[1.3.1] Ferram Aerospace Research: v0.15.9.1 "Liepmann" 4/2/18

[andrehsu]

I do love this mod but I have to say I think it makes launches too easy, for example most first stages can make it to orbit and there is barely any use for upper stages. This is not like this in real life because in real life it takes a lot more delta v to reach orbit yet in ksp it takes less. This wouldn't normally be a problem with the stock ksp as the atmosphere is much thicker and compensates for this, however FAR makes it more realistic so therefore I think it makes it easier, maybe too easy, for example I never have any use for solid rocket boosters to increase the payload capacity.

Is there any way to increase the difficulty or would I just need to decrease the efficiency of all engines in the game?

You can use this mod: http://forum.kerbalspaceprogram.com/threads/52882-0-24-Kerbal-Isp-Difficulty-Scaler-v1-3-4-1-7-17-14

This mod has a preset that make it so that the Delta V needed to reach LKO in FAR is equivalent to the Delta V to reach orbit in stock kerbin

[marach]

I do love this mod but I have to say I think it makes launches too easy, for example most first stages can make it to orbit and there is barely any use for upper stages. This is not like this in real life because in real life it takes a lot more delta v to reach orbit yet in ksp it takes less. This wouldn't normally be a problem with the stock ksp as the atmosphere is much thicker and compensates for this, however FAR makes it more realistic so therefore I think it makes it easier, maybe too easy, for example I never have any use for solid rocket boosters to increase the payload capacity.

Is there any way to increase the difficulty or would I just need to decrease the efficiency of all engines in the game?

KIDS

/10char

[Jasmir]

Nerfing the engines makes no sense. It is so easy, because the needed DV is so low. And this is caused by the 1/10th scale of the whole solar-system.

So, if you want more challenge, the scale is the thing you should adjust. (RSS)

[MainSailor]

If it is related to the previous craft file issues, go into the craft file and search for "FARBasicDragModel." Then go and copy the entire block the module is defined in, including the curly brackets {}. Then Go and use the replace tool to replace that block with nothing in the entire file; the issues should be removed.

With 14.0.2 should this still be causing issues? I built a new plane (not an imported, old craft file) and fly it similar to the craft I flew in .23.5 and it doesn't have near the expected performance. If I go to the .craft file and remove the FARBasicDragModel lines and reload the .craft file, it flies as expected. Additionally, Mechjeb reports a drag coefficient of .1 to .14 when the bug occurs, and my drag losses have approached 1500-2300 m/s. When I remove the references, Mechjeb (I assume correctly) shows a drag coefficient of .000 and drag losses are under 200 or so.

If I save and launch the craft again the issue appears, and (obviously) the FARBasicDragModel references are added back into the craft file. It's almost as if for some reason the game 'forgets' that FAR is active and I'm flying through soup again, even though the GUI and stats tools are working correctly and I can right click on the parts and see the lift and drag stats.

I can upload a output_log if you'd like, some time this morning. I looked through it earlier and didn't see any NREs or unusual events, but I'm not sure what I'm looking for.

EDIT: I was using NEAR, but I removed it because I realized I did like the analytical functions of FAR. As well, NEAR I think tended to ignore my carefully crafted wing shapes when applying CoL (as advertised, I know.) I completely uninstalled NEAR and built new crafts to test, so there's no conflict between the two mods.

Second Edit: Output log and some pics.

Output log: https://dl.dropboxusercontent.com/u/147659278/output_log.txt

Pics:

First build, in flight: (note Mechjeb drag coefficient - 0.119) I had drag losses over 900 by the time I reached an apo of 85 km.

Went out, removed FARBasicDrag from craft file, reloaded .craft save in SPH.

Second flight, no other changes in SPH:

Mechjeb now reports .000 for drag coefficient, and FAR stats show lift and drag applied to parts. I think I had less than 200 m/s drag losses, but I just did a max throttle to get to orbit quickly.

Edited July 23, 2014 by MainSailor
Added log and photos.

[andrehsu]

If I am not mistakened, the new petal shroud in the latest kw rocketry release isn't shieding the payload corretly. When you right click on the item inside the the fairing, the isShielded = false, and when in flight, things fall off the lander that is inside the fairing

[dshriver]

I'm having an issue with both NEAR and FAR. when ever a piece of debris is put on rails I see a massive slowdown of the game. It only happens during lunch, and only with the first piece of debris. I see a debug message that says something like debris of ship name at 0.6 atm was placed on rails. Then the game slows down dramatically. This continues for about a minute and then the performance returns to normal. I tested this with a vanilla install of .24 32bit. I tried it with both near and with far and see the exact same issue. I don't see this problem unless I have near or far installed.

[KerbMav]

Try playing during supper?

Only for the first part of debris as in it does not get any slower for the second and third?

What happens when the performance gets back to normal, is the debris deleted (2.5km range in atmosphere)?

If so, can you ascend slowly, stage the first time and try to continue flying in the atmosphere until the spent stage is deleted/performance returns to normal - then decouple something else and check if performance goes down again?

After that, can you stage something just attached to your craft to be jettisoned before launch while still sitting on the launchpad - to get your first part of debris - and check performance then and when you finally launch and stage normally?

[ferram4]

@MainSailor: It is possible to cause the issue if you are using a lot of other mods that change PartModules or if you constantly add new mods that change the behavior of existing parts, changing the number of PartModules. This is caused by a stock bug in the PartModule loading code that was added in 0.24 and it is out of my control completely. Basically all the hotfixes have been code for me to try and work-around the bug and remove as many situations that could cause it, but I can't do anything. Honestly, 0.24 dropped a week too early and really should have stayed in experimentals longer to fix this and other issues, but the hype was too much. We'll have to wait for 0.24.1 to hopefully fix it.

@andrehsu: Aware of it, that's because it not only doesn't include any of the keywords in its title to load either of the shielding modules, it'll need the cargo bay one since it doesn't act as a fairing.

@dshriver: Where is your output_log.txt? Post it, or I can trace that back to a bug. Anyway, NEAR and FAR doesn't do anything to crafts once they're off-rails, so the lag is likely due to something else entirely.

[MainSailor]

@MainSailor: It is possible to cause the issue if you are using a lot of other mods that change PartModules or if you constantly add new mods that change the behavior of existing parts, changing the number of PartModules. This is caused by a stock bug in the PartModule loading code that was added in 0.24 and it is out of my control completely. Basically all the hotfixes have been code for me to try and work-around the bug and remove as many situations that could cause it, but I can't do anything. Honestly, 0.24 dropped a week too early and really should have stayed in experimentals longer to fix this and other issues, but the hype was too much. We'll have to wait for 0.24.1 to hopefully fix it.

I completely understand it's behavior in .24...please don't think I was blaming you at all. I've also had problems with KW under .24 as well.

I'm trying now to reduce the number of MM scripts I have installed or reduce extraneous parts. Off hand, do you know if there's a way to force a particular script to run last? I know there's a way to append FINAL in the MM script but I'm not sure how it works.

I appreciate your assistance, I know I've been a pain both here and in the NEAR thread.

[ferram4]

Let's say you have a bloc like this:

@PART[decoupler_ftr_4m]
{
    CoMOffset = 0, 0, 1.12183
}

If you add :FINAL to the end, it'll run in the last group, like so:

@PART[decoupler_ftr_4m]:FINAL
{
    CoMOffset = 0, 0, 1.12183
}

Things in the FINAL group will still be done alphabetically inside that group, but they'll end up being the last set applied.

[TeiwazVIE]

So I'm having a "funny" issue with FAR.

It seems that I can't get my aircraft to land - it's just not losing velocity. Not even with a cut engine. Even better: doing an almost 90° cork screw seems to basically accelerate it infinitely (if it wouldn't lose structural integrity), which results also in gaining height.

I really don't think that I invented the "super plane" with just stock tiles...

That's the plane:

As I said, it's all basic stock parts.

Any tips? Flareing doesn't help btw.

[NathanKell]

You have very little wing area and very little wetted area on the fuselage. You probably have a quite high wingloading, maybe 1.5-2x what the space shuttle does, and the space shuttle lands at 200 knots (~110m/s). You have the weight of an F-15 and the wing area of an F-104, basically.

Add flaps and add spoilers, increase your wing area, and lengthen your fuselage even if you don't put fuel in the extra bits.

[mdgates]

I absolutely love the mod, and have had many hours of good fun. But for my current playthrough, I made the difficult decision to disable FAR. This playthrough is 100% horizontal takeoff and landing. With Ferram, the engineering burden of developing planes that would actually be flyable is just too great. More so when you consider that I'll be dropping tanks and engines on the way up.

Great mod, but too realistic for a spaceplane playthrough. :-P

[ninjaweasel]

What should Terminal velocity be at KSC with FAR loaded?

I had an issue before that I somehow solved, but on my new install it's appeared again. Default Terminal velocity at sea level is about 90 m/s. First time I installed FAR, Terminal velocity on the ground was over 500 m/s. I was getting some weird flight characteristics, so I re-installed and it was around 400m/s.

I built a Mun lander before installing FAR that could comfortably get to the MUN. After installing FAR, I could get it to the edge of the Kerbin atmosphere before the boosters even burnt out! I built a MechJeb window to monitor losses and I got about 9m/s of drag loss on a 220T, 3.75m rocket with 2 x 2.5m boosters over its entire ascent! (i.e. established in a proper orbit on a little over 3K Delta v)

While the idea of going to the Mun on a single tank of gas is amusing, I don't think this is working as intended! I haven't tried this time but last time it was behaving like this, I was able to get an air-breathing craft on an escape trajectory following a couple of sub-orbital hops. =) How could I have possibly screwed up the install to cause this?

[NathanKell]

See, I don't understand that at all. I find it *so* much harder to make planes without FAR than with FAR. What engineering burden do you mean?

[peter.alfredsen]

What should Terminal velocity be at KSC with FAR loaded?

I built a Mun lander before installing FAR that could comfortably get to the MUN. After installing FAR, I could get it to the edge of the Kerbin atmosphere before the boosters even burnt out! I built a MechJeb window to monitor losses and I got about 9m/s of drag loss on a 220T, 3.75m rocket with 2 x 2.5m boosters over its entire ascent! (i.e. established in a proper orbit on a little over 3K Delta v)

That sounds about right for a craft that has a good-ish aerodynamic profile. I'm sure you could do better though. IIRC, I'm able to get a proper orbit at 70km for quite a bit less than 3000 delta-v using a wing-shaped rocket with high TWR, fins galore and optimal ascent profile. It does take some doing though, I usually just go two-booster and a main with fairings and slip some cones on the boosters.

[Renegrade]

You have very little wing area and very little wetted area on the fuselage. You probably have a quite high wingloading, maybe 1.5-2x what the space shuttle does, and the space shuttle lands at 200 knots (~110m/s). You have the weight of an F-15 and the wing area of an F-104, basically.

Add flaps and add spoilers, increase your wing area, and lengthen your fuselage even if you don't put fuel in the extra bits.

He seemed to be complaining that the plane had no drag, not that it was crashing or something.

Which ties in with an issue I've been chasing - people noticed in the NEAR thread that planes were going crazy fast as you may recall - part of the issue is believed to be lack of trans/supersonic drag.

However, I switched back to FAR (I actually like the analysis module, I just think a few minor documentation/labelling changes might make it more non-aerospace engineer friendly), and I discovered that the planes there were almost as fast.

One of the things I've done is a PURE install - just FAR, and KSP 0.24/x86. Even there, a single engine turbojet craft screams along at mach 2.5.... At 300m altitude. (Scary business at 400 kilopascals of dynamic pressure).

That's just KSP and FAR, nothing else, no modifications, tweakings, or setting changes aside from like resolution and a change from Kerbal days to Earth days. Also here's the Craft File (if desired), but it's very basic design.

Formerly this sort of craft only goes about Mach 1.4-ish in previous versions of FAR at that altitude.

[Kniben]

The drag when using FAR is indeed extremely low compared to fully stock or previous releases of FAR. It would be awesome if the amount of drag would be balanced around stock KSP drag! As it stands now, it breaks gameplay by making it too easy to leave the atmosphere (atleast the one on Kerbin). Maybe the reason it is like this is that it is based on reality, but it makes little sense to use real world values for a scaled down world!

[LostOblivion]

@MainSailor: It is possible to cause the issue if you are using a lot of other mods that change PartModules or if you constantly add new mods that change the behavior of existing parts, changing the number of PartModules. This is caused by a stock bug in the PartModule loading code that was added in 0.24 and it is out of my control completely. Basically all the hotfixes have been code for me to try and work-around the bug and remove as many situations that could cause it, but I can't do anything. Honestly, 0.24 dropped a week too early and really should have stayed in experimentals longer to fix this and other issues, but the hype was too much. We'll have to wait for 0.24.1 to hopefully fix it.

Thanks for clarifying! Been looking for an educated statement on this gamebreaker for days!

The only question I have now is, am I safe moving my career onto 0.24 now (not being able to use old craft), or should I wait for 0.24.1 in case this bug corrupts my in-flight crafts permanently?

[mdgates]

See, I don't understand that at all. I find it *so* much harder to make planes without FAR than with FAR. What engineering burden do you mean?

I mean, with stock KSP's unrealistic physics and soupy atmosphere, I just strap together a plane with five airfoils and five control surfaces, then shift the wings until the center of lift is just ahead of the center of mass. This is true for just about any payload up to some large weight limit, and the craft is not very sensitive to payload aerodynamics.

With FAR, it's half an hour of fussing around to get the stability & control derivatives in order, then half an hour of tweak-crash-repeat before getting to orbit. Then I decide for the next mission I need to lift a Science Lab, and it's time to start over.

I find FAR planes more fun, more rewarding, but WAY too frustrating for me to develop the 40 of them needed for a KSP planes-only playthrough.

[ferram4]

@ninjaweasel: Those numbers are correct. In fact, I've seen higher terminal velocities for heavier launch vehicles. I think you're just expecting the atmosphere to be made out of pudding rather than air like it really is. Perhaps you need fewer boosters?

@Renegrade: Yeah, supersonic drag of wings got changed to account for 3D effects and sweep better, which means that drag is lower. It's expected, I simply haven't nerfed the engines to hell like I should have. Next update will have thrust nerfed a bit to balance it.

@LostOblivion: I think (don't quote me on that) that 0.24.1's changes will probably save the crafts, but I'm not sure, having (obviously) not had the opportunity to test it myself. I'd say upgrade, but make a backup of your save and crafts just in case.

@mdgates: I see you take this far more seriously than I do. I only use those tools when I'm flying a decent test rig (built only using the CoM and CoL indicators) and find it lacking in some way, which, while it involves more crashes and hair-raising landings, is certainly faster and more entertaining than judging by the numbers all day.

[Master Tao]

@mdgates

I'm with NathanKell on this one, but I definitely see where you're coming from. Stock aerodynamics is very forgiving of aircraft shape and that's really helpful until you develop a feel for how real airplanes are shaped. I probably enjoy FAR more because I grew up playing realistic flight simulators – MS Flight Simulator, FlightGear, and X-Plane.

One thing really made a difference for me was the Wikipedia page on wing configurations. It has a lot more discussion of how the shape and control surfaces affect flight than even this tutorial. After skimming it, I spent a few days playing around with different wing configurations to see how they performed as space planes. That led to a lot of space planes that I found much easier to handle.

[Renegrade]

@Renegrade: Yeah, supersonic drag of wings got changed to account for 3D effects and sweep better, which means that drag is lower. It's expected, I simply haven't nerfed the engines to hell like I should have. Next update will have thrust nerfed a bit to balance it.

Ah, I see. Thanks for the response, I was getting a bit worried there

Oh um one other thing I noted there, the swept-wing version of that plane is actually faster than the delta wing (they're otherwise identical).. I thought deltas were supposed to have better drag characteristics at supersonic speeds..? (although, visibly speaking, the delta looks like it has a lot larger surface area..)

[NathanKell]

He seemed to be complaining that the plane had no drag, not that it was crashing or something.

Indeed s/he was. That's what I was responding to: aircraft with very high wingloading won't slow down easily at all: think dart rather than paper airplane. The way you make a dart slow down is add more fins. Etc.

Re: your response to ferram, the swept wing version has lower induced drag and a smaller wetted area. In real life those wings would break, probably, which is more or less why we don't use high AR wings supersonic IIRC.

[Renegrade]

Indeed s/he was. That's what I was responding to: aircraft with very high wingloading won't slow down easily at all: think dart rather than paper airplane. The way you make a dart slow down is add more fins. Etc.

Ah, good point. That reminds me..it seems rather hard to make a decent low wingloading plane with stock parts, especially in career mode for FAR. Any tips there? In stock, you can stick wings on willy-nilly until your plane looks like a hedgehog with wings instead of quills, and a stall speed of 12 m/sec...

Re: your response to ferram, the swept wing version has lower induced drag and a smaller wetted area. In real life those wings would break, probably, which is more or less why we don't use high AR wings supersonic IIRC.

The control surfaces certainly broke a couple of times... had to fly it carefully with trim instead of keys

Maybe I should get a proper joystick...

The revenge of the '104!!

...eventually hit 71.4km apoapsis.