Ask Me A Few More Things

[kicka55]

I wish there would be more concrete statements about how much you guys expect performance to increase over the coming months. I know it's hard but a ballpark of what's theoretically possible on that terrain would be great. Are we talking 30-40% or 300-400%. Because even doubling fps won't make it playable for min spec right now. It drops to single digits frequently on any resolution and we're talking small sub 50 part rockets. So that's even a 1000% increase that would be required under certain circumstances. That would be the first early access in history to make such performance jumps so it is very hard to get the hopes up on that one.

[Overheal]

Looks good Nate. Also looking forward to a barebones game/contracts/science structure. 

Similar to colonies perhaps I missed it but will there be additional features for space stations to differentiate them from just craft in orbit? I mean, a suitably large construction it's not going anywhere, that can build interstellar craft. 

[wazman1943]

This might be a unpopular idea for future expansion to KSP 2 as the Dev Team continues along the Road Map.  Maybe once the first extra star system(s) for interstellar exploration are released.  Future systems are part of a DLC expansion each system with unique features to discover and eventually exploit.

[jebycheek]

Is there any chance that we will have some big improvements on the visual aspect？ let's say, maybe ten times better than what we have now and if it is when it's going to be complete？ take duna as example, the surface texture resolution was so low it's almost just mosaic when landed, that's not good.

Edited April 1, 2023 by jebycheek

[Captain Dax]

Really wish we could have at least some sort of fake rock part, it could be anchored on the ground (using a space bulldozer?)  and make a neat small flat spot for base building.

[PDCWolf]

UI: This is needed badly. The UI has a nice palette, but the design and UX sucks, along with being overly bloated, and oversized. Glad you've already taken notice of this.

Pricing: Yikes. You're already charging $50 with plans to increase price, and the suspicion of DLC. You're overstepping the line. Knowing this data now (since I've already refunded), you just lost me as a 1.0 client. I'll wait for a steep sale since I have to wait years anyways.

Camera: In KSP1 you could click the mousewheel to pan, and alt-scroll to change FOV. Can we get easy controls like those instead of occupying the entire numpad? Since we're here, the VAB camera needs some UX work too.

Multiplayer: I'll eat my words that it is not a thing. However, it is still worrying to see the most basic low-part vehicles, right on the ground, without anything special like docking, plus a still image doesn't tell a tenth of the story when it comes to multiplayer. I wanna see multiplayer working with vehicles like the ones you show on the Duna challenge, which is a much more realistic scenario than goofing around on the KSC

STUFF YOU DIDN'T ADDRESS:

Performance: You need to come out with this already. What are your targets, how much more can you (or plan to) squeeze out? And no, I'm not talking about disabling stuff and calling it "optimization".

Visuals: Is what we currently have really the target level of detail / artistic view for the game?

Multiport-Docking: Irredeemably, 100%, broken. Pls Fix.

 

Edited April 1, 2023 by PDCWolf

[LoSBoL]

11 hours ago, HammerTyme said:

21:9 is supported. That’s what I’m playing on. 3440x1440p 

Where you able to select this resolution in the settings of the game, or did you need to manipulate the settings.json file? 

I need to manipulate settings.json to get KSP2 to run in other than 16:9 resolutions. When doing so KSP2 scales beautifully in any resolution and scales like 21 or 32:9, except for the loading screens which get stretched. 

So the game is supporting it, it just isn't selectable in the settings to me and probably others as well. Which I would like to see in the near future. 

[razewing]

Will there be astronomical objects other than stars? Like black holes, rouge planets and nebulae.

[Klapaucius]

Thanks for the updates. I have only one request, which is to come hang out here a bit more often. I think there are more than a few of us who just find Discord chaotic and would much prefer to dicuss KSP in this type of format. 

I'm having a blast, by the way.

[tecanec]

4 hours ago, PDCWolf said:

Multiplayer: I'll eat my words that it is not a thing. However, it is still worrying to see the most basic low-part vehicles, right on the ground, without anything special like docking, plus a still image doesn't tell a tenth of the story when it comes to multiplayer. I wanna see multiplayer working with vehicles like the ones you show on the Duna challenge, which is a much more realistic scenario than goofing around on the KSC

When it comes to software development, the usual procedure for making a feature is to start simple and expand from there, rather than make the whole thing at once. It allows the devs to debug early so they don't end up relying on buggy code.

However, there are a few things that, if relevant, must be taken into consideration from the start. Multiplayer is one such thing, and there's a good reason why: Keeping two or more simulations synchronized across multiple computers is hard. Even the slightest desync could potentially cause a butterfly effect and must either be avoided or re-synced immediately. And given that the computers can't communicate instantly, said butterfly effect may already be spreading before one of the computers were even aware of the desync, further complicating matters. Everything that could possibly get desynced must therefore be possible to re-sync. If you haven't laid the groundwork for that from the beginning, you may well have to rework so much of the game's code that you might as well remake it from scratch.

But laying the groundwork for something and fully implementing it are two very different things. The multiplayer build shown to us is most likely just there for testing purposes; to make sure that they're laying the groundwork correctly. The fact that collisions are working is actually major, as physics are one of the likeliest things to get desynced because of different approximation errors. As for how much that progress is worth, getting something as fundamental as physics to work is an exponential step forward compared to something more specific like docking. (Docking would certainly be major, both technically and gameplay-wise, but even that's a small step compared to physics.)

Also, keep in mind that the scene in those pictures was likely chosen based on what they were testing and how they were testing it. Even the vehicles were probably just what they had on hand or were able to make relatively quickly. Realistic gameplay isn't really the point here. There's no need to send those vehicles to Duna just to demonstrate that collisions work.

[snctfd]

11 hours ago, JoeSchmuckatelli said:

Will HOTAS support for spaceplane flight be introduced during any part of the EA? 

He confirmed that it's something they're very well aware of in last week's AMA, and hopefully will come sooner rather than later.

[PDCWolf]

2 hours ago, tecanec said:

When it comes to software development, the usual procedure for making a feature is to start simple and expand from there, rather than make the whole thing at once. It allows the devs to debug early so they don't end up relying on buggy code.

However, there are a few things that, if relevant, must be taken into consideration from the start. Multiplayer is one such thing, and there's a good reason why: Keeping two or more simulations synchronized across multiple computers is hard. Even the slightest desync could potentially cause a butterfly effect and must either be avoided or re-synced immediately. And given that the computers can't communicate instantly, said butterfly effect may already be spreading before one of the computers were even aware of the desync, further complicating matters. Everything that could possibly get desynced must therefore be possible to re-sync. If you haven't laid the groundwork for that from the beginning, you may well have to rework so much of the game's code that you might as well remake it from scratch.

But laying the groundwork for something and fully implementing it are two very different things. The multiplayer build shown to us is most likely just there for testing purposes; to make sure that they're laying the groundwork correctly. The fact that collisions are working is actually major, as physics are one of the likeliest things to get desynced because of different approximation errors. As for how much that progress is worth, getting something as fundamental as physics to work is an exponential step forward compared to something more specific like docking. (Docking would certainly be major, both technically and gameplay-wise, but even that's a small step compared to physics.)

Also, keep in mind that the scene in those pictures was likely chosen based on what they were testing and how they were testing it. Even the vehicles were probably just what they had on hand or were able to make relatively quickly. Realistic gameplay isn't really the point here. There's no need to send those vehicles to Duna just to demonstrate that collisions work.

You need to contrast those statements against theirs. We've been told (and now "shown") that multiplayer is a thing, and is working. However, you need to take into consideration that the robustness required of the system is not the same on the example showed, versus possible real uses. For my experience, it is vital that I'm able to use multiplayer to build massive craft in a collaborative manner with other players, for example. The images shown do nothing to ensure that multiplayer (and collisions) "work" in those environments.

Now, you could say I'm getting ahead of myself asking for Multiplayer to be close to finished when the game clearly isn't. To justify this I'll tell you I'm only answering to their official statements on multiplayer, where they mention it's already designed, implemented, and now shown "working". Why the difference between their "working" and my "working working"? Because their statements suggest the latter, whilst these new pictures suggest the former. Are these just the first tests? Then that should be clearly spelled out, lest we end up with another "NOT REAL GAMEPLAY" type discussion again.

[FlazeTheDragon]

i got a sort off question:

Mk2 style fuel tanks hold the same fuel as mk1 sized tanks, but have more drag. This makes them literally useless from a practical stand point.

Is this actually how you guys intend things to be or is this some sort of an oversight?

It seems really weird that a fuel tank specifically designed for planes/ssto's would actually be worse for them.

[Guest]

4 minutes ago, FlazeTheDragon said:

Mk2 style fuel tanks hold the same fuel as mk1 sized tanks, but have more drag. This makes them literally useless from a practical stand point.

They produce lift! That has to be accompanied by drag, otherwise it'd be free energy!

Also they're very useful from a design point of view, as you can combine them in the same stack with other Mk 2 parts.

[FlazeTheDragon]

7 minutes ago, Periple said:

They produce lift! That has to be accompanied by drag, otherwise it'd be free energy!

Also they're very useful from a design point of view, as you can combine them in the same stack with other Mk 2 parts.

Oh yah, you are actually correct, very interesting spot. 

Lift they produce seems to be kinda low tho. 

Ran a little lest and it seems like the mk2 fueltank will produce about ~5x more lift then the mk1 sized tanks... however, thats still only around ~0.6kN at ~160m/s

Maybe there needs to be a bit of a re-balancing in terms of how much lift they produce...? In practice, i feel like there still not worth using.

[Guest]

In my opinion there are some really useful Mk 2 parts, like the cargo bays, the passenger compartment, the twin-engine fuel tank adapter, and the inline cockpit. The fuel tanks are a nice addition to this as you can put them in the same stack. I do often use Mk 1 fuel tanks in nacelles but because of the way drag occlusion works that has its own penalty compared to putting everything in a single fuselage. 

There's loads of stuff in the aero model that needs tuning and some of it is still obviously bugged (like the short Mk 1 cargo bays) but I haven't found the Mk 2 parts to be unreasonably draggy. The main aero issue I have with them is with the pointy Mk 2 cockpit -- if I use that, I find that the planes often become aerodynamically unstable even though the CoM is well ahead of the CoP. 

Because of these things my default Mk 2 spaceplane design has a pointy nose with a Mk 1 inline cockpit, then an adapter to the Mk 2 section that carries all the interesting stuff. I can put a shock cone intake on the nose, enough to feed two Rapiers, or the nose cone adapter to make it very pointy. Or if for whatever reason I want a twin-seat cockpit I can use the Mk 2 inline one with either the short or the long Mk 2 to 1.25 m adapter. 

I do hope they'll eventually do something about the Mk 2 pointy cockpit though! It's like that in KSP 1 as well though so I don't expect it's a really high priority.

[FirroSeranel]

10 hours ago, PDCWolf said:

Pricing: Yikes. You're already charging $50 with plans to increase price, and the suspicion of DLC. You're overstepping the line. Knowing this data now (since I've already refunded), you just lost me as a 1.0 client. I'll wait for a steep sale since I have to wait years anyways.

STUFF YOU DIDN'T ADDRESS:

Performance: You need to come out with this already. What are your targets, how much more can you (or plan to) squeeze out? And no, I'm not talking about disabling stuff and calling it "optimization".

 

For pricing, I, for one, am kind of baffled that a AAA video game in 1999 cost about $60, and a AAA video game in 2023 still costs about $60, after well over 100%, probably closing on 200% inflation. I fully expect $80 or even $100 to become the new normal for AAA games in the _very_ near future. Many are already well above that if you count DLC and micro-transactions as part of the price.

I think $50 is perfectly reasonable, probably going up to $80 at launch. Yes, they were forced to release too early, but they're making good progress in catching up, and have admitted that having feedback from a large player base is already helping immensely.

 

For performance, how exactly do you think that works, that they can just "come out with it already"? They've addressed several times already that performance optimization is a top priority, but at the same time, it's also necessarily going to be an iterative process as more features, parts, and bodies are added. I'm actually extremely impressed by the performance, other than the already admitted issues with fuel flow modeling to multiple engines. If you keep to one engine per tank, for me, and for many others I've spoken to, KSP2 runs much better than KSP1 (modded to a similar level of graphics).

My concern is that I'm not seeing any significant changes in craft physics methodology. I'd hoped for a more selective system of on-the-fly "welding" to reduce computation load in non-critical scenes like ordinary flight, as I don't see how colonies will be possible at all if they calculate the physics for every part in physics range all the time. That said, they did acknowledge that the floppy rockets aren't as popular as they'd assumed, so I'm hoping they'll not just tighten up joints like KSP1 did, but also consider treating core "fuselage" parts to be single solid bodies outside of collisons.

I also wonder if you'd updated your graphics driver when you had poor performance. I had pretty abysmal performance at first, and really hideous dithering effects on the edges of clouds that bled into my craft and just generally looked like hot garbage. I even, to my chagrin, wrote some less-than-kind things about it. And... then I updated my GPU driver, and the performance got much better, and the dithering and transparency bleed went away entirely.

 

Overall, while I certainly understand the frustration, and share in a lot of it, I respectfully suggest reconsidering your tone, as I did after my initial anger faded.

Imagine yourself as a game developer, already having put years of your life and passion into a project. Do you imagine that the message you wrote would be motivating? I don't know that someone who already refunded their purchase then going on to threaten not re-purchasing at all if you don't capitulate to their demands would be anything but upsetting and discouraging.

I don't even work there, and your post makes me more inclined to think "Good riddance, then, you aren't the customer they need," than "Oh, they'd better work hard to win him back." In other words... you can catch more flies with honey than vinegar. Food for thought.

[JinxerH]

18 hours ago, The Aziz said:

Yeah speaking of those... I'd very much like to see more keybinds editable. I'm one of the few who's ever confused by default RCS keybinds and i like to have them exactly on the numpad. ...not only it doesn't save for whatever reason, so despite having it set up I still have to use HJKLwhatever, but also it looks like some of the camera movements are locked to the numpad as well.

I agree with this. Just having the option to rebind every (major) keyboard input would be great. For example the new camera controls use the arrow keys like they did in ksp 1 but at least in the first game they could be changed. It would be a lot easier for ambidextrous/left-handed players who are used to using the arrow keys for controlling the movement of the ship if this extra config option was added.

[FirroSeranel]

55 minutes ago, Periple said:

The main aero issue I have with them is with the pointy Mk 2 cockpit -- if I use that, I find that the planes often become aerodynamically unstable even though the CoM is well ahead of the CoP. 

---

I do hope they'll eventually do something about the Mk 2 pointy cockpit though! It's like that in KSP 1 as well though so I don't expect it's a really high priority.

Yeah, I've gotten the impression that they pretty much just copy-pasted the parts from KSP1 and then did some graphical tweaks. I did extensive cockpit drag testing while participating in the last season of Scott Manley's dogfighting competition, and... the results were very odd indeed.

There were immense differences in drag, both with the cockpits on their own, and with an ore tank occluded into them to make them all equivalent area density, with differences in a drop from 10,000m to ground level of as much as 40 seconds. Some fell straight down, while others would end up with asymmetric drag, with neither of those making any sense considering the cockpits' shapes. What's more, one cockpit type was indestructible, and literally just bounced off of its nose when it hit the ground... though I think that might've been an Airplanes Plus mod cockpit IIRC.

[jclovis3]

3 hours ago, PDCWolf said:

...We've been told (and now "shown") that multiplayer is a thing, and is working... 

... where they mention it's already designed, implemented, and now shown "working". Why the difference between their "working" and my "working working"? ...

Where did they say it was implemented and working? I read that the devs were goofing off in multiplayer. That is not to say that it works as intended yet. We are playing KSP2. That doesn't mean it works as intended yet. Multiplayer is a feature requiring many parts, including how to handle time warp, if that will even be allowed. What they have said and demonstrated is that work is progressing on these parts. Getting some form of communication between multiple clients to bring craft together into the same environment is progress from only a single player experience. The design they have is a plan, not an implementation version. What is implemented is not the whole, but just enough to verify data paths between clients. If there is any peer to peer connection between clients, bypassing any server authority, there needs to be sufficient security in place to ensure that one player can't inject code into the stream that can hurt another players experience or worse, gain unauthorized access to their system. A goof session such as the one described can be used to generate analytics for finding performance issues in the communication stream early on before things get more complicated. So do not infer that a feature is working just because some small aspect of that feature is being tested in a game environment. Everything has to be tested in some way as it is being developed. That is, after all, the whole purpose behind an Early Access release.

[StrandedonEarth]

20 hours ago, Nate Simpson said:

 

All hail Shai-Hulud!

[PDCWolf]

5 hours ago, FirroSeranel said:

For pricing, I, for one, am kind of baffled that a AAA video game in 1999 cost about $60, and a AAA video game in 2023 still costs about $60,

No.

Since 1999, videogames have become the most popular medium, bigger than movies and music, combined. Highest selling games from '99 barely broke 10 million sales. Minecraft to date has sold 238 million copies. That's a 183% increase in sales. Inflation is not only offset by how big the medium has gotten, it's beaten. Further on, you need to remember that digital distribution has cheapened production costs (whilst we stopped getting cheaper digital editions) Remember when games where cheaper on steam than disc? I remember.

Edit for sources. Revenue growth in $B from resetEra.

 

3 hours ago, jclovis3 said:

Where did they say it was implemented and working? I read that the devs were goofing off in multiplayer. That is not to say that it works as intended yet. We are playing KSP2. That doesn't mean it works as intended yet. Multiplayer is a feature requiring many parts, including how to handle time warp, if that will even be allowed. What they have said and demonstrated is that work is progressing on these parts. Getting some form of communication between multiple clients to bring craft together into the same environment is progress from only a single player experience. The design they have is a plan, not an implementation version. What is implemented is not the whole, but just enough to verify data paths between clients. If there is any peer to peer connection between clients, bypassing any server authority, there needs to be sufficient security in place to ensure that one player can't inject code into the stream that can hurt another players experience or worse, gain unauthorized access to their system. A goof session such as the one described can be used to generate analytics for finding performance issues in the communication stream early on before things get more complicated. So do not infer that a feature is working just because some small aspect of that feature is being tested in a game environment. Everything has to be tested in some way as it is being developed. That is, after all, the whole purpose behind an Early Access release.

All I want is clarity, pretty much the same that's needed with all their comms. "Multiplayer is [at this point], [this is still left to do], so we expect it'll be here somewhere around [date]". I'm sure their design document is full of ideas for creation, design, implementation, or already in testing, and posting lose images and statements doesn't help someone pinpoint at which point they are. Collision works? Great. Does it work on a realistic use case or you haven't gotten that already? You're showing me a synchronized use case, but you talked about being able to do both sync'd and desync'd, have you done that yet or is that still not a thing?

Clarity, transparency, compromise.

Edited April 1, 2023 by PDCWolf

[the_tunnel]

only answering discords questions was a great disappointment to me I had looked at all the questions on the forum and got hyped for what would be answered and when none were answered i thought who was involved in picking the questions and whoever it was need to be talked to so nobody gets left out.

When will the next ama be?

[RocketRockington]

On 3/31/2023 at 11:12 PM, PDCWolf said:

Multiplayer: I'll eat my words that it is not a thing. However, it is still worrying to see the most basic low-part vehicles, right on the ground, without anything special like docking, plus a still image doesn't tell a tenth of the story when it comes to multiplayer. I wanna see multiplayer working with vehicles like the ones you show on the Duna challenge, which is a much more realistic scenario than goofing around on the KSC

It was posted on April 1st and showed just about nothing.  Maybe don't eat those words just yet

[Datau03]

2 hours ago, RocketRockington said:

It was posted on April 1st and showed just about nothing.  Maybe don't eat those words just yet

The multiplayer screenshot was posted on March 31st Seattle time.