Speaking of Science... we need a 'Kerbilopedia'!

[Vl3d]

Lets just agree that we don't know if the game will have any testing area or simulation mode. And we also don't know about the new science.

[Pthigrivi]

Man, quite the stickler. Obviously nothing is confirmed and nothing is certain. They might remove Donk from the game or come up with completely different solution for radiators. The fact that they say "players learn basic flight concepts in a virtual simulator that can be accessed anytime the game is paused" leads me to believe that the final game will probably include a virtual simulator that players can use whenever the game is paused . Thats about as reasonable as any other assumption one could make about the game. 

[Pthigrivi]

44 minutes ago, Vl3d said:

I initially preferred something like this:

Dedicated testing grounds and facilities at the KSC, some of these can be abstracted but I really like testing subassemblies during design, before launch (off-road testing areas to simulate terrain on other planets, tethers or weights to simulate gravity on other planets, pool to test craft boyancy, dark room to test levels of sunlight exposure for solar panels and battery recharge, aerodynamic tunnel for aircraft, vacuum chamber to test flight in another atmosphere, some kind of pivot for RCS strength and ballance testing in aerodynamic scenarios - for example to prevent rocket flipping when launching, G-force centrifuge) - these would also help a lot when playing without reverting.

In my experience landing simple probes on most planets is pretty trivial, but I do really like being able to use Hyperedit to test more complex landings like safely dropping big base components and rovers and ascent vehicles in tight clusters on a place like Eve. For that kind of thing you really want to know not just the dV but how it performs given different reentry profiles, how maneuverable it is on decent, when to drop fairings and heat shields so they float away rather than crashing on previously landed modules, how to align docking ports between rovers and bases given local gravity, etc. For that kind of thing having a virtual simulator that lets players do test descents and ascents starting on other bodies before you put all that equipment in orbit would be really nice. 

[Bej Kerman]

44 minutes ago, Pthigrivi said:

Man, quite the stickler. Obviously nothing is confirmed and nothing is certain. They might remove Donk from the game or come up with completely different solution for radiators. The fact that they say "players learn basic flight concepts in a virtual simulator that can be accessed anytime the game is paused" leads me to believe that the final game will probably include a virtual simulator that players can use whenever the game is paused . Thats about as reasonable as any other assumption one could make about the game. 

Doesn't mean you have a free pass to say something is coming to the game with little to 0 evidence solely because it has tenuous, extremely tenuous links to an existing dev post. You could say KSP 2 is getting aliens because they've already made a character animation system for the Kerbals, but that is obviously not a reasonable assumption to make - just saying "nothing is certain" as quoted here does not make the assumption about aliens any more reasonable.

Edited June 22, 2022 by Bej Kerman
Grammar

[Vl3d]

@PthigriviI agree with you, Hyperedit is very useful. I have personally never used it because I like to launch testing craft in Kerbin orbit and then test reentry (with rocket power for greater speed if needed). Then I send a test mission to Eve - if it fails I save / reload. Normally I would have to send another one but it's just too time consuming. And if I want to test a craft designed for another planet without an atmosphere I send it to the Mun.

I just think Hyperedit makes testing too easy (the simulation is 1 to 1). Simulated testing should be somewhat more error prone. Kerbal-real-life testing is more valuable and it's more true-to-method in a realistic mission profile kind of way. I like to look at this list to remind myself that failure is part of the fun: https://en.wikipedia.org/wiki/List_of_missions_to_Mars

Edited June 22, 2022 by Vl3d

[SciMan]

4 hours ago, Pthigrivi said:

Man, quite the stickler. Obviously nothing is confirmed and nothing is certain. They might remove Donk from the game or come up with completely different solution for radiators. The fact that they say "players learn basic flight concepts in a virtual simulator that can be accessed anytime the game is paused" leads me to believe that the final game will probably include a virtual simulator that players can use whenever the game is paused . Thats about as reasonable as any other assumption one could make about the game. 

My personal interpretation is that we get something in the middle of what you and @Bej Kerman are saying. It's not what either of you are saying, but it's also exactly what both of you are saying.

So, I'm sure we don't get "nothing", some form of simulation will be in the game, otherwise the shader wouldn't have been showcased.

But I'm also sure we don't get "full simulator able to be used at any point in the game, that lets you take risks with no effective consequences".

What I think we get is a "simulation mode" in the same vein as an airliner pilot would have access to a flight simulator. At any point while you're playing the game, you could enter this feature, and it would give you the ability to practice a few things (docking, landing with rocket thrust or with wings, ascent, driving a rover so it doesn't flip, etc.).
Not with your craft that you constructed, but with pre-set craft available only in that simulation.
In other words, an "interactive tutorial". Seeing as the post it was supposedly mentioned in was one that focused specifically on the tutorial part of the game, I see no reason that we'd see something entirely unrelated like a "let me see if I can land MY CRAFT wherever I am orbiting in a simulation first" type feature (that's not related to a tutorial in any other way than the fact that you "can" use the word "simulation" to describe both, to an extent).

Why don't we get the full simulator? Easy: Despite it being trivial to implement, what would it add that quicksave/quickload does not already add?

Additionally, spaceflight is very difficult and complicated, there is a lot of real risk involved, and no amount of IRL flight simulator testing can fully capture how something handles IRL, because it is impossible to create a perfect representation of what is actually produced, without just actually producing the thing (this is why we build prototypes).
For a recent IRL example, look at what happened with the Boeing CST-100 Starliner, they simulated that thing a whole lot, and yet the first test flight of the thing was a gong show that was almost a complete failure including loss of vehicle (which almost happened twice, once because it almost used all its RCS fuel, and again because if it hadn't been reprogrammed the SM would have banged into the capsule when it was jettisoned).

EDIT: As for "making statements without evidence and then misrepresenting them as fact", I'm not going to comment on that, I don't think it's my place.

EDIT 2: Perhaps we even get a "precision landing" simulator, where the goal is to land a vessel within a certain distance of a target marker on a certain body (and maybe there is a version of this for every body that is capable to be landed on, with them unlocked as you discover distant planets and moons.

Edited June 22, 2022 by SciMan

[Pthigrivi]

Haha well either way Im positive we're all being adequately pedantic about this. The only OT aspect is whether the slate of possible starting points for a player-usable simulator might be upgraded by exploring new locations. Given that the overall challenge of somewhat precisely landing much more complex vessels for building and maintaining colonies already ups the ante I personally don't see a reason to inject artificial error into that kind of simulation. 

Edited June 22, 2022 by Pthigrivi

[Gargamel]

What we really don’t need is yet another thread that covers all the various things everybody wants in a game.  
 

Let’s stick to the topic of a KSP2 kerbalpedia. 

[Vl3d]

1. Speaking of science.. I've always been curious. Is there anything in Kerbalism Science Only that you don't want in KSP2?

I think the science mechanics in Kerbalism are so much better than stock.

Spoiler

The science system in Kerbalism is very different from how you know it in KSP. Science isn’t generated at the click of one button any more, most experiments will take time to complete. Some will just take minutes, others will need years. The good news is that experiments will keep running in the background on unloaded vessels while you’re busy with other missions, and they will generate a constant stream of scientific value while they’re running.

While this doesn’t sound like a big thing, it will change the way you build vessels, it will change the way you plan your missions and it will force you to make tough decisions. You will have to choose which experiments to run, because you won’t be able to run them all at the same time. Sometimes you will have to delete valuable data to be able to collect new science. And you will have to come up with engineering solutions for problems you never had before. Or when was the last time you had to sustain a base with crew for months, submerged at least 100m deep on the ocean floor?

Transmission
Science results can be flagged for transmission home and they will be sent to DSN at the first opportunity. The transmission happens over time, even when the vessels are unloaded. Transmission times can range from mere seconds to years, depending on the size of the file transmitted and the transmission rate of the connection.

Samples
Some experiment results are not transmissible, these are considered samples and need to be recovered or analyzed in a lab. Samples are stored in slots that can be flagged for analysis, and then will be analyzed over time in a laboratory. As the analysis proceeds the sample will be slowly converted into transmissible data.

Samples have mass. Some Experiments like the Goo Container contain just a few grams of sampling material (mystery goo) that is used up while the goo observation takes place. Once the sampling material is used up, the experiment cannot be rerun. Other experiments will collect samples from the atmosphere or the surface. A surface sample for example will collect 25kg of mass that should be accounted for when it needs to be hauled off the surface.

2. Also, do you want to still use action groups to run experiments or do you like the features of mods like [x] Science (list based experiment triggers) and ForScience (notifications)?

[SciMan]

I have no experience with Kerbalism, not even the "science only" version of it, so I can't comment on that. I am pretty sure that we're getting other elements of Kerbalism in KSP 2 (radiation concerns from radiation sources on the same vessel at least), but I'm not sure about science changes.

I do like both [X] Science and ForScience, and it would be excellent to have the features of both in stock KSP 2. However, I also like the ability to tell experiments when to start collecting data rather than having them collect it all the time if I've already been somewhere and I'm just passing thru.
So IMO this could be improved by combining features of [X] Science and the stock ability to tell experiments when to run via action groups.
What I mean is that you'd be able to use the action group as a "master experiment power toggle" type control, but as long as the experiment is enabled it would automatically collect new results as the situation warranted it.
However, it can be improved even further if all probe cores become able to store experiment data for later transmission (as if you added an experiment storage container to them). And if the data does get sent, you should be able to choose where it goes (depending on what vessels or facilities are available that can accommodate such data), not just "always transmit to KSC", because you might have an orbiting science lab (or a nearby colony) that you want to send the results to instead of sending them all the way back to KSC.

EDIT: Additionally, rather than have any single experiment need to be run several times to deliver the full amount of science data, make it so that instruments or experiments which would otherwise need to do that instead be able to be run continuously, this would give incentive to put such sensors and/or experiments on a small lander that can be landed where needed and then subsequently left there until data collection and relay to "wherever" is complete (or forever, if the player doesn't have need of that specific combination of instruments and experiments any more on that planet).
:END EDIT

Anyways, back to the experiments:

In my personal ideal case, certain experiments could run continuously to collect data (with the appropriate sensors creating maps of things like resource distribution (resource scanners potentially of several types depending on what resource you're looking for), temperature on the surface (infrared scanner), general type of terrain to be expected (radar or laser rangefinder sensors), or just general surface features of interest (perhaps a multispectral scanner). However, that multispectral sensor might instead be a "combination sensor".

So, combination sensors, what do I mean when I say that?
Well it's pretty straightforward. I mean a single part that incorporates the functions of multiple discrete sensors, therefore it's a "combination" of "sensors".
I guess you could call them "instrument packages" for lack of a better term.

So you'd have things like instrument packages that specialize in:

• Planetary imaging (IR, Visual, Multispectral imaging sensors), it would look sort of like a bundle of camera lenses, early tech versions might incorporate visual elements that signify it uses film to record the images. Potentially some versions could be on an arm that is able to be swung away from the main body of the craft to get a better view of whatever it's trying to image (like the sensor boom on the Voyager 1/2 probes, or Cassini's camera boom).
• Planetary Spectroscopy (IR, Visual, Multispectral non-imaging sensors), it would look like a different kind of bundle of camera lenses, maybe incorporating prisms to convey that it's trying to split the incoming light based on what wavelength it is. Again, perhaps a version on a boom that can be extended (but starts off stowed for launch).
• Celestial imaging (Low-resolution IR, high resolution Visual and Multispectral imaging sensors), It would look like the imaging portion of the Hubble space telescope, since that's pretty much what this sensor would be, and again early tech versions might use film. The player would be able to service this experiment on-orbit to upgrade it to digital image sensors if they had launched one that had film-based sensors earlier, because Hubble was serviced several times. Note: The reason I specify a "low-resolution" IR imaging sensor in this payload is intentional, the "high-resolution" one would be it's own telescope that looks like JWST's primary mirror assembly and sun-shade.
• Celestial spectroscopy (IR, Visual, Multispectral non-imaging sensors), it would look like a slightly different space telescope, and just like the planetary spectroscopy package it might incorporate externally visible prisms to convey that it's doing spectroscopy.
• X-ray optical observations (both imaging and spectroscopy), it would look like the Chandra X-Ray Observatory's optical segment.
• Electromagnetic science (electrical fields, magnetic fields, and plasma sensors), it would look like a combination of long telescoping antennas to sense radio/plasma waves, and a telescopic boom for a magnetometer.
• Surface-based atmosphere observations (thermometer, barometer, microphone, gas spectrometer (aka atmospheric analyzer), potentially an ISRU experiment based on MOXIE), this would probably look like just a generic box with some greebles on it, maybe a visible air filter element and/or intake fan of some sort, along with exhaust louvers. If the conditions are right, this unit could also provide a probe-scale ISRU that is only able to harvest atmospheric resources (so if you build a hopper powered by cold gas thrusters, this could be the thing that acts as the gas compressor to enable it to "refuel" the tanks with whatever atmosphere is present).
• Surface-based physics observations (accelerometer, seismometer, mass concentration characterization (aka "gravioli" detector), this would look like a generic box with greebles on it, connected to a dome that can be lowered to the ground (the dome contains the seismometer, and the dome is intended to prevent things such as wind or sandstorms from impacting the seismometer and causing false readings).
• Surface-based geology observations (laser-ablation chromotrography, alpha particle spectrometer, potentially other things) this would look like a box (potentially of unusual shape) with some optical-looking greebles on it, take the sensor platform of the Curiosity rover for reference.
• Surface based sample collection observations (mass spectrometer, microscopic observations, imaging of the excavated material), this should look like a polygonal box with a robot arm attached to it that has a rock drill on the end, look at Curiosity's robot arm and sample analysis mechanism for reference.

There's probably other combinations of instruments that could be thought up, so maybe instead of this whole "pre-set instrument packages" thing it should be "we give you several different 'instrument package' parts and you get to pick what you want to put in them based on how many slots the part has, how many slots the instrument or experiment takes up, and what you're doing, with their appearance changing based on what sensors and experiments are installed".

Edited June 23, 2022 by SciMan

[Vl3d]

Also check this out:

But still I think the Kerbalism science system is the best..

https://kerbalism.readthedocs.io/en/latest/science.html

[JoeSchmuckatelli]

27 minutes ago, Vl3d said:

Also check this out:

But still I think the Kerbalism science system is the best..

https://kerbalism.readthedocs.io/en/latest/science.html

I wasn't aware of all those science mods.  OTOH, I'm guessing that someone on the Dev team has looked at all the community modding stuff for 'good ideas' - way back when.

I know we're less than a year out from the currently anticipated ship date; so I'd also have to assume that how they're doing science in KSP2 is already baked in.  That said - I can hope they've done something interesting like the Kerbilopedia idea.  (That kind of thing would give writers like Just Jim a bit of longevity, too, I'd think!)

[mcwaffles2003]

10 hours ago, Vl3d said:

1. Speaking of science.. I've always been curious. Is there anything in Kerbalism Science Only that you don't want in KSP2?

I think the science mechanics in Kerbalism are so much better than stock.

  Reveal hidden contents

The science system in Kerbalism is very different from how you know it in KSP. Science isn’t generated at the click of one button any more, most experiments will take time to complete. Some will just take minutes, others will need years. The good news is that experiments will keep running in the background on unloaded vessels while you’re busy with other missions, and they will generate a constant stream of scientific value while they’re running.

While this doesn’t sound like a big thing, it will change the way you build vessels, it will change the way you plan your missions and it will force you to make tough decisions. You will have to choose which experiments to run, because you won’t be able to run them all at the same time. Sometimes you will have to delete valuable data to be able to collect new science. And you will have to come up with engineering solutions for problems you never had before. Or when was the last time you had to sustain a base with crew for months, submerged at least 100m deep on the ocean floor?

Transmission
Science results can be flagged for transmission home and they will be sent to DSN at the first opportunity. The transmission happens over time, even when the vessels are unloaded. Transmission times can range from mere seconds to years, depending on the size of the file transmitted and the transmission rate of the connection.

Samples
Some experiment results are not transmissible, these are considered samples and need to be recovered or analyzed in a lab. Samples are stored in slots that can be flagged for analysis, and then will be analyzed over time in a laboratory. As the analysis proceeds the sample will be slowly converted into transmissible data.

Samples have mass. Some Experiments like the Goo Container contain just a few grams of sampling material (mystery goo) that is used up while the goo observation takes place. Once the sampling material is used up, the experiment cannot be rerun. Other experiments will collect samples from the atmosphere or the surface. A surface sample for example will collect 25kg of mass that should be accounted for when it needs to be hauled off the surface.

2. Also, do you want to still use action groups to run experiments or do you like the features of mods like [x] Science (list based experiment triggers) and ForScience (notifications)?

I play with kerbalism frequently and playing without it is just painful and unsatisfying. The fact that I dont have to search for a module, right click it and tap a button the moment I proc a condition, but instead can just passively activate it in a side panel while being made to hold a condition is so much greater. Also, the way kerbalism works feeds well into how things like SCANSat works and I would difinitely like to see something like SCANSat be stock so I can survey good landing spots for high resource gain.

[Catto]

"yeah we should"