Sput42
-
Posts
72 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Bug Reports
Posts posted by Sput42
-
-
17 hours ago, leudaimon said:
I think a similar problem was already raised, but couldn't find it.
I'm playing RP-0, and got my first orbital reentry. The problem is, apparently the atmosphere is generating heat, but not aerodynamic drag. As you can imagine, this is not a nice combination. My Mk1 capsule explodes before the ablator runs out, no matter what periapsis I aim for (tried several altitudes from 90-50km). I believe no drag is occurring because my speed is not reduced at all, my apoapsis remains the same (145-160km) during the whole process.
In an earlier craft I tried a reentry of a probe using heat shields, and also failed, but as it was coming from high orbit, I thought that the heat shield simply did not have good enough specs for this reentry. Much earlier (possibly before some update, as I've been going for a couple of months in this career), I managed to reenter a similar small probe from LEO even without heat shields, using high drag wing brakes.
Is there any reported issue with FAR or Deadly Reentry? Maybe I should try a fresh install? My current install is from CKAN, and I'm on linux x64.
Many people don't realize that real-world reentries are much steeper than you'd expect. They're also really hard; it's not like in stock KSP where you point your craft into the atmosphere and let it go down on its own.
The atmosphere is not really going to slow you down above 70 km; it's just going to heat you up over time until you melt. If you look at the lunar-rated Apollo reentries, they aimed for a Pe around 40 km (some even lower). They go in very steeply and get slowed down at several gees. But since g-forces cannot be too high for the crew to survive, after the initial slow-down the capsule must go up again before it goes too deep. That's done by aerodynamic properties (because RCS won't cut it with the forces involved): capsules have an off-set center of mass (in KSP/RO, this is toggled by "Descent Mode" in the capsule's context menu). Rotation along the z axis shifts the CoM and thus changes the angle of attack, and subsequently aerodynamic lift changes the descent angle. So once the g-forces reach a tolerable maximum (around 60 km or so in the initial descent), the capsule rotates such that it starts going up again -- but not too far, because you also don't want to slip out of the atmosphere again! In KSP/RSS/RO, you generally want to keep it in between 60 and 80 km until you are slow enough for final descent. Once you're above 80 km, aerodynamic pressure will not be sufficient to steer you back down no matter how you rotate, and you bounce off the atmosphere for another orbit which may be fatal if you have a life support mod enabled and limited resources in the capsule (which will be out of its service module by then).
So in short: aim for a low periapsis, switch capsule in Descent Mode before hitting the atmosphere and rotate to a neutral-ish position, wait until you see your g-forces raise to 3-4g or so, rotate capsule such that it goes up again, once you are nearing 70 km steer down again so you stay below 75 km or so. Continue to control your descent path to keep g-forces on a healthy level. Once you're slow enough, just let it fall down.
I recommend quicksaving and practicing until you got the hang of it.
-
On 2/26/2016 at 8:29 PM, stratochief66 said:
Originally, the LMA collider would damage the LEM when it was decoupled. You may already have the believed fixed mu, but here it is anyway just incase.
I decoupled the chute cover in orbit with no problems, so the catching might just be fun aerodynamic dynamics. Try it in orbit and see if it works well without the air interfering.
I get the same flip from SM/CM decoupling as well. Perhaps it is a mild collider issue? Dunno, unfortunately not the type of issue I can fix. Someone capable of viewing and manipulating collider model things might.
I also get the LMA bump when separating the LM. My guess is that it is residual collider issues with either the LMA base or the Saturn IU ring. The LMA base is constructed from 2 half circle fairing pieces that were deprecated from FASA long ago. I haven't looked at FASA in stock recently, but perhaps you (or someone) could create the RO-LMA base from something else, particularly if FASA in stock uses some other existing part for their LMA.
That said, I just look at the existing craft file, so some of those issues could be baked in. I really don't know which part features/functions are refreshed from stored configs or are baked in at craft construction.
Well, unfortunately it didn't fix the issue. I believe it stopped the explosion, but it still had an ejection force way too high. So pretty much what chrisl is seeing.
The other bug I had around that time (end of January) was that SM and CM didn't properly decouple. I didn't have a problem with spinning, but the SM would decouple and then immediately latch on again, but slightly skewed. No way to get those separated properly other than reloading and trying again a few times, or rotating quickly enough to get the thing thrown away before it could latch on again.
That said, I didn't get around to play KSP since end of January. I did try your (not really) fixed .mu and ran a mission with it, with the above results. But I can't say if things have improved in later updates.
-
17 hours ago, NathanKell said:
@Sput42 as it happens, Radiernick fixed it for us. Replace your copy of the file (in the FASA folder) with this one: https://drive.google.com/file/d/0B5cCJeD9k76sd2hrdTRMTmlrX0k/view
Well. Now it no longer explodes, but it still is a rather violent separation, sending both parts of the spacecraft tumbling away at a good amount of speed. I guess that one could be fixed by lowering the ejection force though...
I assume the fix was removing/fixing a collision with the fairing base that caused the original explosion? However, that still doesn't affect the decoupler itself. I wonder if this thing should be a docking port instead, or a stack separator rather than a decoupler. Or it needs its parameters tuned at least. Or the special part used in stock FASA ported to RO.
Anyway, at least nothing got damaged, the spent stage was accidentally put on its intended collision course with the Moon, and just 2.5 m/s of course correction also sent Apollo back onto its intended path. So I can continue the mission at last!
-
On 1/18/2016 at 0:01 AM, NathanKell said:
Are you having this problem after having a fully up-to-date FASA install as well as RO? The exploding comes from collider issues in FASA which should be fixed now, it requires no RO changes to resolve.
Finally got around to launch another Saturn V, with up-to-date RO (10.8.0) and FASA (5.44). Got into orbit ok, did the transposition dance - and when triggering the LEM decoupler, the whole thing exploded again. So this issue isn't fixed for me.
Is there anything I can do to (help) debug this?
-
20 hours ago, NathanKell said:
Are you having this problem after having a fully up-to-date FASA install as well as RO? The exploding comes from collider issues in FASA which should be fixed now, it requires no RO changes to resolve.
Yes, I did install 5.44, although the craft that's in the savegame is from an older version. I assumed that bugfixes in parts would automatically propagate; but I can give it a go with a new game. More practice for the gravity turn won't hurt, eh?
I assumed that the RO version of the Saturn V is different from the FASA one, because I can't even load the one shipped in the FASA Sandbox into the VAB (just the capsule shows up, and it looks completely different).
EDIT: ... and the RO version doesn't have floats and uses a different decoupler, correct?
-
So I was giving RO's FASA Saturn V/Apollo stock craft a spin. Launch went ok, transposition went ok - but when I try to decouple the LEM from the third stage, the whole thing violently explodes.
Last time I was trying out this stock craft was still with KSP 1.0.4 (and I think RO 10.2, but I may be mistaken), and I distinctly remember that there was an explosion effect when decoupling the LEM that most likely shouldn't be there. But at least there was no actual explosion involved back then. But now the whole thing gets ripped apart, and fragments fly away spinning in all directions.
Tried editing the savegame and lowering the ejection force of the decoupler in question (stackSeparatorMini), but no change.
Checked the FASA mod, and noticed in the ChangeLog a bunch of changes related to collider meshes and exploding decouplers, also affecting their version of the Saturn V/Apollo. Had a look at their craft file (which I assume isn't tuned for RO/RSS...), and funnily enough they don't even seem to have a stackSeparatorMini in there...
So... any chance to get this fixed? Or maybe even a workaround for now, so that I can continue the mission?
Since NathanKell now is involved in both RO and FASA, can we expect an update of the RO version based on the seemingly more advanced FASA one? Thinking of, for example, the floats for the capsule... and of course non-exploding decouplers
-
I'm currently experiencing a bug with vessels becoming unfocused during ascent... what I mean to say is that the camera angle adjusts downwards from the ship as if it were focused on the lower stage of the ship (and simply fixing it with manual camera controls (MM+drag) doesn't work either). This effect is cumulative and at the end of my ascent to orbit my vessel is usually not more than a blip in the top right corner of my screen, if I can see it at all. Has anyone else experienced this? Besides the fact that it ruins immersion it's also hellishly annoying when you have instruments on the upper stage that needs to be interacted with.
I've had a look around in this thread and googled a bit with no luck (yet I could have missed something). Any help would be appreciated.
Edit: now the bug is becoming game breaking as vessels spin uncontrolably and blow up when they get out of range.
I had this issue when using the Mk16 parachute with RO. All other parachutes seem to work fine. Not sure if this bug has been fixed recently; haven't tried out Mk16 in a while...
-
Sorry I've been super busy with stuff in RL lately
No worries, it just sounded like people were already toying around with a newer version, so I wondered if there was an alternative download location that I missed or somesuch
Take your time, I know how it feels to be swamped in RL stuff...
-
Have there been any updates to the 64 bit version since beginning of August? I don't see any relevant commits on Github, but maybe I missed a memo somewhere
-
Hi RSS is very good but when i try to fly with Saturn V FASA Mod it's impossible to go to the moon
I can just orbit around the earth
In 0.90 Kerbal ISP Difficulty scaler mod can resolve this problem but how in 1.0.4 ????
If any1 can reach Jupiter or Saturn orbit please explain me
I managed to get the Saturn V/Apollo all the way to the Moon, landed, and got back to Earth with some fuel to spare in both the LEM and the Service Module. Have yet to try reentry to complete that mission (loading up the game as I write this...).
Since you seem to have trouble: Did you follow the actual mission profile? I used NASA material to tweak my trajectories to the point where I managed to be very close to real life in my replay. You have to have a surprisingly shallow ascent path if you want to get into orbit with enough fuel left. First stage should burn out at around 70km height at a speed of around 2.5 km/s. Second stage should get you near an apoapsis of 190 km and a speed of around 7 km/s. Third stage should only use up around 500 m/s or so for circularization, and then have enough fuel left to get you all the way to the Moon (aim for a translunar injection profile that brings you close to escape velocity, i.e. around 10.8 km/s if I recall correctly; that way you need less fuel from the Service Module for the capture burn). Aim for a 110 km orbit.
Once you're in lunar orbit, you should have plenty of fuel in the LEM to get you down and back up again. You can increase this if you use the SM to lower your periapsis to about 15 km before detaching the LEM. Once you're ready to depart to Earth, you should have at least 1200 m/s left in the SM (with no LEM attached, of course), and that is plenty to bring you back home.
Also note that in the stock Saturn V from FASA, all upper stages have their fuel and resources disabled, so you don't drain them prematurely. You have to right-click the parts and re-enable resource flow before using them.
Also, don't forget to activate your CO2 scrubbers and fuel cells, otherwise your crew won't survive the trip and you'll run out of power.
Good luck!
-
Speaking of example vehicles... let me take the opportunity to thank everyone involved for the stunning Saturn V/Apollo replica that comes with FASA and RSS/RO. I'm currently re-enacting the Apollo missions just using the replica, without any tweaks on my side (other than adding a launch tower for looks). I mostly use real NASA sources for figuring out the numbers (such as ascent trajectory, orbit heights, burn times etc.), and they work just fine with the provided Saturn V. There's exactly as much fuel (water, food, oxygen...) in the various stages as is needed for the NASA-defined mission profile. Not only that; but the recent updates setup everything correctly - ensuring that staging works as it should, that fuel and other resources get drained in order, that I'm not using the RCS fuel from my capsule when I still have a S-IVB, and all that... just works out of the box.
So, yeah. Thank you!
-
Could anyone here give me tips on making a Manned Lunar Orbiter? I've managed to get into orbit (albeit with F12 menu as I somehow underengineered the rocket.) I messed around with a procedural fuel tank with 2.5m diameter and a Poodle engine, and when I made the tank even 41m in length it wasn't enough DV to go and get back from Munar orbit. Though I might be being an idiot with the DV map, as I'm counting the Geostationary transfer, which may be my problem.
Anyone have tips for making rockets?
Heh. Again I don't know how SMURFF works in relation to Realism Overhaul... but keep in mind that the original Saturn V + Apollo was *massive*. 110m high, 10m wide, weighing 3000 tons, to get three men to the Moon. A tad more than your measly 41m x 2.5m rocket
While getting into low Earth orbit is "easily" doable considering you need about 10 km/s of dV, getting to the Moon and back requires another 7 km/s or so... and that does not include the lander and its fuel to get to the surface and back (another 5 km/s). So in total we're talking about 22 km/s of dV, plus all the additional hardware for landing, plus of course you need to lift all that from the ground. Twice. So, yeah. A rocket getting you to the Moon and back will be *much* bigger than one that just goes into LEO. Space travel is hard, and there's a reason why Apollo was prematurely canceled and noone went back to the Moon since Funnily enough, getting to Mars does not require much more delta-V (although landing and take-off are more challenging). The Falcon Heavy is half the size and mass of the Saturn V, and able to get 14 tons to Mars (but not back)...
Realism Overhaul actually comes with a Saturn V replica that seems to be pretty close to the original. I'm currently on a mission to recreate the Apollo flights using that thing, and so far I could just follow official NASA material for flight trajectories etc, with the fuel just being enough for every step (as it should be), and travel times etc. pretty much in line with the real missions. I have yet to land on the Moon to see if the LEM behaves like reality as well, though
-
Okay then, will have to make insane 10km/s rockets that I can't comprehend, is it even possible with 2.5m or 1.25m parts?
Orbital speed in Low Earth Orbit is around 7.5 km/s or so, so the 9.1 km/s of launch dV seem to be largely plausible to me. Probably more than that, if you're not hitting a near-optimal ascent trajectory.
I haven't played RSS without Realism Overhaul, so I can't say anything about that combination. But at least with Realism Overhaul, building something small that goes into orbit isn't particularly hard, and doesn't require huge parts - real life fuels and engines provide much more delta-v than the parts in the stock game (mostly because the parts in the stock game have been gimped so achieving the measly 3500 dV you need to get into Kerbin orbit is still some sort of a challenge). You can build a multi-stage sounding rocket that manages to bring something Sputnik-like into LEO...
-
If you want R-P0 with RVE, you can go to the commit and apply the changes to the areas Nathan did to the contracts, but make sure you have Notepad++ it will make it a ton of easier.
Thx for the reply; I guess it'd be even easier (and more useful) to just grab RP-0 as a whole from Github. NK tends to ship the DLLs in the repo, too.
But my main question was if pingopete would be willing to update his Linux64 branch for RVE itself; I'm sure there were lots of neat things added in the past month that I'd love to tinker around with
-
Hey Pete, I was wondering if you have made progress with the Linux64 branch in the past month, and feel like pushing an update to Github
RP-0 seems to have fixed its incompatibility with your mod (in Git, not released yet), so soon I'd be able to install RVE again to bask in its beauty... -
---EDIT : So I tried to mess with the thermal values in the debug window, putting Convection Factor from the default "6" to "1" and now I was able to do a clean reentry, with a perigee of 60km, and peak heating of ~2700 degrees at 65km altitude. But now I'm wondering if I'm hiding a bug by tweaking another variable... I don't want to make reentries at other speeds/altitudes unrealistic...
--------------
Just wondering what Im doing wrong.. Coming back from a Moon Orbit (after a very fun moon landing mission in first person with RPM and ALCOR Pod, what a treat), My pod+heatshield just blows up at around 80km altitude. I feel like this shouldn't happen? I'm using the regular RO install with FAR, no DRE.
At 130 km :
Mk 1-2 Pod
4m Heatshield
Orbit speed : 11014 m/s
Apo : 598 Mm
Peri : 80.655 km
Here's a screenshot a second before blowing up
Am I missing something? is it a bug?
Thank you very much! this is really frustrating
I guess rather than changing around heat values, you should try out different reentry trajectories. I've only just started playing around with RO/RSS, and I had a similar issue as you - coming in with a periapis of around 60-70km, then hardly slowing down at all until I blew up at 80 km or so. I remember from playing stock KSP with DRE that this is actually a problem with going in too shallow; contrary to intuition, coming in shallow means you're likely to blow up simply because you spend lots of time in your plasma bubble without really slowing down, while at the same time the thin air doesn't allow for much convection. So going in steeper should solve that problem - although come in too steep, and you get unsurvivable g-forces or don't have enough time to slow down enough before hitting the ground. There's a reason real-life reentry corridors are pretty narrow
I subsequently researched into "the real thing", i.e. NASA data from the Apollo missions, to see how reentry looked like in real life. Turns out they aimed for a much lower periapsis, at around 38 km, so they hit thicker atmosphere quickly to slow down. Once they reach about 2g or so, they steered the capsule up again (by rotating around the offset center of mass, which changes the aerodynamic properties of the lifting body and gives quite some amount of controlling your vertical trajectory; in RO you can do this by enabling "Descent Mode" for your capsule). The capsule goes up more or less ballistically at this point, reducing g-forces back to zero, however not leaving the atmosphere again. Once it falls down again into the thicker atmo, the capsule is slow enough to go all the way down without exceeding maximum g (although one can still somewhat influence the impact point by doing the rotation thing again).
Here's a fun video by NASA explaining the process in detail (although withouy giving many numbers):
Disclaimer: I haven't actually tried this myself yet (still on the TLI part of my Apollo mission recreation), but I would expect Realism Overhaul to behave closely enough to the real world that this should be a way of surviving reentry without tweaking variables.
RealHeat and Deadly Reentry are very much a requirement however; DRE is not making things more deadly anymore, but may tweak things to behave a bit more realistically.
-
Just out of interest, are there any configs/packs for this? I know pingopete has his for RSS, but I was wondering if there were any for stock
Proot is working on a new version of KSPRC making use of this and Scatterer. Looks gorgeous. Maybe you can get access to some pre-release builds if you ask nicely?
-
Scatterer has released a new version (0.0175) a couple days ago, and it's not yet available in CKAN, so I thought I'd mention it here. I take it that Scatterer does not (yet) have native CKAN support to be picked up automatically.
-
I have switched my KSP installation from Windows to Linux and now I'm trying to install RVE. So I downloaded the Linux master package. But it is incomplete as the readme states. I have to download another .zip for the eve files (the Overhaul zip) and a third for the scatterer. In the end, installation is very complex, error prone and doesn't work for me. Probably I did something wrong.
The "Overhaul" zip contains only one folder for GameData, where the plugin folder is empty... So EVE won't function.
And instructions on how to install it for Linux?
For example:
The Linux zip has this folder structure "RVE" -> "EVE" (NOT EnvironmantalVisialEnhancements". And EVE has the very same files as "BoulderCo".
The Overhaul zip has a GameDate folder with "BoulderCo" in it and a folder "EnvironmantalVisialEnhancements" (NOT EVE), which contains shaders but no plugins.
The 1.04 master zip (for Windows only?) has a folder "EnvironmantalVisialEnhancements", which contains shaders AND plugins. And an RVE folder like the Linux zip. But no BoulderCo.
?????
I am confused.
If I recall correctly, the EVE repo contains a file called "release-x86.zip" or somesuch. This is the actual release containing the plugins - download, unpack, and put into GameData.
-
Fundraising is indeed a bit OP, one shouldn't go above 25% at any times if u want to keep things realistic-ish. As for mechjeb... I consider it normal to use it, after all real rocket's weren't steered by hand like in stock KSP. There's an alternative called kOS, but ain't it too complicated?
Well, first of all, it's a sandbox game, so the definition of "cheating" is anyone's own choice. Personally I never use MechJeb, because I consider flying my spacecraft to be the core gameplay. For me that would feel like playing a flight simulator and letting the autopilot do everything. Manually getting that Saturn V into the perfect orbit, under the constraints of RSS and RO, feels like a real accomplishment to me, and I wouldn't let any autopilot spoil that for me. I do, however, use Kerbal Engineer to get all the flight data, because - much like in a flightsim - knowing your height above ground, or your remaining fuel, makes sense to me.
Others see the designing of rockets as their core gameplay, and let MJ do the work of actually flying them. And some, for example, like building giant space stations and use MJ to do the shuttle flights for the material. Different goals, different styles of play, I guess.
EDIT: Regarding kOS - that's actualy scripting, you're writing programs to control your rocket. It's a completely different kind of gameplay, but also one that I'd find much more rewarding than using pre-made scripts or something like MJ.
-
It works on a 32 bit system...with MANY mods. Just be sure to have the opengl tweak.
Ah, maybe the master branch does. I was playing around with the Linux64 branch, which has all the updated eyecandy. But had to remove it again due to incompatibilities with RP-0 until pingopete fixes that issue.
My current install clocks in at around 6 GB *without* RVE, but I'm also using 8k textures for RSS.
-
Hi, can you tell me how did you got RVE working ? It keep crashing on loading, and when it don't, it's all .... up. ^^
Sounds like you're trying to run it on a 32 bit system.
-
Aaah. Maybe this is what I saw last time I tried out RVE, when I reported the cities being visible underwater. Maybe it was daytime.
-
Sure I can Running Gentoo Linux.
Now, the bug report you linked me to is pretty interesting. My version of curl is much newer, but by default Gentoo builds it without asynchronous DNS. I'll enable the adns USE flag and see if it helps with CKAN... I'll report back.
So after having recompiled my libcurl with async DNS support, and using CKAN for quite some time, I can happily report that I didn't experience any crashes anymore. I'm a bit unsure why this also seemed to have fixed the crashes that occured while not downloading anything, but I'm not gonna complain...
Thanks for the pointers to that bugreport! Would've never occurred to me to look into that direction myself.
[WIP][1.0.5]* RSS Visual Enhancements (RVE)
in KSP1 Mod Development
Posted
Old Linux partitions? Linux is my main OS, and I intend to continue running KSP on Linux because that way I don't have to reboot for KSP fun...