procedural parts or in the style that we have right now?

[έķ νίĻĻάίή]

so, I was thinking about procedural parts and I was thinking that they should be implemented to ksp 2.

What do you think?

[Gydra54]

A procedural part system with preset parts in the part selection menu would be ideal IMO, if they could make them as visually appealing as what we'd get with preset parts. I don't care either way though, I'm sure a procedural parts mod will be made eventually anyway.

[έķ νίĻĻάίή]

Huh ok

1 hour ago, Gydra54 said:

A procedural part system with preset parts in the part selection menu would be ideal IMO, if they could make them as visually appealing as what we'd get with preset parts. I don't care either way though, I'm sure a procedural parts mod will be made eventually anyway.

Such as e.g procedural airships I guess...?

[Xd the great]

15 hours ago, burn boi said:

Such as e.g procedural airships I guess...?

More like procedual fuel tanks.

[shdwlrd]

Don't forget procedural wings. Too many parts are needed to make large custom wings. Selectable sizes for the similar wing shapes, tiny to large. But really, you can say that for these other part categories, fuel tanks, structural, utility, heat management. (The static radiators anyway.)

There comes a point where there is too many Legos in the pile and it gets difficult to sort through them all.

[έķ νίĻĻάίή]

Yeah, procedural wings would be great . Like simple planes!

[AVaughan]

Personally I like procedural parts.   If they aren't in the stock game, then I hope a mod adds them.

[Francois424]

I vote procedural, but a lot of things are better as we have now (Lights, Landing Gears, Wheels, Docking ports etc).

But for tanks, wings, fuselage parts (including girders) and engines?  most definitely.  Having the capability to pick the fuel in said tanks would go a long way too (procedural fuels?  Fuel switch?)
And since we're on to the fuel subject here, might as well give me that pump mod that automatically handles fuel transfers between crafts depending on priority, but I'm getting off topic.

 

[SpaceFace545]

I think the only procedural parts we need are wings, heatshields and better fairings.

[Bartybum]

If procedural fuel tanks are a thing I'd definitely want them to have procedural 3D models for them too. The current procedural tanks mod is extremely ugly

[Idleness]

Procedural I think would be better.  I wrote a post about the idea of the tech tree unlocking the allowable dimensions of procedural parts.  Could be based on several factors like radius of tanks, different shapes if you wanted a rectangular tank for something, the total volume of a tank, etc.  Anything that gives me the full functionality I want - but with fewer parts - is helpful IMO.  Would also be cool to add automatic sizing too if using procedural parts.  I believe there's actually a mod for this called Smart Tank or something?  But, if I could tell the game, okay here's my science satellite, here's the tank STYLE I want, and here's the engine - now make the tank whatever size it needs to be to hit 1.2 starting TWR.  This could be done multiple times in a build (for multiple stages) so you can get more efficient stages and building for efficiency becomes a rewarding task.

[paul23]

Procedural would be too easy. It's hard as it should be to make "nice looking" ships. Function over form, and in real life you also have a scrape by using refurbished models. Not every tank should be available in every form: that's illogical.

[Idleness]

19 minutes ago, paul23 said:

Procedural would be too easy. It's hard as it should be to make "nice looking" ships. Function over form, and in real life you also have a scrape by using refurbished models. Not every tank should be available in every form: that's illogical.

I hear what you're saying.  Perhaps a better idea, WRT auto-magically determining how big to make a tank for specific characteristics, they could instead implement standard parts like they have now but the contents can be optimized (in the sorts of fashion) without having to launch, fly, revert, try again 14 times to get it right.  You can sort of do this in stock now with the D/v and TWR for each stage but it's clunky and I rely on mods to help me.

[έķ νίĻĻάίή]

I think that, good looking parts would reign supreme over bad looking procedural parts

[Francois424]

6 hours ago, έķ νίĻĻάίή said:

I think that, good looking parts would reign supreme over bad looking procedural parts

Agreed.  Tho one does not prevent the other from existing.

[Incarnation of Chaos]

I'm going to be that guy; Why not both?

Iv'e used stock wings even when iv'e had Procedural wings available, heck i had to use stock control surfaces because Procedural ones never worked reliably. And you can always tie Procedural parts to the progression system to avoid making bigger tankage/wings than you should be able to at that point in the game.

The only thing not having Procedural Wings/Tanks/Fairings does is give the player less options, and it doesn't take away from the ones you already have in the slightest. Now Procedural Engines/Command Pods/etc. are completely out of the question; not just for balance reasons but because it's apperently rather difficult to code. So unless KSP2 is different in this regard i don't think it's happening.

[Idleness]

I guess I'm just spoiled.    I've only played with mods for a long time now - because I think when I build a rocket in stock it looks like a pile of... dog waste - when compared against using modded parts/textures.  Though I'm getting the urge to start a kOS run through and I'm less concerned about the look of vehicles, and more focused on what I can make them do.

[paul23]

6 hours ago, Incarnation of Chaos said:

The only thing not having Procedural Wings/Tanks/Fairings does is give the player less options, and it doesn't take away from the ones you already have in the slightest. Now Procedural Engines/Command Pods/etc. are completely out of the question; not just for balance reasons but because it's apperently rather difficult to code. So unless KSP2 is different in this regard i don't think it's happening.

But giving less options is a charming point of ksp. That you can make with a select few options amazing crafts and compare said crafts.

 

As for your second point: procedural engines "etc" wouldn't be much harder to make. An engine is just a matter of a few numbers: fuel consumption, mass flow rate, total temperature and total pressure (later 2 are just magic numbers for a given combustion and shouldn't change based on engine shape, they are only influenced by the materials and how well the burn process occurs) and mechanical nozzle shape.

Those determine the effective exhaust velocity (= specific impulse * 9.81) at different places in the atmosphere)

v_e = v_actual_exhaust  + (p_e - p_a) * A_e / (m_flow)

p_a the atmospheric pressure (here we notice why an engine performs worse at sea level than in space) and A_e the exit area of the nozzle bell (and m_flow the total mass flow). - Oh and obviously the total weight (larger bell = more material etc).

 

with p_e the pressure at nozzle exit, and v_actual_exhaust the actual velocity at nozzle exit. For supersonic nozzles (all engines except juno I think?) this is quite easy: You have two areas of importance, A_e (exit area) and A* (critical area, smallest area, here the flow reaches the speed of sound, before mach 1 a nozzle that becomes smaller increases flow velocity, after mach 1 increasing the size increases velocity but for ksp one could consider the mach 1 point as magical and only design the rest of the bell procedural).

Those last two are a bit finecky to calculate, as those depend on the exit mach number. And the exit mach number needs to be numerically approached (formula can't be solved algebraically) but even though that's not trivial (the rest is), since it only has to be calculated once for a given engine - and for procedural parts an accuracy of 1 promille is enough this still isn't a problem.

 

The thermal formulas are a bit hard to write down without latex so I'll just point you to the Source (NASA)

[Bartybum]

On 11/22/2019 at 12:54 AM, paul23 said:

Not every tank should be available in every form: that's illogical.

I don't care much for procedural tanks, because I do like the Lego system we have at the moment, but in the real world engineers can design and build a tank to whatever size they need, so it's not illogical at all.

Edited November 27, 2019 by Bartybum

[mcwaffles2003]

Would having procedural tanks as  a final tech tier be a good compromise?

 

Ultimately they're good for reducing part count, which is huge, but I just don't want them to undermine the early game challenge of having limited sized tanks 

[marce]

In an ideal world: both

Start with a couple parts with a specific size [and resource for tanks].
Then later on allow to research procedural parts which allow to choose size[, shape][, resource] as needed.

If no procedural parts are provided I think at least a scaling option (like TweakScale) should be in which also reduces part count (and clutter) in the editor.

[Incarnation of Chaos]

On 11/26/2019 at 7:14 AM, paul23 said:

But giving less options is a charming point of ksp. That you can make with a select few options amazing crafts and compare said crafts.

 

As for your second point: procedural engines "etc" wouldn't be much harder to make. An engine is just a matter of a few numbers: fuel consumption, mass flow rate, total temperature and total pressure (later 2 are just magic numbers for a given combustion and shouldn't change based on engine shape, they are only influenced by the materials and how well the burn process occurs) and mechanical nozzle shape.

Those determine the effective exhaust velocity (= specific impulse * 9.81) at different places in the atmosphere)

v_e = v_actual_exhaust  + (p_e - p_a) * A_e / (m_flow)

p_a the atmospheric pressure (here we notice why an engine performs worse at sea level than in space) and A_e the exit area of the nozzle bell (and m_flow the total mass flow). - Oh and obviously the total weight (larger bell = more material etc).

 

with p_e the pressure at nozzle exit, and v_actual_exhaust the actual velocity at nozzle exit. For supersonic nozzles (all engines except juno I think?) this is quite easy: You have two areas of importance, A_e (exit area) and A* (critical area, smallest area, here the flow reaches the speed of sound, before mach 1 a nozzle that becomes smaller increases flow velocity, after mach 1 increasing the size increases velocity but for ksp one could consider the mach 1 point as magical and only design the rest of the bell procedural).

Those last two are a bit finecky to calculate, as those depend on the exit mach number. And the exit mach number needs to be numerically approached (formula can't be solved algebraically) but even though that's not trivial (the rest is), since it only has to be calculated once for a given engine - and for procedural parts an accuracy of 1 promille is enough this still isn't a problem.

 

The thermal formulas are a bit hard to write down without latex so I'll just point you to the Source (NASA)

KSP should give you limited options at the beginning; i agree. Forcing you to work with what you have encourages lean missions and planning. But you should have room to grow; which procedural parts allow.

Also i said that it was difficult to code, not calculate. You have to balance it with stock engines; make sure that procedural engines don't break after a certain point etc. 

This means calculating the performance of any individual engine is simple enough, but coding the entire system is work.

[paul23]

8 hours ago, Bartybum said:

I don't care much for procedural tanks, because I do like the Lego system we have at the moment, but in the real world engineers can design and build a tank to whatever size they need, so it's not illogical at all.

Not really, the tank must be small enough so that the pumps can extract the majority of the fuel.. Size is also dictated by the prevention of sloshing. And the ability to cool/pressurize the tank.

[blorgon]

Another possibility would be to "lock" the procedural adjustments into specific intervals (much like how engine plates and fairing attachment nodes are implemented right now).

At the very least, it's very clunky and annoying selecting between several very similar looking models in the parts list where the only difference is their diameter and length.

I do like the idea of limiting part selections in career mode, and either having procedural be an "advanced manufacturing" tech upgrade, or limiting procedural to sandbox only.

Not only would having procedural tanks and wings streamline the building process, but it'd cut down on loading times quite significantly.

I hope to god we get procedural STRUCTURAL AND UTILITY parts, too. Procedural trusses would be an absolute godsend for station building.

[mcwaffles2003]

31 minutes ago, blorgon said:

I hope to god we get procedural STRUCTURAL AND UTILITY parts, too. Procedural trusses would be an absolute godsend for station building.

Never knew how bad I needed this