Jump to content

Come back old ASAS - all is forgiven!


ComradeGoat

Recommended Posts

My observation while testing this thing was that now you actually need control fins to compensate for missing torque. I see the zepto rockets above have no fins. You used to not need them, now you need them. There's a reason for control fins in real life.

All old designs will need a bit of tweaking. The old ASAS and SAS modules now weigh a lot more, and are completely unnecessary on smaller craft now. Control surfaces (fins especially) are now necessary to compensate for changes in the torque system. Put some fins on your rockets.

I thought that it was the fins as well. so I slapped a bunch on my rocket and it still controls terribly. I think it's safe to assume its a bug with how varied everyone's experiences are.

Link to comment
Share on other sites

Just to clarify I'm talking about my rockets... not planes. thrashing back and forth as the stupid new ASAS leans one way... gimbals flip full the opposite way... overshoots and flips back. I have a stack of 8 reaction wheels now on Munbug. Still unflyable.... its the 7 thousand (and rising) people who have downloaded my ship I feel sorry for as I certainly won't be able to make a working version under this awful system.

My Gemini is also wobbling like crazy. Haven't tested the rest as it's depressing watching hundreds of hours of development go down the tubes.

There is something other than the new ASAS going on here because I have just tried out your Gemini and Munbug I rockets; Munbug I flies really well and I placed it into an 80km orbit with no problem (and I am a dreadful pilot). Gemini did wobble like crazy on the pad but once launched the ASAS smoothed the flight out nicely (it did seem to be lacking power however and ran out of steam at around 12km).

Link to comment
Share on other sites

My observation while testing this thing was that now you actually need control fins to compensate for missing torque. I see the zepto rockets above have no fins. You used to not need them, now you need them. There's a reason for control fins in real life.

All old designs will need a bit of tweaking. The old ASAS and SAS modules now weigh a lot more, and are completely unnecessary on smaller craft now. Control surfaces (fins especially) are now necessary to compensate for changes in the torque system. Put some fins on your rockets.

Unfortunately, that is not true. The new SAS doesn't do anything, it was removed.;.;

Link to comment
Share on other sites

My 0.21 experiences so far, on a fresh stock install and very basic rockets that I threw together in the VAB specifically for testing:

1) Flying a manned ship isn't bad, but can be a little frustrating. It makes no attempt to stop a roll and the only way to get it to lock onto the heading I want is to turn SAS off and back on. Otherwise, it likes to drift back towards the previous heading, even if my last steering correction is much closer to the current heading. The command authority from the pod is definitely weaker than 0.21.

2) Flying a probe with an ASAS part also isn't bad, though a little more frustrating due to the weaker reaction wheels.

3) Flying a probe without an ASAS part is nothing short of a micromanaging nightmare. Since probe cores don't have the "SAS-equipped" tag in the VAB, that's understandable, except that turning on SAS definitely does something, just nothing helpful. It feels like "utrafine" controls, where even in a 2 ton ship, you have to hold keys for several seconds to see any change in movement. It will not, however, stop a very slow tumble even in vacuum, so it's not acting like the pre-0.22 SAS, it's just seriously reducing command authority. It feels like all it's doing is fighting my commands, but not anything else.

I kind of have to question the "ASAS built into capsules but not probes" decision, especially given that probes don't even have the pre-0.22 SAS functionality any more. This means that either we micromanage the launch, or we consider a part with a flight computer mandatory. I launched probes without ASAS in prior builds, and while SAS wasn't as helpful as ASAS, I could do it, but there's no way I'm doing it again in 0.21. The weaker control wheels are completely understandable, and the 2.5m probe core finally has enough rotational authority that adding an unmanned capsule for more rotPower is no longer a given.

I also have to question the fact that the 1.25m ASAS (now named Inline Advanced Stabilizer) got a reaction wheel but the 2.5m ASAS didn't. Since the flight computer is redundant on a craft with a capsule, the part will only get used on probe craft, and any probe flying a 2.5m stack is probably going to need more torque than a probe flying a 1.25m stack. This also means that if we want to add stock reaction wheels to a 2.5m craft stack, we have to add the 1.25m reaction wheel or a command pod.

Overall, I'm not sure that changing the way ASAS worked at the same time as reducing the rotational power of capsules/probe pods and removing SAS functionality from probe cores was a good idea. We're looking at two or three changes on almost every rocket, so it's hard to say which parts of which changes are the most negative.

Other notes: The PDD-12 Cupola Module no longer outclasses even the Mk1-2 Command Pod, it's now barely more powerful than the Mk2 Lander-can.

Link to comment
Share on other sites

As someone who is avoiding updating for reasons other than the new SAS system (I am anxious to try it so I may update my stock backup install), I think the problem here is that, from what I've heard, the SAS controllers are not being aggressive and are too heavy (how heavy can a computer controller be?). Personally, I can't wait to try out my heavy rockets with additional torque from reaction wheels. That might solve my loss of control problems on things like my station core. But what I am hearing is that those ASAS controllers have now been configured to ease into to not create oscilation. This is great but according to people, it is not doing what it has to to hold headings. Gentle it should be but there ought to be a balance. Now to go update the stock backup and see how these things handle for me.

Link to comment
Share on other sites

It's such a shame that the new SAS ruined the update for me. I was really excited for the new parts (mainly the new parts :D), buildings and terrain. But now flying a rocket to space is like wrestling a snake that just wants to go anywhere but the where you want it to go. :)

Link to comment
Share on other sites

When I get 0.21 loaded, I'm going to try two quick designs with reaction wheels and the new SAS systems; one conventional design (with the axis of thrust pointing through the CoM and the the CoM below the CoL) and one that my old model rocketry experience says is unstable near the end of its burn but worked in 0.20. If the former works like a charm and the latter pinwheels like it would here on Earth I promise not to complain.

-- Steve

Link to comment
Share on other sites

Didn't C7 (think it was him during a live stream) say that the whatever-they-are-called parts that replaced the SAS modules apply torque on the part they are attached to or something like that? It seems fine to me but it could be slightly stronger imo.

Link to comment
Share on other sites

that's unreasonable, the SAS should have NO problem holding my perfectly symmetrical rocket using the reaction wheel and 3 fins i put on it, yet it doesn't, because its too weak

and GIMBAL. gimbal alone should be more than enough to keep a rocket steady. RCS is for keeping you straight while you're not firing the engines.

Link to comment
Share on other sites

Overall, I'm not sure that changing the way ASAS worked at the same time as reducing the rotational power of capsules/probe pods and removing SAS functionality from probe cores was a good idea. We're looking at two or three changes on almost every rocket, so it's hard to say which parts of which changes are the most negative.

I think I heard somewhere that was the cardinal rule of balancing or something. That changing two or more thing that effect each other at once may overdo it.

Link to comment
Share on other sites

Planes, and even SSTO's are entirely possible with the new SAS. I think that some problems here are from porting old crafts, some from misunderstanding on how the new sas works (torque forces...) and some might be attributed to bugs :D. I can definitely say that from my experience, the planes do not hold the axis stable, every time you try to steer the ASAS releases ALL of the axis. And yes, I have tried without the joystick, and I HAVE set my dead zones. On the plus side, the new ASAS is FAR from useless as you can see here:

Brand new install, no mods, new, freshly designed craft - successful SSTO in the orbit (and landed without problems as well).

screenshot0_zpsd43ab286.jpg~original

screenshot1_zpsb7268bdb.jpg~original

Link to comment
Share on other sites

I would bet money this is a problem with copied craft files only. Has anyone with the problem created a craft from scratch and see if it has a problem?

This is what im thinkin too....

ANYONE?

Link to comment
Share on other sites

I think I've figured out the issue. And it is the same issue as was found before: SAS does not play nicely with other SAS. The issue here is that now all pods have SAS pre-installed, meaning ASAS and the like are redundant and will in fact fight with your command pod's SAS. If you plan to use an external SAS of any sort, disable the command pod's SAS first thing.

Link to comment
Share on other sites

I think I've figured out the issue. And it is the same issue as was found before: SAS does not play nicely with other SAS. The issue here is that now all pods have SAS pre-installed, meaning ASAS and the like are redundant and will in fact fight with your command pod's SAS. If you plan to use an external SAS of any sort, disable the command pod's SAS first thing.

Nope, still sucks.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...