Jump to content

KSP2 - Science Progress


Cpt72Bug

Recommended Posts

Experimental modules have to be my favorite part of KSP1. As a Molecular Biologist / Entomologist I'm really curious to see what direction they will go in terms of the included projects for KSP2. Hopefully, on top of more physics and geology scanning parts they integrate more biology scientific projects akan to those we see currently on the international space station. 

 

Interested in hearing what your thoughts are in what may be included for science experiments. Also if you have any references to current released footage I'd be happy to see it posted here.

 

 

Dr. Bug

Link to comment
Share on other sites

Ive made several threads on the topic and I think its just easier to post them here than to re-write a wall of text. In short, I think experiments could be made vastly more useful to the gameplay experience outside of points and be much more fun to interact with.

Top hope for me would be an interactive index of planet properties that fills out as more science data is collected to help guide players and mimic the actual purpose of science.

Link to comment
Share on other sites

Yup, I think there's a "realistic" way to incorporate Science into the game and a "gamey" way, but either one would probably be better than what we have.

"Realistic" is as @mcwaffles2003 suggests - let the science actually discover... science! By doing experiments, we'd uncover information that would make future missions easier. It would be great to discover things like SOI boundaries, atmosphere altitudes and density, gravity, and have those all inform future missions. One thing I love about the mod SCANSat for example is that you can send a preliminary mission to map a planet, which then makes all future missions easier because you now have a great map!

One "Gamey" suggestion I've read on this forum would also be things like having a variety of currencies discovered by science so that certain experiments help unlock specific parts of the tech tree. I'm not as much in favor of this since it would make the game more linear when it should be anything but that.

Regardless, I'm also interested in how the Science gameplay will actually function. I don't want a return of the "click-a-menu-button at the right precise time" game mechanic. I'd rather there be something like Kerbalism, where an experiment is armed and ready, taking time, EC, and data bandwidth in order to transmit. It makes ship designs more interesting because the systems all have to support the science much more directly.

Link to comment
Share on other sites

This is an interesting topic because I don’t think I’ve talked about what experiments I would actually like to see. I’m mostly hoping for more in-depth physics experiments , but I suspect that as we colonize the solar system, exo-geology and exo-meteorology will both be relatively common areas of study. I’d like to see an experiment collected via a mesh of ground stations measuring the speeds of ground waves to generate a map of the internals of a planet, or an experiment that you put on a plane that records its trajectory and telemetry as it passes through extreme weather events. The actual nitty-gritty can be abstracted a bit, but I’d like to see some high-level representations of the data. 
 

For physics experiments, I’d love a long-running experiment that functions a bit like LISA, detecting really small gravitational waves in space. And I’ve always wanted discovery of other stars and planets to be completely fleshed out to an unreasonable degree, with spectrum images and light curves and lots of data, alongside an intuitive interface where you can click on a few absorption lines and then a box pops up telling you what chemical species is causing that. Highlighting cyclic dips in light curves can reveal orbital periods of planets (or secondary star) or the period of the star if it is a variable star. Observing stars over a long time will give parallax data which gives distance which… Well, I don’t think even 1% of that full system will be in the base game, but I’m going to mod it in for sure. 

Edited by t_v
Link to comment
Share on other sites

On 10/22/2022 at 10:06 PM, MagicCuboid said:

Yup, I think there's a "realistic" way to incorporate Science into the game and a "gamey" way, but either one would probably be better than what we have.

"Realistic" is as @mcwaffles2003 suggests - let the science actually discover... science! By doing experiments, we'd uncover information that would make future missions easier. It would be great to discover things like SOI boundaries, atmosphere altitudes and density, gravity, and have those all inform future missions. One thing I love about the mod SCANSat for example is that you can send a preliminary mission to map a planet, which then makes all future missions easier because you now have a great map!

One "Gamey" suggestion I've read on this forum would also be things like having a variety of currencies discovered by science so that certain experiments help unlock specific parts of the tech tree. I'm not as much in favor of this since it would make the game more linear when it should be anything but that.

Regardless, I'm also interested in how the Science gameplay will actually function. I don't want a return of the "click-a-menu-button at the right precise time" game mechanic. I'd rather there be something like Kerbalism, where an experiment is armed and ready, taking time, EC, and data bandwidth in order to transmit. It makes ship designs more interesting because the systems all have to support the science much more directly.

The Realistic suggestion sounds great. Actually getting useful data to use in future flights, such as airbreaking potential of certain atmospheres. I agree though the more gamey one will limit the nonlinear potential. 

 

On 10/22/2022 at 10:37 PM, t_v said:

This is an interesting topic because I don’t think I’ve talked about what experiments I would actually like to see. I’m mostly hoping for more in-depth physics experiments , but I suspect that as we colonize the solar system, exo-geology and exo-meteorology will both be relatively common areas of study. I’d like to see an experiment collected via a mesh of ground stations measuring the speeds of ground waves to generate a map of the internals of a planet, or an experiment that you put on a plane that records its trajectory and telemetry as it passes through extreme weather events. The actual nitty-gritty can be abstracted a bit, but I’d like to see some high-level representations of the data. 
 

For physics experiments, I’d love a long-running experiment that functions a bit like LISA, detecting really small gravitational waves in space. And I’ve always wanted discovery of other stars and planets to be completely fleshed out to an unreasonable degree, with spectrum images and light curves and lots of data, alongside an intuitive interface where you can click on a few absorption lines and then a box pops up telling you what chemical species is causing that. Highlighting cyclic dips in light curves can reveal orbital periods of planets (or secondary star) or the period of the star if it is a variable star. Observing stars over a long time will give parallax data which gives distance which… Well, I don’t think even 1% of that full system will be in the base game, but I’m going to mod it in for sure. 


Weather phenomenon and random weather events would be a really cool angle they can possibly source many science experiments from that I didn't think about. Thank you for this!

 


 

Link to comment
Share on other sites

I'm fine with gamey science, but I desperately want it to be something other than just "science points". What I really want is for there to be a reason to go certain places, aside from "it was there" or "it had a higher research multiplier". I want to have to get exotic particles from a dead kraken on one of Jool's moons to unlock future tech.

Link to comment
Share on other sites

I think that most experiments should be "sensor-type" experiments.
There would be exceptions, like surface samples for instance would be a "instantaneous" result (and we really need a way to collect those without the presence of a crew).
All the experiments give you "Data" or "science points", or "whatever they come up with to represent scientific data in a general sense" (the name doesn't make a difference to me personally).
The data would require time to be analyzed by the scientists back at KSC (or in an orbital science lab, if the signal can reach it, no you don't have to bring the actual experiment back unless it's something physical like a surface sample or goo pod or materials bay).

Sensor data would be generated over time, not instantly. Physical experiment data like surface samples would be gathered instantly, and you would get only a vanishingly small amount of data from transmitting it without a "sample analysis" experiment to go with the "sample collector" experiment (similar to the sort of instrument used to analyze soil samples that's carried by Curiosity).

To generate sensor data, you would need 2 things:
Power to run the sensor(s),
A place to put the data.

There are 2 places to put the data:
A data recorder,
An antenna that has a valid connection to any of (a crewed science lab, a colony with a research lab, or KSC itself).
Antennas are the primary means of moving data around. You "can" return data recorders to "within recovery range" of anywhere that would accept an antenna connection, but this doesn't gain you anything over just transmitting the data. It might be the only way to do things for some missions tho, due to mass, power budget, or tech limitations.

Put another way, data recorders act like "a battery for sensor data".
All probes and command capsules would have a pretty small data recorder built in, enough to store the data of one instant experiment, or the data from most of an orbit of a sensor experiment.
There would be a separate data recorder part (in several sizes) that would be able to add on to that capacity, if needed (and if you have more than one sensor going and expect to lose signal, you're going to need that extra data recording capacity).
Later tech data recorders would be smaller and lighter than early tech data recorders, eventually going to solid state data recorders that are packaged in a radiation vault (still lighter than a rad-tolerant magnetic tape recorder, because it's just a few chips in a thin lead box).
Data recorders "charge" or store data when the antenna can't handle transmitting the data back to somewhere else.
This can be because the antenna doesn't have enough bandwidth, and/or it can be because the antenna doesn't have a valid connection.
But if the antenna has a valid connection, and the antenna is not using all the bandwidth that it can carry, the data recorder will fill in the extra bandwidth to allow the antenna to operate at full speed, until there is no more data in the recorder.
So if you have a high-bandwidth experiment like a video camera, you should really be using high bandwidth antennas, tho these would obviously take a lot of power to run.
But if you have a low-bandwidth experiment like a thermometer, you'd probably want to save power by choosing a lower bandwidth antenna.

Because no matter what bandwidth of science data signal is being carried by the antenna, it uses a fixed amount of power to transmit.

However, it CAN still makes sense to use a high bandwidth antenna even on small probes, because there's a trick that reduces the power needs. You need a data recorder AND the antenna tho.
It's called "Data burst mode". This makes a data recorder and an antenna work together.
The data recorder stores data until it's mostly full, and then (using a lot of power but only for a short time in the high bandwidth antenna) the antenna will transmit the data in the recorder until the recorder is empty. The cycle then repeats. If you're familiar with what a "relaxation oscillator" is, this is basically one of those in a way.

There are 4 factors in play with the antennas (both IRL and in my proposal):
Range, Power usage, Bandwidth, and physical Size
For the same Bandwidth and range, the smaller the antenna is, the more power it's going to use, because this means the signal is less directional (more power gets "wasted" by not being pointed at the antenna on the other end, so you need more transmit power to compensate).
For the same Bandwidth and Power usage, the smaller the antenna is, the shorter its range will be (again, the signal spreads out more with the smaller antenna, so it's not "loud" enough to be received as far away as with a larger antenna).
For the same Range and Power usage, the smaller the Bandwidth you need, the smaller the antenna can be (it takes a longer time to pick out a smaller signal from the background noise, and smaller antennas that use the same power don't emit as much of a signal in the right direction for it to get picked up by the antenna on the other end).

 

"OK", you're going to say, "What does this gain the player?"
Glad you asked.
Having this many different variables in play allows the player a lot of choices for how they want to solve their data transmission problems. Having this many choices means that there's a lot of solutions that can be made to work.
It also allows you to simulate things like the Galileo mission, where the main high-gain antenna didn't deploy right and they had to use the included data recorder to store data that would have otherwise been transmitted as it was gathered, but the low-gain antenna didn't have the bandwidth to do that.

Link to comment
Share on other sites

1 hour ago, SciMan said:

I think that most experiments should be "sensor-type" experiments.
There would be exceptions, like surface samples for instance would be a "instantaneous" result (and we really need a way to collect those without the presence of a crew).
All the experiments give you "Data" or "science points", or "whatever they come up with to represent scientific data in a general sense" (the name doesn't make a difference to me personally).
The data would require time to be analyzed by the scientists back at KSC (or in an orbital science lab, if the signal can reach it, no you don't have to bring the actual experiment back unless it's something physical like a surface sample or goo pod or materials bay).

Sensor data would be generated over time, not instantly. Physical experiment data like surface samples would be gathered instantly, and you would get only a vanishingly small amount of data from transmitting it without a "sample analysis" experiment to go with the "sample collector" experiment (similar to the sort of instrument used to analyze soil samples that's carried by Curiosity).

To generate sensor data, you would need 2 things:
Power to run the sensor(s),
A place to put the data.

There are 2 places to put the data:
A data recorder,
An antenna that has a valid connection to any of (a crewed science lab, a colony with a research lab, or KSC itself).
Antennas are the primary means of moving data around. You "can" return data recorders to "within recovery range" of anywhere that would accept an antenna connection, but this doesn't gain you anything over just transmitting the data. It might be the only way to do things for some missions tho, due to mass, power budget, or tech limitations.

Put another way, data recorders act like "a battery for sensor data".
All probes and command capsules would have a pretty small data recorder built in, enough to store the data of one instant experiment, or the data from most of an orbit of a sensor experiment.
There would be a separate data recorder part (in several sizes) that would be able to add on to that capacity, if needed (and if you have more than one sensor going and expect to lose signal, you're going to need that extra data recording capacity).
Later tech data recorders would be smaller and lighter than early tech data recorders, eventually going to solid state data recorders that are packaged in a radiation vault (still lighter than a rad-tolerant magnetic tape recorder, because it's just a few chips in a thin lead box).
Data recorders "charge" or store data when the antenna can't handle transmitting the data back to somewhere else.
This can be because the antenna doesn't have enough bandwidth, and/or it can be because the antenna doesn't have a valid connection.
But if the antenna has a valid connection, and the antenna is not using all the bandwidth that it can carry, the data recorder will fill in the extra bandwidth to allow the antenna to operate at full speed, until there is no more data in the recorder.
So if you have a high-bandwidth experiment like a video camera, you should really be using high bandwidth antennas, tho these would obviously take a lot of power to run.
But if you have a low-bandwidth experiment like a thermometer, you'd probably want to save power by choosing a lower bandwidth antenna.

Because no matter what bandwidth of science data signal is being carried by the antenna, it uses a fixed amount of power to transmit.

However, it CAN still makes sense to use a high bandwidth antenna even on small probes, because there's a trick that reduces the power needs. You need a data recorder AND the antenna tho.
It's called "Data burst mode". This makes a data recorder and an antenna work together.
The data recorder stores data until it's mostly full, and then (using a lot of power but only for a short time in the high bandwidth antenna) the antenna will transmit the data in the recorder until the recorder is empty. The cycle then repeats. If you're familiar with what a "relaxation oscillator" is, this is basically one of those in a way.

There are 4 factors in play with the antennas (both IRL and in my proposal):
Range, Power usage, Bandwidth, and physical Size
For the same Bandwidth and range, the smaller the antenna is, the more power it's going to use, because this means the signal is less directional (more power gets "wasted" by not being pointed at the antenna on the other end, so you need more transmit power to compensate).
For the same Bandwidth and Power usage, the smaller the antenna is, the shorter its range will be (again, the signal spreads out more with the smaller antenna, so it's not "loud" enough to be received as far away as with a larger antenna).
For the same Range and Power usage, the smaller the Bandwidth you need, the smaller the antenna can be (it takes a longer time to pick out a smaller signal from the background noise, and smaller antennas that use the same power don't emit as much of a signal in the right direction for it to get picked up by the antenna on the other end).

 

"OK", you're going to say, "What does this gain the player?"
Glad you asked.
Having this many different variables in play allows the player a lot of choices for how they want to solve their data transmission problems. Having this many choices means that there's a lot of solutions that can be made to work.
It also allows you to simulate things like the Galileo mission, where the main high-gain antenna didn't deploy right and they had to use the included data recorder to store data that would have otherwise been transmitted as it was gathered, but the low-gain antenna didn't have the bandwidth to do that.

Yes, but what experiments do you want to see in the game?

Link to comment
Share on other sites

3 hours ago, t_v said:

Yes, but what experiments do you want to see in the game?

Many of the ones that we already have and some that have been introduced with mods.

Barometric and temperature sensors to attach to probes which can produce graphs like this as they gather data at different altitudes

Spoiler

Duna_Atmosphere_T&P.png

A magnetometer to map a planets magnetic field

A micrometeorite detector to discover the comparative safety of an orbits region

A scintillator to map the strength of radiation belts at different altitudes

Spoiler

radiation-fields-01.png

Space telescopes to discover new planets and star systems revealing planetary orbits and general information about the stars that they orbit with high tolerance assumptions about those bodies composition and atmospheric properties

SCANsat like sensors to produce global maps like these which would reveal terrain maps, biome maps, and somewhat inaccurate (very general) resource maps:

Spoiler

687474703a2f2f692e696d6775722e636f6d2f6b

cTbNjiN.png

A multitude of core sample drilling experiments of varying mass and length as well as seismic impact experiments to more accurately prospect a regions resource density increasing the accuracy of the prior maps with greater and greater accuracy.

These last few could be done by having a perfect map established with layers of blurring masks over top of them with each experiment removing a certain radius to a depth of a set amount of the masks.

 

I can go on... but finally, a single screen with which to view all this data like an evolving atlas, with each body taking its own page, each stellar system its own chapter, the whole kerbal  cosmos, the book.

 

Edited by mcwaffles2003
Link to comment
Share on other sites

The possibility of revamped science is entirely behind my hope /wish for an auto-updating 'Kerbilopedia'. 

My thought was that the game would incorporate a Kerbilopedia of the places the player knew about and visited. With each science experiment, or landing in a new place, the respective page would update.

Examples:

  • Player performs a Temperature check at the pad at KSC.  The page about the KSC updates with information about the temperature range. 
  • Player discovers and lands in a new biome, say 'Kerbin's Desert, does a temperature check, a barometric pressure check and performs a crew report.  Each of these gets added to the sub page for Kerbin, Biomes, Desert - with some informative text. 
  • Player takes a sample on the Mun.  Again, this is displayed on the Mun's particular page with some kind of 'data' made up by the writers of KSP2 

Of course, this is highly unlikely to materialize, as it is effectively a 'game within the game' and would be a whole 'thing' that the developers would have to build and incorporate. 

 

...but it would be kewl! 

 

Link to comment
Share on other sites

 

The thing about how you would have to repeat experiments at places - like a temp check at the pad - for reduced science output could be handled (in my Kerbilopedia example, above) by a 'confidence' number.  First time: you get say 67% confidence.  Meaning you 'get 67%' of the 'science points' - which is reflected in the KPedia via the 'confidence number'.  Once you've done all the tests, the 'confidence number' reads in the 99% range.  

Tells the player they've gotten all the science points they're likely to get from that.

 

The really cool thing - from the "Using KSP to inspire the next wave of Space Scientists" is that it could give them an idea of what's out there, and why.  i.e. you might have a very small temp range at the surface and very different readings at certain altitudes, which could build a chart like this:

profile.jpg

Thus, as the player goes and builds upon their Kerbil's knowledge... very subtly they'd be learning about our own solar system and how stuff works.

Again... that would have to become a 'game within the game' - but incorporating it would make KSP absolutely next level.

DLC?  Would buy.

Link to comment
Share on other sites

@JoeSchmuckatelliI don't think it would be that goliath of an effort to create as it may seem. All of the data to be found in a compendium would already exists in tables in the game. I think it would just largely be UI development based around how that data is displayed and accessed. Effectively the entire compendium would have all the tables/maps/etc already there but just at the games start everything would be presented as blank. Every time you complete an experiment one of those blank spaces would simply reveal the data that was hidden before. This is easy to to with written data in a table form, but for maps I was thinking something like this.

The top image would be the real data while the bottom is a representation of what the user could see given some whacky satellite scanning strangely rectangular portions of a sphere:

Spoiler

cTbNjiN.png

sg95F43.jpg

Originally, the portion of the achieve UI that would hold this map wouldn't be seen until the body is discovered. Not even a blank page, just no page. Once the body is discovered (possibly with an orbital telescope) it would show up as a grey sphere representing the player knows something is there, but there is no clue as to what it is. In the bottom image to the right of the red rectangle is the portion with maximum blur. This is what the body would be covered with using a level 1 scan (advanced telescope from a distance or perhaps a crew report from orbit). Inside the red square would be a level 2 scan (Orbiting probe with a low tech camera). Inside the yellow box would be a level 3 scan (Orbiting probe with a more high tech camera). Finally, the green square, a level 4 scan (which wouldn't show up as a box but perhaps a 20km radius circle around where a site was prospected. Just make it so if you do a level 3 scan in a region that has only had a level one scan it bypasses the level 2 masking layer as well.

It's not a perfect representation, but I think it gets the point across. All I needed was the hard data (top image), then layer copies of that image with a simple blur mask overlayed that I made in photoshop in 5 minutes. Something similar is done in SCANsat and the amount of layers or the ways of uncovering them can very to the devs preference. Furthermore, just have tabs on the page for what kind of map data you would like (altitude, biome, resource) and represent them accordingly.

Another example from an image I posted earlier:

Spoiler

Duna_Atmosphere_T&P.png

9gYM2we.png

Once again, the top image is the full data that would be unknown to the player, stored as a table (not an image) of data points that is then graphed out. The bottom image is a representation of what a player might see (without the printed explanations). Basically the game would record through which portions of the atmosphere the probe was recording what data, and upon transmission that data would unlock the according sections of the spread sheet and allow them to be graphed.

I could be wrong, but I don't see how these things would eat up a sizable portion of dev time compared to the rest of the game that has been built but it would add an insanely immersive science simulation aspect to the game that would have us all digging through data and actually doing science. To top it off, if you don't wanna dig through data... just avoid going to that portion of the game, just like how if you only play sandbox you will never have a reason to go to the control center

Link to comment
Share on other sites

KSP2, I gather, is supposed to have a Lua scripting interface geared for, I suppose, a more entry level modding system short of a full blown C# based mod.

Currently, everything @JoeSchmuckatelliwrites about could be done in kOS, but maybe the Lua scripting interface would be a good way to prototype the idea in KSP2 as a demo to the devs, and if they don't see it as a priority the demo becomes an installable mod.  If it takes off enough and the devs still shows no interest, perhaps the prototype could be perf-bumped via a port into C#

Link to comment
Share on other sites

On 10/29/2022 at 5:34 PM, JoeSchmuckatelli said:

The thing about how you would have to repeat experiments at places - like a temp check at the pad - for reduced science output.....

I would like some more comments on my ideas about the little details I went into about the antenna system, I thought that was a rather engaging gameplay mechanic that gives you several ways to solve any one communications problem.

Anyways, about the quote (couldn't figure out how to put that little beginning statement above the quote that I picked out):

Well, that's why I think all "sensor" experiments should be able to be run in "data-logging" mode, or in other words continuously.
For the thermometer at the launch pad example, not only would that give you more "basic" data to work with (to for example build an idea of what the "average temperature range" is at that location), but it would also allow you to track time-varying trends in the data, which if you have any experience with meteorology then you'll know that having a good network of sensing stations that can be polled for data at any time is one of the central pillars that holds up that whole arm of science with regards to weather forecasting.
So a player might start with just pressure and temperature data, but then they might come back later with a better instrument package that adds in wind speed, wind direction, humidity, and potentially even can make the communications antenna do double-duty as a lightning detector (Lightning bolts, being electric arcs, are a source of very powerful EM radiation, and that includes wide-band radio noise).
How does it make the antenna do that? Easy, antennas are the thing used to send or receive radio waves, and like I just said, lightning bolts put out a lot of radio waves. If there's also a microphone on the weather station, a relatively low-power and low-capacity microprocessor can be used to measure the delay between the lightning strike and when the thunderclap arrives at the weather station, and from that divide by the local atmospheric speed of sound to get a distance to the lightning strike. With a network of such sensors, you can build a map of where lightning has struck recently, and this is done IRL on many weather radar sites (look thru the options on the display to find it, it should be there and not too hard to find).

So what does that mean for KSP 2? Well, to me it means that you could gain either science or get paid to put out a network of weather sensor stations around Kerbin, maybe for max science return or payment you'd need to put down X per biome, with Y meters distance between the sensors. Ideally they would be placed in a way that forms at least one triangle, so just dropping them out the back as you fly overhead in a straight line wouldn't be ideal (you can't triangulate with a straight line unless the sensors can also tell what direction the data came from).

But even more in line with what KSP 2 is all about, you can measure planetary surface weather (and solar weather as well) using satellites and probes. IRL, we technically have a weather satellite around Mars right now (it also does mapping and communications relay duties, NASA seems to use all 3 capabilities pretty much equally).
And that probe we have in Mars orbit has been critical to our ability to further explore Mars with probes. It will also likely be critical to the manned exploration of Mars (or replaced by superior equipment), whenever we're ready to do that.

So basically, the point is this. If you can run a sensor for longer than "spit out one measurement and then don't do anything again until the player interacts with it", the player should get rewarded for that.
However, obviously, sometimes that's not possible because of constraints of communications, time, or power, so the sensor type experiments should retain the ability to be operated in "one-shot" mode as all experiments operate in KSP 1 (this would use a lot less power and not take any appreciable amount of time to transmit the data, but it would also severely limit the amount of science data you'd gain from it to say 5% of the total compared to an experiment that is allowed to run to completion).

If you're getting the idea that I want science in KSP 2 to work more or less a lot like it does in Kerbalism in KSP 1, well that's exactly correct. The main reason I don't use Kerbalism in my KSP 1 games is because it "adds too many things and makes KSP not play like KSP" to me (and yes I realize that that is an opinion, and therefore totally subjective, I'm not trying to start an argument about Kerbalism here).

Experiments that return data over time are good, some experiments can't do that (surface samples are the only one that comes to mind at the moment, and no I don't think the materials bay experiment should be an "instant" experiment because when we sent a "materials bay" into orbit IRL using the Shuttle, it did indeed "run over time" before it was recovered and brought back for study, if you want to look up that experiment it was called the LDEF, or Long Duration Exposure Facility).

Link to comment
Share on other sites

@SciMan

Personally, I enjoy that real antennas take into account things like atmosphere for antenna transmittability for specific frequencies as well as some frequencies can have higher bandwidths with lower range, not to mention the use of gain and its interaction with power consumption. I also really enjoy the idea of having cones of connectivity. I get that all this maybe that may be too niche and more of a mod territory.

As far as the interaction between acquiring data, data storage, and data transmission, I like the idea of having data buffers like hard drives and holding experiments as such (very similar to kerbalisms manner of science data storage).  As for data completion, I think instead of taking multiple samples with the same equipment at the same area to get the full amount of science points feels a bit grindy, forcing players to repeat projects they have already done leaving people with a "can we be done with this and just move one" type of feeling. At least that's my take for very simple experiments like gathering temperature but I do think there should be a tech buildup over time for more involved experiments that have a more central role in other gameplay mechanics.

Maybe an attribute that could be added to science experiments in general could be SNR (signal to noise ratio) where experiment reports could show a graph with error bars or lightly colored wide lines with a darker colored thin line running through the average.  The true data that exists in the game wouldnt be shown, but a noise algorithm could generate a "false" data readout that would appear correct in line with how science is actually not 100% precise. So basically the real data would always be within the error bars/lightly colored line but you wouldn't see exactly where it is.  Kind of like this crude representation I've made:

Spoiler

fZ2CkWC.png

The yellow line is the apparent value which results from the average of the light blue highlighting, which is the error range, and the dark blue line is the actual value the player wouldn't see. Using experiments with greater SNR would result in a tighter blue highlighted area the conforms to the unseen actual values of the blue line.

Basically, I just don't want experiments to feel like pointless grind and the player can feel like they're actually getting something out of it besides squeezing the water(smidgens of extra science points) out of a rock (experiments). At least that's what stock KSP 1 science felt like to me.

Link to comment
Share on other sites

That's true - KSP 1 science is very grindy, especially with being able to collect from all biomes.

My ideal science progression would be milestone driven - in the very early stage you would only need to do obvious tasks, so that simply launching your first rocket with some kind of telemetry part would unlock some parts. Same with suborbital and orbital flights. And afterwards you would need to collect data under more specific circumstances to unlock more parts, e.g. to unlock the first set of colony parts you might have to both do some kind of specific research on a space station for X amount of time plus collect some measurements from a number of different celestial bodies. 

Link to comment
Share on other sites

Just had an idea about making it less grindy. Maybe there are a not small number of experiments that can only be run in certain biomes?

For instance, robotic sample collection. IRL, you can't really get that much science done with only a sample scoop, especially if your land in a rocky riverbed. Your sample scoop simply won't be able to pick up those large rocks, even if your lander or rover is perfectly capable of landing in or traversing over such terrain without issue.
For a rocky riverbed (or other rocky terrain), you'd want something like a rock drill, maybe the kind that extracts a core, maybe the kind that just pulverizes the rock and collects the created rock dust in a small container for conveyance to the sample analyzer system.
If it extracts a core, it's kinda hard to analyze it on-site with robotic equipment, but you could store it and eventually return it to KSC to be analyzed there. If it creates rock dust, that's a lot easier to analyze, probably the best instrument to do that would be a gas chromatograph and mass spectrometer.
Or you could do what Curiosity did, and mount FREAKING LAZER BEAMS that vaporize a small amount of the surface along with a specialized camera that can categorize the spectra of the (probably turned to plasma by the laser) rock vapor, to determine what it's made of.

That laser DOES work pretty much anywhere since with enough heat everything turns into plasma that can then be analyzed, but there's a tradeoff in that you can't really analyze the physical structure of the rock or sand or dust you shot the laser at, all you can do is figure out some of the chemical makeup of it. Additionally, if the sample had any complex chemicals in it (maybe organic in nature?), the laser beam would certainly have caused those chemicals to be decomposed by the intense and highly localized heating, so in your spectrogram you'll be sensing the decomposition products not the chemical itself, and that means there's going to be an amount of uncertainty about what exactly that chemical compound could have been, because you'll only be able to detect "what elements and what relative abundances", not so much "what configurations were the atoms and molecules attached to each other".
Water is easy enough to find with this method, but true evidence of life (complex organic chemicals) needs another experiment that collects samples a different way and doesn't expose them to such harsh heat.

That's rocky terrain covered, what about more sandy or dusty terrain? Well obviously, scoops work a lot better here, and rock drills don't work quite as well (you can switch drill heads to make it work better, if for instance you have a rock grinding drill bit and a simple auger drill bit you'll have better coverage).

Even the Dmagic science experiment mod in KSP 1 has 3 different ways of "sampling the surface to find out what it's made of", and that should be reflected in KSP 2's science system I think.

The days of "just one kind of surface sample" should be long gone, because it's not that simple, different samples take different tools to collect, and each one will tell you different things about the planet.

And the same goes for a bunch of other kinds of experiments and sensors too.

There should even be a tiny ISRU experiment (however for the sake of the game it would be fun if it could also do double duty as a truly probe-scale ISRU, enabling a rocket-powered surface hopper to be made entirely unmanned, that might not be needed if rovers are a lot better in KSP 2 than in KSP 1 but I'd still love the option of a micro-scale low-output low power consumption ISRU suited to be put on just about anything (but it can't synthesize life support resources, so that would become the limiting factor of your vessels unless they're robotic, always gotta consider the tradeoffs).

Link to comment
Share on other sites

I like the idea that, just like engine and tank and payload optimization and choices, there is (and should be) options for antenna-data-battery. Now, moving around data is less exciting in-game than moving around planetary samples in cargo holds (they feel more physical than just numbers of bits), but nonetheless they're a HUGE part of real spacecraft design. It should have an easy option that works well enough for players who don't care...just timewarp through the system inefficiencies, or make very chonky data relay satellites that are like 20 stock all-in-one modules all slapped onto one spacecraft for enough bandwidth. But there should be options for us optimizers!

For planetary science I *loved* the biome aspect, but can appreciate that making each biome on each planet's contributions somewhat unique could be really exhaustive to implement. I wonder if there's a way to double up on the systems...all the interesting ideas people have mentioned here, and a bit of the old "chug out homogenous science points per biome"...like, to research a new colony module, it could cost "Soil samples from 3 different 'highlands' and 1200 science points"...so each research requires some bespoke, some generic currency. Maybe even some colony modules requiring planet-specific researching. A regolith processing module might need to be re-researched on each planet, requiring "soil and temperature samples from 6 local biomes", incentivizing landing, initial setup, exploring, diverse sampling, then colonizing.

Possible grindy addition: planets could have hidden discoveries that have a chance to show up...take a sample on a watery planet's beach, bring it back to a module for analysis, it has a chance to "drop" that beach biome's unique discovery. Want it less grindy? Survey satellite data can direct you to locations with higher "anomaly ratings" where the chance of the discovery is much higher. More advanced survey satellites can convey what kind of anomaly, informing specifically which experiment to bring....so a well-prepared explorer can hone in and target down a unique discovery in a single sampling expedition, while a casual explorer can just be pleasantly surprised.

Link to comment
Share on other sites

Okay, so much to unpack here. Now I love KSP1 and mostly played career, but let's start with the 3 biggest problems with the science implementation: the grind, the grind, and the grind. There are a lot of great ideas above but more than anything we need to keep our eye on the ball here and simplify this equation. We're trying to get new and old players from the first flight, up to orbit, do a few munar missions, and then press on to interplanetary and setting up colonies and stations. If there is an incentive to do more than 6 total science missions for the Mun or Minmus something is wrong. We've fallen into the same repetitive biome-hopping grind trap that meant most players smurfed science from KSOI and never left. 

In designing a game like KSP you're trying to maximize the sphere of solutions with the simplest set of rules and least possible micromanagement. What we're ultimately trying to do is connect up 3 things with the fewest possible steps: maneuvering vessels to new places, increasing the amount of useful information the player has in-game, and unlocking parts. Now, I love those  atmospheric charts, but for most players they just don't mean much. They aren't telling players what they actually want to know: can my vessel survive re-entry at its present trajectory, and where will it land. Rather than barometers and thermometers giving you a graph they should unlock automatic overheat prediction and trajectory factoring drag. The graph is much, much more opaque than data the player can easily see in engine, in map mode. You don't need an encyclopedia. You need flight visualization tools
 

On 11/1/2022 at 6:19 PM, mcwaffles2003 said:

As for data completion, I think instead of taking multiple samples with the same equipment at the same area to get the full amount of science points feels a bit grindy, forcing players to repeat projects they have already done leaving people with a "can we be done with this and just move one" type of feeling.

This is a critical point. Get the grind out. All data collection should transmit at 100%. It's bad enough each body has a dozen different barely discernible biomes, the transmission rate encourages players to visit each multiple times.  And I hear what @SciMan and others are saying about data and transmission but I think even that can be simplified without losing core gameplay. You don't need separate hard drives if you have probe cores. The probe cores just hold whatever information has been collected until they reestablish communications and then transmit it automatically. Each transmission array is rated for a given distance (this also should be much easier to understand in the VAB), and has its own power requirements to operate. So if you've got a probe core, an antenna, and solar or batteries to run it, everything transmits at 100% with zero micromanagement. It doesn't need to be more complicated than that. 

Edited by Pthigrivi
Link to comment
Share on other sites

5 hours ago, Pthigrivi said:

This is a critical point. Get the grind out. All data collection should transmit at 100%. It's bad enough each body has a dozen different barely discernible biomes, the transmission rate encourages players to visit each multiple times.  And I hear what @SciMan and others are saying about data and transmission but I think even that can be simplified without losing core gameplay. You don't need separate hard drives if you have probe cores. The probe cores just hold whatever information has been collected until they reestablish communications and then transmit it automatically. Each transmission array is rated for a given distance (this also should be much easier to understand in the VAB), and has its own power requirements to operate. So if you've got a probe core, an antenna, and solar or batteries to run it, everything transmits at 100% with zero micromanagement. It doesn't need to be more complicated than that. 

This is why I push so hard about the in game encyclopedia thing. It wouldn't be a grind if it became an undefined useful goal. I'm okay with science generating points to push the tech tree along, as a game mechanic, it makes sense. But just because science accomplishes a simple game mechanic goal doesn't mean that should be its only use. Now personally, I wouldn't mind if the comms management got a little more complex, but I understand if that isn't for everyone.

That said, Kerbalism's mechanic of capturing science passively over time, in my opinion, is a real necessity. It pushes players to be able to maintain a condition as opposed to momentarily reaching it which leads to more requiring more creativity, ingenuity, and engineering... which I believe is a good thing as that seems to be a core goal in this game, as well as taking out the scramble to click all the "take data" buttons which feels a bit empty to me.

Link to comment
Share on other sites

On 10/22/2022 at 2:07 PM, Cpt72Bug said:

Experimental modules have to be my favorite part of KSP1. As a Molecular Biologist / Entomologist I'm really curious to see what direction they will go in terms of the included projects for KSP2. Hopefully, on top of more physics and geology scanning parts they integrate more biology scientific projects akan to those we see currently on the international space station. 

 

Interested in hearing what your thoughts are in what may be included for science experiments. Also if you have any references to current released footage I'd be happy to see it posted here.

 

 

Dr. Bug

Reported 

Spoiler

With a nomination for Thread of the Month!  I've really enjoyed the contributions by the the forumites.  Thanks for starting this! 

PS - others; please join me!  Click the three dots on the OP and report with a nomination for TOTM!

 

Edited by JoeSchmuckatelli
Link to comment
Share on other sites

14 hours ago, Pthigrivi said:

(...)

In designing a game like KSP you're trying to maximize the sphere of solutions with the simplest set of rules and least possible micromanagement. What we're ultimately trying to do is connect up 3 things with the fewest possible steps: maneuvering vessels to new places, increasing the amount of useful information the player has in-game, and unlocking parts. Now, I love those  atmospheric charts, but for most players they just don't mean much. They aren't telling players what they actually want to know: can my vessel survive re-entry at its present trajectory, and where will it land. Rather than barometers and thermometers giving you a graph they should unlock automatic overheat prediction and trajectory factoring drag. The graph is much, much more opaque than data the player can easily see in engine, in map mode. You don't need an encyclopedia. You need flight visualization tools.

(...)

This touches upon the biggest problem with Science in the game: its purpose.

Initially we were all in awe with KSP; you can build your rocket, launch it, and land it somewhere else. Then the realization came: now what? Well, Science! Of course just doing Science gets boring as well if there's nothing to do with it. So we get the tech tree. Obviously we want a challenge; welcome to Grindworld!

My only wish for Science in the game is that its purpose is more fluid and natural. I don't have high hopes, given how the Dev team emphasizes that they want to keep a lot of things the same, but I wish it does.

In my perfect world, the solar system is randomly generated (within some boundaries). You'd still have Moho, Eve, Kerbin, Duna, etc, but you don't know their exact orbits; they need to be discovered first, and measured. Moons? Only after putting a probe in orbit. Atmospheric parameters? Only after measuring it. And you will need that data for the right heatshield and parachute properties. Maybe sandbox mode would have everything fixed just like KSP1, but Science mode should just be that; being forced to discover what's out there.

Link to comment
Share on other sites

1 hour ago, Kerbart said:

Atmospheric parameters? Only after measuring it. And you will need that data for the right heatshield and parachute properties. Maybe sandbox mode would have everything fixed just like KSP1, but Science mode should just be that; being forced to discover what's out there.

Love that - would make the educational (and discovery) parts of the game phenomenal.

 

(We know they're not, purposefully, making an educational game.  But KSP turned out to be such - and thus its legacy as an educational game should carry forward into KSP2.  Even if doing so requires a DLC - they really should do Science up right.)

Link to comment
Share on other sites

×
×
  • Create New...