Kerbal Space Program

[InnerPartisan]

You need *convincing*? Oh, how I envy thee.

 

[Absalom]

You need *convincing*? Oh, how I envy thee.

? Ive spent 200+ hrs with it and mods man

 

[Multi-headed Cow]

 

[Hellraiser]

And 0.22 is released:



Not featuring glow in the dark kerboctopi though. Wait, what if that is the space kraken?



Also in bad news the major code overhaul has not yet been integrated into the main build. So apart from bugfixes the performance shouldn't be improved much yet. Here's hoping it goes through QA soon and makes it into 0.22.1

 

[Absalom]

So I've fucked around with the new update with the last couple hours (about to play more) but I can confirm the Career mode is groovy but christ it needs tweaking. As expected.

I appreciate the challenge of using limited parts, but at this point I don't know where to go from here; and the parts have a ridiculous arbitrarily "science" value

You pay a shit ton of science points, for example, for a fucking ladder. A fucking ladder!

 

[Multi-headed Cow]

You pay a shit ton of science points, for example, for a fucking ladder. A fucking ladder!

 

[EG]

i can't seem to be able to reach the damn mun

Who needs a Mun?

 

[Absalom]

Also, a BROS Pro-Tip: For most science shit, you gain far more from retrieving a mission than by transmitting data. No, the game doesn't tell you this anywhere. In fact, im not sure if there's ever a point where you want to actually transmit data since science points are so scarce.

You're going to want to do multiple missions repeatedly to get all different types of science! points available and to get lesser values of already known data. Also, theres generally data triers for flying in atmosphere, flying in high atmosphere, LKO, HKO etc and for doing EVERYTHING in every biome possible. Yes you're going to grind yourself retarded doing this.

Really in need of general tweaking but I'm kinda grooving with it.

A final BRO-tip: launch a mission at the beginning of career mode with just a command pod and get ll soil samples/data reports available then immediately retrieve it to get like 35 off the bat

 

[Hellraiser]

Well the record holder from a few days ago managed to max out the tech tree in 3 flights. Granted that was with very careful planning and things like doing a Mun and Minmus flyby mission with the starting parts. So they could have tweaked it a bit and over did it.

And the less science from transmitting is mentioned in the patch notes and there's a x% icon under the transmission icon in the experiment result window showing you how much less you get. Also the transmitting log shows the max amount and transmitted amount, but that scrolls by quickly so it can be easy to miss.

 

[Absalom]

I knew that percentage was fucking up my data somehow but I didnt know if it was conditional or what until I tried recovering.

And sure some aspie can munchkin the mechanics to death but I'm a p good kerbal player and I'm struggling. The struggle does make it interesting tho

 

[Hellraiser]

Well I just finished updating so I'll post my impression soon enough. But it is likely they went for overkill in making tech harder to unlock after the few streamers played the min-maxing game.

 

[Hellraiser]

Okay I did five launches now, well nearly five since I need to finish the last mission of those and it is definitely not that easy but so far I haven't been stuck really. I ironmanned everything and managed to do a lot of science unlocking quite a bit of parts. Although I must confess that I did kind of rush towards the materials bay to make things easier. Biggest problem so far is the fact that docking ports are so deep into the tree. Not sure how I am going to land on the Mun without them, although I could do a soviet style spacewalk rendezvous. Also life without fuel lines and asparagus staging is hard. Still I am looking forward to doing interplanetary missions without the LV-N and Ion Engine for once. Also I manage to build rockets without struts that hold themselves together. Definitely a different and new experience.

Also the lead dev confirmed they reduced the science gained from experiments between the "3 launches for whole tree" experimental build and the release one. It would have been easier if I was willing to strand kerbals on the Mun or Minmus and could transmit soil sample data back lovelessly though. The incentives for doing manned return flights are good but I think they're a bit of an overkill with 80% data loss on the goo canister in Mun orbit IIRC.

Well I'll play some more and see how it goes, I have a plan on how to build a damn big but still proper looking rocket with what few structural parts I have.

 

[Hellraiser]

Absalom somebody mentioned that there was some digging and math posted on the KSP forum regarding transmitting data back versus recovering the vessel with it. It may seem obvious in hindsight after you read it, but despite the diminishing returns redoing the experiment multiple times and sending the results back each time gets you nearly 100% data back give or take some rounding down/up error of half of a science point.

Now I'm not sure how redoing experiments affects diminishing returns and the science pool. That is if partial data transmission counts as one experiment or X% of the experiment for diminishing return purposes, where X is the percentage of data that survives beaming back to Kerbin. So basically if you keep getting say 20% of 50 science for an experiment each time you do it and send it back until it reaches 50 science or if you get 20% of 50 first, then 20% of the diminished science value diminishing with each attempt.

I think I'll do a test with the materials lab in LKO to see which of these two scenarios happens (or I can find the original topic), if it is the first one than outside of electricity limitations or doing experiments that require kerbals anyway (like soil sample collection), there is little advantage in not using data transmission. So probes start looking pretty damn good especially when you lack the parts to do insane return missions from Moho, Jool or Eeloo for ridiculous science.

EDIT: Yes, dimminishing returns do apply for each repeated transmission.

                        Science Points Recived
test number          SC-9001              Mystery Goo
                recover  transmit      recover  transmit
totals          384.8    384.6        198      197.9

1              275      55            110      44
2              78.6      47.1          48.9      34.2
3              22.4      40.4          21.7      26.6
4              6.4      34.6          9.7      20.7

The total pools are similar give or take a fraction of a point due to rounding for transmission values. You need three transmission for goo to reach the value from a single sample return. This is worse for the materials bay as it has a lower data return rate. People at the moment are suggesting to have the science pool be divided into a total transmittable science pool and recoverable. That is so that the limit for transmitting soil sample data is lower than the total pool for returning samples from say the Mun. Basically so that you need to return samples to max it out because transmission only deplete the science pool by X% or something

 

[Hoaxmetal]

So, did they improve the game's performance?

 

[Hellraiser]

So, did they improve the game's performance?

Apart from fixing the obvious bugs like the one causing increased SoI/scene transition loading times from last update? No.

The old code optimization (which is supposed to improve frame-rates by at least 30%) is still in internal testing as there are still a few bugs in it apparently, so it didn't make it into the latest update. Depending how long bug fixing takes expect it either for 0.22.1 or 0.22.2. Other than that Unity still has an old shitty single-thread PhysX which the game ises and there is no 64-bit version so until that changes there won't be improvement in those areas before Unity gets its shit together.

 

[Absalom]

Except trying to transmit 5 diff experiments 3 times each while say, in the upper atmosphere is an incredible pain in the ass Hellraiser

My original point stands, while very cool R&D is still a work in progress by the devs

 

[Hellraiser]

Yes, time constraints are an issue for some cases. This applies not just to atmospheric flight but also to say doing flybys of Jool's moons. And you are right, this does need tweaking. A proper transmission strength system with smaller total science pools for just transmitting some experiments rather than returning them (goo/materials/sample study mostly, stuff like temperature and pressure is numerical data anyway so returning it for bonus points makes no sense) would be better. Still this means that sending a one way trip Eve lander probe is not such a bad idea, although you will need to click more to get the equivalent of a return mission science yield.

Luckily the devs are open to suggestions on improvements here and this system will be improved. One thing they want to do that was mentioned is somehow fit space stations into the science gaining mechanic. A rockomax-size zero-g lab module would be very nice to have in the game if you ask me, but I am wondering how they want to make it work since the current experiments don't favor long-term orbiting things so a new mechanic would be needed. At the same time the devs are not fond of giving stuff for X units of time since you can just timewarp and get stuff with no effort.

 

[Absalom]

I like the space station rockomax idea and if they don't want time warping, then either have a minigame mechinc where you can only do it once every day where you can decide which experiments with who (with stats making a difference maybe) or something like that.


OR have it be VERY slow as in it takes years to get all the data since you're going to timewarp for other missions neway

 

[Hellraiser]

Or docking with science supplies for the lab. Think about it, lab starts with a full supply, you set it to continuous experimentation where it slowly uses up those supplies over a few months and converts them into science every day or week or so. But eventually it runs out so you have to resupply it somehow. Requires some effort, doesn't rewards for simply time warping and you can abandon the station if you don't want to keep supplying it. This could also work for surface research bases, although the issue there is that land docking sucks without mods like KAS.

 

[Absalom]

Or docking with science supplies for the lab. Think about it, lab starts with a full supply, you set it to continuous experimentation where it slowly uses up those supplies over a few months and converts them into science every day or week or so. But eventually it runs out so you have to resupply it somehow. Requires some effort, doesn't rewards for simply time warping and you can abandon the station if you don't want to keep supplying it. This could also work for surface research bases, although the issue there is that land docking sucks without mods like KAS.

Indeed. It could even go hand in hand with a life support system that I hope they add (It may be in an upcoming feature, I don't keep up with KSP devs like you do)

Land docking shouldn't even be a thing, KSP just needs to implement resources that can be moved between different vehicles, this could also make rovers and land bases much more attractive (Esp with SCIENCE! etc)

 

[EG]

Moon and back with a quarter tank, baby. *Dances*

 

[Burning Bridges]

It's been a long time since I have fired up KSP. I see that clouds are still not modelled?

 

[Absalom]

It's been a long time since I have fired up KSP. I see that clouds are still not modelled?

Yep. They decided to work on career mode first

 

[Hellraiser]

Burning Bridges pretty much. Development focus for at least the last two update was for career mode, with the previous one mostly putting in place the system to track progress which now the research and development system is using and in the futures missions will. There has been polish done with the planets, but outside of eye-candy the only significant change were procedural craters on the Mun, which will get added to other planets and moons in the future whenever they decide to do another save-breaking update. Weather is still on the agenda, just not yet the focus, as the least developed part of the game is career mode.

 

[Burning Bridges]

Good, everyone has their own priorities. Personally I didn't care about the career mode and have doubts it is worth it. I'll come back later when they have made the next pass on the planets.