[1.2] Real Solar System v12.0 Dec 8

[etheoma]

It would be a good thing for you to provide a save file with a single craft on approach to Eve. That would be the easiest way to try to reproduce your problem. (stock or B9 parts on the plane please)

I have a setup that I can try that with but I don't have the time to put together a plane and launch it and send it to Eve.

Edit:

And regarding your other post, any density changes are coming primarily from FAR utilizing pressure and temperature, which is why I mentioned Eve's temperature curve below (Edit: Above! I meant above. Previously). It affects density. If there really is a bug in RSS it can ONLY be in the temperature or pressure curve. RSS doesn't have anything else that can affect density except those two things. FAR does everything else including actual density implementation and application of the forces that are destroying your plane. You might have been playing with FAR for a year and a half but I've contributed code and support to RSS for as long as it's been around and I know how it works internally.

They had to do work on the atmosphere to account for the difference is curvature between stock EVE and real eve and all the other planets.

[etheoma]

edit: oops, just saw your older post where you said it was turned off.

Sounds like FAR ripping your plane apart with its Aerodynamic Failures feature. Even back when I used FAR I disabled that because I felt it was unnecessarily and unrealistically harsh. That said however, if your plane is losing control then it's more likely to be your design than 'a bug with the atmosphere'. If you're deorbiting in any sort of space plane then you need to design carefully, especially if it's an unpowered reentry.

Front page. Links are there to exactly what you're looking for. Also to 6.4x Kerbol system.

I have used multiple ships to try this. Also I have another video encoding at this moment to show you that on kerbin I can fly though air densities of 0.010 at 8kms which is 10 times the air density of which I was flying over EVE and 3kms faster.

And also another take of me descending into Eve but a complete take so you can see the ship getting knocked about for no reason where I eventually blow up with a air density of 0.000 so less than 0.001 and probably less than 0.0005 because FAR probably rounds 0.0005 to 0.001

Edited November 29, 2014 by etheoma

[Starwaster]

They had to do work on the atmosphere to account for the difference is curvature between stock EVE and real eve and all the other planets.

I have no idea what you're trying to actually communicate there. Just in case it wasn't clear btw, 'they' happens to include me.

In any case, the more I look at your video, the less likely it is that this is a problem with Real Solar System. You might not think that that fractional density of 0.001 is a problem, but that's about the equivalent of 45,000 altitude over Earth, and you're doing over Mach 25 in it. In a space plane. Maybe you could pull it off in a less sophisticated design, but nothing that you're going to take off in again.

By the way, you do understand don't you that Eve is Venus in Real Solar System? Its atmospheric pressure is 90 times that of Earth's? That's why I say your 90 km altitude is the equivalent of 45km Earth.

And doing a physical time warp isn't helping you either. Doing so causes math errors in everything that's going on in your reentry including FAR's calculations for application of force.

I'm not at all surprised that you're breaking up there

Edit: Also, if by some miracle you make it down to the planet, those Kerbals are never leaving that planet. Not in that plane.

Edited November 29, 2014 by Starwaster

[etheoma]

I have no idea what you're trying to actually communicate there. Just in case it wasn't clear btw, 'they' happens to include me.

In any case, the more I look at your video, the less likely it is that this is a problem with Real Solar System. You might not think that that fractional density of 0.001 is a problem, but that's about the equivalent of 45,000 altitude over Earth, and you're doing over Mach 25 in it. In a space plane. Maybe you could pull it off in a less sophisticated design, but nothing that you're going to take off in again.

By the way, you do understand don't you that Eve is Venus in Real Solar System? Its atmospheric pressure is 90 times that of Earth's? That's why I say your 90 km altitude is the equivalent of 45km Earth.

And doing a physical time warp isn't helping you either. Doing so causes math errors in everything that's going on in your reentry including FAR's calculations for application of force.

I'm not at all surprised that you're breaking up there

Edit: Also, if by some miracle you make it down to the planet, those Kerbals are never leaving that planet. Not in that plane.

Dude I have video on the way to show you that you can go though Kerbins atmosphere in 4x physical acceleration at a air density of 0.010 at 8kmps.

And as I said I have done this 20 times without phyical acceleration, I was just bored at that point because I had tired 40 times +

And in the video when I tired to enter Eves atmos again the Frac desity was less than 0.001 or showing 0.000 which I think means =< 0.00049

Edited November 29, 2014 by etheoma

[Starwaster]

Dude I have video on the way to show you that you can go though Kerbins atmosphere in 4x physical acceleration at a air density of 0.010 at 8kmps.

And as I said I have done this 20 times without phyical acceleration, I was just bored at that point because I had tired 40 times +

And in the video when I tired to enter Eves atmos again the Frac desity was less than 0.001 or showing 0.000 which I think means =< 0.00049

Just because you got away with it one place doesn't mean you can get away with it anywhere. Math errors make bad situations worse. The first time you ever do a physics time warp on a new KSP install it even warns you that there can be consequences. Is it really so surprising to you to actually see that in play? The more you time warp the worse the problem gets. Fewer and fewer data samples and updates occur causing math errors to creep in. Not only that but KSP's code mixes doubles and singles allowing for math errors even when not time warping. When you combine the two the problem gets worse.

You can see it at play in anything that depends on measuring time delta (the amount of time between updates) including generators and solar panels losing power during time warp when they should be gaining.

You can see it time warping in Deadly Reentry when a shield depletes in seconds when it should have lasted all the way down to landing

That's also the reason why you aren't allowed to physics time warp faster than 4x.

[etheoma]

Just because you got away with it one place doesn't mean you can get away with it anywhere. Math errors make bad situations worse. The first time you ever do a physics time warp on a new KSP install it even warns you that there can be consequences. Is it really so surprising to you to actually see that in play? The more you time warp the worse the problem gets. Fewer and fewer data samples and updates occur causing math errors to creep in. Not only that but KSP's code mixes doubles and singles allowing for math errors even when not time warping. When you combine the two the problem gets worse.

You can see it at play in anything that depends on measuring time delta (the amount of time between updates) including generators and solar panels losing power during time warp when they should be gaining.

You can see it time warping in Deadly Reentry when a shield depletes in seconds when it should have lasted all the way down to landing

That's also the reason why you aren't allowed to physics time warp faster than 4x.

Dude really I said I have done it 20 times without it and "Everytime." the same thing had happened with or without time acceleration.

And after spending 1 - 3 hours trying to get it down every time I'm not doing it again at regular speed.

And went WAY beyond what I was doing on eve 10 times the air density 25% faster and with time acceleration.

failing apart at much less extreme circumstances is a bug by any definition.

Also I recently overclocked my machine and dw I tested it with prime95 for like 10 hours no bug and also the reduction of part count my timer hardly ever went yellow never mind red which means most of the calculations were being done perfectly you can check out how much was being skipped your self.

http://youtu.be/GQkG8LreZCI

next week I can do one without time acceleration to just prove the point even futher but can tell you the results are the same I have personally tested it plenty of times on Eves atmos same thing happens

Sometimes you can without time acceleration SOMETIMES manage to get down to 18km and then blow up travelling at 20 - 40ms then end up in your pod in space travelling near the speed of light. <--- that is certainly not normal behaviour, Dat da space kraken.

Is the space karken part of the real solar system mod because I don't like it XD

Also I have another video of a non-phyical accelerated decent already recorded if I really need to upload it as prove.

and the curve changed helped now I can get down to 92 before my ship get blown up, Small steps

Edited November 29, 2014 by etheoma

[Pro7ech]

Hi everyone, I am stuck with this. The game is still running, responding and music is going one. Anyone has an idea about what is going on?

Edited November 29, 2014 by Pro7ech

[Pro7ech]

double post

Edited November 29, 2014 by Pro7ech

[Einkleinermensch]

It's all OK, you must to wait a bit. The RSS loads now this way, instead of "freezing" the game with black screen like in previous versions.

[Pro7ech]

Ok, found what was going on. It is related to the memory limitation of the 32bit version. It will stop loading but wont crash the game.

[NathanKell]

Changelog

v8.3

*Added many new biomes from Felger (Luna, Mars/Deimos/Phobos, Venus, Jupiter and its moons, Saturn and Titan, Uranus, Pluto.

*Includes DDSLoader from Sarbian, and full DDS support.

*Allows specifying textures via GameDatabase rather than via direct loading (remove GameData/ prefix and remove extension, and make sure the image you are referencing is not under a PluginData folder)

*Update loading to include AppRoot when loading locally (thanks TriggerAu)

*Log on not finding LaunchSites.

*Add support for VertexPlanet and for useHeightMap in LandClass (thanks Starwaster)

*Support more light-shifting (can now modify Sun.AU and Sun.brightnessCurve)

NOTE: To take advantage of the new DDS support, DELETE your old RealSolarSystem folder, then install this, then install textures from the NEW links in the opening post.

[Starwaster]

Dude really I said I have done it 20 times without it and "Everytime." the same thing had happened with or without time acceleration.

And after spending 1 - 3 hours trying to get it down every time I'm not doing it again at regular speed.

And went WAY beyond what I was doing on eve 10 times the air density 25% faster and with time acceleration.

failing apart at much less extreme circumstances is a bug by any definition.

I'm not going to argue with you any more about your use of phys time warp, but when you get odd things happening to you when using it you have nobody to blame but yourself for persisting in using it.

That's number one.

Number two: Stop treating RSS Eve as Eve. What you can or can't do with stock Eve has nothing to do with what you can do here because this is Real Solar System and Eve is VENUS. Average pressure on its surface is 90 times that on Earth. You'd have to go a kilometer underwater to encounter such pressures on Earth. The altitudes you're dealing with in your descent when you come apart is the equivalent of flying on Earth or Stock Kerbin at 45-50 km with FAR except that you're travelling at orbital velocities. I don't know why you expect to do that safely in something as fragile as a space plane but you can't.

Number three: Stock Physics.

Sorry but I'm done here. I have to move on to other things (Deadly Reentry, Ioncross, Stock Drag Fix, in that order).

[ANWRocketMan]

Changelog

v8.3

*Added many new biomes from Felger (Luna, Mars/Deimos/Phobos, Venus, Jupiter and its moons, Saturn and Titan, Uranus, Pluto.

*Includes DDSLoader from Sarbian, and full DDS support.

*Allows specifying textures via GameDatabase rather than via direct loading (remove GameData/ prefix and remove extension, and make sure the image you are referencing is not under a PluginData folder)

*Update loading to include AppRoot when loading locally (thanks TriggerAu)

*Log on not finding LaunchSites.

*Add support for VertexPlanet and for useHeightMap in LandClass (thanks Starwaster)

*Support more light-shifting (can now modify Sun.AU and Sun.brightnessCurve)

NOTE: To take advantage of the new DDS support, DELETE your old RealSolarSystem folder, then install this, then install textures from the NEW links in the opening post.

Thanks for all the hard work @Nathan! I'm assuming we will see at least some memory usage reduction with the .DDS textures? I'm running at 2.5GB now but it can jump to 3GB every now and then so it would be nice to have some overhead.

[Guest]

I think you need to use Open_GL or DX11 mode for it to matter, but if you use one of those, it will.

[ANWRocketMan]

I'm using OpenGL mode. Thanks.

[Kejchal]

Just found problem under Linux. Custom Biomes not loading RSS biomes at all.

Renaming KERBAL.png , KERBAL.att to Kerbal.png and Kerbal.att solved problem. Same with other planets.

[NathanKell]

...I thought that was fixed. ARGH. Repacking.

EDIT: Repacked. Try again...

Edited November 30, 2014 by NathanKell

[etheoma]

I'm not going to argue with you any more about your use of phys time warp, but when you get odd things happening to you when using it you have nobody to blame but yourself for persisting in using it.

That's number one.

Number two: Stop treating RSS Eve as Eve. What you can or can't do with stock Eve has nothing to do with what you can do here because this is Real Solar System and Eve is VENUS. Average pressure on its surface is 90 times that on Earth. You'd have to go a kilometer underwater to encounter such pressures on Earth. The altitudes you're dealing with in your descent when you come apart is the equivalent of flying on Earth or Stock Kerbin at 45-50 km with FAR except that you're travelling at orbital velocities. I don't know why you expect to do that safely in something as fragile as a space plane but you can't.

Number three: Stock Physics.

http://i.imgur.com/WelQ6Zol.png

Sorry but I'm done here. I have to move on to other things (Deadly Reentry, Ioncross, Stock Drag Fix, in that order).

Ok before you go I have 2 things as well and I'm going to use capitals because you didn't listen the first 3 times

I HAVE TRIED 20~ TIMES WITHOUT TIME ACCLERATION.

I mainly did it with time acceleration to shorten the video without having to do post editing.

Second And to that point if you were trying to say that Frac SL Density scales with the sea level air density you would be right to a point but it is only at MAX 3.6 times the density of the same Frac SL density on kerbin which I left a over 10x safety margins in Frac SL densities so its a moot point.

You see that it shows near 30 Frac SL density which means that it does not scale from 1 at sea level if your right that its 90 times kerbin atmos then it means that at best taking the frac SL density at 10km even is only 3 times out, and that's being kind considering I'm 10km up.

And when did I say that it wasn't only a problem with my instillation, Conflicts etc.

Edited December 1, 2014 by etheoma

[Starwaster]

Density and pressure are not the same thing at all.

[Yaivenov]

Is there a fix for the bouncy water somewhere that I've missed? Seems that every splash down my craft sits there and keeps bobbing rapidly preventing me from recovering it.

[ANWRocketMan]

I haven't noticed that problem for a while now actually. Somehow. Not sure why. My landing speed is quite low(4m/s) and my capsules land at an angle. They don't bounce as they used to. Not sure why not though.

[Fizwalker]

~Snip~

I'm going to jump in here for a moment. The solution to determining whether or not the issue is with RSS or FAR is pretty simple because FAR is not a dependency of RSS. Get your lander/space plane/ship into orbit around Venus (represented by EVE), exit the game, remove FAR and then reload and descend into Venus's atmosphere. If it's an RSS problem, the ship will disassemble. If not, it will continue merrily on its way and RSS isn't at fault. While running with both mods, one cannot be sure of where the problem is.

Only thing I'm going to ask etheoma, is of the numerous times you've attempted this, how many of them did you try without FAR installed? (I see the button for FAR in your screenie which is why I ask. )

On a different note, Nathan, Starwaster and crew. Thank you all very much for all of your hard work! I love the mod!

[Yaivenov]

Just a quick note for others using mechjeb and having trouble with RCS attitude control. Apparently it's something with the built in space capsule RCS thrusters. With them active Smart ASS goes nuts trying to find and hold an orientation, wasting huge quantities of propellant. However when I manually turned off the capsule thrusters and used some surface mounted thrusters everything went back to working normally.

ETA: Also a general question: With the new star system mods being put out is RSS going to switch over to one of those methods instead of altering the Kerbol system (and being short a few planets and moons, Earth being called Kerbin, etc.).

[Starwaster]

I'm going to jump in here for a moment. The solution to determining whether or not the issue is with RSS or FAR is pretty simple because FAR is not a dependency of RSS. Get your lander/space plane/ship into orbit around Venus (represented by EVE), exit the game, remove FAR and then reload and descend into Venus's atmosphere. If it's an RSS problem, the ship will disassemble. If not, it will continue merrily on its way and RSS isn't at fault. While running with both mods, one cannot be sure of where the problem is.

Only thing I'm going to ask etheoma, is of the numerous times you've attempted this, how many of them did you try without FAR installed? (I see the button for FAR in your screenie which is why I ask. )

On a different note, Nathan, Starwaster and crew. Thank you all very much for all of your hard work! I love the mod!

I did that myself actually. Finally got around to trying it both with FAR and without. I posted a screenshot of stock drag only showing my craft at 59.x km and 222 m/s. Using FAR there was a definite increase in disassembly frequency. I'm not necessarily 'blaming' FAR per se, but it is not kind to space planes attempting a Venus reentry. Reentry speeds will be too high for the increased pressure that you find on Venus. And the density changes caused by temperature fluctuations don't help either, but the problem persisted even after removing the temperature curve entirely. Without that, it uses stock temperature changes resulting in constant density increase with pressure rather than the fluctuations. Which btw is implemented by FAR. RSS only provides the data used by FAR to determine pressure and density.

I suspect a space plane with substantially fewer wing parts and wing surface area would fare a lot better, but like I said before, I have to move on to other modding matter and no longer consider it to be an RSS matter at all

Venus is Venus, not Eve and it has atmospheric properties that are different enough from other planets that one must plan carefully for. At least if you're using FAR.

Just a quick note for others using mechjeb and having trouble with RCS attitude control. Apparently it's something with the built in space capsule RCS thrusters. With them active Smart ASS goes nuts trying to find and hold an orientation, wasting huge quantities of propellant. However when I manually turned off the capsule thrusters and used some surface mounted thrusters everything went back to working normally

That's not an RSS matter though. And it's kind of a known issue when incorporating multiple thruster into a single part that ordinarily would be represented by multiple discrete parts. Theoretically it could be gotten around by using multiple sets of uniquely named RCS thruster transforms coupled with multiple ModuleRCS declarations. (one for each set of transforms. Like 'thrusterTransform = starboardTransforms' where the starboard grouping is named 'starboardTransforms'. Theoretically.)

Edited December 2, 2014 by Starwaster

[Guest]

Well, usually (discounting cases like Apollo CSM or LEM, which are welded) thrusters on the capsule are mean to be used mostly for reentry attitude control. Normally, you'd fly with them off and only enable them after decoupling your other control systems.