Developer Insights #11 – Engine Exhaust Visual Effects

[Pthigrivi]

The man's design and modeling skills are just

[Lewie]

Absolutely breathtaking!!!!

who else is PSYCHED that this is gonna be STOCK?!!! 

[AtomicTech]

Me!

Also me!

and not to forget the most important, me!

[starcaptain]

Friend of mine is sus that the blendshapes don't conform to real world maths. He sent these citations from Rocket Propulsion Elements by George P. Sutton.

As usual, I expect my (or my friend's) pedantry can be solved by mods but nonetheless I'm delighted to see the game's progress nevertheless.

Edited September 17, 2021 by starcaptain

[Anth]

All great information. I love the diagrams of how it all works together. And the skill of the developers in regards to explaining how things work is amazing.

I can understand a thing. But to explain it is another.

Jon Cioletti great work.

 

I have said this in previous developer insights...but keep these coming please. They are amazing!

[SOXBLOX]

UUOUU. Nertea is part of the team now!?

Somebody should give his account that "KSP2 Team" tag...

[Kapitalizing Every Word]

I can't wait to see KSP2 it looks AMAZING and I love that the devs keep us updated.

[Admiral Fluffy]

Looks great! Keep it up!

Spoiler

OR ELSE

Oh yeah i forgot....

(flashy warning)

Spoiler

ITS

 

Edited September 18, 2021 by Admiral Fluffy

[funnelton]

10 hours ago, Nate Simpson said:

@Nertea is in fact now part of our team (yay!), but since we love doing things backwards, he joined JUST as Jon, Aaron, and Paul were putting the finishing touches on this system. I suspect there will be an ongoing conversation about tuning, as there always is. But we're all very happy with this system in its current form.

OOOOOH

hey also random question where did everyone get those ksp 2 animation style avatars??

Edited September 18, 2021 by funnelton

[mcwaffles2003]

I really like how this looks and it really contests with waterfall. The outer boundary of the plume looks very natural and soft which is really cool to see. A question I have though is, do the mach diamonds really stay in place and expand as ambient pressure goes down? I would have anticipated they would slowly fade away from the nozzle and elongate as the furthest diamonds would each in turn fade away as pressure decreases.

 

Also, I know this is an edge case.... but what happens when the vessel goes into atmospheres with far greater pressures than that on kerbin? Will the plume compress even more with more diamonds or is the configuration at kerbin sea level a minimum?

Edited September 18, 2021 by mcwaffles2003

[rextable]

Veieiriery interdasting!

Out of pure morbid curiosity: do you plan to implement engine startup and shutdown transitions/effects?

There would be 'much awesome' to see engine bell giggle/wobble on startup as the exhaust plume stabilizes. Not to mention floaty globules of propellant blobbing around in the vacuum of space upon engine shutdown. Stuff like that.

I'm not too demanding am I?! 

[Admiral Fluffy]

11 hours ago, funnelton said:

hey also random question where did everyone get those ksp 2 animation style avatars??

This, it’s locked now though.

Now I gotta find something to keep this on topic

Are you going to use other systems for some other kinds of engines?
Like, what will the exhaust look like on the Orion drive spacecraft? 

Edited September 18, 2021 by Admiral Fluffy

[funnelton]

5 hours ago, Admiral Fluffy said:

This, it’s locked now though.

Now I gotta find something to keep this on topic

Are you going to use other systems for some other kinds of engines?
Like, what will the exhaust look like on the Orion drive spacecraft? 

darn

also didnt they show the exhaust of the orion drive before?

[neistridlar]

This looks really promising. Definitely leaps and bounds ahead of KSP1!

Like @mcwaffles2003, I too was wondering about underexpanded plumes. It seems only the overexpanded side is represented here. Granted underexpanded engines are very limited in real life use, due to flow separation, this issue is handled differently, at least in KSP1. If nothing else I think an underexpanded plumes would be a nice cue to the player that they are using a suboptimal engine for the current conditions.

An other minor issue: In the video demonstration, the speed of the low throttle exhaust seems ridiculously slow. I may very well be wrong, but it is my understanding that exhaust flow speed for chemical rockets should always be supersonic, and that exhaust speed is directly linked to specific impulse of the engine. In other words, the exhaust speed should always be pretty darn fast.

[mcwaffles2003]

22 hours ago, neistridlar said:

This looks really promising. Definitely leaps and bounds ahead of KSP1!

Like @mcwaffles2003, I too was wondering about underexpanded plumes. It seems only the overexpanded side is represented here. Granted underexpanded engines are very limited in real life use, due to flow separation, this issue is handled differently, at least in KSP1. If nothing else I think an underexpanded plumes would be a nice cue to the player that they are using a suboptimal engine for the current conditions.

An other minor issue: In the video demonstration, the speed of the low throttle exhaust seems ridiculously slow. I may very well be wrong, but it is my understanding that exhaust flow speed for chemical rockets should always be supersonic, and that exhaust speed is directly linked to specific impulse of the engine. In other words, the exhaust speed should always be pretty darn fast.

I'm guessing that the particle animations might be tied to the total length of the plume so if the plume shortens the apparent particle speed decreases. Just my guess as to how plumes have been implemented.

Also, to add onto what I said earlier, I think that the final video isn't a full representation of an ascent seeing as this shows the elongation of the mach diamonds which isn't shown in the video animation:

On 9/17/2021 at 5:00 AM, Intercept Games said:

Scaling of shock diamonds

https://i.imgur.com/fUYMjik.png

Though it may be the case that the shock diamonds length will only scale with throttle, as I believe that is the one variable changing in the animation above, and is independent of ambient pressure, which intuition leads me to believe this is false but I know of no footage showing supersonic exhaust in varying ambient pressures and the outer plume kind of gets in the way for videos space x footage. It would make sense, to me, for shock diamonds to extend with increased throttle since higher throttle -> higher mass flow -> higher chamber pressure -> higher exhaust velocity and the mach diamonds length is tied to exhaust velocity and ambient pressure.

Another thing I've noticed is that the entire volume of the inner plume glows brightly whereas shock diamonds shine the brightest at their tips where flow contracts and compression is at its highest before fading after flow decompresses and has enough time to emit the heat it has acquired as light:

 

A lot of what I've said here is largely speculation/educated-guessing so if anyone especially knowledgeable on the subject here disagrees, please let me know. And to they employees who made these I hope you don't take my comments as unappreciative because what you've made so far is beautiful and beyond expectation. Just, if there is a chance, something I say here could help then I feel I need to speak. Please keep up the good work and I can't wait to play your game  

Edited September 20, 2021 by mcwaffles2003

[Synonym Toast Crunch]

One question I have is about how it'll handle multiple engines at once.
https://imgur.com/ykdMwO9
As we see here on the Falcon 9, the outer eight plumes bow out from the center as pressure from each plume forces them outward from the center, with the bowing increasing as the atmospheric pressure decreases. Will this effect happen in KSP 2, or will we see nine engines cones expand out and clip into each other with no effect on each other's direction? Is there a way to set a bounding area around each plume and assign a value that determines how much pressure it creates (that falls off with distance), and use that to determine how much effect that can have on any other plume that enters its bounding area, possibly using a parametric deformer on the plume to deform it in different directions?

[RocketSimplicity]

Looks amazing, I wonder how they’ll manage different engine exhausts overlapping at low pressures as the exhausts aren’t opaque. Looking miles ahead of KSP 1!

[AtomicTech]

Honestly, before last week, I thought Nertea's name was pronounced Ner-ee-tah for some reason.

[MechBFP]

1 hour ago, AtomicTech said:

Honestly, before last week, I thought Nertea's name was pronounced Ner-ee-tah for some reason.

That is the sound those guys in Quake 2 who shoot rail guns at you say the first time you see them. 

[JoeSchmuckatelli]

Looking gorgeous!

Really appreciate the detailed work you are putting in to communicate artistically what is going on behind the scenes (art telling the story of the game's approximation of science).  You guys build confidence (and excitement) with each reveal.

(insert thumbs up emoji (((b/c I don't know how))))

[Guest The Doodling Astronaut]

Looks awesome, just came back from the deserts of Utah. So I didn't see this until now

[Nirgal]

Nice but haven't we already seen engine exhausts?

I seem to be the only person who's disappointed, we waited 2 weeks to see something we've already seen before, just in more detail.

In my opinion, it would have been much cooler to see something about colony construction, boat / dock functionality etc. 

[crowsnose]

On 9/17/2021 at 2:10 PM, AtomicTech said:

HOLY 

  Hide contents

DERPING DRES(es), JUMPING JOOLS, KICKING KERBINS, DANG DUNAS, EVIL EVES,

NERTEA JOINED!?

My reaction exactly!!!!

[AtomicTech]

2 hours ago, Nirgal said:

Nice but haven't we already seen engine exhausts?

I seem to be the only person who's disappointed, we waited 2 weeks to see something we've already seen before, just in more detail.

In my opinion, it would have been much cooler to see something about colony construction, boat / dock functionality etc. 

Spoiler

I may or may not mildly agree....

 

[Bej Kerman]

On 9/21/2021 at 11:29 AM, Nirgal said:

Nice but haven't we already seen engine exhausts?

I seem to be the only person who's disappointed, we waited 2 weeks to see something we've already seen before, just in more detail.

In my opinion, it would have been much cooler to see something about colony construction, boat / dock functionality etc. 

Because, Nirgal, Take Two doesn't want you knowing about the thing they want you to preorder. Why? Pfft, customers shouldn't know about the thing they want you to spend 60 pounds on before there's a blooming product to be bought. Rather, let's pretend to make consumers know about how the game is coming along by spouting technical java to give the illusion you know more about the game when really you just know more about what the game is going to be running on. I've nothing against the devs for putting the tech stuff out there, I just want some confirmation that what we're seeing isn't just a really advanced version of Waterfall topped with KSPIE/FFT. I just want some gameplay - it can't tarnish the game's rep much to see an in-dev build of the game working. Seeing the game all in its alpha state won't lead anyone to believe it's all shambles, it's just gonna confirm that it's in its alpha state. Besides all that, I'm just a bit miffed that gaming culture has shifted from "here's a 2 hour demo, please pay between 15 and 30 pounds for this game" (which Elliptic's Rodina does, which I respect) to "here's 12 minutes of gameplay accumulated over 2 years, please pay 60 pounds and maybe we'll deliver the wanted product". Now I know there's a lot of trust in Nate to deliver alright, but it'd be blindfully ignorant to dismiss the thought that the game may flop, and either way you shouldn't be preordering!