KUAR

did T2 refuse to do any work in case their lovely paintwork got wet?

Sorry to say, Westinghouse, but if it is cancelled the chance that any of that will happen is next to zero surely.

At risk of deviating from the topic, have you ever tried running scrum for a hybrid software/hardware product which is subject to waterfall procurement and agile development (internally)? Really struggling to mesh the competing frameworks!

For me, the fun is about discovery. I don't think we need more planets, but I regret that we can see the current ones as clear as day from the KSC, mission control and the tracking station. I wish that we started out not knowing anything more than Kerbin, Kerbol, Minmus and Mun. Everything else needs to be discovered. We can get a rough, grey-ish indication from a space telescope; we can get more information by flybys; more by putting a scansat into polar orbit. That way, every mission adds something to our picture of the cosmos. The concept could be extended to Interstellar. Rather than identifying a specific planet, perhaps we only know that there's a star our there. We need to send a probe to figure out what planets there are and their orbits, and whether they have atmospheres.

Upvoted for the "Babe" reference

This is listed in "known issues" on the release notes

I have also seen this bug

I have also seen this bug

That's good and I agree...but figuring out the new drag cube maths will be interesting!

This is honestly the only way I see it working and I (perhaps naively) think that it should work. Perhaps it's an extension of sub-assemblies: you can define an assembly which is a rigid body, with a single resource storage/generation/specification characteristic, etc. etc. There would of course have to be limitations - decouplers/docking ports couldn't be in the middle of your assembly for instance, and perhaps it's progression-locked (rigid sub-assemblies up to X tonnes/Y parts, upgradeable like the VAB size/part count limits). It could also be limited to a certain percentage of your overall craft's mass or part count. I can see that it would be complicated to implement with e.g. aero model, parts manager UI, etc. etc. but I see it as being the best option. A small number of large monolithic parts doesn't interest me as it'll kill variety. If they're going down that route then serious investment in part development would be required.

As a person who generally errs on the more positive and optimistic side for KSP2, sadly I've been a little underwhelmed at the imagination that the new science mode appears (at first glance) to exhibit. Then, positively shocked at the clunkiness of the work in progress reentry heating animation! I expect it will be made a little more slick before release but right now it looks like the graphics design apprentice did it as homework. Sorry, I know there's probably a committed developer on the other end of it, and I'm sure there's complexities I've missed...but that's how it looks to me.

I disagree. I'm fine with being punished for forgetting to do this at the right time. It's realistic - we've lost control of the probe. Allowing us to deploy them after that is cheaty like the OP implied. Now, there's an argument over whether some models of panels have some exposed area even when folded away - that's fair.

Better - scrap the current method and have a kerbal placement dialog after hitting the launch button. It's something I forget to check far too often. I'm launching an unmanned rescue mission only to discover all the seats are already occupied by hitchhikers...

Before I start, I'm of the view that KSP2 should be aiming to differentiate itself from KSP1 and with the first two milestones it feels like the aim is to create a remaster rather than a sequel. It feels like the team is a little afraid of changing too much look and feel - although perhaps they have their hands full with the under-the-hood changes that are needed for the later milestones. I'll also deliberately avoid commenting on specific parts, and gameplay mechanics that conceivably could be in announced milestones. That's a separate thread I think - e.g. "how should colonies work". That said, here's a list: - Take the most popularly-installed mods from KSP1 and bring them into core. The ones on my mind right now are procedural parts; docking port alignment; KER; etc. Specifically, I can't see a good reason why not to have procedural tanks, even if it's of a set diameter and variable length. Adds some graphical challenges for the part modellers but as a gameplay mechanic and general tidy-up of the parts catalogue, invaluable. - More reasons to progressively explore planets with different science experiments to enable future exploration/expansion. For example, off the top of my head - we can use a drag prediction tool so we can get the right parachutes config...but only if we've measured the atmosphere composition at the relevant altitudes. We can EVA...but we need to have measured the temperatures, radiation etc. so we can "have a suitable spacesuit". I'm of the camp that says it doesn't have to make perfect scientific sense. It's a game and they rarely are completely logical. But, if you could imagine yourself convincing a five-year-old that "we need to do X so that Y can happen" then that's good enough. - Mission planning tools. I've mentioned parachute predictions earlier - it's disappointing that we need to guess the number. Equally, landing stability calculations - will my craft balance OK with this much fuel left on this angle of slope (ok so perhaps we have CM indicators for that, so never mind). I'm not sure how realistic re-entry will be but do we need to identify how much ablator we need? Heat/solar management predictions. I know we can go get a notebook and calculator for some of this, but some in-built tools would be really cool. - Ability to rotate docking ports once docked, or they snap on docking to the right orientation if you're close. If I'm building in orbit, I want to be able to get the modules of my space station lined up right. Aah dammit, just generally more support for orbital construction. I'm now moving away from the things that have bothered me, and am now into the realms of pulling ideas out of thin air... - Auto-adjustment to achieve a given CM or CT. If you're creating a non-symmetric design, when you thrust up it can start to rotate (obviously). Tools in the VAB to offset certain masses or thrusts to achieve alignment (might need to assume a certain fuel load) would be handy. Later, Auto-adjustment of engine thrust per-engine (or per symmetry set?) to maintain heading - we can do it with RCS, reaction wheels and gimbal but not dynamic thrust adjustment. Perhaps only certain engines can support it for "technology" reasons... -

I can' believe we ever used to build wings from multiple fixed panels and it's revolutionary* (now one of the wings doesn't think it needs to be upside-down in the VAB). *no pun intended

I don't think those are necessarily incompatible. I don't work in the industry but I do work around Agile software engineering. I can see how 80% of big games out there are basically the same and how KSP2 is novel in many ways. That's why we love it and why we're excited about it. Platformers, first-person shooters, open-world exploration, and real-time strategy. They've been done many times before. I reckon an experienced developer could put a pretty good estimate in terms of sprint points on how long it would take to develop a new level in COD 14 (Modern Warfare 5) for instance, or to develop a new diplomatic feature in Civilisation 8. In contrast, how long is it going to take (for example) to develop a new manoeuver system which (unlike KSP1) double-integrates the acceleration over time in two orders, taking into account staging and mass changes, and god knows how many other external dependencies, to figure out a predicted orbit after burn; present it in a user-friendly way; and provide tools to support execution? When you think about the engineering challenges, re-designing that particular system so that it supports long and slow interstellar burns (DAWN-like) is actually a real headache with a lot of dependencies and unknowns, which hasn't been done in gaming before. The KSP1 equivalent now looks trivial. And that's just one example - I'm sure there's others. The mechanics for interstellar, time-warp enabled multiplayer while simultaneously supporting automated colony/resource routes are a bit mind-bending. And I suspect they've been re-building all the base systems to support that. So, I'm not surprised that estimating on this job is an absolute nightmare. We had an equivalent in our business recently - on a multi-million pound programme, where we're supplying subsystems into to a high-profile larger system, we under-estimated by a factor of three the complexity of doing one particular work package because we didn't understand the problem well enough. And a supplier to us, to whom we'd passed derived requirements, fluffed it even worse.

My views are that it's got to happen soon as people are "played out" in Sandbox right now. There's no point waiting for things to be perfect before adding features, just stable enough that you're not trying to build a house on shifting sand.

It's looking great! I actually played a relatively complex mission and most of the issues I was running into were design issues, rather than bugs. I'm getting a weird bug where after timewarp some of the stacked parts aren't centered however a save and reload tends to fix that. But, that was nothing compared to the multiple reminders that I need to consider the relative positions of centre of mass and centre of drag - both into and out of Duna's atmosphere with my crewed lander! I now remember the frustration of ship design rather than just the frustration of bug avoidance... Performance is better but still not great and weirdly worse in map view?

Thanks, quite familiar with the Discord tracker - and so know there's a hotfix planned. But like you say, past track record of number of bugs per patch is dodgy ground for an extrapolation!

I've not seen any evidence that they're hotfixing anything other than the registry bug. If you've seen it, I'd be excited - but like many others, I've actually put KSP2 to bed until at least v0.2.0

The last stated plan was 6-week intervals between releases. I've no idea if that's still the target. If so, that would place 0.1.5 on or about Friday 13 October.

I'm very interested in anything which adds progression, and limits visual interest until you are there in-person. It's why I've mentioned in the past that I'd really like a "simulation mode" where you could check if something would land/take off from, e.g., Eve without spoiling the visuals by warping to re-entry and skipping the journey. The more you probe/scansat/science your way the more accurate your simulation mode can be. I really like the idea that unmanned missions have limits/drawbacks but I agree, an optional setting. Flying solely off telemetry might be fun initially but does take away from the point of building pretty spacecraft.

Chris, thanks for an informative and interesting Dev Blog. I'll be honest, my initial reaction was "sounds like an oversimplification" but having considered it a bit more I think you're on the right track. Some people are complaining that by not modelling internal conduction you're allowing designs that weren't possible before. However genuine systems would always be able to factor in active cooling to radiator panels, so I think that's OK. In fact KSP1 had just such a system which led to the same results with a far higher calculation overhead. I think that perhaps, with some refining of parts' thermal generation and emission rates in different scenarios (sunlight/shade/atmosphere/vacuum) it is a pretty solid basis. People harp on about realism but wouldn't recognise it if it slapped them in the face - KSP1 heat model was far from realistic in some ways and people were quite happy to learn the rules of the game in those cases. If they want more details, then they can wait for modders to get their hands on it. As for was the blog worth the man-hours and the subsequent grief from committnegatives, I think that for your own sanity though sadly perhaps the time is better invested in developing.

Ah yes, now I do remember that. I just never used it really, found it awkward and preferred the modded one.

I'm with the others above, I reckon you installed a mod and got so used to it you forgot it wasn't stock! I know the one,it is good and definitely a valuable feature