Stock versus Modded Game

[Conarr]

This isn't another one of those posts. I just wanted to comment that I learned to play the game pure stock and then started using mechjeb and other mods. Just out of curiosity, I created a purely stock install a while back and I'm doing one mission on the modded game and the same mission on the stock game - where appropriate. Obviously there is no Kethane equivalent in the stock game right now.

I will admit, my skills are a little rusty. I can't make orbital insertions with the precision mechjeb can manage. I most certainly can't limit acceleration to terminal velocity or g limit as well either. But I can do a pretty fair approximation at least. So far I've been able to land on the Mun, plant flags, etc in both modded and stock games. Has anyone else tried to do anything similar, run the game modded and stock?

[stupid_chris]

I have. It's normal to not be as precise as MechJeb, no one can be. MechJeb is a computer. This is why real space agencies only work on computers, it reduces the human error risk. Of course the machine failure still exists but it's much more reliable.

You're probably not rusty, just not used to what a non MechJeb insertion looks like ^^ When I downloaded the mission pack that goes around the spaceport I remember that I almost forgot to do my gravity turn, at least the popup reminded me

[Temstar]

I can't make orbital insertions with the precision mechjeb can manage

How precise is mechjeb 2 these days anyway? Previously it didn't achieve that precise of a circular orbit. I routinely achieve a AP/PE difference of less than 50m by hand, is that considered precise?

[Guest]

How precise is mechjeb 2 these days anyway? Previously it didn't achieve that precise of a circular orbit. I routinely achieve a AP/PE difference of less than 50m by hand, is that considered precise?

+/- 50 metres? That's highly accurate. MJ gets in the ballpark, maybe not that precise but close enough. MJ2 has been highly accurate for me so far in both launches and landings.

Edited June 5, 2013 by Jack Wolfe

[rodion_herrera]

MechJeb can get really precise, esp. if the maneuvers prior to the one you just did were also accurate--if you started off in bad form (manually), don't expect MechJeb to compensate for that.

I've only recently begun to use MechJeb again, but mainly to fine-tune my stack back at VAB--the ship specs feature is indispensible for that. But I've gone to Eve/Gilly and back (didn't land Kerbals at Eve though--that's still a goal of mine down the road, esp. if I do plan to do it purestock) with a high degree of precision using just stock KSP, some tables, MS Excel, and a little luck.

[Guest]

One thing I haven't bothered checking is the eccentricity of the orbits I get with MJ. I got so used to the Orbiter MFD doing it for me...

[1096bimu]

I didn't get this game to practice any sort of "skill", It's a sandbox game with a science and engineering theme and that's what I'm gonna do, build stuff and test them out. Everything that can be done by a few lines of code will be done by the computer.

[rodion_herrera]

One thing I haven't bothered checking is the eccentricity of the orbits I get with MJ. I got so used to the Orbiter MFD doing it for me...

But you can see that visually in Map view, by checking how the AP-PE nodes "rotate" or "flip around" when you circularize. Granted, it's not a precise value, but it's quite good to figure out if you're circular or not.

[Conarr]

+/- 50 metres? That's highly accurate. MJ gets in the ballpark, maybe not that precise but close enough. MJ2 has been highly accurate for me so far in both launches and landings.

I have a mental block, I still don't trust MechJeb to handle the landings.

[rodion_herrera]

I didn't get this game to practice any sort of "skill", It's a sandbox game with a science and engineering theme and that's what I'm gonna do, build stuff and test them out. Everything that can be done by a few lines of code will be done by the computer.

Ironically, a lot of the real-world manned missions were actually "saved" by piloting skill. Case in point--if Neil Armstrong allowed the PGNS/AGS to continue with its planned descent, the AP11 LM would have landed in an area that would have potentially tilted the LM to the point that it would have toppled over. The PGNS/AGS already wanted to land because it felt it was running out of fuel, but thanks to Neil's handling (and best guesstimate on fuel left), he managed to fly the LM horizontally towards an area that was flatter and where there were fewer obstacles. Lots of other cases in other manned missions.

I have a mental block, I still don't trust MechJeb to handle the landings.

Yep so in my example above, PGNS/AGS in the LM was basically the MechJeb equivalent.

[Guest]

But you can see that visually in Map view, by checking how the AP-PE nodes "rotate" or "flip around" when you circularize. Granted, it's not a precise value, but it's quite good to figure out if you're circular or not.

I've noticed that, and assumed it was because the values were dead close. Good to know I guessed right!

I have a mental block, I still don't trust MechJeb to handle the landings.

I consistently get within 5~10 metres of the desired landing site if I start from an orbit of over 110km. I typically use 120km as a rule of thumb.

[Fel]

Well... MechJeb has a direct data feed to the PRECISE orbital information currently available and "can feasibly" do realtime calculations that have an insane degree of accuracy. (Doesn't mean the developers care enough to do them, just that it can feasibly do them).

It is more than a "Flight Computer", it is a "Total Awareness Unit" from CyberDyne Systems. O.O

*Just meaning, you can't really compare your "skills" against something that can theoretically have near 100% accuracy... while you're limited by the little information that is present.

Edited June 5, 2013 by Fel

[NeoMorph]

Ironically, a lot of the real-world manned missions were actually "saved" by piloting skill. Case in point--if Neil Armstrong allowed the PGNS/AGS to continue with its planned descent, the AP11 LM would have landed in an area that would have potentially tilted the LM to the point that it would have toppled over. The PGNS/AGS already wanted to land because it felt it was running out of fuel, but thanks to Neil's handling (and best guesstimate on fuel left), he managed to fly the LM horizontally towards an area that was flatter and where there were fewer obstacles. Lots of other cases in other manned missions.

Yep so in my example above, PGNS/AGS in the LM was basically the MechJeb equivalent.

Yeah... and P66 is the Translatron.

[MR4Y]

I like how people forget that Mechjeb just assists flying a mission. And I'll agree with some people that it's not perfect. I had rockets that I can bring to orbit manually but Mehchjeb can't do the same thing. Mechjeb is by no means perfect. It can even do two completely different things with the exact same ship.

It's just like auto pilots like the one from the Lazor System. That thing can't land a single space plane, even the ones that come with the mod.

I only use MJ for things I tried and failed many times, like docking.

[NeoMorph]

I kind of like the 1.9.8 version of MechJeb. It's not as capable at MJ2 but it does the job. It doesn't automate things like rendezvous and docking but it gives me all the info that I need.

[Temstar]

I consistently get within 5~10 metres of the desired landing site if I start from an orbit of over 110km. I typically use 120km as a rule of thumb.

120km orbit? Is this a Mun landing? It seems pretty high, I normally start powered descent from a descent orbit somewhere between 10-20km.

Manual precision landing is actually not that hard, I say about the same difficulty level as docking and same as with docking you get better at it the more you do it. It's a lot like flying a helicopter and that's certainly possible.

[FluxCap]

1234567890

Edited June 5, 2013 by FluxCap

[Weatherman159]

Well the latest MJ I used for landing on the Mun scared the hell out of me. I mean, it landed perfectly and probably the most efficient way too, but it also had me going "gonna crash gonna crash gonna crash" for the entire landing, so, I might go back to my less efficient manual landings.

[Akalaël]

I had MJ installed in the 0.19.

0.20 came, and I didn't put that computer on my game.

Actually, I can do the same things than this tool, and I am playing the game, not MJ.

Of course, I can't land so precisely than MJ, my bests landing are 500m away from my targets. Doesn't bother me.

I launch my rockets better than it did (especially the big ones, and without this annoying search to be in the right plan... you know, tend to go to the right, overshoots, goes to the left, oversoot even more, ect ect ad nauseam)

I dock better than MJ, no monopropellant wasted (I need, say, 20l of mono for the final approach, and that's it) and no time waiting in the lower orbit as recommended by the help encounter module.

I recon that MJ was a sort of teatcher for me. It shows me tricks that I wasn't aware of : near perfect ascending path, how to get an encounter.

But the thing is that I learned all the manoeuvers in KSP before I get MJ installed. So I perfected my skills with it, and no longer needs it.

Well, no longer for the driving help at last.

I'm missing all the infos it was provided. I installed engeenier redux to have the delta-V of my rockets, and to have basics infos on my fligths, but I really regret the MASS of data you could get.

Of course, in the orbital map, you can have some of the infos you need, but I like being in the staging mode and perform my manoeuvers. Simple exemple : how to circularise your trajectory in the staging window ? That momment when your AP is flipping with your PE is transcribe on the numbers you can get from MJ, but you have to be in the orbital map in the basic game to see it. I deplore that fact.

I wish sometimes the devs will implement more data to be read in the game, so that I can get rid of all those addons.

Cheers

Edited June 5, 2013 by Akalaël

[Uncle Mort]

I kind of like the 1.9.8 version of MechJeb. It's not as capable at MJ2 but it does the job. It doesn't automate things like rendezvous and docking but it gives me all the info that I need.

I thought I was the only one!

I prefer the cleaner, simpler displays on 1.9.8 over MJ2 and, as you say, it does the job.

[NeoMorph]

I thought I was the only one!

I prefer the cleaner, simpler displays on 1.9.8 over MJ2 and, as you say, it does the job.

I think the MJ2 transparent displays look cool and flashy but they are not really easy to read which is a definite negative score on my end. I like the information to be easy to read so I can make an informed decision. I've played with MJ2 and it's a VERY powerful guidance computer. It can play the whole game for you. But that isn't what I want. I want something like the Apollo guidance computer.

Now the one reason I'm going to have to play with is due to using hardware. Telemachus has a function to interface with MJ2 and my hardware will grab the data from the Telemachus system (hopefully).

But for pure keyboard control I use MJ1.9.8. As you say, it cleaner and simpler... and is easy to use. I do the Hohmann Transfers and docking manually. The only extra thing I use is the Lazor docking camera.

[Karriz]

I definitely need the information that Mechjeb provides (delta-v, speed, altitude), otherwise it feels like I'm flying blind.

When designing a rocket, I need to know if it's capable of getting anywhere in the first place. Without Mecheb's delta-v display I can only guess.

[NeoMorph]

Well I'm amazed at the makers of MechJeb... Hidden away in the Settings of MJ2 is an option to make it have the skin of MJ1... so I that's one big plus for MJ2 in my book. I'm not a fan of transparent menus you see.

Edited June 5, 2013 by NeoMorph

[whiterafter]

I'd just like to say here that I only use Lazor System. Even then, I use that only for informational purposes (physical parameters, phase angles) and stuff like that. I haven't used MechJeb at all.

I think some people use MechJeb so "confidently" that they're not cheating by using it, that they use it so much they practically are cheating. I've been to Duna and back without any mod but Lazor System (used informationally).

I may eventually get MechJeb, but I'd probably make a restriction that I'd have had to already do the landing on my own (Apollo 11) and then use MechJeb for all the other landings and such (Apollo 12,14,15,16,17).

I am really interested in its landing feature but I'd have to use that with caution.

And I'm not a purist, but this is just how I've sort of introduced myself to the game. I sort of felt hurt when I found out people had effectively made it into orbit without doing anything with MechJeb, when I spent 2 weeks trying to get into orbit the normal way.

[NeoMorph]

Apollo 11 landed WITH THE EQUIVALENT OF MECHJEB TRANSLATRON.

If something is going to push my buttons it's this myth that Neil landed the LEM manually. He landed it under computer control with human doing the equivelant of saying to the computer "Don't land in the crater, go over there". The thrust was under P66 landing program with manual guidance input... THAT'S where manual is mistaken for totally manual.

At MET 102:43:20 (430 feet) Armstrong flicked a spring loaded toggle switch with his left hand, entering the rate-of-descent mode (P66). Now the computer controlled the spacecraft's thrust to maintain a rate-of-descent commanded by the ROD switch. A flick upward slowed the descent by one foot per second; a flick downward increased the descent rate by the same amount. Using the joystick, Armstrong tilted the LM to null out horizontal velocity and bring the LM to a safe area for touchdown. After some "possibly spastic" control motions because dust kicked up by the exhaust plume distorted his perception of translational velocity, at MET 102:45:40, Armstrong landed the spacecraft safely in the Sea of Tranquility.

Yup... Armstrong was using the equivelant of a MechJeb Translatron to help handle the thrust so he could concentrate on choosing a good clear landing area.

Go and have a read here... http://www.klabs.org/history/apollo_11_alarms/eyles_2004/eyles_2004.htm - it's from one of the guys who made the apollo guidance computer.

I guess Armstrong was cheating then heh.