Jump to content

Zeiss Ikon

  • Posts

  • Joined

  • Last visited

Everything posted by Zeiss Ikon

  1. I'm sad to say that I spent most of my game time this weekend redoing the test install of RP-1 on 1.8.1 -- I had hopes, since a couple of the mods were updated since my last such install resulted in the game crashing to desktop sporadically. Nope, new install crashes to desktop sporadically, even with KIS/KAS being left out, which also means at least some of my previous non-test save's designs don't import correctly; also some of the engines have reverted to more familiar, but less convenient models, so I can no longer mount three DerWent jet engines close together in the tail, as they're back to the large centrifugal compressor. Doesn't matter much; the game will only rarely even let me finish editing (I've regained the habit of saving after every change, which is bad because I then can't just revert to what I had an hour ago when i realize I've been on entirely the wrong path -- or, if I save with a new name each change, I wind up with dozens of very similar designs and difficulty recalling, after a crash and waiting ten minutes for the relaunch, which one was closest to working correctly). I was reminded, however, that no way, no how, can one get 90 units of Sounding Payload and a parachute suborbital with the first gen WAC Corporal engine. Extended tank to overburn crushes the TWR, as does adding another stage. That means too much time at too slow a speed, resulting in too much gravity turn. That's a contract that either has to fly on an RD-100/A-4 (for which starting level avionics take literally years to build), or wait for the XASR-1 (or even the AJ10-27).
  2. Well, glad to know it's not some weird bug on my system, anyway. Thanks for responding promptly, too!
  3. I'm on Kubuntu 20.04(.3, I think) and I've been getting nagged by my system updater for a week or more about `ckan` having an upgrade -- but every time I try to install the upgrade from the repository (the one on the CKAN page, I checked) I get an error about "file size doesn't match" and then it asks if the mirror is in process of update? I'm no Linux guru, but I think this means that the file on the repo has incorrect information, perhaps that information (size, checksums, and hashes) didn't get updated when the most recent update got uploaded? Not critical to me, I haven't been playing much lately, never mind fiddling with the install, but I can't be the only person getting this message...
  4. If I could get it stabilized, I'd like to try a career with only procedural engines. We've got tanks, fairings, and aero surfaces (and likely won't ever have procedural pods, though we do have procedural avionics). The idea behind procedural engines is to reduce being locked into historic-like vehicles ("Oh, yep, today I put the first Kerbal in orbit with a rocket that looks suspiciously like an Atlas -- the alternative was one that bears a striking resemblance to an R-7."). I'd like to explore HTP/Kerosene engines in 1958-equivalent, as a replacement for peroxide/ethanol, which was an evolution of LOX/ethanol (instead of the LOX/kerosene evolution we actually got). SPE can do this -- if I can ever get 1.8.1 RO to settle down enough. Hypergolics in different sizes seem reasonable, too (the Titan II that launched Gemini used hypergolics), and good for interplanetary because the propellants don't boil off.
  5. I'd be happy if I could get my 1.8.1 RSS/RO/RP-1 game to run right. Sometimes I can play for an hour or so -- and sometimes it crashes the first time I try to stick a part on a rocket or aircraft. And don't even think about trying to run with Simple Procedural Engines. I did manage to complete construction of Eagle 2, my current RP-1 career's first rocket plane (crewed sound barrier was flown with three Derwent jets, launched from the runway). A couple small drop tanks to get my burn time up to 6 minutes, and I should be able to complete all the rocket plane missions at least until I get the cockpit upgrade for higher altitudes than 30 km. Sorry, no pictures, just about ready to take one, and the game crashed.
  6. I spent an hour or so fighting with Simple Procedural Engines. The mod won. In an RSS/RO/RP-1 1.8.1 install that will normally run until memory leaks bring RAM usage up to 10.5+ GB, if I install Simple Procedural Engines (freshest version downloaded direct from GitHub), I can induce a crash to desktop just by putting a bare B class engine in the empty VAB. If it doesn't crash the game as soon as it's placed, it'll do it either immediately when I select the "configure" option from the PAW or before I can make and save any configuration changes. Sure would like to be able to use engines (or engine families) other than the Aerobee and RD-100 in the pre-kerolox stages of the game. Not to mention having the option a little later to try kerosene/peroxide or kerosene/nitric acid (both nice high-density combinations with decent Isp, allowing small tanks, good mass ratios, and reasonable performance with zero boil-off).
  7. I'd have to think at least twice about naming a vessel Poseidon -- I've seen (most of) the first movie, and I hear the remake was much the same. I also see you've missed out one of my more common naming schemes in RSS/RO/RP-1 -- Alpha for Aerobee series sounding rockets, Beta for A-4/RD-100 boosters, and so forth. Beta 1 Mk. 1a, for instance, flew the first bio-capsule and film camera recovery suborbital missions and (so far) the early downrange distance milstones in my current campaign. A couple careers ago (I'm not a very good funds optimizer) Gamma flew the first orbital mission (roughly Jupiter/Baby Sergeant class).
  8. Actually, it's not torque reaction that makes a real Corsair (or R/C model) want to roll over -- it's a yaw force known as P-factor, cause by angled airflow through the propeller when you rotate for takeoff (at least in the real world; I don't know if FAR models to that level of detail and I'm pretty sure the stock game doesn't), which then brings dihedral into play to cause a roll to the left. This is exacerbated by slipstream rotation acting on the tail surfaces, turning the airplane to the left as well (and again, I don't know if FAR models this, but stock surely doesn't). In the stock game, it's almost certainly just torque reaction, similar to a reaction wheel. The Corsair was often called "The Widow Maker" because, due to the very high engine power, it had far more of this P-factor than whatever aircraft new pilots were transitioning from, and required a large right rudder pedal force to counteract this, applied as soon as back stick was brought in to lift off. Letting the aircraft build up more speed on the ground helped (rudder was more effective, and the angle of attack to lift off was less), but in theater, short fields were the rule and the aircraft was also usually flying heavy (with fuel and munitions) when launching for a mission, which increased the necessary angle of attack. In game, aileron trim is most of the answer -- but you'll need to add a bunch before starting your takeoff roll, and take some out once you're in the air (the faster you fly, the more effect you get from a given amount of trim). If you have FAR, you may also find you want some rudder trim -- or simply some right-yaw taps during takeoff.
  9. After upgrading from Ubuntu Mate 16.04 to Kubuntu 20.04.1 (kernel 5.4.something, 64-bit) -- i get the exact same results. Something triggers (clicking too soon after another operation in VAB, for instance, or changing from Space Center to Launch Pad to launch a rolled-out rocket) and the sound stops playing, then after few seconds I'll get just a little snippet of sound and the entire window is gone. Just in case anyone has been avoiding helping because I haven't included enough information: AMD FX8350 (8 cores/8 threads, 4.2 GHz max clock), 16 GB RAM, nVidia GTX 750 w/ 1 GB RAM on PCIe x16 w/ nVidia 440 Linux drivers. Main save has the above list of mods, plus some parts mods (British Rockets, for instance). I'd like to use Simple Procedural Engines, but if I install it, I get a crash one or two clicks into configuring a procedural engine, every single time. Sometimes it'll crash as soon as I've gone into the VAB and Spaceplane Hangar enough times to get the Space Center deglitched (launching from Brownsville or White Sands) -- and sometimes it'll let me play for a couple hours, fly an X-Plane contract and launch a couple rockets (currently atmospheric and suborbital, game date is in 1954). Anything to look for, attach, etc.?
  10. Last night (late, like right at bedtime) I test flew Eagle 1. RSS/RO/RP-1 in 1.8.1, this is the first aircraft in this career. Initial mission: crewed Sound Barrier attempt, X-Plane contracts, and likely Flying Low science around Brownsville, Texas (there are seven biomes accessible on about a two hour flight, and I think I can add enough fuel to this to give that endurance. Still have to figure where to mount that huge Early Film Camera, though. Inspired by the Curtis Ascender fighter prototype, as well as the Vought Cutlass carrier fighter. All starting parts, except aluminum separate structure tanks instead of steel. Stock "Swept Wing" with B9 Procedural control surfaces. the inner trailing edge parts are locked, outers serve as ailerons only (pitch control is on the Standard Canards). Delta Deluxe winglets for the tip fins/rudders. Three Derwent jet engines -- seems to me the radial compressor sections used to be bigger. And yes, that's a JU-87 Stuka cockpit; everything else in the starting parts node is worse. The aircraft is transsonic, sustained level flight, at 5 km altitude. Service ceiling with the Derwents ought to be a little above 7 km. Pilot(s) will be on oxygen at that altitude, of course. It's a little cranky -- tends to pick up a side crab and tip off one way or the other in roll -- but the autopilot handles it fine. A live pilot just has to pay attention all the time. Takeoff is a little more, um, "fun" -- requires near full elevator deflection to keep the nose up below 140 m/s Should be really interesting to land (no flaps, though I may test them) -- haven't tried that yet; images are from KRASH simulation.
  11. Further testing suggests there may be a memory leak somewhere in the install. When I first start the game and load the save, htop reports about 8.5-8.7 GB RAM in use; by the time it crashes, it's at 10.5+ GB -- and it always crashes before it hits 11 GB, any time I've had htop open. Not sure why, since there's 4+ GB remaining free.
  12. Title says it -- I followed the instruction page very carefully, aside from leaving out a couple mods I don't like or don't expect to use (Atmospheric Autopilot, for instance -- KCT has combined VAB and SPH build queues so there's no longer a significant advantage to flying planes -- you have to choose to build planes instead of rockets rather than in parallel with rockets, and you have to choose which building to upgrade on start since you only get one upgrade point to begin). I'm running Ubuntu 16.04 (yes, I know, EOL in five months -- am I the only one in the computer world who doesn't like having to learn a new interface every couple years?), 64-bit, with directly downloaded KSP freshly installed, without Making History or Breaking Ground, for the RP-1 installation. Here's my directory of mod folders: This is a "light" install that I built for testing -- it's hard to elimination test, because RP-1 isn't really playable with just the core mods, Real Solar System, Realism Overhaul, Realistic Progression 1 (parts shortages) -- so I can't be sure the issue isn't already present at that time. By the time I get to the bottom of Section 2 on the installation page, the problem has started to appear. I do know that any attempt to edit a procedural engine from Simple Procedural Engines (the latest version, compatible with 1.8.1) will produce the crash reliably (so I haven't installed it in this test install), but otherwise, it just takes longer. The issue is that, at random-seeming times, in editor or during a flight, the game will crash to desktop (especially annoying due to the startup times for this heavily modded install). In editor, it's usually after duplicating a part or building on too many parts before saving; in flight, it typically happens around staging (so far, when dropping an expended Tiny Tim to let the WAC Corporal sustainer continue to near-space). This sounds like a load issue or garbage collection error. I'm not out of RAM; I have 16 GB physical RAM installed, separate VRAM on the GTx750, and according to htop I'm only using about 9.5 GB (including all open background tasks -- paused BOINC tasks, SeaMonkey browser if open, etc.) and virtually no swap (a couple hundred MB, which I think is minimum). My processor is AMD Fx8350 (8 cores, 8 threads, 4.2 GHz maximum clock, as optimized as I can make it for performance reasons). I'm happy to reproduce the crash and then post a log in my Dropbox share, but I'll need specifics on what file that is.
  13. Okay, I found the binary download -- and despite being a version intended for 1.8.1, when I try to actually edit an engine in the VAB in my 1.8.1 RSS/RO/RP-1 career, the game will instantly crash to desktop. This may not be due to this mod; I get this a lot with this 1.8.1 install (works fine in stock with Making History and Breaking Ground, but I find I can't play stock any more after almost two years playing RP-1 -- what is Jeb doing riding the VERY FIRST ROCKET ever to fly, instead of letting a basic radio transponder and a couple instruments do it?). I'm off to post in tech support thread, I presume I'll get no help there (as usual).
  14. @soundnfury I'm just setting up a new 1.8.1 RSS/RO/RP-1 install, and want to use SPE -- but downloading and installing doesn't seem to work the same way as other mods I've gotten from GitHub. Usually, I get a .zip file with either the mod folder itself, or ~/GameData/[mod folder] and (presuming CKAN doesn't find it) I just drag/copy the mod folder into Game Data and Bob's yer uncle. What I can get from GitHub in this case, however, doesn't show up correctly when I drag it into Game Data. The only .zip I find is source code, and I'm not programmer enough to build from source. What am I missing?
  15. This depends on how old the toilet is. Jet-assisted siphon bowls came along when the effort started to reduce the amount of water used for each flush. When I was a young fellow, no one cared if you used 19l (5 gallons) of water to flush the bowl, so the simplest design was preferred; the water just poured in from the rim until the level rose enough to start the siphon, then enough additional was included in the cycle to keep the bowl running long enough to remove all the waste. Next to no momentum involved. You can spot these older bowls by two marks: first, they'll invariably taller than modern ones, because the height was needed to give enough head inside the bowl to reliably start the siphon; and second, they don't have the small jet hole in the front side of the lead-in to the trap. Once people started wanting/needing to reduce water consumption, there was a generation of really bad toilets that would clog at the slightest provocation (sometimes on just a few sheets of tissue). These had been modeled after pressure-flush bowls from public restroom designs, but there wasn't enough head in the tank to reliably start the siphon if there was material in the bowl -- unless you mounted the tank on the wall, a meter or so above the bowl. This produced very positive flushes, much like the pressure flush type, but eliminated the ability to store items on the tank lid. Finally, the modern pressure-assisted siphon bowl came along, starting with premium brands and working its way down to the point where the cheapest stool you can buy today flushes better than the best one did back in the 1990s. These use an improved pressure jet design which, yes, uses momentum to get the siphon started, and will reliably flush even a very heavy load with less than 3.8l (1 gallon) of water -- I've seen bowls that were rated as low as 2.5l in the last couple years. Hint: if you have one of these modern low-flow toilets, don't try to save even more water by putting bricks or a full jug in the tank -- they will not work if you cut their water use by even as much as 10%.
  16. Wow, seriously? I just barely (like, last week) got 1.7.3 running with RSS/RO/RP-1 and Principia. Kerbalism and Test Lite, this time (instead of [X] Science and Science Alerts Re-Alerted, TAC LS, and Test Flight). By the time I get anywhere in an RP-1 career, I have to stop updating Principia because they've stopped supporting the base KSP version I'm on...
  17. Even if you figure these asteroids are only about 50% solid (i.e. similar to a pile of pebbles, lightly vacuum welded together, hence why they break up and don't produce a crater on the ground), the overall density should be close to twice that of water or solid ice -- so a 1.5 m spheroid at 5.5t is pretty reasonable.
  18. So, this is basically a Dyson bladeless ring-fan, with as much power as the design can handle? I seem to recall discussion (somewhere) leading to the conclusion that this is a terrible way to build a fan, in terms of energy efficiency. Which would imply it's also not a good propulsion system.
  19. Not sure, but there might be a 12 step program for that...
  20. If you're trying to decrease build times, either spending KCT upgrade points at the VAB to increase build rate, or optimizing parts choice and tooling everything you can will cut the number of game days from clicking the "launch" icon in VAB until you can actually roll out for launch. VAB upgrades also slightly reduce rollout times. If you're concerned about the player time while the vessel loads onto the launch pad after KCT is done with it, probably only a computer upgrade to a CUP with higher single-thread performance, or switching from a platter hard disk to SSD is going to help...
  21. It's still not going to pass for a flight sim, but it's better than the stock aero, hands down.
  22. The Space Shuttle had a specific system for this, Terminal Energy Management (TEM). They used the only available strategy, with very limited drag brakes (split rudder only), gear that couldn't be retracted in flight, and a drag chute, also of limited effect: they reentered on a path to be a little high and hot, then used what amounted to S turns in the last minute or two of descent to dispose of excess energy. And then they landed on long, long runways...
  23. I saw a fireball over northern Idaho back in 1980 -- greenish, bright, and traveling south to north. We figured then it was probably a satellite burning up after its orbit decayed. Those don't generally get bright enough to be visible in daylight, though, never mind to flare a camera sensor in daylight... @Vanamonde Average meteoric entry speed is something like 20 km/s, as I recall -- compared to ~1 km/s for an SR-71 going all out...
  24. @nepphhh I have the Kerbalism RO config -- I followed the RO/RP-1 installation instruction page very carefully, after wasting a week of evening spare time and getting it wrong once. I'm now getting data, it just takes a long time to complete each experiment -- but as long as data from multiple flights is added together (it is) and I get the partial Science to spend before completing the entire data set (I do, for instance, 0.1x completed for Supersonic Flight Analysis was good for 2.5 Science points), so i can start R&D working, it works well enough. At this point, around January 1952, with half a dozen launches under my belt, I've got most of the temperature and pressure scans completed for the biomes around Canaveral, and for Space Low; telemetry similar, about 2/3 done with Superonic Flight Analysis, and after three flights on "Space Biological Science" contracts, about 2/3 of the bio sample data -- and I haven't even built an A-4/RD-100 class vehicle yet.
  25. Are those small yellow blocks the Thuds are mounted on Lf/O tanks? If they aren't, the Thuds don't have any fuel to draw from. Fuel ducts will pull Lf/O from their supply end to keep tanks on their output end filled as long as there's supply -- but that assumed there's something for the duct to keep filled.
  • Create New...