Jump to content

Do you think KSP 2 will eventually end up more optimized?


RandomGrape

Recommended Posts

8 hours ago, Scarecrow71 said:

CBT or HDRP, which have no defined timelines.

I believe it was CBT which would succeed the PQS+ system currently implemented. I believe there was some mention of a timeline for this new system but I might be hallucinating here... worst case scenario would be a few years.

HDRP is a render pipeline which would not have an effect on the custom generated meshes of the terrain systems. Although KSP2 graphics is fantastic, the game might still be using the deferred default render pipeline (which would be horrible) but at least it means an easy transition to hdrp would significantly increase framerate. The reason I believe the game is not currently using URP instead of the deferred render pipeline is because of the large amount of lights that can be rendered on terrain (URP as of the blog post could not support more than 8+ point lights in one area). The deferred renderer is the likely other option as it isn't very difficult to create custom shaders to create good graphics, e.g lens flares, however it is more realistic to believe that ksp2 already uses hdrp as ksp2 can use 100% of gpu while games with deferred renderer cant normally.

Link to comment
Share on other sites

Given that part count and the 1000000000000000000000000000+ parts  "promise" came up again I think this (and the discussion that followed) is relevant:

On 4/9/2023 at 11:01 AM, Master39 said:

 

Please read my whole message. You'll discover I've already addressed both your points.

 

Point is, if you listen to the reply Nate gave to Shadowzone he treads around the thing without committing to anything specific. He outright states that you'll be able to make the game slow down with high part counts, only saying that the number of parts needed for that is going to be higher.

 

kerbal_space_program_2_2.jpg

I see this thrown around a lot, but actually look at that ship, a crew cabin, something behind it, 4 rings, 2 big habitats, maybe something else in the middle, let's say 10 parts and we've already finished the crew section.
Let's say the long structure is not modular and made up of 5 segments, let's make it 10, we're at 20 parts and 2/3 down the ship, 6 containers in a 6x symmetry (I don't see the ones below, but, let's pretend), another 36 parts, we're at 56, then we have 3 tanks in another 6x simmetry, 74 parts so far, 82 with the bigger tanks, 83 with the single-part huge engine.

Now, on the station, I see 160 parts in panels alone, and with the truss structures, if they're as small as they appear, we can get up to 250/270 parts easily. To that we add 24 huge spherical tanks, 24 containers, 8 big habitats, 9 rings, 2 smaller inflatable modules, 1 orbital VAB (the whole assembly the big ship is attached to) and, why not, an additional bonus 100 parts for whatever that thingie at the end of the rings is plus antennas, RCS thrusters and all the small bits not present in this image, another 168 parts to the count.

We have 524 parts.

And we can cut half of those out if we have procedural trusses, solar panels, or if we scrap the ISS look and do the right thing by bringing a nuclear reactor to our fusion fueled colonization ship shipyard which is sitting in jool orbit.

 

It may not be a huge difference, but we're sitting at half of what people is saying was promised with that image, and half of that are aesthetics (again, solar panels on a fusion-power capable station in Jool orbit)

 

The most important thing I take away from counting the parts is that barely a third of the parts are actually functional, the rest is all piling on of smaller parts to get to the desired size and a ridiculous amount of solar panels in a 8x symmetry (seriously, it looks like trying to use the OX-STAT to make the ISS arrays). The ability of welding truss segments together, or them being procedural is going to dramatically reduce the part count of everything, the same goes for solar arrays. 

 

Not saying that the game shouldn't be able to deal with 1000 parts ships, only that optimization goes both ways, a lot of effort goes into reducing the amount of parts needed. The orbital VAB alone is going to do a lot for that, given that most user examples of 1000+ ships posted here are on the launchpad and including the rocket needed to take those contraptions to space (BTW I get the challenge of launching an already assembled ISS from the ground, but docking ports are a thing, no need to launch a single 1000 part rocket when you could launch 3 400 parts ones).

 

TLDR: In the famous promise Nate said they were working  on allowing the player to build as big as the things shown in trailers. The "thing" shown in trailers was barely 524 parts, and that counting the spam of solar panels, with procedural solar panels you could easily cut another 150 parts from there (or just put a single nuclear reactor and cut all of the solar panels).

Link to comment
Share on other sites

4 hours ago, Master39 said:

TLDR: In the famous promise Nate said they were working  on allowing the player to build as big as the things shown in trailers. The "thing" shown in trailers was barely 524 parts, and that counting the spam of solar panels, with procedural solar panels you could easily cut another 150 parts from there (or just put a single nuclear reactor and cut all of the solar panels).

If it was about the fact that for players 500 parts is the limit and more is not necessary, then the developers should have said this directly. I think the reaction of the fans would not be too happy. After all, that interview was in 2019, KSP2 was supposed to be released on the last generation consoles with not the highest performance. In general, it would be better if the developers spoke directly more often so that later we would not be engaged in a long decoding of what they actually said.

Link to comment
Share on other sites

4 hours ago, Master39 said:

 

TLDR: In the famous promise Nate said they were working  on allowing the player to build as big as the things shown in trailers. The "thing" shown in trailers was barely 524 parts, and that counting the spam of solar panels, with procedural solar panels you could easily cut another 150 parts from there (or just put a single nuclear reactor and cut all of the solar panels).

 

http://forum.purdueseds.space/pspodcast/episode2/ has Nate in it, where he states:

Quote

~1:14:00 - ~1:17:30

"In KSP1 if your ship has 1000 parts; your computer no matter what isn't having a good time. That's one of the big boulders that we're breaking apart: making sure that framerate/performance does not suffer. The scale is so much bigger than KSP1 [...] with so many ships there's so much more to maintain"

 

Link to comment
Share on other sites

8 hours ago, PDCWolf said:

http://forum.purdueseds.space/pspodcast/episode2/ has Nate in it, where he states:

I'd really like anyone to click on that link and listen for themselves who's talking about what, I'd begin maybe 30 seconds earlier,  1:13:30 to really get the full context.
 

Spoiler

The misquote is a mesh between a statement from the interviewer (speaking the part up until "good time"),  and the start of a reply from Paul Furio (not that matters much but that's definitely not Nate Simpson), who changes topic immediately to speak about performance in general and relative to the background simulation of multiple ships/colonies, not committing nor repeating any actual part count.

 

Edited by Master39
Link to comment
Share on other sites

4 minutes ago, Master39 said:

The misquote is a mesh between a statement from the interviewer (speaking the part up until "good time",  and the start of a reply from Paul Furio (not that matters but definitely not Nate), who changes topic immediately to speak about performance in general and relative to the background simulation of multiple ships/colonies, not committing nor repeating any actual part count.

I think as part of increasing transparency, developers should finally announce some kind of performance targets. For example, for an equivalent PS5 computer, 30 fps per 1000 parts. Although, with such evasions from answers, it is somewhat naive to expect this.

Link to comment
Share on other sites

3 hours ago, Master39 said:

I'd really like anyone to click on that link and listen for themselves who's talking about what, I'd begin maybe 30 seconds earlier,  1:13:30 to really get the full context.
 

  Reveal hidden contents

The misquote is a mesh between a statement from the interviewer (speaking the part up until "good time"),  and the start of a reply from Paul Furio (not that matters much but that's definitely not Nate Simpson), who changes topic immediately to speak about performance in general and relative to the background simulation of multiple ships/colonies, not committing nor repeating any actual part count.

 

It's a pre-recorded interview, and they surely (like any business doing PR) had time to review the questions or at least know beforehand what topics were gonna be discussed. How you interpret what they say is up to you, but they are both there in the interview, answering and following up on what the host said.

Now, regarding what was said, or implied: this is definitely not the "1000 parts promise", for sure, but it's the closest we've had, and it is yet another one of the probably hundreds of quotes relating to their performance/partcount targets, all of which had been nebulous as best, and have, at the very least, been drawn from what the community desired to craft said statements. You must remember that the current narrative of "1000 part ships", "reworked core systems", "new codebase", and so on, all comes from stuff they've either directly said, heavily implied, or at least lightly implied from their answers. They're 100% responsible for the expectations they let build up without managing and in some cases even continue building up, and are also responsible from what backlash comes from those.

If you ask me for a new car and I said "well, I'm working on something with four wheels and five doors", and then showed up at christmas with a hotwheels, you wouldn't be happy.

Link to comment
Share on other sites

5 hours ago, PDCWolf said:

It's a pre-recorded interview, and they surely (like any business doing PR) had time to review the questions or at least know beforehand what topics were gonna be discussed. How you interpret what they say is up to you, but they are both there in the interview, answering and following up on what the host said.

The amount of "we can't talk about that" and "no comments", and the average (low) quality and seemingly improvised nature of most of Nate interviews makes me think otherwise. Here they're talking on a podcast that was new at the time, not CNN.

 

17 hours ago, PDCWolf said:

has Nate in it, where he states:

Implies it's Nate talking.

 

17 hours ago, PDCWolf said:

~1:14:00 - ~1:17:30

"In KSP1 if your ship has 1000 parts; your computer no matter what isn't having a good time. That's one of the big boulders that we're breaking apart: making sure that framerate/performance does not suffer. The scale is so much bigger than KSP1 [...] with so many ships there's so much more to maintain"

Implies it's only one person saying all of this, and that person being Nate. Conveniently you've also cut out all the discussion about the background simulation.

That's misleading at best.

 

5 hours ago, PDCWolf said:

Now, regarding what was said, or implied: this is definitely not the "1000 parts promise", for sure, but it's the closest we've had, and it is yet another one of the probably hundreds of quotes relating to their performance/partcount targets, all of which had been nebulous as best, and have, at the very least, been drawn from what the community desired to craft said statements. You must remember that the current narrative of "1000 part ships", "reworked core systems", "new codebase", and so on, all comes from stuff they've either directly said, heavily implied, or at least lightly implied from their answers. They're 100% responsible for the expectations they let build up without managing and in some cases even continue building up, and are also responsible from what backlash comes from those.

There's a whole argument to be had about the gaming community making up Devs promises and then getting angry at things they've imagined while they over-hyped themselves to oblivion.

A big part of it is that anyone trying to deflate hype and debunk fake lies is seen as an enemy twice at first then they don't blindly believe at everything the hype machine says, pointing out that nobody ever talked about something, or confirmed anything. And then after the game released when they say that the thing they where hyped about was never even mentioned by any actual dev.

Then someone makes a list of lies on Reddit, or Crowbcat makes a video, both of which will be 90% wrong, but still saying it automatically puts you in a "you're the enemy" position.

I'm still not saying that it's the case here, I'm more than open to receive a definitive piece of evidence, a recording or post from Nate saying "100000+ parts", but weirdly enough while everyone seems to agree that that was promised, the only examples are 2 interviews in which the devs clearly evade the question and are very careful at not committing to anything.

 

 

5 hours ago, PDCWolf said:

If you ask me for a new car and I said "well, I'm working on something with four wheels and five doors", and then showed up at christmas with a hotwheels, you wouldn't be happy.

Interviewer: "Mr. Dev, will your car have 5 doors?"

Mr. Dev: "Well, doors, uh? It's a thing we're working on... You see, you can't have a car without doors, and we're working on this new type of hinges that will optimize the door utilization, allowing us to have a number doors on our car."

The community: "You've heard that? They said the car will have 10 doors! And that hinges must be needed to flap the doors and fly, so fully autonomous flight capabilities too"

The car releases, it has 3 doors, the community is mad because it can't fly.

Link to comment
Share on other sites

Whether they like it or not, they have to make it performant otherwise it won't get through Sony's certification process when they start working on a port.

Or figure out a workaround, you know how you can lose frames once your city in cities skylines gets larger? PS5 edition solved it by literally slowing down the simulation to keep the framerate stable. ish. But I don't think slowing down time is an option when there's multiplayer in question. So optimizations.

Link to comment
Share on other sites

2 hours ago, Master39 said:

The car releases, it has 3 doors, the community is mad because it can't fly.

Shortly before the release of the car, the developers announced that the product is not yet ready and will be released in early access. The roadmap states that the doors will be added first (while the car will drive without doors), then the fourth wheel will be added, the third step will be the addition of headlights and parking lights. In the fourth step, the developers will add comfortable seats, and at the very end, the gas tank will be increased from 20 liters to 150 liters, which will allow the driver to drive much further. In the future, the developers plan to release a car for trips out of town. This is more like the real situation.

2 hours ago, Master39 said:

I'm still not saying that it's the case here, I'm more than open to receive a definitive piece of evidence, a recording or post from Nate saying "100000+ parts", but weirdly enough while everyone seems to agree that that was promised, the only examples are 2 interviews in which the devs clearly evade the question and are very careful at not committing to anything.

Adding more zeros in an attempt to reduce to absurdity doesn't help much. In interview to Shadowzone, Nate made it very clear that they are aiming for an acceptable frame rate for ships of more than a thousand parts.

Link to comment
Share on other sites

7 minutes ago, Alexoff said:

Nate made it very clear that they are aiming for an acceptable frame rate for ships of more than a thousand parts.

Nope, not even close, he pretty much evaded the question, just as I said.

Quote

"1000+ parts ships?"
"The CPU [rant about being CPU bound ]  ...making sure people on 'normal hardware' are able to build things on the same scale and complexity as what we've shown in the trailers without a noticeable itch in the framerate... [rant about optimization, physics LOD] you'll still be able to bring it to its knees because we're not going to put a hard constraint"

 

To me it sounds as someone that doesn't want to say that 1000 parts ships are still going to be a problem, but I'm sure I'm biased. Sure enough it doesn't sound like a implying a "yes" to the question, let alone actually promising 1000 parts ships.

Link to comment
Share on other sites

14 minutes ago, Master39 said:

Nope, not even close, he pretty much evaded the question, just as I said.

This is very close to the situation with KSP2. After all, in the phrase about the car, it was about the game in general and its advertising from Nate, in which he always shied away from specifics. I don't know what fans can be happy about, what was made of murky promises? Tutorials and the release of the game at least in some form after three years of polishing? Better performance than the KSP1 is just one of many.

 

Link to comment
Share on other sites

1 hour ago, Alexoff said:

Tutorials and the release of the game at least in some form after three years of polishing?

I highly doubt they had 3 years of polishing. Something happened in between.

Link to comment
Share on other sites

4 hours ago, Master39 said:

The amount of "we can't talk about that" and "no comments", and the average (low) quality and seemingly improvised nature of most of Nate interviews makes me think otherwise.

No, those two sentences are the exact ones they have to use when asked about stuff they can't talk about, by a myriad of reasons. In fact, those are the phrases they should've used when the community mentions "1000 parts" or when they were asked in the interview.

4 hours ago, Master39 said:

That's misleading at best.

Did you listen to the podcast? Since you ask about context, we can move a minute back, starting from around 1:12:00 and end after the full quote. Nate and Paul are (were, since Paul got fired) a team, and it seems to me you're assuming they went in blind and got assaulted with questions and were incapable of saying no to this one question in particular, or for Nate to be incapable of stopping Paul. Well, not only do I not believe any of that, but they're also a team representing the same company and product, so what one ways, unless the other interjects and denies, goes.

Interviewer: You kinda touched on this earlier. Have any features from mods inspired features in KSP2?

Answer (Nate): Yes, the easiest answer is visual fidelity, it needs to feel epic, and there've been a number of visual mods for KSP that have raised the bar regarding what's possible. [...] Eve and Scatterer [...] at least show what the minimum should be and we want to exceed that drastically. We talked about parts mods as well, and when you're making a game that has a bunch of interstellar class engines, Nertea has set the bar very high. We need to be about as realistic and detailed. Is there anything else that pops into your head Paul?

Paul at 1:13:35: We're working with some graphics engineers not only to make our game more beautiful but again performant, we've got some numbers in this week for what our expectations are on machines and holy **** it looks great and performs great. [...]

Interviewer: That's awesome, specially when you have these massive arrays of rigidbody parts. If you have a 1000 parts craft in KSP1 your computer is not gonna be having a good time, no matter what computer it is.

Paul again: That's one of the big boulders we're breaking apart on the engineering team, making sure that framerate performance does not suffer,  I mean look, the scale of KSP2 is so much larger than KSP1, with so many more orbital bodies and potentially so many more ships and colonies doing autonomous background systems there's so much more to maintain, there's so many more systems that are just living in simulation, so we've gotta make sure the thing you're seeing on screen is behaving in a physically accurate and interesting and educational manner that makes sense and is still fun gameplay, but then all these things in the background are still doing what they're doing, is something is in some geosynchronous orbit and you back to it a year later it has be in the right place considering where it is in time, no matter how many times you're timewarping, no matter how many other colonies you have, how many other ships you have or are being built. So making sure all of that feels consistent while the thing that you're doing right now: to have fun or explore or build or launch or blow up in some spectacular fashion is also also there and awesome and feels tactile and realistic, like that is the number one challenge for the engineering team right now.

4 hours ago, Master39 said:

Interviewer: "Mr. Dev, will your car have 5 doors?"

I mean if you just change my example to say whatever else, yeah, sure. That's not what happened, and you have the quote up there to read, and years of evidence of them promising a finished, performant product all over.

2 hours ago, The Aziz said:

Whether they like it or not, they have to make it performant otherwise it won't get through Sony's certification process when they start working on a port.

Or figure out a workaround, you know how you can lose frames once your city in cities skylines gets larger? PS5 edition solved it by literally slowing down the simulation to keep the framerate stable. ish. But I don't think slowing down time is an option when there's multiplayer in question. So optimizations.

Sony is not half as demanding as you make it sound though, even for first party titles (Bloodborne comes to mind), heck, even the KSP1 port on the PS4 was painful.

Just now, cocoscacao said:

I highly doubt they had 3 years of polishing. Something happened in between.

Yes, but they've decided to not say anything happened, so since we're working with their textual words to discount "1000 part ships" promises and others, we can't quote them saying something happened, we can only quote them saying KSP2 will come out performant, or if you ignore the last year, a full product and performant.

Link to comment
Share on other sites

18 minutes ago, cocoscacao said:

I highly doubt they had 3 years of polishing. Something happened in between.

Some say that the game was restarted from scratch. If this is so, then management has made very serious mistakes. By the way, where is this management now? But since the developers didn’t say anything like that and didn’t lie to us, it means that this didn’t happen.

Link to comment
Share on other sites

8 hours ago, PDCWolf said:

Did you listen to the podcast?

The day it released, that's why I was able to spot your BS before even opening the link.

 

8 hours ago, PDCWolf said:

Since you ask about context, we can move a minute back, starting from around 1:12:00 and end after the full quote. Nate and Paul are (were, since Paul got fired) a team, and it seems to me you're assuming they went in blind and got assaulted with questions and were incapable of saying no to this one question in particular, or for Nate to be incapable of stopping Paul. Well, not only do I not believe any of that, but they're also a team representing the same company and product, so what one ways, unless the other interjects and denies, goes.

Don't try to spin things around.

You said:

On 7/8/2023 at 6:51 AM, PDCWolf said:

has Nate in it, where he states:

And then proceeded to quote half of the question form the interviewer and half the reply from another person, very much not Nate Simpson.

That's misleading and dishonest.

I can understand an argument built on interpretation of vague answers, but straight faking quotes banking on the fact that nobody is going to bother with verifying them disquilfies you from the  discussion in my eyes.

[snip]

Edited by Vanamonde
Link to comment
Share on other sites

47 minutes ago, Master39 said:

And then proceeded to quote half of the question form the interviewer and half the reply from another person, very much not Nate Simpson.

Are you saying that an unknown person who had nothing to do with the development of KSP2 came into the room and answered the question instead of Nate? And Nate didn't even react to it? Can we then say that only Nate is authorized to answer questions about the game, and all other AMAs are unofficial and are not any promises or reliable information?

Link to comment
Share on other sites

9 hours ago, Alexoff said:

Are you saying that an unknown person who had nothing to do with the development of KSP2 came into the room and answered the question instead of Nate? And Nate didn't even react to it? Can we then say that only Nate is authorized to answer questions about the game, and all other AMAs are unofficial and are not any promises or reliable information?

perhaps Nate can only say things about design and some other person related can describe the technical side where Nate doesn't have much control on? What is your whole point here?

 

10 hours ago, Alexoff said:

Some say that the game was restarted from scratch. If this is so, then management has made very serious mistakes.

This would be incredibly unlikely, data miners across reddit have brought up the very large amount of content the ksp2 keeps disabled likely as the features aren't complete yet (colonies, interstellar,  perhaps experimental optimisations). Whatever you heard is not only a plain lie,  but I doubt others have even thought of the idea and you are bluffing.

give us some sources.

Link to comment
Share on other sites

48 minutes ago, GradientOGames said:

perhaps Nate can only say things about design and some other person related can describe the technical side where Nate doesn't have much control on? What is your whole point here?

So it was about the fact that more than a thousand parts in an interview were not personally stated by Nate, but by another developer, and therefore this phrase became irrelevant. Which is completely wrong.

51 minutes ago, GradientOGames said:

This would be incredibly unlikely, data miners across reddit have brought up the very large amount of content the ksp2 keeps disabled likely as the features aren't complete yet (colonies, interstellar,  perhaps experimental optimisations).

The presence of some textures and codes does not say anything about the degree of readiness. The hl2 files also have a lot of cut content but we haven't got Borealis ship now

53 minutes ago, GradientOGames said:

Whatever you heard is not only a plain lie,  but I doubt others have even thought of the idea and you are bluffing.

This is a pretty popular idea if you've seen KSP2 gameplay videos from 2019. Since it is difficult to explain why the game has changed so little since then.

55 minutes ago, GradientOGames said:

give us some sources

What source? I kind of wrote that it's not true. There is no source of falsehood, this is a newly born attempt to justify why the game, after 6 years of development, is still at the technical alpha stage. I think you should read what I write more carefully.

Link to comment
Share on other sites

KSP2 is definitely going to end more optimized. Not as much as some might like, but it will. And since this thread is turning nasty and repeating arguments already being fought elsewhere, it's time to move on. 

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...