Nate Simpson at Space Creator Day talks about KSP 2.

[GGG-GoodGuyGreg]

3 hours ago, LoSBoL said:

People tend to think they are part of a majority and some are thinking they can speak on behalve of a majority.

The majority however is silent, and you don't know what they are thinking because they are silent.

No discord, Reddit, steam or this forum have a majority of the playerbase, there are probably more than 100.000 sales, you won't find the majority on either of above platforms.

We don't know nothing 

That’s true, tens of thousands must enjoy the game in silence and that’s why they don’t have the time to comment.

Lucky them. 

[LoSBoL]

22 minutes ago, GGG-GoodGuyGreg said:

That’s true, tens of thousands must enjoy the game in silence and that’s why they don’t have the time to comment.

Lucky them. 

You can fill in blanks any way you would like to, and so many do. The facts are however, that they remain blanks, and yo, me or anyone else don't know.

Like I said, we know nothing.

[PDCWolf]

10 hours ago, Periple said:

The extensible radiators in KSP1 cool the entire craft, placement didn’t matter. It only matters for the radiator panels which would only cool the part they’re attached to. Your “droolfest” was already there.

The heat needs to convect part to part to reach radiators, which diminishes cooling efficiency the further the cooling part is from the heat generating part. Placement did matter.

8 hours ago, Superfluous J said:

To be frank, no I will not wade waist deep into the cesspools that are the KSP Discord or ANY game's Reddit. And I don't need to, as they're also tiny microcosms compared to the entire user base. These kinds of places always will be.

The subreddit has 1.5 million followers, and it was a massively active place before KSP2 dropped as well. Even then, it's got a magnitude more active people than here, thus, less deviation from what can be considered a general opinion about the game. This is also true for the Steam Hub for both games, which is probably the most active place where the game is discussed, and has like 250k followers for the base game and about 180k for the sequel.

If you don't want to get out of the forum, that's fine, but that'll never change the reality that the 1.5 million in the subreddit and the almost 500k between both hubs are more representative than the 20 to 50 users that keep posting in this forum.

8 hours ago, Superfluous J said:

It perfectly fits that 80% haven't purchased the sequel, though. They're not buying it because the content of the game is not worth the money to them. It's not like the game's got any more content than it did when it was released, and it's still got a lot of bugs.

I mean, that's gotta be the biggest part and I agree with that statement, but again, you'd know how big the demand for knowing what happened between the first release date for a full game in 2020 and the EA release in 2023 is if you dared to get out of this echo chamber. It's almost ubiquitous to any discussion of the game (even here if you want to only take the forum into account).

2 hours ago, cocoscacao said:

Never mentioned simulation. Just downgrade and simplification. There's a difference. Downgrade is when you lose a certain functionality.

We lost functionality. Where you place radiators has no relevance anymore, separating parts with more than a single joint doesn't matter anymore, separating heat emitting parts from each other and not having them all linked to a single part doesn't matter anymore.

The system has been downgraded, and thus the experience around it has been degraded, as it's been relegated to a simple "look at number and include part with bigger number". I guess it really is the latest form of copium to just bash the first game without really haven't paid attention to how it works.

2 hours ago, cocoscacao said:

Yes.

Well, not for me. I have no plan to pay $50 if all systems are gonna be either carbon copies, or downgrades, and if new systems are also reduced to simple implementations. That'd be a failure to justify the pricetag.

Edited November 1, 2023 by PDCWolf

[Lisias]

22 minutes ago, GGG-GoodGuyGreg said:

3 hours ago, LoSBoL said:

People tend to think they are part of a majority and some are thinking they can speak on behalve of a majority.

The majority however is silent, and you don't know what they are thinking because they are silent.

No discord, Reddit, steam or this forum have a majority of the playerbase, there are probably more than 100.000 sales, you won't find the majority on either of above platforms.

We don't know nothing 

That’s true, tens of thousands must enjoy the game in silence and that’s why they don’t have the time to comment.

Lucky them. 

YOU may know nothing, but some of us can infer something from available data.

Most clients are silent because most clients are not playing the game. We may not be able to know for sure how many people bought KSP2, but we can use SteamCharts as a heart bit for the current user's interest on it and the last week we had daily peaks of about 400 users (versus the historical low of 40 last month. So, yeah, we have about 10 times more active users in the last week than the worst score, and about twice the peak of the previous week.

So we know that the game stats are improving for now - but the peaks are already dropping, yesterday we had a peak of 200 against ~320 one week ago.

We also know that KSP2 had a peak of 11K concurrent users at launch (an impressive number) on Steam (what happened just 10 months ago). Since Steam is known to have about 75% of the online userbase, we can infer they sold about 16K copies on the very first day (given or taken, as we don't know exactly how many users fire KSP2 outside steam launcher).

We also know that only 1 on 26 customers complain, the others 25 just walk in silence. So it's safe to assume that for each complain you find about a bug, at least 20 others customers had quit playing the game.

From the initial of 16K to about 600 (estimated peak this week after the last patch, i.e. 450/3*4), we have about 26 leaving users for each one still around. DAMN, that 1 to 26 heuristics appears to work!

Obviously, I'm abstracting refunds and new sells, but yet we can have a very good general idea of what's happening.

 

23 minutes ago, LoSBoL said:

Like I said, we know nothing.

YOU don't know nothing. Don't project your ignorance over the rest of us.

[Guest]

2 minutes ago, PDCWolf said:

The heat needs to convect part to part to reach radiators, which diminishes cooling efficiency the further the cooling part is from the heat generating part. Placement did matter.

I'm afraid you're mistaken about this, as this is only true for the fixed radiator panels. The deployable radiators (thermal control systems) cool the entire vessel.

"Deployable Thermal Control Systems (TCS), take heat from every part of the vessel (as if plumbed in with a cooling fluid loop)."

https://wiki.kerbalspaceprogram.com/wiki/Radiator

 

[LoSBoL]

27 minutes ago, Lisias said:

Most clients are silent because most clients are not playing the game.

Asumption, you don't know.  Majorities are often silent, you claiming that they are silent because they arent playing the game is nothing more then an  assumption and not a fact.

27 minutes ago, Lisias said:

We may not be able to know for sure how many people bought KSP2, but we can use SteamCharts as a heart bit for the current user's interest on it and the last week we had daily peaks of about 400 users (versus the historical low of 40 last month. So, yeah, we have about 10 times more active users in the last week than the worst score, and about twice the peak of the previous week.

I can name at least 10 reasons as to why people may not playing, You can't derive any conclusion on why people aren't playing if you can't say for certain how many people fall in which category. So there goes gut feeling and assumptions again.

27 minutes ago, Lisias said:

We also know that only 1 on 26 customers complain, the others 25 just walk in silence. So it's safe to assume that for each complain you find about a bug, at least 20 others customers had quit playing the game.

You said it yourself, an assumption, not a hard fact. Different products, different markets, different people,  different way of handling complaints, and if there are a majority?  You don't know.

27 minutes ago, Lisias said:

YOU don't know nothing. Don't project your ignorance over the rest of us.

If you claim you know, while there is no data in which you can derive your conclusions for certain, that's ingnorance.  I'm not that ignorant to claim I can. 

Edited November 1, 2023 by LoSBoL

[LoSBoL]

33 minutes ago, Lisias said:

YOU may know nothing, but some of us can infer something from available data.

Dataset not specified enough to make conclusions, assumptions and gut feeling take over, you don't know.

[PDCWolf]

42 minutes ago, Periple said:

I'm afraid you're mistaken about this, as this is only true for the fixed radiator panels. The deployable radiators (thermal control systems) cool the entire vessel.

"Deployable Thermal Control Systems (TCS), take heat from every part of the vessel (as if plumbed in with a cooling fluid loop)."

https://wiki.kerbalspaceprogram.com/wiki/Radiator

 

The functionality is there and a part including an abstraction for plumbing simulation and others not is actually more depth.

10 minutes ago, LoSBoL said:

Dataset not specified enough to make conclusions, assumptions and gut feeling take over, you don't know.

For the entire playerbase of KSP2 (people that have actually bought it, so around ~300k), you need a sample size of ~15k for 99% confidence in your results. Who do you think would be closer as representative? The 50 people in the forum bashing the first game to have a reason to defend the second, the thousand active on reddit on peak hours, or the tens of thousands active on Steam? So yeah, the majority is not playing the game. The majority is also not on these forums, they're on those other two places.

Your "don't know" argument is poorly thought out and doesn't reflect any sort of reality of how polling a group works.

[LoSBoL]

16 minutes ago, PDCWolf said:

you need a sample size of ~15k for 99% confidence in your results.

Indeed, but you don't have a sample size of 15k.

And Reddit doesn't give you a 15k sample size, it gives you Reddit user samples, not the complete KSP 2  communities sample. So the sample is no good in the first place.

The question who to believe, Reddit, forum, steam, discord , is none of them. 

And the majority isn't playing the game, I never said it did. Why you don't know? Guesswork.

Same for stating that the majority of KSP2 playerbase is on Reddit, again guesswork based on assumptions.

You might be poling some groups, youre not able to pole the group in which you can derive your 15k sample.

Edited November 1, 2023 by LoSBoL

[Guest]

39 minutes ago, PDCWolf said:

The functionality is there and a part including an abstraction for plumbing simulation and others not is actually more depth.

I'm confused, I thought you said it was a "droolfest" when you can just spam radiators anywhere on the craft to solve the heating problem? Because KSP1 totally lets you do this with the TCSs!

If KSP2 has both kinds of radiators, they will work almost the same way.  The only change is that placing a radiator panel on a part that's next to the part you want to cool won't work anymore, you will have to place them on the part you want to cool.

How is it that "attaching a radiator panel to a part next to the part you want to cool" is intelligent, deep, realistic, simulation, whereas attaching it to the part you want to cool is dumbed-down, droolfest, arcade, gamified, downgrade? I just don't see that as a change that's worth making a fuss about, especially if the reduced computational cost lets you simulate thermals in the background for all of your craft and colonies at all warp factors -- something that KSP1 can't do.

Edited November 1, 2023 by Periple

[PDCWolf]

27 minutes ago, LoSBoL said:

Indeed, but you don't have a sample size of 15k.

And Reddit doesn't give you a 15k sample size, it gives you Reddit user samples, not the complete KSP 2  communities sample. So the sample is no good in the first place.

The question who to believe, Reddit, forum, steam, discord , is none of them.

Nah, you just can't find a sample that agrees with your view and thus are ready to think everyone else is just a groupthink. Again, the subreddit has 1.5 million followers. If you wanna assume they're a hivemind and not viable as a sample because reasons when they represent almost 30% of the entire market share this game has the hope of achieving, then there's nothing to talk about, you're not interested in representativeness, just pushing a point that's completely not based on reality.

6 minutes ago, Periple said:

I'm confused, I thought you said it was a "droolfest" when you can just spam radiators anywhere on the craft to solve the heating problem? Because KSP1 totally lets you do this with the TCSs!

If KSP2 has both kinds of radiators, they will work almost the same way.  The only change is that placing a radiator panel on a part that's next to the part you want to cool won't work anymore, you will have to place them on the part you want to cool.

How is it that "attaching a radiator panel to a part next to the part you want to cool" is intelligent, deep, realistic, simulation, whereas attaching it to the part you want to cool is dumbed-down, droolfest, arcade, gamified, downgrade? I just don't see that as a change that's worth making a fuss about, especially if the reduced computational cost lets you simulate thermals in the background for all of your craft and colonies at all warp factors -- something that KSP1 can't do.

It's an option, you're not forced to skip the entire system. Meanwhile KSP2 skips the whole system for you.

[Guest]

2 minutes ago, PDCWolf said:

It's an option, you're not forced to skip the entire system. Meanwhile KSP2 skips the whole system for you.

I don't understand. Can you elaborate? 

The only thing the KSP2 thermal system changes is that it removes thermal conduction between parts. The only effect on gameplay/craft design is that placing a radiator panel on a part next to the part you want to cool won't work anymore, instead you'll have to place it on the part you want to cool.

How does this change make the difference between "skipping the entire system" and not? It strikes me as completely trivial.

[GluttonyReaper]

1 hour ago, Lisias said:

Most clients are silent because most clients are not playing the game. We may not be able to know for sure how many people bought KSP2, but we can use SteamCharts as a heart bit for the current user's interest on it and the last week we had daily peaks of about 400 users (versus the historical low of 40 last month. So, yeah, we have about 10 times more active users in the last week than the worst score, and about twice the peak of the previous week.

Not saying I disagree, but also... this doesn't feel that surprising? Given there hasn't been a major update in 10 months, and that many (most?) of the EA buys were likely KSP1 players, a game which still exists and is more feature-complete than KSP2, it doesn't seem like a massive revelation that people aren't playing at the moment. I genuinely don't know what kind of numbers are typical for an Early Access sequel (given there aren't many of those regardless...), but this doesn't seem super crazy.

1 hour ago, Lisias said:

We also know that only 1 on 26 customers complain, the others 25 just walk in silence. So it's safe to assume that for each complain you find about a bug, at least 20 others customers had quit playing the game.

"Quit" is a bit strong. There are very few (even complete) games that people will play for 10 months straight (or even a month straight). I doubt it's as dramatic as everyone else actively quitting the game in a fury, never to return - it'd be more accurate to say "not actively playing". There's plenty of games sitting on my hard drive(s) that I like or am ambivalent to (which I suspect is where most KSP2 purchasers sit), but that I'm not particularly in the mood for playing right now this minute.

9 minutes ago, Periple said:

How is it that "attaching a radiator panel to a part next to the part you want to cool" is intelligent, deep, realistic, simulation, whereas attaching it to the part you want to cool is dumbed-down, droolfest, arcade, gamified, downgrade? I just don't see that as a change that's worth making a fuss about, especially if the reduced computational load lets you simulate thermals in the background for all of your craft and colonies at all warp factors -- something that KSP1 can't do.

Not only that, this is something that KSP1 never actually explains. In fact, it does the opposite - the main heat generating parts (drills/converters) have a "required cooling" value displayed in the VAB IIRC... and all you have to do to maintain the correct temperature is add enough active radiators, which have their own "cooling rates" displayed, to match this. The weirdly complex heating system is entirely a background thing which the game never encourages you to engage with, not that I personally think it would even add that much if you did.

[LoSBoL]

7 minutes ago, PDCWolf said:

Nah, you just can't find a sample that agrees with your view

What is my view? Please share your...

...assumptions

You are filling in blank spots you know nothing about.

7 minutes ago, PDCWolf said:

Again, the subreddit has 1.5 million followers.

Yes, the subreddit has 1.5 million followers, and you are assuming a lot from that.

 

8 minutes ago, PDCWolf said:

If you wanna assume they're a hivemind

I'm not the one making assumptions, and no, I dont wanna assume they're a hivemind.

9 minutes ago, PDCWolf said:

and not viable as a sample because reasons

The reasons are a given, you equalize that 1.5 million redditors, are the KSP2 database, including being the silent majority.

11 minutes ago, PDCWolf said:

when they represent almost 30% of the entire market

  Assumptions, and you don't even know how large the market is, youre assuming such.

[Vl3d]

I would have thought that engines overheating while operating and then exploding would have caused more controversy than simplified heat flux.

[GluttonyReaper]

1 minute ago, Vl3d said:

I would have thought that engines overheating while operating and then exploding would have caused more controversy than simplified heat flux.

Good old KSP1 v1.0, where you had to attach your NERV engines to the biggest fuel tank possible as a heatsink, or risk them exploding on long burns  

[PDCWolf]

Just now, LoSBoL said:

-snip-

And you think those assumptions (some of which aren't assumptions) are not grounded on reality. They are.

11 minutes ago, Periple said:

I don't understand. Can you elaborate? 

The only thing the KSP2 thermal system changes is that it removes thermal conduction between parts. The only effect on gameplay/craft design is that placing a radiator panel on a part next to the part you want to cool won't work anymore, instead you'll have to place it on the part you want to cool.

How does this change make the difference between "skipping the entire system" and not? It strikes me as completely trivial.

KSP2 removes the Skin/Core system, and doesnt remove "part to part convection", it removes the whole part dynamic from the equation, and you're failing or refusing to see how that affects the whole dynamic. The whole vessel receives and radiates heat as a single part, that wasn't the case in one, as every part would generate and dissipate heat at different rates. Even if the DTCS in KSP1 does abstract a plumbing system to draw heat from all parts, it doesn't outright delete heat like it would in KSP2. This means in KSP2:

• Craft design doesn't matter. Any dissipating part will be equally effective anywhere.
• Since part to part convection is not a thing, that means craft design just got not only stupid easy but outright unimportant. In KSP1, mounting all your heat generators to a single part would easily make that part overheat, so even with the DTCS there was some nuance to craft design where not all your drills or converters could be attached to a single part, unless you outright spammed the one part that trivialized the whole system.
• Not using the DTCS also meant there was a whole new world to craft design, as heat wasn't magically sucked from all parts.
• This also gave you the chance to create realistic structures where a heat generating part is built away from the main craft, and piped through a multiple part system (specially with docking and thanks to part to part convection) so the heat would dissipate into a number of parts instead of magically heating up a cockpit on the other side of the craft.
• For bases on hostile planets (with mods, as god forbid they actually make the game challenging), isolating your base from the ground was a proper challenge, contrary to any leg of your base conducting heat into the whole of it.
• The Skin/Core system also meant parts could build up and keep heat without dissipating it effectively, as cooling the surface wasn't directly translated to cooling the core.
• In KSP1, I could NOT use DTCS and thus have to play with the heat system in a different way, in KSP2 there's no playing with the heat system, only attaching magic parts anywhere and be done with it. It's literally just a part tax that doesn't even influence design.

Just to be clear, I'm not saying the system in KSP1 was perfect or even good. My point is entirely based on the new system being a clear downgrade. There's less simulation at play, so there's less stuff that will happen, and there's diminished point to the whole system as it is trivialized by magic heat dissipating parts that don't care about spacecraft design. This was mentioned multiple times to Nertea (and I'm sure explained better than me) on his devblog.

[Vl3d]

4 minutes ago, GluttonyReaper said:

Good old KSP1 v1.0, where you had to attach your NERV engines to the biggest fuel tank possible as a heatsink, or risk them exploding on long burns

Sounds great! And also don't forget kerbals dying from radiation instead of having life support. How does this compare to simplified heat flux?

[Guest]

23 minutes ago, Vl3d said:

I would have thought that engines overheating while operating and then exploding would have caused more controversy than simplified heat flux.

I have a list of things I'd like to be more accurately (or just generally better) simulated too, they would have much bigger gameplay effects than this change to the thermal system:

• Joints -- Unity joints are just painful to work with. If there's one area where a bespoke system could have been worth it, despite the complexity, it's here.
• Wheels / landing legs / suspension -- Unity's solution doesn't play well with time warp and save/load either. A bespoke system would just be better here.
• Aerodynamics -- I believe that a FAR-like aerodynamics system would make for better gameplay, assuming it had proper support from the design tools. Even little improvements like ground effect would open up new possibilities!
• Comms -- If antenna orientation was simulated (with auto-orientation tools of course), many craft designs would be interestingly impacted. You couldn't just plunk a dish anywhere on a big craft, you'd have to put it somewhere you could point it at Kerbin (or a relay). Relays would require two antennas, each of which would have to be able to track a target. That would make designing relay satellites actually interesting.
• Liquid surfaces -- It was notoriously hard to make a seaplane in KSP1 because it would tend to faceplant when touching down. A more accurate simulation of landing on a liquid surface would make all kinds of amphibian craft more interesting to design.
• Life support -- nuff said
• Radiation -- ties with life support
• Wind -- This would make atmospheric flight a lot more interesting and would make for considerably different base, lander, or craft designs for planets that experience very strong winds.

Really, tweaks to the thermal system would be pretty far down my list of things that need richer simulations!

14 minutes ago, PDCWolf said:

KSP2 removes the Skin/Core system, and doesnt remove "part to part convection", it removes the whole part dynamic from the equation, and you're failing or refusing to see how that affects the whole dynamic. The whole vessel receives and radiates heat as a single part, that wasn't the case in one, as every part would generate and dissipate heat at different rates.

I'm sorry, but you're mistaken about this. The whole vessel does not receive/radiate as a single part. Fluxes on each part are simulated separately. The only change is that conduction between parts is no longer simulated. Quoting from Nertea's article (emphasis mine):

Quote

KSP2’s thermal system will use a model based around managing heat fluxes – the amount of energy applied to something over some amount of time. Every process we want to model boils down to applying a heat flux to a part. This flux can be positive, causing a part to increase in temperature, or it can be negative, causing a decrease. Here is a sampling of fluxes we care about from our user stories:

• Positive flux from things going on in a part, such as drilling, resource production and powerful engine reactions.
• Negative fluxes from things going on in a part, like heat radiators and other heat sinks.
• Positive flux from stellar sources in sunlight.
• Positive or negative fluxes from warm or cold atmospheres, fluid bodies or surfaces.
• Positive fluxes from high velocities in atmospheres (reentry).

Using this model, we can sum up all the fluxes affecting a part and communicate a single value to the player. If the sum of all fluxes is positive, the part heats up. If it is negative, the part cools down. In the absence of any fluxes, a part is stable. If the part’s temperature increases above its maximum, well, you get an explosion. I’ll illustrate with a few cases! 

(Also, I have to nitpick -- it is conduction, not convection, which is a totally different process.)

Edited November 1, 2023 by Periple

[LoSBoL]

15 minutes ago, PDCWolf said:

And you think those assumptions (some of which aren't assumptions) are not grounded on reality.

 Yes, and I elaborated on why that is the case. 

And my view according to you as I asked? is that an assumption or is that one of the (some of which  aren't assumptions)?  

Quote

They are.

Opinion based on assumptions.

[PDCWolf]

18 minutes ago, Periple said:

(Also, I have to nitpick -- it is conduction, not convection, which is a totally different process.)

My bad.

19 minutes ago, Periple said:

Really, tweaks to the thermal system would be pretty far down my list of things that need richer simulations!

I'd agree with this on principle, I didn't really expect to be discussing something like the heat system, but I expect a sequel to go deeper rather than simpler.

19 minutes ago, Periple said:

I'm sorry, but you're mistaken about this. The whole vessel does not receive/radiate as a single part. Fluxes on each part are simulated separately. The only change is that conduction between parts is no longer simulated. Quoting from Nertea's article (emphasis mine):

Then you're missing how important part to part conduction is. I hope you realize the problem by looking at the 3 part spacecraft in orbit example, and the one right below them:

• A part can explode by itself without affecting anything else.
• A part can cool another  on the other side of the craft even though there's no part to part conduction, and thus that hot part doesn't affect its neighbors.
• As a bit of a secondary consequence of the downgrade, not having a Skin/Core system also means heat can instantly be dissipated to the environment for free.

Now think of how that lack of dynamics compares to KSP1, where a heat part might not only take itself out but its neighbor/s, and where you place heat generators vs dissipators influences the design (bar usage of the one magic part).

The one reason this kind of stuff isn't headlights-to-the-eyes evident is because SQUAD was really afraid to put consequences for anything in the game. Normally, having a Kerbal EVA right after you re-entered a capsule means the little guy should melt. Having a nuke producing heat should degrade the quality of life in the cockpit as coolant is sent there unless you had extra. Living in a hellhole where you can't touch the walls should mean the science or resources produced should have a drop in quality or amount. Mounting 17 engines to a single plate should mean a piping hell that kills your performance with excessive mass. Flying your SSTO inside a fireball should vaporize the kerbals inside before the structure (or worse). But god forbid the game actually has any semblance of difficulty, and now in KSP2 any semblance of depth.

[Vl3d]

7 minutes ago, PDCWolf said:

But god forbid the game actually has any semblance of difficulty, and now in KSP2 any semblance of depth.

It might not have depth, but it's striving for generous girth.

[Guest]

21 minutes ago, PDCWolf said:

Then you're missing how important part to part conduction is.

Maybe? Or could it be that you're missing how unimportant it is? You did have a pretty fundamental misunderstanding about the system, if you believed that parts had been removed altogether and the entire craft is simulated as single part!

As an aside, has correcting this fundamental misconception altered your view of the system at all? Because I do agree, if the entire craft was simulated as a single part, that would be a significant change for the worse!

21 minutes ago, PDCWolf said:

I hope you realize the problem by looking at the 3 part spacecraft in orbit example, and the one right below them:

Let's see:

21 minutes ago, PDCWolf said:

• A part can explode by itself without affecting anything else.

Exactly like in KSP1. If a part heats up past its limits, it will explode, by itself, without affecting anything else. I think we've all seen this happen when re-entering a bit too hot, as solar panels, antennas, goo containers etc. explode... or even the retrograde parts of the spacecraft itself. (It could cause a structural failure of course, if a component in the middle of the craft blows up, like when you're coming in too hot in a spaceplane.)

21 minutes ago, PDCWolf said:

• A part can cool another  on the other side of the craft even though there's no part to part conduction, and thus that hot part doesn't affect its neighbors.

Exactly like the deployable radiators in KSP1.

21 minutes ago, PDCWolf said:

• As a bit of a secondary consequence of the downgrade, not having a Skin/Core system also means heat can instantly be dissipated to the environment for free.

I think the skin/core system can be abstracted away with zero impact to anything. If I understood it correctly, it was a representation of a part's ability to soak heat flux, which can be represented as a single value (thermal capacity).  

21 minutes ago, PDCWolf said:

Now think of how that lack of dynamics compares to KSP1, where a heat part might not only take itself out but its neighbor/s, and where you place heat generators vs dissipators influences the design (bar usage of the one magic part).

Can you describe a vessel where this would actually happen? I can't think of any way to do this. Literally, any way, even if I tried. It's always the heat flux directly on a part that kills it.

For example, can you think of any way you could connect a Mk 2 part (high temperature tolerance) and a Science Jr (low temperature tolerance), and have the Science Jr explode due to the flux from the Mk 2 part? I can't. In a re-entry, if the Science Jr was exposed to the atmosphere, it would blow up first because of that; if it wasn't, the conducted heat from the Mk 2 part wouldn't be enough to blow it up before the Mk 2 part itself blew up. 

As to the dissipator placement, once again, the only change with KSP2's system is that if you have a dissipator that only affects the part it's placed on, then placing it on a part next to the one you want to cool won't work anymore, you'll have to place it directly on the part. As far as I can tell, that is the only gameplay/design change the KSP2 system imposes. Is it really that significant?

Edited November 1, 2023 by Periple

[Pthigrivi]

1 hour ago, GluttonyReaper said:

Good old KSP1 v1.0, where you had to attach your NERV engines to the biggest fuel tank possible as a heatsink, or risk them exploding on long burns  

This is a good example of why I'd actually prefer the game assumes proper coolant plumbing rather than focusing on adjacencies. I often use NERVAs on big sky-cranes for delivering modules and rovers to the surface and you want them way out on outriggers, while you want your fuel mass tight to the CoM for better maneuverability. Assuming proper coolant plumbing allows much greater design flexibility to do things like that. 

Edited November 1, 2023 by Pthigrivi

[Guest]

I think my previous message was a bit long, so I'll try a tl;dr. @PDCWolf, can you describe, or show, an example of how the thermal system in KSP2 would change the way you would have to design a vessel, compared to KSP1? And by "an example," I mean an actual vessel you're designing, not describing the system in general terms. 

(I've done this exercise, and the only difference I can find is the one I've been repeating: if you have radiators that only cool the part to which they're mounted, then you have to attach them directly to the part you want to cool, rather than an adjacent one. And, once more, if that really is the only difference, then in my view it is a pretty trivial one -- and well worth it, if the reduced computational cost allows simulating thermals for all craft and bases, whether they're focused or not.)

Edited November 1, 2023 by Periple