Unixsystem

I tried using search and didn't come up with anything. I like the utility and UI for the Breaking Ground robotic parts, but I find them way too flimsy and buggy to be useful for any heavy duty tasks (Starship style flaps on extremely large diameter ships, large cranes for lifting more than a few tons, etc). I know you can lock the parts to allow autostruts to traverse the parts and make them stable for transport, but as soon as they are unlocked they just turn into noodles and flop around no matter what I do. I've tried screwing around with the part.cfg files doing things like increasing max torque or the amount of spring in the joints, but nothing seems to make a meaningful difference. Likewise for things like rigid attach or Kerbal Joint Reinforcement. Is there any existing mod/config edit I can try that may at least alleviate the problems, or are the expansion parts just slaves to the physics engine and limited to smaller capacity loads?

Been playing with this mod on my main campaign save for a few weeks now and I absolutely love it. It feels pretty OP and "game-y" at times but even so it makes science collection way more satisfying and I get a much more incremental feel for upgrading my rockets and space program versus my typical strategy of design a good rocket, fly it a few times until I unlock new parts, then start from scratch building around new engines or a new tank diameter. If I could request one thing it would be upgrades related to radiators, both max cooling and core heat transfer. With Nertea's atomic engines mod, it becomes borderline impossible to cool the high end nuclear engines after short burns with a couple fuel flow upgrades, even with dozens of the highest end Heat Control radiators. Being able to beef them up a bit would both make them scale with upgraded engines while also making it so that even non-nuclear craft could use one or two small rads to cool other equipment and save on space/part count. Otherwise I love the mod and really hope to see compatibility with fuel switchers and similar mods in the future.

Different person, but I just downgraded from Restock/Plus 1.1.2 (where R&D wasn't working) to 1.0.3 and now it appears to be working again. 78 other mods installed including most of Nertea's other stuff, so it's definitely specific to Restock 1.1.X.

I wasn't using them as radiators per se, just using them to physically cover up sensitive parts on the windward side of the craft since they have a higher base heat tolerance. With Heat Controls curved rads it actually doesn't look too stupid or add too much to the part count and it does help with high speed reentry. Again I know it's such a niche use case that I don't expect anyone to add it to the SR computation, I was just mainly curious if there's any way to figure out the actual speed of a given craft before reentry becomes a dice roll. That way I could at least clip a couple ablative heat shields into the fuel tank or something. Then I could aim my supply ships straight at Kerbin's atmosphere from Duna and then forget about them knowing that SR would take care of the rest.

There's no actual issue with the mod, everything appears to work as intended. It would just be nice if in the VAB GUI where we have the terminal velocity stats for each stage if we also had stats for the maximum reentry speed before risking the stage burning up. That way we could see how much ablator and/or radiators we need to ensure survival (assuming rads are even counted at all). That said I know you're maintaining like half of the useful utility mods at this point and ultimately it's a really minor complaint that really only helps in a handful of niche use cases.

Is there any way in game to tell what your chance of burning up on reentry is? SR works well for dumping interplanetary transfer stages back into Kerbin for recovery without needing to manually land every one, but they inevitably burn up. My ship designs generally don't accommodate heat shields and it seems like radiators don't actually help with SR's reentry calculations (or maybe I'm just not adding enough, I can't tell). Some sort of added column in the VAB popup that mentions "maximum entry velocity for 100% reentry survival" would be neat, although I'm not sure if this mod is actually be actively developed or just maintained at this point. Edit: I should note that I'm using a Starship style reentry profile that can survive pretty easily when done manually. I know rads generally aren't great for reentry heating but I feel like using them to line the wildward side like makeshift tiles are the closest I can get to a heatshield without just making the whole thing look goofy. Maybe what I'm looking for is just beyond the scope of the mod, it just gets real tedious to land them by hand every time.

This mod seems extremely well made. The engines and tanks both look great, the engines all seem relatively well balanced to one another and the addition of boil off is a neat feature to take into consideration in ship design. All that said, I can't for the life of me find a scenario where the added ISP outweighs the added engine/tankage weight needed to take advantage of it. I'm still relatively early in my campaign, so is it just a matter of the benefits becoming more noticeable as the ships get bigger and bigger? Even getting into 2.5m+ parts it still seems like its always easier and more effective to just strap a traditional LOX engine on and get basically the same delta-V. I know that this is an issue IRL as well so I know it's less of a mod problem and more of a physics problem, but from a pure gameplay perspective where are the actual use cases where it definitely makes sense to load up a bigger H2 rocket over a smaller LOX one?

I hope. They delay it further. As promising as the project sounds, what they showed seemed pretty early and release within even a year, let alone 6 months, sounded like a stretch assuming that they intended a finished product with all the features that they promised. Sure, they can always patch things after the fact, but it's always going to be easier to optimize and bug fix a game pre release than after it already live and you can't really tear out major gameplay/technical systems in order to make them run better. I want KSP2 to be the game that completely replaces KSP1 on every level (content, performance, buginess, UI, etc) and isn't that bad sequel that everyone pretends never happened after the fact. That just can't realistically happen on a roughly one year development cycle. Take your time, do it right. Delays in game development are almost always good news for the finished product.

Sandbox mode and Career mode are totally difference experiences, and adding an extensive narrative that gets in the way of actually playing a career game would make that mode substantially less appealing to me. You could always add an additional mode that is career + story, but my worry is that then the career mode without story wouldn't include the alien aspects. I definitely would like a more goal oriented, scavenger hunt style way of playing, I just would rather that it not be bogged down in narrative. I suppose that I should define that when I hear "narrative" I'm assuming that includes things like cutscenes, extensive dialog, scripted events and (relatively) tightly defined objectives/goals. Many of those things wind up being both very time consuming and very expensive from a development perspective, and at least for me that would be time and money that would be better spent almost anywhere else. Not to mention that while the writing in KSP has always been mildly clever, I don't know that the world of KSP really lends itself to a really engaging narrative. Could be done, certainly, but again it requires throwing resources into something that is completely outside the bounds of the reasons that many people play the first game for. Now, if by "narrative" you mean some text backstory/early game optional handholding to lead you to the first anomaly, followed by something like a few paragraphs of text for each subsequent anomaly explaining what it is and vaguely hinting at where you should go next, I'd be totally on board with that. Sort of the same way that the built in missions in KSP1 have events that will cause the KSC mission control to pop up and explain what happened/what's next. That would be easy to implement and could help give more of a driving force to finding cool stuff and is more along the lines of what I was originally thinking.

I disagree. One of the biggest appeals of KSP is the sandbox nature of the game and the freedom to set any goal you want. If they bake in a real set narrative then you essentially get railroaded into doing what the game wants rather than what you want. By making any sort of hypothetical alien artifacts/planet a completely optional "side mission" of sorts, you give something for players to hunt for and be rewarded if they want that sort of experience without hampering those who don't care and just want to do their own thing. That said, I could see an optional extended tutorial option at the start of a career game that sort of holds your hand in building ships, getting to orbit and maybe your first Mun landing, and in that they could possibly introduce the idea of an alien scavenger hunt (maybe the "narrative" to the tutorial is that Kerbals want to get to the Mun to investigate a strange signal, for example). That way you give a newer player a concrete goal to progress towards as they learn how to expand out further and further, but again it wouldn't interfere with those of who are more focused on building optimized space planes or the perfect orbital colony or what have you.

Actual live aliens would massively increase the amount of work to get the game out the door, while adding basically nothing to the core draw of the game (building, exploring, colonizing). There are loads of space sims and 4X games that cover alien diplomacy, I'd rather Kerbal stick to being Kerbal. That said, and expansion of the anomoly system could be really fun. Doesn't need a full narrative or anything to that extent, but some sort of progression of finding weirder and weirder anomonlies, maybe where one leads you to the next, could be really fun. Cap it off with some sort of unlock for finding a certain number/all of them like a hidden planet that is otherwise invisible on the map, and you've got some really neat endgame objectives. Pretty sure something similar was planned early in KSP1's development, but with the bigger focus on deep space and surface exploration in KSP2 it feels like a natural fit.

I should clarify that I don't necessarily mean so much actual realistic, live development updates, and more that I'd expect a pre-planned media rollout strategy. The initial announcement had quite a bit of fanfare, so in a lot of other circumstances I'd expect them to have a schedule of media releases planned out. For example, one month they'll have a text post going over the propulsion systems we saw in the trailer, largely covering the same info we saw in interviews with maybe one or two extra tidbits. Next months, maybe a handful of screenshots to show the planets of the Kerbin system and the face-lift that they've gotten. The month after, maybe a post about the new resources and how they will be managed (in general terms, I'm sure that even they haven't ironed out the specifics). Even if they were just posting some basic info with a few new screenshots every few weeks, it would be something to let those following the game latch onto and keep the hype train rolling. Again, I'm not saying that the lack of these things is really a problem, it's just somewhat surprising how big of a splash they made for the announcement, and how open they were with the community, followed by nothing. Hopefully they are just saving up all the new things they are working on for another big media circus in ~6 months.

While 2 months isn't all that long in game dev terms, it is a bit strange to have this big media push and then immediately go dark. I'm hoping that the reason for announcing it at all was to gauge the reaction and get feedback from the community in an effort to better tune and balance the game, and their outreach to modders and Youtubers somewhat supports that. The more cynical side of me still thinks it's awfully weird not to even have any sort of basic monthly "here's what we're working on" status update to keep a low level hype train rolling. It's possible that they are just heads down trying to work with and implement the books of feedback that they got from the community, or maybe they planned to continue providing media and hit some development snag and don't have anything new to show. Either way, I think 2 months is way too little time to get too worried about the state of the game, especially given that it was only just announced and isn't coming out for potentially a year or more.

Really unless you get awfully close to it, I doubt that Kerbol is even rendered as a polygonal object. At great distances it most likely acts solely as a light source with a 2d texture acting as the flare that you actually see in the sky. At worst I'd imagine that from Kerbin's orbit or further if it is rendered in 3D at all it's likely represented by no more than 1 or 4 pixels on any common resolution. Once you get to Moho or closer it may actually start to be noticeably polygonal, but even then it's so difficult to get close that I'd have to imagine that it's the least detailed object in the solar system.

I think there may be some fundamental misunderstanding about how rendering works. I'm not an expert by any means, but my understanding is that with traditional rasterization (aka 99% of all games) everything starts from the geometry that is being rendered. So if you're landed on a planet, you have polygons for all of the terrain, KSC buildings, and your ship whereas in space you essentially only have your ship, and then potentially a planet or two which will be very low poly due to level of detail scaling. Once the GPU has the actual polygons in place, it will then look at the available light sources and shade the polygons that will be visible from the current camera position. So in a typical scene the only major light source is going to be the sun, which is treated as a point source essentially infinite distance away. This means that in space, the game only needs to render lights coming from one source, on a handful of geometrically simple objects, which most of the time is only taking up a small fraction of any given frame (the rest is just skybox, which is basically free from a rendering perspective). If you add lights to your ship that ramps up the calculations quite a bit, but you need a lot of lights to really start making a difference. On land, you also need to compute the lighting for all of the terrain and ground scatter around your ship, but again the terrain in KSP is so simplistic compared to modern games that while more work than a space scene, it's still trivial for most modern GPUs. In terms of your comments about the dimensions of these calculations, they are always the same regardless of where you are. Just because you as the player have no real reference frame for where exactly you are l, the game always knows your exact coordinates, rotation, and location relative to the sun. The calculations are equally as intensive on Kerbin as they are out floating out past Eeloo. If the game can render your location on the map screen, then it knows where to place the Sun to properly light up your ship. EDIT: The tl;dr here is that graphical performance in a visually simplistic game like (stock) KSP is going to be primarily based on the number of polygons on screen at any given time. In space, that will almost always be less than on a planet, and generally in order to stress a decent GPU you would need to load up so many parts that you would start hitting CPU limits long before any graphical ones.

Rendering in space is vastly easier than rendering something like a first person game set in a detailed ground environment. The game's skybox is just that - a 2d box (or sphere) set at infinite distance from the objects in the game, and it requires essentially zero graphical horsepower to render. That means that when in flight you have at most 2 polygonal things to render: the ship (or ships) in flight and a nearby planet if present (more distant planets are so small on screen that they usually amount to at most a few hundred pixels). Even high part count ships in KSP are extremely geometrically simple by modern hardware standards, so calculating lighting and shadows from a point light source like the sun is relatively simple. Adding dozens of lights onto the craft itself can and will result in a much more dramatic performance hit, but you need to really go ham with light sources to tax a modern GPU. Planets don't begin to render shadows until very low altitudes, and even when they do the landscapes are extraordinarily simplistic and the textures are low resolution (finally being updated to higher quality but still nothing to write home about). Turning on terrain scatter can increase this quite a bit, and if you start adding mods like EVE, Scatterer and high resolution texture packs, you can absolutely start becoming GPU bound, but in vanilla you need to be running a very low end GPU to hit the graphical limits before the much more common CPU ones. In response to your last comment, I assume you are inferring that he's saying that the physics problem is unfixable and as a result KSP2 doomed, but that's not the case. KSP1 was built by amateur game devs and was not optimized for the level of performance it demands given the type of game it has become. At this point fixing KSP1 would essentially require tearing apart the entire engine and rebuilding from scratch, which would be way beyond the scope of an update for a game this old. KSP2, however, is being built from scratch, and so they can build the entire physics engine around the unique problems that the first game constantly runs into. It remains to be seen how successful they will be, but they've made it clear that performance is a key development focus so I'm cautiously optimistic.

It seems to me that most of the potential problems are 90% solved by improving the base performance of the game. KSP1 by all accounts seems to be a game that is built on an extremely rocky foundation and even so we've seen modders add most of the features KSP2 is promising, usually only limited by the fact that you can only build ships/stations/bases so big before the game starts to fall apart. With the ability to start essentially from scratch and build the engine to support large part counts (and MP) from the beginning, most of the remaining work is just on the design end to make it actually be fun. Not a trivial issue, but a lot easier than wrestling with flaky technical issues and it's an issue that they have copious mods to look at and see what works and what doesn't. I certainly don't think it's a guarantee that KSP2 will love up to its potential, but assuming that they aren't outright lying about their work rebuilding the engine I don't see any reason that what they are promising is unreasonably ambitious. That said, healthy skepticism is always good with an unreleased game.