release to today, if you filter by 2 hours playtime (the refund window), 57% of the people that bought the game, or hadn't opened it till then, left a negative review. And from the total (2000 reviews in that timeframe), 500 of them kept their review negative, or made a negative review.
I'm not sure the point of this even. Not playing more than 2 hours is barely enough to get a feel for any of the changes. You might barely hit tech tier 2 if you rushed in this time, maybe a couple of launches. This is precisely the correlation I would expect to see of any game. If you come in with negative impressions and don't give it a chance, of course your impression won't change. On the other hand, if you filter reviews above 2 hours, that is players who legitimately invested in the patch to try it out seriously, you find reviews are 79% positive. That's a massive turn around compared to the 57% lifetime value.
This is mirrored in the many posts I've read from players who actually played into the exploration experience: reception of the new features is positive. The team just needs to keep up the good work and more sour takes will turn positive when they give it a legitimate chance through future updates.
-
7 hours ago, PDCWolf said:
You can absolutely like the update and have fun and hundreds of hours in it, but reception is a measurable fact and oh boy. Reviews barely budged up to mixed from mostly negative, and there was another influx of negatives with that. The subreddit started automatically (by humans, not bots Dakota) downvoting KSP2 content again, media coverage was null. Sure, the player numbers jumped, but only to a quarter of the peak, and rapidly went back to <5% of original players, and half of KSP1. Also, the constant "complaining about complaining" and dwindling of activity tells you that even here the reception wasn't that good.
There are multiple ways to interet reception. Static metrics like a Steam rating is just one and it is plagued with it's own issues. The best way to understand is to read what people are actually saying and the response in posts following FS was extremely positive.
It should come as no surprise that many still did not update their Steam reviews as the game is still not finished. This is a huge factor for many who rate on Steam. You have to dig deeper to isolate feedback specific to the patch itself and thaf was quite positive. There is no arguing the 0.1 release was botched. Their marketing failed heavily in that regard. But FS showed that they are moving in the right direction. Keeping their heads down in face of all of the unreasonable vitriol makes sense as no amount of communication will alleviate emotions originating out of pure impatience.
6 hours ago, Scarecrow71 said:I agree that quality should be provided over quantity. I also agree that you can't just go fast-fast-fast because people want you to. But at some point they need to move with haste.
I see this as a direct contradiction. You can either go fast, or carefully. At least on that linear spectrum. Unless you want to add "cost" on as a point to a triangle, there are only ever two options. For a game targeting a life matching or exceeding the original, carefully will pay dividends over the lifecycle compared to short term haste.
Despite what the marketing had us believe about 0.1, my thinking aligns with @MechBFP in that development proper started much later than many wish was the case. I find the payoff for quality comes much later in the development cycle than the team is in now. Once many of the past mistakes are wrapped up, I would expect content pace to move faster as they will no longer be held back by the technical debt of the early alpha mistakes.
-
8 hours ago, Pthigrivi said:
If fans are going to throw a fit every time their expectations aren’t met its best to just not say or release anything until its fully ready.
This is honestly probably the most accurate picture of the situation as I see it. For Science felt like it came out of the blue and really nailed a lot of what they were going for with the update. It was remarkably well received and yet there are still people who downplay its success and the accomplishment of the team to move the game in the right direction. I can't really blame the devs and team as a whole at IG to not want to engage with the community right now when the overarching atmosphere is one of extreme impatience.
I personally have always been one to champion quality over speed. I'm happy to see the devs focusing heavily on quality in their current approach and this shows in the results of the FS release. If they keep this up, each milestone will win favor back bit by bit.
-
I'd say I already committed pretty seriously. When they said that we'd be able to play our exploration save through all later updates I was all in. I was feeling an itch to revisit KSP1 just before FS! release was announced as imminent so I held off on the revisit of 1 to pick up 2 again. Everything about the mission structure was just miles ahead of the repetitive contracts of the original so I went full on with completing the core story missions.
I still have a few side missions but my completionist side had me collecting as much science as I could along the way. I'd estimate between missions and scanning I got roughly a quarter million. This is good as I saw recently that colonies plans to fit in at tier 4/5 of the tech tree so I can dive right in once it comes out.
I'm on a bit of a break for a few other games that have been on the back burner for a while. I'm sure a few good bug fixes might pull me back in to knock out another 10k science collection or a mission or two. I expect that colonies will offer much to do such that knocking out the current array of side missions and science collection won't detract from the new experiences to explore.
-
2 hours ago, Icegrx said:
KSP2 currently plays like an incomplete test demo.
You mean the release version or the current version with For Science? Because if you mean the latter I have to hard disagree. It still has bugs, sure, but the exploration mode was a blast to play for over 100 hours before I got burnt out collecting a quarter million science and decided to detour to a few other games while I await colonies. The missions were great and the overall experience was far more compelling than KSP1. I still think I have a few dozen more hours clocked on the first than KSP2, but I have zero interest in going back to the original over just continuing off with the Moho mission I have lined up next.
The lack of communication is disappointing, but mostly because they hit it out of the park with FS and I'm really excited to start hearing how the design theory on colonies is turning into practical implementation.
-
8 hours ago, herbal space program said:
but in my book that would still involve resources on some level, i.e. some base resource has to be present on the body you're occupying to enable production of specific things. Even better in my book would be if the game requires you to set up your colony within some distance of said resource, e.g. near water deposits in some permanently shadowed Munar crater, which doesn't seem like it would require a whole lot of new stuff to implement, but I guess we'll see.
Colony placement was something called out before so it's a good theory. Even a basic true/false would suffice in my view: Mun and Minmus can only produce oxidizer, Duna can produce oxidizer and methalox, Eve can produce these and hydrogen as a simple subset of examples.
If the thermal management aspects of design are compelling enough on their own, this might be a far more interesting aspect than the fuel you get out from the effort. I'll likely enjoy building several in various places just for the fun of it and I expect they are counting on that too to keep it interesting until full resources are added later.
-
If it makes sense for the long term, sure why not allow conversions. I wouldn't ask for any more than is absoluately necessary of the interim solution. The goal should be to release colonies in a state that is fun new content that players can experiment and feedback on knowing there will be changes for the future final version refined by the resources update. If this will include conversion options then including those now gives the chance to benefit from the feedback on this as a gameplay option.
The Comprehensive Colony Communications Archive
in KSP2 Discussion
Posted
I don't think the parts are what are holding them back. As you noted, new parts and modules aren't challenging. The fundamentals of the system including improvements to performance to support them aren't something easily released early. The requirements are quite extensive and why I have always felt the original 6-8 month timeline was a bit optimistic. A lot has to be done if they are aiming for it to be complete to the extent that they can fully shift new development to interstellar.