cantab

Probably Smokescreen, I had it do it as part of the Realism Overhaul pack. Opening Smokescreen's config I could see the particle count periodically drop and that was exactly in sync with the severe stutter.

Yeah, since you can't deplete the ore you can run mining equipment off fuel cells. In a decent deposit one unit of LFO into the fuel cell will drill and refine more than one unit.

I'm playing with no reverts or quickloads*, so testing becomes important. I'll test re-entry vehicles, atmo landers, and orbital spacecraft around Kerbin, and if need be estimate performance (eg speed under parachute) on other bodies. Probe missions need this the same as Kerballed ones, since a critical failure means flying a whole new mission. I try and remember an abort action on Kerballed rockets, and failing that I have Vanguard EVA parachutes so I can just bail. So far I've yet to kill a Kerbal, though I've stranded a few and had to rescue them. (*Blatant bugs excepted. And I'll revert unmanned launch failures, since I'm in Science mode and it saves me time.) For a particular example, for my Kerballed Serran (New Horizons mod world) landing and return I had already returned a probe from the surface which itself took several missions to accomplish. I then ran analogues to Apollos 8, 9, and 10 - an orbit and return, a test of the landing ship in Kerbin orbit, and a near-landing that is intentionally aborted close to the ground. Each of those missions taught me things I needed to know to make the real deal, the actual landing, a success.

That message pops up all the time when I choose to convert science results into data. The data still gets made and put in the lab though, provided there is in fact enough space.

Quite. A craft can be fine in stock and persistently overheat in FAR, I call that a FAR bug. After all it's Ferram Aerospace Research, not Ferram Thermodynamics Research, and I'm of the view that it should leave the heating the heck alone.

Doesn't the LV-N generate electricity?

In real life all planets that formed orbiting their star are going to orbit in more or less the same plane and same direction. A retrograde planet must either be actually a sub-brown dwarf that formed by cloud collapse itself, or have been captured from another solar system. As far as gameplay goes, a very distant object in a retrograde orbit requires little more delta-V than the same object in a prograde orbit. A small comet-like or asteroid-like planet in an appropriate orbit would make much more of a gameplay challenge. I'd also like to see a real outer irregular moon for Jool. Jool's SOI is *huge* and even Pol is close to the planet and essentially a regular moon. A new Joolian moon way far out would add some new interest and gameplay challenges. You'd probably want to encounter it directly from your interplanetary trajectory where-ever it is on its orbit, and transferring between it and the inner moons could be lengthy if you minimise delta-V.

Learn Kopernicus configs and remove Mun and Minmus from orbit of Kerbin. Problem solved. As for me, well I keep in mind that real space programs haven't always gone at breakneck speed. For me things are likely to slow down a bit once I start doing interplanetary missions, but that's no real problem. Also, all I need to send out probes is probe cores, solars, and at least one science experiment. In any case my current save has a Mun science station. I have to timewarp to make that make science. And it's incredibly profitable, I fed it data from orbit and two landings and I'm expecting 1500 bonus science total from that, with the option to run more landings if I like. But to be honest I don't want to keep doing the same thing over and over again, going somewhere new is more appealing.

On probe cores, I on the other hand don't use the OKTO2 for probes. I use it for automated control of larger ships, but for a small probe OKTO2+Reaction Wheel is just way too much torque, an OKTO or HECS handles better and saves a part too. The one probe core I *don't* use is the 2.5m RGU. It's half a ton and a potential weak point in my stack. If I want to control something big I'll mount a small probe core on it, usually via a radially-attached small nosecone. On the LV-T30 Reliant, it does have its advantages. It's a noticeably higher TWR than the Swivel, and a bit more absolute thrust than the aerospike. I've done things like run a mixture of Reliants and Swivels in a cluster so I get extra thrust without having no gimbal. The Reliant would also be a good choice for a liquid booster, though I tend to use solids anyway. On decouplers, you remind me that I *rarely* use the TT70, the one with the strutwork. I just prefer my boosters flush to the core. The one time recently I did use the TT70 was to hold the upper wing of a biplane, it looked a bit nicer than using a girder; of course it was never meant to decouple in that application.

Things may differ in other countries, but in Britain the 6700K is still overpriced. Currently you're looking at4790K - £270 5775C - £300 5820K - £305 6700K - £330 To me it's very very hard to justify the 6700K at prices like that. If large RAM capacity is important and the Core i5 6600K isn't a strong enough CPU I'd be going with the 5820K. If 32 GB of RAM will be enough and fast stock clocks are important I'd go with the 4790K still. The 5775C is an interesting option if you know your use cases will benefit from its large L4 cache and you plan to overclock.

kerbalmaps.com will tell you *where* the anomalies are, but not *what* they are. ScanSAT will do the same. To refer to anomalies I usually use their general locations. For example "the anomaly on Kerbin's northern ice cap". Those who know the anomalies will know the one I mean, those who don't know are saved the spoilers.

I wonder how those drag figures compare with FAR.

At the current price, maybe. I've played it loads and would easily have gotten my money's worth at any price - and would happily have dropped more after getting hooked by the .18.3 demo. But £30, while a typical price for "big name" games, is on the expensive side for "indie" stuff. Looking in the top sellers in what Steam call "indie", few other games are as expensive or more so than KSP there. But maybe it doesn't matter much. I expect most people who want to get KSP right away will pay £30. Meanwhile those who plan on "waiting for a sale" would probably still wait for a sale if the regular price was £20 or £25. For that matter, I wonder how meaningful "standard" price is for games on Steam at all.

I think it would run to most of the parts list though

Edge isn't Firefox or Chrome. Those two have had a reputation for being RAM hogs.

Yeah. The problem is realistically Kerbin re-entries wouldn't cause significant heating. Squad have a few choices: Implement realistic re-entry heating and accept that it will be a complete non-consideration unless the player does something extreme. Implement a "fudge factor" so things get hotter at lower speeds. This is what they did. Reduce temperature limits on parts. I'm not sure if this would even work though, Kerbin re-entry speeds might still be too slow. Make Kerbin bigger. This has other impacts on gameplay. The big one in my view is a significantly larger planet means a significantly longer launch to orbit. That's something that can't really be balanced out by changing engine performance or tank masses.

I've not been able to find an exact comparison, but here's some testing that may be instructive: http://uk.hardware.info/reviews/5766/3/amd-fx-6300-vs-intel-core-i3-4330-budget-gaming-cpu-deathmatch-performance-in-games http://www.techspot.com/review/943-best-value-desktop-cpu/page5.html In both sets of tests the Core i3 is beating the AMD FX in many of the games tested. The second set pits a 3.7 GHz Core i3 against an FX-8320E overclocked to 4.6 GHz, and still the i3 beats the FX in every single gaming test. (Though note it's a slightly stronger Core i3 than yours, having a bit more CPU cache.) As for graphics card, there's no reason I know you can't pair the AMD card with the Intel CPU or the nVidia card with the AMD CPU. The 960 is the stronger graphics card but also the more expensive one.

Actually plasma strongly absorbs laser light, causing the plasma to heat up further and impeding the laser beam from propagating. It's a big problem for laser weapons in atmosphere, where unwanted plasma forming somewhere along the beam prevents it effectively hitting the target, though the issue can be mitigated by simply using a brief high-power pulse instead of a continuous beam. When the plasma instead forms in contact with the target that can be useful, as in the https://en.wikipedia.org/wiki/Pulsed_energy_projectileIn space the big issue is that a hot plasma will quickly expand and cool. That I think would nix the idea of a cloud of solid particles as defence. However a shield of plasma from the start might be confined using magnetic fields. Which I believe was suggested already in the thread. Keeping the plasma energised and dense enough to be useful will be a challenge.

This may be something all mod developers need to watch out for. Under Linux filenames are case sensitive, under Windows they are not.

With KSP as an interest I'd go for an Intel CPU always. Even if it's not actually a major use I'd still be inclined to pick the i3 over the 6350. Single-thread performance helps everything, and the fast Core i3's have often benchmarked better in games than the FX-6### series. What does strike me though is the cost of the motherboard on your Intel build relative to the processor. While you don't want to absolutely cheap out you could easily save 20 or 30 bucks with a less expensive board, especially as unless you anticipate a future processor upgrade you don't need to restrict yourself to overclocking-friendly Z-series boards but can instead consider less expensive B- and H-series ones.

Looking closely, I think he's put low drag devices, possibly air intakes, on the tanks then rotated them backwards. In newstock aero, and I think in old FAR too, that gives even less drag than a regularly-oriented nosecone.

Probably a hundred firefox/chrome tabs at once. I've heard of people doing that.

The "jitter" on the hold prograde and similar is a known issue, those functions are a bit crude and struggle with ships that have either much more or much less torque for their size than average. The rotation on regular SAS is weird though.If you use FAR, check you haven't left the Flight Assistance tools on!

Have you thought of putting in some sort of indicator that the Flight Assistance tools are active? More than once I've ended up chasing "phantom torque" on spacecraft that turned out to be because I'd left FAR's Wing Leveller on at the end of my previous playing session and the setting is remembered. With the FAR GUI closed there's no indication I've done that. Maybe a simple highlight or effect on the FAR icon in the toolbar would help.

I put that to use on my latest Mun lander. The front and back of the 1-Kerbal lander can have the door and window so ports couldn't go there. The 45 degree sides have the solars and anyway the stock RCS ports are inefficient placed at 45-degree angles. So on the left and right sides I put a quad and two linear ports each to get full RCS translation.Though frankly if I had a five-way port I wouldn't have needed to do that.