brdavis

OK… so you have two sphere, each with a radius of 100,000… that's 100,000 what? Meters? Kilometers? Something else? If you are using G = 6.673e-11, then all your units had better be in kilograms, meters, and seconds (or combinations based on those). If they start at rest at a distance of d = 2.5e+10 meters apart… then that's a separation between their centers of 2 * r + d = 2 * 1e+5 meters + 2.5e+10 meters = 2.5e+10 meters, close enough. On that scale, the radius doesn't make any difference. Now… if you release those from rest, they should "fall together", gaining velocity the whole way down, right? That sounds like a problem with escape velocity or energy conservation… you are applying an equation (often called the "Vis-Viva" equation in physics) that considers energy conservation but deals with the case where one mass is much much more massive than the other. Is that the case here? You may or may not have yet another problem - how are you entering numbers into your calculator? It may not matter here, but scientific notation has a different order of operations than the "normal" functions like multiplication… so 1000 / 100 = 10, but 1*10^3/1*10^2 = 100,000, and (on my calculator), 1e+3 / 1e+2 = 10 again. Do you see the difference? Do you know which your calculator is doing? Yeah, that Vis-viva equation is handy - it's how I figure out my transfer orbits in KSP all the time. -- Brian Davis

Probably the backup plan, if the entire TDRS constellation dropped out, would be "come home immediately"… assuming everybody wasn't dead. You could still use ground transmitters to talk to them however. The bigger problem is the "Kepler syndrome" (i.e., an orbital debris cascade). My son asked me while watching the movie "would this happen?", and I had to answer with a firm "No, but…". Yes, you can get to a state where the debris multiply rapidly (exponentially)… and that's a serious concern... but that doesn't mean anything like instantly. The movie seems to imply everything going to kibbles and bits in a single orbital period or so - not even the right order of magnitude. An orbital debris cascade would take many multiples (as in "hundreds" or more like "thousands") of the orbital period. And it wouldn't change spacecraft communication in the least… not unless such a debris cascade in LEO would impact Geo-sync orbit, which is where the multiple (yes, they have back-ups) TDR Satellites are located. Another significant goof here is if the debris stream (from a recent break-up) encountered the shuttle one orbit #1, the chances are incredibly remote that it would encounter it on orbit #2, #3, etc. After all, if the break-up was in the same orbit as the shuttle, it wouldn't have a velocity relative to it… and if it was in a different orbit than the shuttle, they only intersect in two places. So unless you have the break-up event occur in an orbit inclined to the shuttles, but with an identical period, this is not a regularly once-an-orbit repeating phenomenon. Still bigger problem? Given the likely encounter speeds, they literally never would have seen it coming. Let's say they had a relative encounter speed around 5 kps… if you were lucky enough to see it at a distance of a mile (think about this… given the size of the chunks, do you think you'd spot them at a distance of a mile?), you'd have 0.3 seconds from catching a glimpse to impact. No chance… especially since anything on-coming to you at those speeds would have nearly zero drift velocity across your field of view. Yeah, the fire in space bit bothered me - because while you *do* get fires in space (Mir had some *serious* issues in this regard), they do not apparently look like that. But, hey, it will visually give the viewer the idea of "danger from fire". At least they didn't resort to "hearing explosions in space". I appreciated that. -- Brian Davis

First… it's a movie, so I didn't expect much in terms of scientific accuracy. Second, if you've not read the Bad Astronomy blog on this one, it's probably worth a read. Third, if we really must manufacture rationals for plot points… as she is grappling with the Tiangong, it is clearly undergoing re-entry with some atmospheric drag (because, um… that would happen. Somehow. Maybe magic). If that's the case, she would have potentially had a high drag coif than the Soyuz, leading to a separate (both are decelerating, but she is decelerating more due to a lower areal density). But… let's be honest, that's really sort of stretching it for a reason. And Tiangong shouldn't have been deorbiting anyway. And the relative closing velocity would have killed her. And the off-axis thrust from a fire extinguisher (a pressurized fire extinguisher on the ISS?) would lead to even worse yaw/pitch issues than shown. And hypervelocity impacts would have a *terrible* time coupling in that much angular momentum to a system like the Shuttle or even the Canadarm. And the lack of blog when a guys head was essentially drilled through. The fire on-board the ISS would have generated internal pressures more than sufficient to do structural damage without impacts (remember Apollo 1?). ISS and Hubble aren't even remotely close (wildly different inclination… and from KSP, we all know different inclination are easy to transfer between, especially from a "low" orbit, right?), and Tiangong isn't there either. No way would you use something like an MMU to be moving anything like that fast with respect to nearby objects. And… And… and… Clooney didn't have to let go in the first place. They were *at rest* with respect to the ISS. At that point, there were lots of possible solutions. Yeah, I'm a physicist. There's a lot of science it gets wrong. And (except may, perhaps, for me… a little bit…) it really doesn't dent the movie much. Visually, it's great… and it's still so far outside most people's understanding of reality that it's a fun "oh, I never thought of that" ride. -- Brian Davis

As a college professor teaching 200-level physics and 100-level astronomy... Yeah, KSP is fairly awesome. The biggest problems I have at this point (remember, still not release-ready) is the lack of heating effects and aerodynamics. A patched-conic approximation for orbits is wonderful; it's all you need unless they added a "double-planet" or Lagrange points. A few weeks ago I downloaded the 0.21 demo... within two weeks my 200-level students were doing a Kerbal-based lab, calculating the thrust of the LV-909 based on the measured performance, taking the changing (and unknown) mass of the spacecraft into account, demonstrating the impulse approximation, calculating the mass of Kerbal itself based on the measured orbital properties, and doing the calculation for a transfer orbit (no, I did not tell them about, or let them use, the maneuver nodes). Yeah, as a teacher, and physicist... I like it. It wasn't even hard to set up a lab (set up and copied a saved game, so they wouldn't need to do a launch... fun, but not needed for the lab). I need to figure out how to author tutorials, so I don't have to keep restoring 'normal' games for the next lab. -- Brian Davis

OK, first post... I've not been playing long (like, about 2 weeks) and then 0.22 comes out. So, first... Wow. I'm already using the Demo 0.21 version in teaching university-level physics. And loving it. Now if I just didn't need to sleep... can we add that to the research tree? Three things I've noticed... 1) The new SAS does seem to have some issues, but I'm not sure if it's the SAS or the way I've been treating it previously (not enough experience). Having it auto-set the mode (so I don't have to keep tapping "F" to unlock it and change position) seems to be a problem (overshoot/undershoot). 2) The LV-909 doesn't produce power... Did I just completely miss that in my previous experience, only noticing it now when power (for transmission of data) is so critical? 3) I still have trouble with "phantom throttle" - throttle off, hit "esc" to pause, come back later, unease... and the engine throttles back to full. Hitting "X" at this point cuts the engine, after which is mysteriously just ramps right back up to 100%. Throttling down to zero will shut it off again, but "X" will not in this mode. Is this just me, or a known problem (had it on 0.21 as well). Running under OSX (with nothing else running) A thought on the science value transmitted vs. "hard returned" (but not too hard)... yeah, it's a little odd. I *want* to radio data back, because the #1 rule is usually "let the next guy know what killed you" - the actual sample is better, but as it stands there's not a lot of reason to "back up" these things with transmission. Maybe have the transmission give you 70% (or higher... I agree things like thermometer records should be 100%), and you gain an additional 30% if (not when) you land? Sleep is for the weak, right? -- Brian Davis