Phelan

Ahhh, thank you, the "start the engines/spool up first, then start the reactor" is the kind of thing I meant with "how are they supposed to be flown"

Okay, my mistake. But your "don't want an RGU, don't use an RGU" completely misses the point. What should I use when I build a 2.5m drone ship with as few parts as possible that's also intended for vacuum usage only?

I simply tried to put together 'packages' that at least give the stats of the 2.5m wheel. I also never went with the HECS2, so no idea why you keep bringing it up. You might as well bring up the ion engine in a talk about cheap, discarded upper stage engines. Now look at your own list there and compare the ones that I actually mentioned myself, namely the "flat" 1.25m core compared to the 2.5m one (forget about the 0.625m ones for the time being). Which of my 'packages' don't have that?

No, I never included the price of any batteries or additional reaction wheels to the 2.5m probe core, I only pointed out that at least some extra batteries would be needed. But for the actual price/weight comparison, I only added batteries&reaction wheels/service bays to the smaller cores and compared these "packages" with the 2.5m core all by itself.

Did you actually read my OP? The entire post I mean? Because I laid out the calculation for full functionality as well as just pro/retrograde hold and showed that the 2.5m core is always worse than the alternatives in both price as well as weight.

Heh, same here, though with keyboard instead of joystick. Along with Trajectories and MJ's spaceplane guidance (just the navball indicator for the alignment during the final approach), it's pretty easy. If coming from an equatorial orbit that is, something like a 30° incined orbit makes it a good bit harder. Ah well, I guess for the more "routine" flights, I'll just use an overdesigned spaceplane that'll have enough fuel left for a powered approach, and do the more critical flights "by hand".

Throttled down you mean?

I've simply installed NFE via CKAN, but considering that the engine does have a reactor module, I'd say yes. But cooling for that would be the weird part, afterall, the reactor's heat is precisely what's used to heat up the air for propulsion, and even when I dial the reactor back to 1-2% and go full throttle, it overheats. Doesn't make sense. I'll try some radiators, but yeah... they absolutely should not be required, on the contrary.

Does anyone have a flight plan for an unpowered landing (a la spaceshuttle)? It'll obviously need to come down much steeper since there won't be any means to maintain speed, and it'll probably depend much more on the craft itself (weight/drag/lift/..), but I figured before I start fiddling around with the numbers, I could try asking first

How are the nuclear engines supposed to be flown/work? I've tried fiddling with the reactor power and thrust (both up, RP down to 1-2% with full thust, low thrust with full RP, both tuned down), and the reactor keeps overheating within minutes at best unless I throttle down so far that even the pre-built M2X Zephyr can't stay in the air. (and yes, I've tried the copy-and-paste patch for the functionality.cfg in the OP since I'm using NearFuture stuff, no change as far as I can tell, at least it still keeps overheating way too quickly)

Please list the better functionality of the 2.5m core over the (flat) 1.25m one.

I know that it can be "fixed" with mods, and depending on how little one cares about it being "nice", rather easily. But that's a bit like saying that the vanilla game would be fine with just the 1.25m parts - you can afterall mod all the parts you want into the game. No. When there is a noticeable fault, it should be pointed out. Even if it's just lowering the weight/cost of the 2.5m core to something more sensible compared to the 1.25m one (whether with again more sensible "built-in parts" like SAS wheel and batteries, like I listed in the calculations in my initial post, or just a much lighter and cheaper part).

With the switchable part functionality, it shouldn't be too hard to add a switch so you could select during construction which functionality you want for a given probe core (with upper limits depending on the core/reserached techs/...) and to adjust the other stats of the part accordingly. So that after researching the 2.5m probes, you could use them with just SAS and prograde/retrograde hold and have it weigh and cost very little. Without the "on-construction-switch", I'd probably just add two or three more parts. And don't forget that my criticism applies to the 1.25m and 2.5m cores even more - no difference in functionality there, but the 2.5m one is heavier and more expensive than the 1.25m plus service bay. It doesn't make any sense.

That last sentence is my whole point. Yes, the smaller ones require fairings to actually "fit" for larger vessels, but if you just take the probe core and the fairing, you have almost the same functionality - but you pay less (both in terms of money as well as weight), you only have that extra part (not good if you want to build "small" vessels and keep the part count down). Essentially, my problem is that the kerbals who build the larger probes are just plain stupid since they build bigger, heavier, more expensive probes with almost no advantages compared to the smaller ones. It's basically just a 1.25m probe core with an empty, very heavy and expensive (and otherwise unusable) shell/fairing/dedicated service bay around it.

We need an option to collect bulls.. taurusfecalium on other planets... Maybe not with a drill, but rather a "0.1 units (~10kg) per surface sample" thing or so?

Agreed

So you read my "So you never really need to set up a good relay network the way it was intended, which is a shame" as an argument, and then say that your "So you never really need to use these features the way they were intended, which is a shame" was not an argument? How does that work?

With you. I thought you'd recognize your own argument there and how I turned it around 180° Overall, okay, I should have phrased my OP a bit differently. Larger probe cores do have usages - but I only very rarely need the functionality. As an analogy - are 3.75m tanks and engines good and useful? Sure. Do you use them to launch everything once you've unlocked them - or do you use 2.5m stuff even then and enjoy the lower cost/weight/...?

So you never really need to set up a good relay network the way it was intended, which is a shame. See the problem? You can play without a relay network if that's what you want, but I can't play with lighter/cheaper/more functional probe cores if I want to.

You mean attach a tiny probe core between two 2.5m parts and move it into the 'lower' one? Even with KJR, unless the 'top' bit is relatively light, that's a bad idea for stability reasons (and I in part want probe cores to land the lower stage again, meaning that tiny probe would 'hold' the entire payload). Adding struts for stability or using the tiny strut cube for surface attachment would kind of ruin the idea of not wanting to add another part.

Ah, right, my mistake. Though I've never had the need for a rendezvous/docking "out of control range". Some basic relay setup is what I do anyways before anything more complex than a "controlled crash" ;P ...and that's why the medium and large probes' reaction wheels should be stronger when compared to even the small reaction wheel. The main thing still is that if you only really want a basic probe core for anything bigger than a .6m probe, you have to add another part (service bay), or use something that'S far heavier and more expensive than what you need.

Uhhh... "huge reaction wheels"? 1.5 strength... over even the 0.6m reaction wheel's 5 strength? And what do you actually need to land? Retrograde hold I'd say is enough if you know what you're doing (which the cheap HECS 1 has), the single most difficult thing after that is thrust control, and in that, all cores are equally useful/useless when in "limited control" mode.

Am I the only one who keeps using smaller probe cores (the 0.6m ones or, if I need max. SAS functionality, the 1.25m one) inside service bays instead of the larger cores? The reaction wheels are close to useless anyways (if you need the control of a reaction wheel, it's easy and cheap in every way to add even just the 0.6m wheel inside the service bay again), and the only advantage of the large cores that I can think of is part count and height (both of which don't make up for the increased weight and cost, let alone the additional use one gets out of having a service bay anyways). Yes, I know they have the best anomaly detection, but I'd argue that that's a rather tiny advantage again. Besides, it doesn't make much sense in the first place. Sure, the maneuver/target alignment functionality of the late cores is great and deserves to come with a price, but at the very least the 2.5m core shouldn't be worse than the 1.25m one. Starting off with large cores and later developing smaller ones would mess up the parts progression of the tech tree, so how about just making the later ones weigh less, flatter, and maybe cost a bit less? Or give them larger batteries/stronger reaction wheels/... to reduce the need for additional parts? Just for comparison: 2.5m core: weight 0.5t, cost 3400, negligible reaction wheels and battery storage, so you either need an 0.2t battery as well or surface-mounted one, which probably would require a service bay. 1.25m core: weight 0.1t, cost 2250, same functionality; put inside a 2.5m service bay (0.3t, cost 500), it already costs and weighs less than the 2.5m core - and now you already have the service bay for a nice actual reaction wheel and those battery packs When you only really need prograde/retrograde hold, it gets even more ridiculous with a Probodobodyne HECS inside a 2.5m or 1.25m service bay, especially since three small reaction wheels have the same strength as the 1.25m one (and 1-2 are usually enough already): 2.5m stack: service bay, HECS, 1 small reaction wheel, 2x400EC battery packs: weight 0.49t, cost 2850 - and now you have a better reaction wheel, more batteries and some storage space at the same weight and lower cost than the 2.5 probe core.

I'll 'generate' a fresh, minimal one for the exception to show up, one moment (and stop reacting so quickly, you're making me feel bad for my reaction times to bug reports ) output_log.txt: https://de.scribd.com/document/367333270/Output-Log

KSP 1.3.1.1891 (Win x64), RoverScience 2.3.3 (basically, just the version that CKAN shows right now)