Jump to content

AVaughan

Members
 View Profile  See their activity

  • Posts

    662

  • Joined

  • Last visited

 Content Type 

Everything posted by AVaughan

  1. AVaughan
    It's been a few years while since I played with USI life support, so my recollection of it's exact behaviour is slightly fuzzy, but I used to regularly make one "mistake", that turned my kerbals into unintended tourists. (The first time it happened was a real what the heck moment). I tend to start building a new vessel with my Kerbin return capsule, then I turn that into a lander for for whatever body I am heading for, then add extra hab etc to the transfer stage. So I would get to my destination, decouple the lander, start the deorbit burn, and suddenly all my kerbals would become tourists. Easily worked around provided I remembered to make sure that the root part is on the transfer stage, and/or use a separately launched lander. (I also started to always add a probe core to every lander as well, which means that I probably still have control even if they all become tourists).
  2. AVaughan
    20 m/s is 72 kph (or 45 miles per hour), which is quite a respectable walking speed for anything.
  3. AVaughan
    I'm pretty sure that contract is doable with a 20 ton pad.
  4. AVaughan
    @AmpCat Do you have kopernicus installed? There has been a discussion about performance problems in 1.7.3 related to kopernicus in the kopernicus thread. It might be relevant.
  5. AVaughan
    More parts means that both the physics calculations and the rendering takes longer. Doubling the number of parts probably roughly doubles how long the cpu and gpu takes to render the craft. But doubling the number of parts probably also increases the physics calculations time by around 3-4 times. So going from 45 to 224 part is probably quite noticeable, unless the 45 part craft limited by the frame rate cap, and cpu and gpu were idle part of the time. (Also note that much of ksp is single threaded, so don't expect ksp to ever max out all 10 cores).
  6. AVaughan
    From memory on my install it was more like 90 days. (But my memory isn't the these days, so I could easily remember wrong). Regardless you will want to spend funds on increasing both VAB and R&D rates. The tiny tim isn't actually necessary for a first rocket. (I'm pretty sure I skipped it). Did you remember to tool the tanks? Not tooling the tanks makes the rocket more expensive, and that will make it take longer to build. Which version of RP-1? From ckan, github release or github master? (I'm using github master from a few weeks ago. Possibly there have been balance changes since). Edit: Also the best place to ask is on discord. https://discord.gg/8u3nrav scroll down toward the bottom for the ro and rp-1 support channels. Edit2: Just tried in a new game, minimal rocket, procedural tank (smooth cone 300x800) for the nose, 2 procedural tanks 300x800 for the body, 300mm sr core areobee. (I culdn't be bothered adding fins or a launch clamp, since I'm not actually going to launch it). Before tooling 161 days. After tooling 78 days.
  7. AVaughan
    I don't have an RO install, but my 1.6.1 rp-1 install is working fine, including avionics limits. Maybe just try installing rp-1?
  8. AVaughan
    @Sammakko78 Start by checking that your mods are upto date and installed correctly. If that doesn't fix things then we will need more info. For a start at least your KSP version, and a list of all the mods you have installed (including their versions). See
  9. AVaughan
    @AmpCat That sound like a well known issue of KSPs electrical system at high time warp. See Streetwind's comment
  10. AVaughan
    Was that image from a direct ascent concept? (I think that the early Soviet concept level plans were direct ascent). Edit: It might also have been from an Earth orbit rendezvous concept were the lunar mission was assembled in Earth orbit, then did TLI, capture, descent, ascent and return without a Lunar orbit rendezvous. That would also explain the return vehicle "docked" to lander.
  11. AVaughan
    Regarding used crew Dragons. I wonder what the market is like for orbital space tourism? Used booster, new second stage, used crew dragon, 5-7 passengers at maybe $10-20 million per seat? Life support should be adequate for them to stay over 8+ hours in space. To me that would be more attractive than Blue Origin's or Virgin Galactic's sub-orbital flight with only a few minutes of weightlessness, even if it was 2 or 3 times the price. (Not that I have even $100,000 spare anyway). This wouldn't need to be a long term program, just for however many years it takes until they have a manned starship version available as an alternative. (That could be a few years, even if starship + superheavy make a success test flight next year). Commercial passenger flights to ISS is another possibility. SpaceX can probably undercut other bidders there, and might be able to re-use crew Dragon that way.
  12. AVaughan
    For the initial crew version, I sort of expecting that they will need to include a capsule (ie something equivalent to a Dragon/Orion/Apollo capsule. For the initial crew versions, I'm expecting max crew size of around 6 - 12). That can serve as part of an emergency escape system during ascent/descent, and even as a re-entry capsule for a lunar return, if for some reason Starship has a serious anomaly. It will cut into payload capacity, but Starships payload capacity would still be large even carrying a built in escape capsule, and I'm not expecting that they can get NASA to man rate starship for Earth takeoff and landings without some sort of emergency escape system, at least not for the next few years. The only alternative I can think of is either transfering crew after launch or docking and pushing a separately launched capsule. For a 50-100 passenger Mars version such an escape system would be less practical, but they could either transfer passengers after launch, (eg as part of the refueling flights), or maybe just not bother to get that version man rated by NASA. (NASA is likely to insist on a man rated vehicle for any NASA contracts that are going to be carrying astronauts, but do SpaceX need a man rated vehicle for Mars colonists)?
  13. AVaughan
    There is no released Kopernicus version for 1.7.2. Use KSP 1.7.1 or 1.7.3, or compile from source (at your own risk if that doesn't work properly).
  14. AVaughan
    So I compared that to the figures I get based on the assumptions I used in https://forum.kerbalspaceprogram.com/index.php?/topic/159887-spacex-discussion-thread/&do=findComment&comment=3600985 and https://forum.kerbalspaceprogram.com/index.php?/topic/159887-spacex-discussion-thread/&do=findComment&comment=3601019. Using the same assumptions, (dry mass 75 tons, payload 100 tons, fully fueled mass 1100 tons, ISP 380) I get a dV of 6850m/s. If I bump the fully fueled mass up to 1115 then I get 6900m/s. (Of course Elon only specified 2 significant figures, and 6850m/s would round to 6.9km/s). But this agrees well enough with my calculated figures that the assumptions in those two posts look reasonable.
  15. AVaughan
    If you are running out of ablator and burning up before you slow down, then lower your Pe. (That way you will slow down faster).
  16. AVaughan
    My personal reaction is "No thanks". (Note I talking about [X] Science, I haven't used Science Alert in the last few years.). Whilst I could use Remote Tech's flight computer to schedule science collection whilst out of communication, for me that is more hassle than it is worth. (If other players want to that is fine. If they do that, then they can just close the here and now window, or ignore its buttons). In a real space missions they probably either collect the data continuously, or have someone to calculate the correct time to make the observation. Stock type science experiments just aren't setup for continuous observations, and using the flight computer to schedule science collection is cumbersome, error prone and tedious. In the same manner I tend to just use mechjeb to execute maneuvers when out of comms. Whilst playing with RP-0, the flight computer caused a couple of epic failures to execute maneuver nodes because it was unable to oriented the ship in the correct direction). Personally, gameplay and a more useable UI are more important to me than realistic simulations of signal delay/comms blackout. (If you want a realistic simulation then remember that real spacecraft flight controllers/engineers have days/weeks/months to plan, program, rehearse and practice any actions that spacecraft need to do whilst out of comms).
  17. AVaughan
    Well one possible idea would be a sort of hybrid of KSP + RPG. Jeb build a rocket in his backyard from junk, launches into space, finds junked bits of other spaceships in orbit or on the planets/moons. From studying them he gets access to new parts. He also gets experience from visiting different planets/moons etc. With the experience he can upgrade his skills and use those upgraded skills + the new parts to design/build better rockets. That might work as a spinoff. (Though the flying to other places would be central to both games, and it might be too similar to work as a spinoff).
  18. AVaughan
    THe most recent release is meant for KSP 1.6.1. The current version probably wont work properly with 1.7.x.
  19. AVaughan
    Was the kerbal carrying anything? From memory the new ground science parts have mass. (I'm not sure how much mass but potentially enough mass to be noticeable on eva). I'm also not sure whether KSP also adds mass if you pickup a Munstone.
  20. AVaughan
    I've only skimmed the anandtech review atm, but Intel is often ahead in games and in single threaded benchmarks. (From what I remember of previous discussions, KSP seems to be mostly single threaded, so single threaded benchmarks are relevant to the discussion). eg look at these benchmarks. I cherry picked those benchmarks as an example, but it does point out that at least in some workloads, Intel may be better. (The reviewer also mentioned out that Intel still has better memory access latency, which is also potentially relevant for single threaded/KSP performance). Overall through AMD seems to have caught up in IPC, and might be ahead in instructions dispatched per watt. (Instructions dispatched per watt probably should consider the motherboard + cpu + ram as a platform, and Anandtech didn't report that). So AMD did better than I expected, and I'm pleasantly surprised. Now hoping to see some comparison benchmarks from (cpu bound) games I actually play.
  21. AVaughan
    Just add a probe core or a docking port to the top of the vehicle. Then you can right click the part -> control from here. Then turn SAS on and set it to radial out.
  22. AVaughan
    Which procedural wings mod are you using? According to CKAN, for RO you should be using B9 Aerospace Procedural Wings - Fork. There are a few different for of proc wings, and last time I installed RO/RP-1 it took me a few attempts and some reading of the relevant mod forum threads to find a version that worked with RO/RP-1/FAR and KSP 1.3.1. That was probably 6 months or more ago. I might even have ended up using a .dll that someone posted to one of the mod forum threads.
  23. AVaughan
    I consider that good enough for most purposes in RO. (Hitting a precise orbit is harder in RO than in stock, since in RO most engines don't throttle down to zero the way they do in stock). If you really want a circular 176 km orbit you should be able to fine tune that easily enough using RCS.
  24. AVaughan
    If the problem is wet noodle syndrome, then I suggest enabling "advanced tweakables" in the settings of the game, then using one of the autostruct options on some of the parts. (No idea whether you have already tried that, but your OP is lacking in details).
  25. AVaughan
    Probably because the most recent release is for KSP 1.3.1. Search for Realistic Progression One. (I believe the mod devs are working on a version targeting 1.6.1).
×
×
  • Create New...