Eclipses!

[SnailsAttack]

I think that solar and lunar eclipses would be really cool. As of right now, any planet that passes infront of Kerbol just appears to pass through/behind it. I don't believe that the implementation would be impossible, but I could be wrong.

This feature is exhibited quite nicely in this video. 

 

Anyways, thanks for reading.

[Blaf]

Not only this, whole vanilla KSP would really use massive graphics upgrade.

[Rocket In My Pocket]

16 minutes ago, Blaf said:

Not only this, whole vanilla KSP would really use massive graphics upgrade.

Not usually a good idea to raise the minimum specs on a game, you're just closing the door on a huge percentage of potential customers with low-end hardware.

Edited January 17, 2018 by Rocket In My Pocket

[John FX]

14 minutes ago, Rocket In My Pocket said:

Not usually a good idea to raise the minimum specs on a game, you're just closing the door on a huge percentage of potential customers with low-end hardware.

It is completely possible to raise the maximum specs without raising the minimum specs.

I support any graphical improvements.

if you use Principia, you get authentic eclipses, at the correct date as well.

[klesh]

3 hours ago, Rocket In My Pocket said:

Not usually a good idea to raise the minimum specs on a game, you're just closing the door on a huge percentage of potential customers with low-end hardware.

 

I would purchase a HD DLC in a heartbeat.   It can have it's own set of min specs and the potato crowd can stick to the terrible graphics.

[Rocket In My Pocket]

8 minutes ago, klesh said:

I would purchase a HD DLC in a heartbeat.   It can have it's own set of min specs and the potato crowd can stick to the terrible graphics.

Is that really what the dev's should waste time on though? (Especially considering the problem has already been solved by modding; which is optional and free.)

Content that everyone can enjoy seems far more important to develop doesn't it? Even from a sales standpoint; a wider potential user-base=more sales.

Keep in mind, my comp is a beast; so I'm not being selfish here.

[klesh]

2 minutes ago, Rocket In My Pocket said:

Is that really what the dev's should waste time on though?

 

 

Yes.  I have several friends who own the game and wont play it because it looks so old.  I have several others that wont buy it because of same.

[Rocket In My Pocket]

12 minutes ago, klesh said:

Yes.  I have several friends who own the game and wont play it because it looks so old.  I have several others that wont buy it because of same.

Yet Nintendo continues to dominate the gaming market with last generation hardware again and again?

The proof is in the pudding. Most people don't care about graphics so long as the game is fun. (Indie games in general also support this, most of them feature graphics that are dated by decades.)

I've also had friends who say the graphics aren't good enough, but they are just deflecting. The honest truth is the game probably just doesn't interest them the same way it does you and me, and that's fine. Now if we're talking about a sequel, then yeah; Most definitely give the game a face lift to "bring it up to par."

Edited January 17, 2018 by Rocket In My Pocket

[pandaman]

I tend to agree with @Rocket In My Pocket on this one.

I started playing KSP because it interests me and i still enjoy doing 'space stuff', not because it looks nice.

Not that better quality graphics and visual treats would not improve my enjoyment, they absolutely would (and do, which is why i installed SVE), and no doubt updated graphics etc would help grab the attention of a few who wouldn't otherwise look.

I don't play ANY game just because it's pretty.  KSPs main appeal (IMO) is the subject matter and gameplay, and that won't change just because it has prettier pictures.

Do I think Squad should improve the visuals and sounds etc?  Yes, definitely, but not at the expense of the minimum specs needed to run it.

By all means let those with the right hardware have the joy of gazing on the magnificent swirling upper atmosphere of Jool through the broken hazy clouds at sunrise on Laythe if they can.  But to deny so many players the challenges and joys of just landing on the Mun just because they don't have a high spec machine would be such a shame when tbe gameplay itself can run without one.

[SnailsAttack]

16 minutes ago, pandaman said:

By all means let those with the right hardware have the joy of gazing on the magnificent swirling upper atmosphere of Jool through the broken hazy clouds at sunrise on Laythe if they can.  But to deny so many players the challenges and joys of just landing on the Mun just because they don't have a high spec machine would be such a shame when tbe gameplay itself can run without one.

I'm not sure how my suggestion for celestial bodies blocking the sun became a whole discussion on redoing the game's graphics from the ground up and ruining the experience for low-end computers. I'm not asking for a full graphical redux, just this one feature that I think could be easily implemented without a high cost for performance.

[Snark]

45 minutes ago, SnailsAttack said:

I'm not sure how my suggestion for celestial bodies blocking the sun became a whole discussion on redoing the game's graphics from the ground up

Because implementing eclipses would require redoing the game's graphics from the ground up. 

45 minutes ago, SnailsAttack said:

ruining the experience for low-end computers

Well, either they'd have to change how KSP works for everyone (which would ruin the experience for low-end computers), or else they'd have to implement two different graphics modes so that players could choose which one to use, based on their hardware.

The latter would seem to be an "everyone wins" from the players' perspective ... except that nothing in life is free.  That latter approach would require implementing more stuff, which means more developer time, which means time not spent developing something else that players would also like to have.  And the feature would only benefit people with high-end machines, which might not be most of the player base-- as opposed to gameplay enhancements that everyone could enjoy.

So, from the developer's perspective:  You can't implement everything-- you have to pick and choose, which means you have to prioritize.  Bang for the buck.  "What feature delivers the most player value for a given amount of engineering input?"  If a feature is expensive to implement, and/or benefits only a small fraction of players, and/or doesn't provide as much perceived "play benefit" as some other feature ... then it becomes less attractive to the developer to implement.

50 minutes ago, SnailsAttack said:

I'm not asking for a full graphical redux

Except that you kinda are, even if you don't think you are.    The game's current graphics model doesn't support eclipses.  You have to get fancy with it, which means making a lot of changes.

50 minutes ago, SnailsAttack said:

I think could be easily implemented

And you base this opinion on how many years of experience as a software engineer working on graphics systems?

(Not criticizing-- just pointing out that technical professions are hard, and people who haven't done it for a living typically aren't in a position to say how hard it would be.)

51 minutes ago, SnailsAttack said:

without a high cost for performance

And you base this on what experience, doing 3D graphics and complex lighting models?

(Again, not criticizing-- but lighting is expensive, in terms of required graphics horsepower.  Without having done that sort of thing for a living, it's hard for the layperson to know just how much performance a given feature would require.  Some things that may seem simple to you are actually quite demanding of the hardware.  Complex lighting and shadows happens to be one of those areas.)

[SnailsAttack]

37 minutes ago, Snark said:

And you base this opinion on how many years of experience as a software engineer working on graphics systems?

(Not criticizing-- just pointing out that technical professions are hard, and people who haven't done it for a living typically aren't in a position to say how hard it would be.)

And you base this on what experience, doing 3D graphics and complex lighting models?

(Again, not criticizing-- but lighting is expensive, in terms of required graphics horsepower.  Without having done that sort of thing for a living, it's hard for the layperson to know just how much performance a given feature would require.  Some things that may seem simple to you are actually quite demanding of the hardware.  Complex lighting and shadows happens to be one of those areas.)

thank you for your unreasonably toned response

[Blaf]

23 hours ago, Rocket In My Pocket said:

Not usually a good idea to raise the minimum specs on a game, you're just closing the door on a huge percentage of potential customers with low-end hardware.

I take this argument more as a trade - on one hand losing potato users, on the other gaining customers for which better graphics is a decision factor to buy the game.

[SnailsAttack]

17 minutes ago, Blaf said:

I take this argument more as a trade - on one hand losing potato users, on the other gaining customers for which better graphics is a decision factor to buy the game.

I don’t think it has to be a trade-off. Non-dynamic, pre-calculated darkening of a certain planetary zone (such as, i don’t know, the night side) seems like it’d be possible without a full game recode/redux.

Even if this was a cpu-intensive feature, it could be disabled with an option in the settings.

[John FX]

16 hours ago, Snark said:

The game's current graphics model doesn't support eclipses. 

And yet we have them without a change to the game's code, just a mod. The current code does not, but it can be made to with a mod.

This suggests an `eclipse module` could be made which is extra to the code, like the asteroid mod which added features and parts but was optional.

Then those with better hardware could have a better experience, and those with a potato can play a lower quality game.

I agree that nothing is for free so the addon could be for a price. I would pay for clouds, eclipses, HD textures, more complex particle effects, better flame exhausts. All these are possible with not a byte of change to the main game's code. We know this because it has been done.

No doors closed, no need to change the base code, coding paid for. All optional.

win-win.

 

EDIT : Also, If you feel the need to say more than once in a post <not criticizing> then you might just be being over critical. Especially if you are a mod...

Edited January 18, 2018 by John FX

[Rocket In My Pocket]

1 hour ago, Blaf said:

I take this argument more as a trade - on one hand losing potato users, on the other gaining customers for which better graphics is a decision factor to buy the game.

Sure, but we already have better graphics options available for those people, through modding.

Look at some of the more famous computer game success stories: World of Warcraft, League of Legends, Minecraft. What do they have in common? Simple graphics that can run on just about anything. If the only thing that can get some people to play KSP is "look, shiny graphics!" I really doubt they'd play for long anyways before getting bored, and moving on to the next, shinier game.

Getting back on topic here...The eclipse shown in the video looks quite nice; if it could be done cheaply and quietly then sure. It's not so amazing or important that I think they should prioritize it though.

[Snark]

3 hours ago, Blaf said:

I take this argument more as a trade - on one hand losing potato users, on the other gaining customers for which better graphics is a decision factor to buy the game.

Sure-- pretty much everything is a trade-off.    The question is this:  is it a profitable trade-off?

Based on my own experience with how much GPU horsepower it takes to really turn up the knobs on graphics F/X in KSP, I'd guess that there are a lot more "potato" users out there than there are "graphics mavens" with top-end gaming rigs.  (And when I say "potato", I don't mean only the people with really bargain-basement, low-end machines-- I'm including people with "fairly average desktop PCs", even ones with slightly-better-than-average graphics cards in them.)

My guess is that this is a trade-off that wouldn't be worth the developers' while-- I think it would be a really bad tradeoff for them, for two main reasons:

• Size of target audience.  They'd be appealing to a fairly small audience, at the cost of alienating a very large target audience.
• Impact.  Switching to a very graphics-intensive model would appeal to the high-end users, sure... but they're not being excluded without it.  I expect that those same people would be happier with fancy graphics, but they'd probably be buying KSP even without it.  On the other hand, going that way would completely exclude all the folks who don't have top-end gaming rigs.

So I think any argument to radically boost KSP's graphics to target folks who only have top-end machines would be a non-starter.  Realistically, they have to target the bulk of potential players.  So if they want to cater to the top end, they'd have to implement it in a way that lets people toggle the settings to whatever suits their hardware.

(Of course, the market isn't static, either.  Hardware continues to get faster and cheaper; the "average" PC graphics card today is a whole lot better than it was five years ago.  So at some point I expect it would be worth their while to give the game a general graphics pass just to keep up with the march of progress.)

3 hours ago, SnailsAttack said:

I don’t think it has to be a trade-off. Non-dynamic, pre-calculated darkening of a certain planetary zone (such as, i don’t know, the night side) seems like it’d be possible without a full game recode/redux.

Again:  do you have enough of a technical background to understand the details of exactly what you're asking, here?  Because my impression is that the game's current lighting model doesn't allow a simple/trivial fix to do what you want.

It's always great to suggest "hey, I think X would be nice from the player's perspective", and to ask whether it seems doable.  But asserting that it's easy to do generally requires providing some technical justification for the assertion.

Note that I'm not saying that I know what you're asking for ("eclipses, for trivial investment of engineering") isn't doable.  Merely that I fairly strongly suspect that it won't (given my technical background), and that you haven't provided any specific technical rationale to indicate that it would be.

 

2 hours ago, John FX said:

And yet we have them without a change to the game's code, just a mod. The current code does not, but it can be made to with a mod.

Sure, but that's basically a red herring that has no real significance.  The game can be made to do this with "just a mod" because it's a remarkably open and flexible game that can be made to do just about anything with "just a mod".  There are a few things that are hard-coded in the core game that are physically impossible to change with a mod-- but most stuff isn't like that.

In implementation terms, "just a mod" is a distinction without a difference.  The only difference between "it's just a mod" versus "it's part of the stock game" (from the developer's implementation perspective) is, "does the code reside in Assembly-CSharp.dll, or does it reside in some other DLL file."  It's just packaging, like Amazon deciding to ship your order in two boxes instead of just one.  Squad's free to put whatever code they want into however many packages they want.  That kind of packaging decision is fairly trivial-- i.e. moving code from one DLL to another isn't much engineering time.

2 hours ago, John FX said:

This suggests an `eclipse module` could be made which is extra to the code, like the asteroid mod which added features and parts but was optional.

Then those with better hardware could have a better experience, and those with a potato can play a lower quality game.

Absolutely.  But this is basically exactly the same as what I said,

19 hours ago, Snark said:

they'd have to implement two different graphics modes so that players could choose which one to use, based on their hardware.

• They could make it part of the stock game, and a user chooses pretty-stuff-or-not by selecting an option from game settings.
• Or they could make it an optional add-on, like Asteroid Day, and the user chooses pretty-stuff-or-not by deciding whether or not to install it.

The development time is exactly the same either way for Squad.  The benefit to the player is exactly the same either way.  The fraction of users who could and would choose to use the pretty stuff would be exactly the same either way.

So they amount to pretty much the same thing.  Yes, there are a few minor differences around the edges-- e.g. the former may be a little friendlier to players (no need to install an add-on), the latter might be a little friendlier to Squad (would have the freedom to rev the version independently of the stock game, and perhaps could charge for it if they wanted).  But by and large, it boils down to the same thing.

2 hours ago, John FX said:

I agree that nothing is for free so the addon could be for a price. I would pay for clouds, eclipses, HD textures, more complex particle effects, better flame exhausts. All these are possible with not a byte of change to the main game's code. We know this because it has been done.

No doors closed, no need to change the base code, coding paid for. All optional.

Sure.  But remember, something like that won't happen unless it seems like a good idea from Squad's / T2's perspective, based on the business model.

Absolutely they could develop it independently, and release it as paid DLC.  The question is, how many people would buy it, compared with how many people would buy some other sort of DLC such as Making History.  A "general" DLC like Making History has, as its target audience, basically every single KSP player.  Whereas a graphics-enhancement DLC would only be able to target the players with high-end machines.  So the question from the developer's perspective is:  if I have a finite number of engineer-hours to spend developing something, which of these would be better worth our while, in terms of return on investment?

If they develop a graphics-enhancement paid DLC, they can only target a relatively small fraction of the KSP audience.  I would imagine that to make up for that, they'd need to charge a fairly high price per copy.  But I would also imagine that the presence of free graphics-enhancement mods out there, like scatterer and SVE, would tend to limit how much they could charge.

So, would it be worth the tradeoff to them?  Not just "could they make money", but "could they make more money than developing some more general-audience type of DLC"?  I dunno.  It's an interesting question.  Gotta say I'm a  little skeptical, but since I don't have any more access to their marketing and sales numbers than you do, I'm not in any position to give an authoritative answer to that.

[Melfice]

Can we not refer to people who are on a lower end of processing and graphical power as "potato users", or similar?

Incredibly demeaning.

[technicalfool]

3 hours ago, John FX said:

And yet we have them without a change to the game's code, just a mod. The current code does not, but it can be made to with a mod.

Scatterer is a pretty neat mod, however it didn't get to where it is today in one step. There have been some awful glitches on its way to the GPU-battering prettiness that it's at now, and I'm not sure whether the dev team really want to deal with the hassle that would come from a whole new round of development, this time with Cg/HLSL. Not when they could be working on core game features and the Making History expansion.

As I recall also, Scatterer's shaders are all based on shaders that are publically available but for noncommercial use. I'm sure @blackrack would be able to correct me if wrong, there. So while it might be possible for a mod author to grab some available shaders, inject them into the game and tweak them until they fit, that's not quite so possible for a commercial venture. Not unless either some actually-free shaders become available, or a whole lot of work gets done to make shaders from scratch.

So while I won't say it's absolutely impossible for a shader and HD-textures update to happen, it's possibly not the highest priority for development right now.

[John FX]

1 hour ago, technicalfool said:

Not when they could be working on core game features

Sorry but this gave me a chuckle.

 

1 hour ago, technicalfool said:

while I won't say it's absolutely impossible for a shader and HD-textures update to happen, it's possibly not the highest priority for development right now.

I have no doubt that a graphical update (which has been called for for years with no result) is very far back in the list of priorities. I was more replying to the effect that it is possible that the core code would not need to change (although likely it would), and funding would not need to be found from main game sales, which seem to be declining. The actual validity financially of doing so and likelihood of it being done is something nobody here can comment on, either through lack of information or an NDA. My suspicion is that the validity of being able to do so is very low.

I am fully aware that nothing will happen unless it is the most profit making thing. As a business it would be daft to do otherwise and I have no delusions about this.

I have previously said on the forum that I apply a little test to features or suggestions to see if they are likely to appear in this version of KSP.

Will doing something else bring more profit for the amount of effort than X, if so X will not be done.

If doing X is worth the expenditure in time etc, will waiting to release X until KSP 2 bring more profit than if released for KSP 1? If so, X will not be done for KSP 1.

Obviously these are dynamic and change with new information.

As such I do not think we will see a graphical update for KSP 1 because I think a graphical update would bring in far more sales for KSP 2. We can however have clouds, eclipses and the rest now by leveraging the very open code via mods, however unlikely it is for that to happen in stock.

Eclipses for KSP 1 is something I would say to download a mod for, IMHO it is wasted energy calling for it to be included in the stock KSP 1.

[FleshJeb]

I think we're all forgetting Rule Zero:

"Never ask Squad to implement a feature that a mod already does better, for it shall be buggier, and less polished."

<Insert long screed (referencing Snark's points regarding financial ROI) as to why John Maynard Keynes was right.>

[SnailsAttack]

4 hours ago, Snark said:

Sure-- pretty much everything is a trade-off.    The question is this:  is it a profitable trade-off?

With the logic path of "is it profitable?" you could probably give the finger to any number of the more content-heavy suggestions on the forums. One example of that would be weather, which is a good, if difficult, suggestion to implement. It'd add a lot to the game and be pretty interesting, but yes, it would take up a significant chunk of dev time.

Eclipses have proven to be possible to code (to some extent) with a simple mod that was never even completed. I'm not sure how powerful the user's computer was or how far something like an eclipse could be optimized, but I'm sure that some visual trickery could be used to lighten the load a bit. And of course it could just be an optional setting, although eclipses could also play a role in gameplay (e.g. solar panels getting blocked)

If the cost of implementation versus profitability is that important, then the devs could just throw in a handful of extra graphical effects or whatnot (bloom, clouds, whatever, along with some bug fixes and other minor features) and advertise them on the game's store page. I've heard that the primary reasons some people don't buy the game is because it looks like it was made in 2008 (which it does) and because the game has a lot of bugs (which it does). The game could use a polishing update.

I guess if you had to you could make it a DLC, but microtransactions are a dangerous path to go down.

[blackrack]

On 1/18/2018 at 6:53 PM, technicalfool said:

As I recall also, Scatterer's shaders are all based on shaders that are publically available but for noncommercial use. I'm sure @blackrack would be able to correct me if wrong, there. So while it might be possible for a mod author to grab some available shaders, inject them into the game and tweak them until they fit, that's not quite so possible for a commercial venture. Not unless either some actually-free shaders become available, or a whole lot of work gets done to make shaders from scratch.

You're right but there is enough resources out there that they could hire one experienced graphics programmer and he could implement all of this from scratch in a few months of full-time work.

[Capt. Hunt]

Eclipses happen every new mun in game.

Edited January 30, 2018 by Capt. Hunt

[Onigato]

5 hours ago, Capt. Hunt said:

Eclipses happen every full mun in game.

This^^^

They may not always happen over the KSC, but because the Mun is in a perfectly circular orbit, beautiful eclipses happen on a very regular basis. They may not have all the scattering effects that some mods add in, but that Munar shadow does cover the ground, and has a pretty big footprint when it does. I don't have any graphical mods installed, and I've seen dozens of full eclipses, some have been rather annoyingly timed and killed kerbals, but the OPs video could easily be from vanilla, graphically.