.22 SAS is now all WIbbely Wobbly (but not timey wimey)

[thescientist]

well...I play in hardmode and I've never had problem with the SAS. I think that the old one used to keep the attitude with brute force but this one is a bit softer, so unstabile/overloaded/extreme crafts have problem because they NEED to have a quick and brute responce to attitude changes.

[Mykill Metal]

I've noticed that if you tap F on the point you want it to stay, it will.

[Mykill Metal]

If your S.A.S. is stronger than your torque from thrust.

[lajoswinkler]

Could this simply be a bias, because experienced players are now forced to use primitive Kerbal rockets? I don't see that much problems. There was a period early in the career mode when I had to use winglets for my rockets and need to rely on the position of center of mass and thrust (just like amateur rocketry) but as soon as you earn IRW, you're fine and you can do lots of advanced designs without problems.

[magnemoe]

well...I play in hardmode and I've never had problem with the SAS. I think that the old one used to keep the attitude with brute force but this one is a bit softer, so unstabile/overloaded/extreme crafts have problem because they NEED to have a quick and brute responce to attitude changes.

The old asas locked an small ship into position just as fast as timewarp. The 0.211 sas was unable to hold an heading unless perfectly symmetric. 100 kg on the side of an 3 ton 1.25 meter probe would cause it to drift while the old one worked much like mechjeb assisted burn except that it did not adjust against nodes but kept the heading, if ship drained fuel unsymmetrical it would keep heading until it became to unsymetric for the gimbal to handle

[ComradeGoat]

Could this simply be a bias, because experienced players are now forced to use primitive Kerbal rockets?

No. This is sandbox mode. Vessels that behaved perfectly well under SAS under 0.21.1 are hunting under 0.22.

It works very nicely for planes, but for heavy lift rockets it is frustrating!

It's not as bad as the 0.21.0 SAS, which was a show-stopper, but having to fight it all the time is not enjoyable.

[ilikemoneygreen]

It's working just fine. It's worked just fine, and works even better now.

OMG! Damion! How you been? It's been such long time.

You should come by more often

[shadowsutekh]

Don't turn it into a chat thread please. It's a discussion on .22 SAS, and remember to stay civil please.

Personally I find that the SAS works far better for everything planes or rockets.

[Psycix]

I think it is less suitable for having it mixed in with a series of small course corrections (like a gravity turn).

It doesn't ramp up as aggressively as it did, and it veers off before settling in. This makes it useless for "hard-locking" a heading for a small amount of time.

[regex]

It doesn't ramp up as aggressively as it did, and it veers off before settling in. This makes it useless for "hard-locking" a heading for a small amount of time.

Control finesse and smart use of the F key address that problem. Manually adjust heading smoothly and slowly, hit F when in the right spot, preferably at the end of a yaw or pitch maneuver. In other words, SAS is an assistant during launch, not an automatic launcher.

[Specialist290]

Control finesse and smart use of the F key address that problem. Manually adjust heading smoothly and slowly, hit F when in the right spot, preferably at the end of a yaw or pitch maneuver. In other words, SAS is an assistant during launch, not an automatic launcher.

Indeed. That's the technique I've been using since 0.21 changed it in the first place; it's just the new way it works has thrown my timing completely off.

Like I've mentioned before, annoying, but I'm sure I'll adjust to it eventually.

[Psycix]

Control finesse and smart use of the F key address that problem.

Actually, my point is that smart use of the F key is impossible/broken, because every time you tap it the SAS allows it to deviate way too much before holding the heading.

During F-key maneuvering, I strongly prefer the aggressive SAS of 20 and before.

it's just the new way it works has thrown my timing completely off.

That's my problem as well. You need to re-learn the entire way SAS responses.

I hope I can adjust to it, but so far I think SAS takes too long to steer more aggressively. Sure, would re-introduce wobble, but wobble can be mitigated with F.

I'm still hoping for an adjustable SAS system.

Edited October 19, 2013 by Psycix

[Aphobius]

So many threads about SAS with people just saying that it's not working and give no solution.

I've made a thread too, but no one wanted to discuss there.

The best solution would be if they would program SAS like a proper PID controller (based on thee angular velocity) and give us access to the values of the gains. That would make everyone happy and it could have some default values for the noobs.

But since everyone is making threads writing random things, crying that it doesn't work how they want and giving no solution, I guess the developers are not even reading these threads anymore.

[regex]

Actually, my point is that smart use of the F key is impossible/broken, because every time you tap it the SAS allows it to deviate way too much before holding the heading.

I haven't experienced this, or I've already adjusted to it. I hit F at the end of yaw/pitch motion, when it is close to or on the heading I desire. It's not really any different than 0.21 aside from the fact that it's actually holding a heading.

In truth, I'd much prefer a selector for SAS functionality where I could switch between 0.20 and 0.21/0.22 so I could have the best of all worlds. Sadly, I don't think that will ever happen.

[Exposure]

Man I wished 0.21 SAS worked like 0.22 SAS. At least 0.22 can actually hold a heading, 0.21 would just keep being indecisive over whether it actually wanted to hold a heading or not right up to the point spontaneous disassembly caused by its waffling occured. :/

This seems like an issue that might be best solved by just letting people customize the SAS since it seems what is a update that breaks things for one person results in SAS working again for a different person.

[boomerdog2000]

I sort of use the new SAS as the capslock key, having it damp the rockets movement enough then cycle it using t once I reach the new heading.

[Specialist290]

In truth, I'd much prefer a selector for SAS functionality where I could switch between 0.20 and 0.21/0.22 so I could have the best of all worlds.

This seems like an issue that might be best solved by just letting people customize the SAS since it seems what is a update that breaks things for one person results in SAS working again for a different person.

I'd be fully in favor of such a thing, myself.

[lincourtl]

I sort of use the new SAS as the capslock key, having it damp the rockets movement enough then cycle it using t once I reach the new heading.

That's the flight style I've adopted. I do agree with others that tapping 'f' doesn't seem to work for this purpose like it did before.

[DMagic]

I updated my SAS tutorial with new info from 0.22. http://forum.kerbalspaceprogram.com/threads/41941-New-SAS-functionality-and-You%21-0-22-Update

I think there are a few things that should be mentioned.

That "steady drift" that happened in 0.21 with unbalanced loads (where your heading would drift a few degrees upon throttling up before locking in place) seems to have been greatly improved now. Fairly unbalanced loads will stay on course and use the available control authority to keep it that way.

Whenever you enter damping mode (the two spinning arrows icon) all controls are brought to zero for a moment. This means that if SAS is using the pitch and yaw controls to maintain heading and you try to roll, the system will zero out the pitch and yaw axes. Control resumes after a second, but that brief interval when the other control axes drop to zero can be enough to throw you off course (and sometimes manual control will be needed to bring you out of damping mode).

The two spinning arrows indicates damping mode, not just manual override, which is slightly different that what is implied in the release notes. If you start spinning out of control the system will get stuck in damping mode and the icon will remain on the two spinning arrows even if you don't provide any manual input.

Control finesse and smart use of the F key address that problem. Manually adjust heading smoothly and slowly, hit F when in the right spot, preferably at the end of a yaw or pitch maneuver. In other words, SAS is an assistant during launch, not an automatic launcher.

This is really good advice. With careful manual control and use of the F key you can prevent most of those oscillations. This can get really tricky with an unbalanced or very unstable rocket, but those are design problems, not SAS problems.

[boomerdog2000]

Whenever you enter damping mode (the two spinning arrows icon) all controls are brought to zero for a moment. This means that if SAS is using the pitch and yaw controls to maintain heading and you try to roll, the system will zero out the pitch and yaw axes.

Are you sure about that? I've been doing some launches into equatorial orbit and only messing with the yaw but my pitch always ends up slightly off, putting me in up to a 2 degree inclined orbit.

[earth]

that is because of the inertial matrix of the ship. Yaw can cause sideways velicity without even making pitch drift. If you know the math, install FAR, because the mod can display that info.

[DMagic]

Are you sure about that? I've been doing some launches into equatorial orbit and only messing with the yaw but my pitch always ends up slightly off, putting me in up to a 2 degree inclined orbit.

That's kind of my point. Even though you only apply manual input on the yaw axis, the pitch and roll axes can still drift. It isn't as if the system dampens the yaw axis and locks the pitch and roll axes, it seems to dampen all three after briefly dropping all non-manual axes to zero control. I might be wrong on this, it might still try to lock on the other axes, but it isn't as effective as it is when the whole system is in locking mode. And yeah, like earth said, control on any axis can cause changes in the other axes, regardless of what SAS is doing.

[RayneCloud]

Personally, this really feels like complaining for the sake of complaining. People have complained about SAS for the last several updates, and C7 has dumped many dozens of hours into the system in an attempt to make it amazing. In my opinion, he's done just that. It's a hell of a lot smarter than it ever was, and a lot more capable. I'm not just saying this because I know C7, he's a good friend, that's true, but he's also extremely capable and has done a great deal of good with the stability assistance system. It's the best it's ever been.

[DMagic]

Personally, this really feels like complaining for the sake of complaining. People have complained about SAS for the last several updates, and C7 has dumped many dozens of hours into the system in an attempt to make it amazing. In my opinion, he's done just that. It's a hell of a lot smarter than it ever was, and a lot more capable. I'm not just saying this because I know C7, he's a good friend, that's true, but he's also extremely capable and has done a great deal of good with the stability assistance system. It's the best it's ever been.

Oh I fully agree. Before 0.21 I almost never used the stock ASAS system; I got to the point where it was difficult to launch manually because I relied on MJ so much.

The new SAS system has been a huge improvement. The changes made for how RCS is more efficiently and effectively used in docking are amazing; that alone will probably save new players an enormous amount of pain in learning how to dock. And while the whole system isn't perfect (as I've discussed in my tutorial), the updates made for 0.22 improve it even more.

I think many of the problems arise because people want to continue using their old designs (which is not to say that there aren't legitimate complaints to be made). The problem with this is that those old designs were built around the specific limits and capabilities of the old SAS/ASAS system. Everything about KSP is a moving target, so I expect to have to change my designs whenever updates are released.

[-RanZ-]

not sure what was wrong with the last asas.. seems to suck now, won't keep a heading like it did in .21