Squad on the recent Aero changes

[Spuds]

During the Squadcast with the team, Mu commented on the recent aero changes in the patches. I took some notes below:

SHOULD NOT BE TAKEN AS AN OFFICIAL STATEMENT. PLEASE REFER TO THE STREAM FOR MU'S FULL COMMENT.

1.0.1/1.0.2 fixed a bug with occlusion in the 1.0 drag; made some parts have less drag than intended

Scott Manley told them that the pods did not slow down as enough; drag was upped to compensate

As a result, lift and other values were tweaked.

You really shouldn't be able to hit high mach levels at sea levels using jets.

Will be changed in the future, but aero is working as intended

Transonic drag profiles are being tweaked and such.

[Adester]

uh more changes to the jets? Don't get me wrong, things need to be fixed and balanced. i had the same problem with minecraft. one is building something, that takes a lot of time... the good free time... and then an update destroys all your work. i don't know how to avoid that, espacialy balancing needs time. but my heart bleeds, when i have to decide to update or to invest in an outdated version.

btw. a few days ago i saw a video of a jetfighter breaking sound barrier on sea level:

[merendel]

btw. a few days ago i saw a video of a jetfighter breaking sound barrier on sea level:

Its possible to hit mach 1 at sea level as that video points out. going for mach 3+ at sea level on the other hand would be rather difficult and dangerous. It specificly mentioned high mach levels which would generaly imply significantly higher than mach 1. Even mach 1 is not generaly done at low altitudes if for no other reason than folks dont like geting their ears blasted by a point blank sonic boom. Even the few jet aircraft that have been made that can reach multipule mach speeds do it at high altitude, I doubt their airframes could handle it at sea level and it would probably do bad things to the engiens as well.

[mcdjfp]

I don't know about others, but my biggest concern was the very moment the drag went away, it was suddenly increased a couple of days later. With enough time to experiment, it seems acceptable (at least for Kerbin). I am quite happy to hear that they are still working on things though, and I think I will want for several major patches/updates before worrying again.

[Drowning Fish]

Did they mention reentry heating at all? As far as I know heat shields are still basically useless.

[mikegarrison]

Various aerodynamic properties scale differently. I think that is the root cause of Squad's difficulties. By choosing to scale Kerbin to a different size than Earth, they pretty much mandated that if you adjust the atmosphere to correct for the scaling in one regime you will mess it up in another regime. They are forced to pick and choose and try to balance things out as best they can, and they will never be able to make everyone simultaneously happy.

[falloutaddict]

they will never be able to make everyone simultaneously happy.

This 1000%

[Kaezan]

Various aerodynamic properties scale differently. I think that is the root cause of Squad's difficulties. By choosing to scale Kerbin to a different size than Earth, they pretty much mandated that if you adjust the atmosphere to correct for the scaling in one regime you will mess it up in another regime. They are forced to pick and choose and try to balance things out as best they can, and they will never be able to make everyone simultaneously happy.

^^^^^^^^^^ THIS! ^^^^^^^^^^

And doing the best they can is more than enough

[spinomonkey]

Ugh, sounds tedious. I don't envy Squad- good job guys!

[sephirotic]

You really shouldn't be able to hit high mach levels at sea levels using jets..

This doesn´t even make sense, since the most advanced high performance military grade real life jet engines barely hit the 150kn thurst rating WITH AFTERBURNER AT ALLTITUDE. Even in real life it would be damn possible to hit mach 3 at sea level with such a absurdly overpowered jet engine that weights only 2 tons but hits 400kn of thrust. Of course your vessel would desintegrate in a few seconds, so the lack of realism is not the drag levels of the original 1.0 aero, but the absurdly high thrust of the Jet engines with a very forgiving damage model due to aerodynamic and gee-froce stresses.

[Spuds]

Mu was talking about hitting Mach 3-4 at sea level, around 1300 m/s. That is unrealistic. Fighter Jets can hit mach 2 at sea level, I believe, but they usually hit these speeds at high altitudes where there is much less drag. Anything larger than an Aeris-3 A sized jet shouldn't be blasting off of the end of the runway at breakneck speeds like they used to.

[NathanKell]

Where there is much less heat. Heat is the limiting factor, not drag. Jet thrust is broadly proportional to dynamic pressure, so outside of heat (and excess intake drag) there isn't much difference between low and high in terms of max speeds. You need to make sure your airframe doesn't melt and your compressor doesn't melt, which means sprints speeds far higher than sustained speeds, and tighter thermal limits below 50,000ft.

[allmhuran]

Where there is much less heat. Heat is the limiting factor, not drag.

Am I right in thinking you're talking about heat inside the engine here, not the heating of the skin of the aircraft? If so, then that seems to translate better to "thrust is the limiting factor", which is equivalent to saying that drag is the limiting factor.

Or do you mean that the only reason why, say, a Lear Jet can't run at mach 3 is because the skin of the aircraft would overheat?

[Browning]

Fighter Jets can hit mach 2 at sea level, I believe

Not even close. The fasted fighter jets don't get much over Mach 1.2 at sea level.

It's been that way since the late 1950s and adding more powerful jet engines does not help very much. The air down low is just too thick to be able to run jet engines at high speed without having problems with compression heating and needing super-rich fuel air mixes.

If you want to go fast at sea level you need rockets, not jets.

[ferram4]

Thrust is the limiting factor... because the engine needs to throttle down to prevent itself from melting. If you could keep the internals cool, there's no compelling reason why you couldn't sustain Mach 3 at SL. Now, I think it would be great if KSP engines modeled that, because then what you'd get at SL is that you can continue pushing the engine far past what you should do, eventually leading to the turbine failing, possibly in a spectacular uncontrolled fashion that destroys the entire region of the aircraft around the engine. It's really not the thrust that's an issue, it's creating the thrust without destroying the engine that is.

Also, yes, Learjets can't handle Mach 3 partly because of the skin temperature. 2.8x the ambient absolute temperature will be enough to cause most aluminum and steel alloys to being losing strength, and then you're in deep trouble.

[MalfunctionM1Ke]

The dragmodel is fine for me, but please Squad... fix the asymmetric flameouts.

[sephirotic]

Thrust is the limiting factor... because the engine needs to throttle down to prevent itself from melting. If you could keep the internals cool, there's no compelling reason why you couldn't sustain Mach 3 at SL. Now, I think it would be great if KSP engines modeled that, because then what you'd get at SL is that you can continue pushing the engine far past what you should do, eventually leading to the turbine failing, possibly in a spectacular uncontrolled fashion that destroys the entire region of the aircraft around the engine. It's really not the thrust that's an issue, it's creating the thrust without destroying the engine that is.

Also, yes, Learjets can't handle Mach 3 partly because of the skin temperature. 2.8x the ambient absolute temperature will be enough to cause most aluminum and steel alloys to being losing strength, and then you're in deep trouble.

Am I right in thinking you're talking about heat inside the engine here, not the heating of the skin of the aircraft? If so, then that seems to translate better to "thrust is the limiting factor", which is equivalent to saying that drag is the limiting factor.

Or do you mean that the only reason why, say, a Lear Jet can't run at mach 3 is because the skin of the aircraft would overheat?

Which is why it's obvious that the Soapsphere is back. You can test that yourself:

A real life F-15 can fly at mach speed at Sea level with 1.12 TWR and around 200kn of Maximum thrust, probably lower due to overheat issues.

The Stock Ravenspeark Mk1 With the new 1.0.2 drag model, two Turbojet engines gets over 900KN of thrust, 5.52TWR and yet it now tops around the same speed of the F15. The drag increase is clearly exagerated. Not to mention I can now land heavy planes as low as 30m/s (that's less than 120km/h) by just planing without stalling. The problem of the Aero back on 1.0 wasn´t the aero, but was, and still is, the overpowered JET engines.

Edited May 8, 2015 by sephirotic

[Endersmens]

Which is why it's obvious that the Soapsphere is back. You can test that yourself:

A real life F-15 can fly at mach speed at Sea level with 1.12 TWR and around 200kn of Maximum thrust, probably lower due to overheat issues.

The Stock Ravenspeark Mk1 With the new 1.0.2 drag model, two Turbojet engines gets over 900KN of thrust, 5.52TWR and yet it now tops around the same speed of the F15. The drag increase is clearly exagerated. Not to mention I can now land heavy planes as low as 30m/s (that's less than 150km/h) by just planing without stalling. The problem of the Aero back on 1.0 wasn´t the aero, but was, and still is, the overpowered JET engines.

30 m/s is closer to 120 km/h I believe. (Doing mental math here)

Now, for a Prop driven ultralight Cub, this is reasonable. (Some can land at near stand still)

But for anything that's using a jet, landing speed should be MUCH higher.

[sephirotic]

30 m/s is closer to 120 km/h I believe. (Doing mental math here)

Now, for a Prop driven ultralight Cub, this is reasonable. (Some can land at near stand still)

But for anything that's using a jet, landing speed should be MUCH higher.

Exactly, the drag is too high, and the lift is also too high. 108km/h to be precise enough for a glide landing with a supposely massive space plane is ridiculous low, I could land a massive Spaceplane wet of fuel gliding on duna which was supposed to have a thin atmosphere. It's just insane. How could they have approve such tweaks from 1.0 baffles me.

[NathanKell]

I don't disagree in the slightest that drag is too high for planes. I do disagree that it's too high for capsules, and I challenge you to compare capsule terminal velocities in KSP 1.0.2 stock, and KSP 0.90 with FAR. At sea level, since the atmosphere up high changed.

[White Owl]

For airplanes, the drag may be slightly too high, but the lift is what's causing the undesired behavior: landing speeds so slow it looks like a blimp. If you set the physics.cfg lift multipliers to 1.0 values while keeping the 1.02 drag values, then airplanes handle much closer to what I'd expect. I have not done one shred of math to figure out what the lift and drag values should be, but simple experimentation and playtesting will bear this out. It's easy to try yourself.

[sephirotic]

I don't disagree in the slightest that drag is too high for planes. I do disagree that it's too high for capsules, and I challenge you to compare capsule terminal velocities in KSP 1.0.2 stock, and KSP 0.90 with FAR. At sea level, since the atmosphere up high changed.

Are you saying that is impossible to make a drag curve and drag characteristic for pods in the current model that satisfies both planes and rockets and that pods gets the priority?

[basic.syntax]

I just got through rewinding a bunch of times over the first 20 minutes of the Twitch stream archive, to squeeze as many words as possible out of the Aero comments, and I think I have a pretty good transcript now, of that early section. I may be posting a more complete summary later, if my OCD and patience holds

11:40

HavesteR lands his airplane nicely in the green hills near KSC, earning this nice one-liner from Maxmaps:

I fly spaceplanes in the same way a protester throws bricks.

Mike (Mu) - "We... fixed a bug in the last week of Experimentals, which... fixed the occlusion on a lot of parts. Thus, the drag dropped significantly. This wasn't quite balanced out by the time we went to 1.0, and a lot of the streamers... Scott Manley for instance... he basically says 'the pods are not slowing down enough. It's very hard to do sub-orbital flights... you're having to run [word unclear] parachutes at high altitudes...' The drag was far too low. So, we upped the drag basically, to what it should have been... before the bug fix. And that also meant tweaking lift values, tweaking the parachutes again.

We've not quite finished with the Aero balances... obviously quite a few people have been critical of it, BUT, it is working as intended at the minute. We do have some other changes, just to the drag profiles... the transonic drag profiles, and some other bits coming. But.. that won't be this week."

Maxmaps - "Yeah, it certainly will not be this week.... we're going skydiving on Saturday, that generally isn't... the... best environment to code..."

Mu - "It's research!"

Maxmaps - "Yeah It's research, of course!"

Kasper - "It was important to get it out there, why we did the Aerodynamic changes. A lot of people were very..."

Maxmaps - "confused"

Kasper - "confused about them. It's just people thought they were bugged in the patch version, when actually they were bugged in 1.0"

Mu - "exactly. There's no way you should be able to do high mach at sea level..."

Maxmaps - "...without annihilating yourself..."

Mu - "...without some sort of, like, warp drive. Thing. because the power levels are quite insane."

[allmhuran]

Thrust is the limiting factor... because the engine needs to throttle down to prevent itself from melting. If you could keep the internals cool, there's no compelling reason why you couldn't sustain Mach 3 at SL. Now, I think it would be great if KSP engines modeled that, because then what you'd get at SL is that you can continue pushing the engine far past what you should do, eventually leading to the turbine failing, possibly in a spectacular uncontrolled fashion that destroys the entire region of the aircraft around the engine. It's really not the thrust that's an issue, it's creating the thrust without destroying the engine that is.

Also, yes, Learjets can't handle Mach 3 partly because of the skin temperature. 2.8x the ambient absolute temperature will be enough to cause most aluminum and steel alloys to being losing strength, and then you're in deep trouble.

Yep, this is exactly where I was going as well.

As far as I know (flight sim experience only) aircraft tend to have a fuel flow limiter, but some aircraft (military only?) let you override the fuel flow to push it beyond the "safe" limit. In KSP we need not include that additional control as an extra complication... we already have the engine overheating mechanic in place. So running engines at full throttle for extended periods should cause the engines themselves to overheat much more quickly than it currently does. This gives the precooler a more significant purpose as well. Something like: with no precooler, start to overheat pretty much immediately at 100% throttle, engine explodes within 2 minutes. With a precooler, start to overheat after 2-3 minutes, engine explodes after about 10. Tweak the balance as required since I just made these up arbitrarily.

Throw in "afterburner" characteristics (massive thrust and fuel flow increase above ~90% throttle) for the turboramjet, put it all together and I think we start to get a more believable model even before we start to touch the atmosphere itself.

While every aircraft is going to be theoretically limited by heating of the skin given ideal engines, it shouldn't be the primary factor for the vast majority of them.

[CodedThreat]

Thanks basic.syntax for typing that out. And what Kasper said is the truth, 1.0 levels were the bug, everything was too 'slippery' for lack of a better word.