On Thruster Mechanics

[koonschi]

I feel like I have to make a statement on how thrusters are meant to work in Avorion since many of you are getting the wrong idea.

 

If you relied on surface area of thrusters, then yes, the new volume based mechanic will feel like a slap in the face. There are multiple factors that play a role in the decisions I made.

• Stacking very thin layers of thrusters doesn't make sense. In any way. I think we can all agree on that. Same goes for solar panels. Although I still want to allow thrusters to be obstructed by other blocks and still work, for the sake of simplicity.
  
• Thrusters should NOT be as strong as main engines. Just look at them, it would never make sense that they are even remotely as strong as engines.
  
• When you want to brake a space ship in newtonian physics, you flip it around, and thrust into the opposite direction. The thruster blocks were never meant to be the main brakes of your ship. They help you, they steer the ship in the right directions, but they will and should never be as strong as the main engines.
  
• Large spaceships do feel sluggish. They have a lot of mass and that mass needs a lot of power to move. That's why you can place turrets all around your ship and that's why you can target independently from turning, either by holding Ctrl or by using the alternative control scheme (which, by the way, used to be the default scheme).
  

 

When I said that I buffed the thrusters, what I meant was, that their power is a lot stronger now based on their power per "value", whereas the "value" is now calculated by their volume and no longer their surface area. It's getting apparent to me now that the wording is misleading. Building large thrusters should be a lot more effective now.

 

If you build thrusters the way that they were actually meant to be built (meaning, as regular blocks and not pancakes), then you should see an improvement of about 50% power in brake and steering performance. I tested this on the ships that I built the way I thought ship building should work. This may be subjective, but it's what I had in mind when I started designing the game.

 

Last but not least: Please also keep in mind that this is the beta branch and I pushed these experimental changes in such an early state exactly because I want to collect feedback.

[Kane Hart]

I just want to say it will take people a bit to get use to the new methods. Only because they been getting use to the new way so it's going be hard on them. That being said I always disliked the thrusters but that is not my fault it's the fault of getting it so easy in so many games. If it takes to long to stop you normally just put actual engine at the front of the ship as well.

 

That being said it would be nice if there was some sort of 3d radar type setup somewhere on the screen so when you flip your ship you can actually target the direct momentum point and cancel out the current velocity.

 

Does not even need to be say a 3d little rader map it could be a simple new icon that shows up on the screen slightly faded a circle if you like to allow you to cancel it out. Maybe simple way to turn it off or on or bring it up when in need.

[BlackWyvern]

I'm sure I'm not the only one who appreciates this post.

 

Being said, while I completely agree that thruster arrays were unrealistic, I feel obligated to put in my two cents, as I really would like to see this game go places. The thruster changes do indeed make sense for larger ships (A capital ship shouldn't be able to spin like a frisbee because they have 100+ sheets of thruster arrays) but it's a huge nerf for smaller ships.

 

• Thrusters should NOT be as strong as main engines. Just look at them, it would never make sense that they are even remotely as strong as engines.
  
• When you want to brake a space ship in newtonian physics, you flip it around, and thrust into the opposite direction. The thruster blocks were never meant to be the main brakes of your ship. They help you, they steer the ship in the right directions, but they will and should never be as strong as the main engines.
  

 

I agree with the subtleties of Newtonian physics, but in my personal experience, your engine simply does not work this way at current. Turning your ship around and thrusting does not brake your momentum, instead it rotates your vector of momentum and you end up moving at nearly the same speed you were going forwards, now backwards. (Get upto about 1km/s, turn around and thrust the other way. Your ship will slide around to the left [usually] and you will end up going the reverse direction at 7-900m/s) I have not once successfully braked a ship via spin-and-thrust. As it stands for me, thrusters are the ONLY viable way to come to a stop. That means having strong braking power, or high resistance along other axis.

 

• Large spaceships do feel sluggish. They have a lot of mass and that mass needs a lot of power to move. That's why you can place turrets all around your ship and that's why you can target independently from turning, either by holding Ctrl or by using the alternative control scheme (which, by the way, used to be the default scheme).
  

 

From a personal standpoint, anything too large to fit through a warp gate is impractical at best. And anything that can't dodge or get out of weapons range of enemies is completely useless. I digress however, I am a fighter pilot at heart.

 

Manufactured turrets do not respect the range stat at all (completely different issue), so even if a railgun says it has a 7.4km range on it, it might only be able to hit something within 4.1. This means that for the vast majority of weapons (The only exception I've come across being range affix cannons) you will inevitably have to get within weapons range of your enemy. There's no 'sit back and snipe' unless you get extraordinarily lucky with drops, or find that one station that can craft with range affixes currently.

 

Due to very large ships being nearly impossible to move, unless you are designed to be a space-whale and shield tank everything (matters a lot more in higher difficulties), you have to be able to get out of that range quickly, and/or dodge incoming weapons fire. This means building a small to medium sized craft that is highly maneuverable and frontloading it with 10-14 high powered alpha-strike turrets for hit and run maneuvers. If a currently maneuverable small-mid sized ship becomes significantly less so, you have two options.

 

1. Either the design has to be completely scrapped and redone sacrificing other stats now because to get back to the same level of agility would weigh the ship down significantly..

2. You're pigeonholed into building a mammoth damage sponge.

 

Neither of those are very appealing prospects. In a game about freeform spaceship building, eventually funneling everyone into the same singular design base is a very, very poor choice.

 

We should be able to have nimble little destroyers and corvettes and still have utility late game. We can throw 12+ turrets onto something like that easily and still hold 60-100k shields, enough for hit and run alpha strikes. Good rails can already three-four shot anything in the Xanion/Orgonite range.

 

I'm not trying to say these behemoth supercapitals with 20-30 turrets on them don't have a place as well. If people want to go big, more power to them. That's what ctrlview and autoturrets are for, cuz they certainly aren't going to be rotating to coordinate their firing arcs any time soon.

 

---

 

I guess what I'm trying to say in a TL;DR is that, while a change to thruster arrays is needed, a flat 70-90% power reduction is not the way to go about it. And while I don't have any particularly -good- ideas on how to change them for the better something needs to be done to pull ships back into balance.

 

[bad ideas:

►Lets make thrusters have a minimum size to start.

►I would GLADLY pay increased power upkeep for stronger thrusters, or even being able to select their power/efficiency ratio.

►Hell, lets even throw in power drain while in use [not passive braking] for even moar thrust similar to spacebaring. If it's not on fire, spewing out ungodly amounts of unburnt petrol, and sitting on the brink of overheating and structural failure, it isn't doing its job. (Minmatar for life.)]

 

Here's the before and after on my actual main ship (+/- minor stats due to different mods in different test galaxies for version handling), not some big WIP junker thing like I posted before. It can barely strafe anymore, and while it can still turn, it's not going to be evading any cannon volleys any time soon.

 

 

 

---

 

I like the idea of the radar thing, but I think a button to rotate retrograde would work too.

[Duncan Idaho]

I see what you are trying to achieve with the Newtonian physics model, but I feel it doesn't work very well with such densely packed sectors. In the real world asteroids are thousands or even millions of miles apart even in the densest parts of the solar system, but in Avorion they are typically less than 1km. You *need* a nimble ship so as not to pancake against every chunk of space debris. In my experience, turning and thrusting to slow down inevitably causes some sidewards motion quickly resulting in a collision. It just doesn't work. You quickly realize that the only sensible course is to stack thrusters until your braking thrust is at least half of your acceleration thrust, and even then it is risky.

 

If you consider the X games, which I have been playing since XBTF, the sectors are similarly packed to Avorion but the speeds are much slower. Instead the player has the means to fast forward time such that they don't get bored moving about but at the same time they don't end up blasting through sectors at suicide speed. While I don't think this solution is a good idea for Avorion, the lesson is important.

[Subquake]

It would be nice if the game would allow to place engine thrusters facing any direction as well, that way you could design a ship to incorporate break thrust into the design on top of maneuvering thrusters.

 

Personally I was very disappointed that you can face the engines only one way no matter how you turn them, because there are many other games, that don't have such limitation and you can place them however you want.

 

(..) When you want to brake a space ship in newtonian physics, you flip it around, and thrust into the opposite direction. The thruster blocks were never meant to be the main brakes of your ship. They help you, they steer the ship in the right directions, but they will and should never be as strong as the main engines. (..)

 

I completely agree with this statement, however this game doesn't have sectors large enough to take advantage of this, instead it puts a burden to the player, and what if you don't want to flip the space ship around all the time?

 

If for instance I have a large ship with huge mass and I have to go to a station nearby, that's only 20km away, I basically have to thrust away 10km and then turn around and start breaking. Saddest part it doesn't quite work that way currently in the game as BlackWyvern described it, but if it did, then first few times it may seem to be fun, but when you are doing it 100+ times, it's starting to become more annoying and a burden more than being fun and realistic.

 

Games should have a balance between realism and having fun and as a quality of life feature having the ability to build nimble space ships, that can break as fast as accelerate and with decent strafing capabilities (for small space craft dodging bullets) is a must in my opinion.

 

If the game would allow us to build engines in any direction, then instead of the game deciding on how you build the space ship, players would have the capability of deciding if they want to build a "normal" space ship or a space ship with increased maneuverability and exceptional breaking power.

[Jerhema]

I agree with the subtleties of Newtonian physics, but in my personal experience, your engine simply does not work this way at current. Turning your ship around and thrusting does not brake your momentum, instead it rotates your vector of momentum and you end up moving at nearly the same speed you were going forwards, now backwards. (Get upto about 1km/s, turn around and thrust the other way. Your ship will slide around to the left [usually] and you will end up going the reverse direction at 7-900m/s) I have not once successfully braked a ship via spin-and-thrust. As it stands for me, thrusters are the ONLY viable way to come to a stop. That means having strong braking power, or high resistance along other axis.

this!

 

Manufactured turrets do not respect the range stat at all (completely different issue), so even if a railgun says it has a 7.4km range on it, it might only be able to hit something within 4.1. This means that for the vast majority of weapons (The only exception I've come across being range affix cannons) you will inevitably have to get within weapons range of your enemy. There's no 'sit back and snipe' unless you get extraordinarily lucky with drops, or find that one station that can craft with range affixes currently.

maybe this thread helps: http://www.avorion.net/forum/index.php/topic,1413.0.html

 

____

 

Finde die Änderung gut. Ich hab immer das Gefühl zu cheaten wenn ich dutzende Papier-Thruster stapele. Außerdem lässt das die Blockzahl explodieren was auch nicht im Sinne des Erfinders ist. Was ich aber noch lieber sehen würde wären gerichtete Thruster(stärkere Thruster, die aber nur in eine Richtung funktionieren) oder wie schon vorgeschlagen die Möglichkeit zusätzliche Haupttriebwerke in entgegengesetzter Richtung zu bauen. Realismus schön und gut, aber das Spiel soll ja auch irgendwo Spaß machen, oder? Und außerdem: was ist an Bremstriebwerken unrealistisch?

 

___

Google tranlator:

Find the change well. I always feel like cheating when I staple dozens of paper thrusters. In addition, the block count explodes, which is not in the sense of the inventor. What I would rather see would be directed thrusters (stronger thrusters, but only in one direction function) or as already suggested the possibility to build additional main engines in the opposite direction. Realism nice and good, but the game should be somewhere fun, right? And besides, what is unrealistic about brake engines?

[lordkabal26]

A nice compromise would be to have engines be able to be rotated to give you the ability to make sudden stops and evasive maneuvers by diverting the power from the rear engines to the engines you would place in key locations along the bow, port, starboard, ventral, and dorsal sections of your ship. Thrusters can still be used for gradual slowdown and rotational changes along the three axis of rotation from the center of your ship.

[Eurgiga]

Earlier posters are correct in asserting that stopping a ship is a pain in the ass well beyond what it should be. It would be far simpler to just give us an "all stop" button that automatically rotates your ship to the reverse vector and stops you rather than the mess of "guess, decelerate, realign, guess, decelerate, realign, guess, decelerate, realign" we have at the moment.

[thedamngod]

A thing I have found, which makes reverse thrust an absolute non-issue for me, is that you not only have to reverse thrust but also accelerate to one side.

 

This means that I turn the ship around and press A/D + W + Space. This is results in a very fast stop. You have to use A or D depending on which side of the ship you have more facing to the current travelling direction. If your left side is facing the direction a bit more than the right side you have to press D. It's the same the other way around, but then you have to press A.

I might have mixed up the effects of A and D, so you might have to experiment/train this a bit. In general this has worked very well for me.

BUT, this was before this beta patch and I have not been able to play this one, yet.

[Duncan Idaho]

A thing I have found, which makes reverse thrust an absolute non-issue for me, is that you not only have to reverse thrust but also accelerate to one side.

 

This means that I turn the ship around and press A/D + W + Space. This is results in a very fast stop. You have to use A or D depending on which side of the ship you have more facing to the current travelling direction. If your left side is facing the direction a bit more than the right side you have to press D. It's the same the other way around, but then you have to press A.

I might have mixed up the effects of A and D, so you might have to experiment/train this a bit. In general this has worked very well for me.

BUT, this was before this beta patch and I have not been able to play this one, yet.

 

It's certainly doable but its damn risky in those crowded sectors. Scrap a roid at 100m/s and say goodbye to your ship.

[AstroOwl]

It's certainly doable but its damn risky in those crowded sectors. Scrap a roid at 100m/s and say goodbye to your ship.

 

In fact, whole "turn around to brake" thing hardly goes together with game where you have to avoid dense asteroid fields where asteroids are dozens of meters apart, catch upgrades flying out of enemy ships by flying over them, etc. This is not a Kerbal Space Program, and it won't be. I am somewhat worried about statements in initial post...

 

I like the idea of game where large battleships are sluggish , battleships shouldn't be too nimble. Situation with stacking thrusters until your multi-megaton-ship turns on a dime was terrible. But the game about space battleships, in my opinion, isn't  the game where you need to do a 180 turn to slow down. If we go full realism, there would be no Space Battleships. There should be some compromise anyway. Like "turning, especially big ships, is hard, but you can still brake at reasonable pace without turning around".

 

I probably should do a longer post with thoughts on it. Once i think about all this a little more.

 

And yes, changes in this patch are great, finally no more thruster-stacking BS. Now the shape of thruster is just a way to distribute its power.

 

And i would certainly distribute most of the power in braking, not turning.

 

[gblnk]

Stacking very thin layers of thrusters doesn't make sense. In any way. I think we can all agree on that. Same goes for solar panels. Although I still want to allow thrusters to be obstructed by other blocks and still work, for the sake of simplicity.

 

I agree with your position on stacking, but unfortunately allowing thrusters to work while obstructed means it will still be a thing.

 

I've been playing around with thrusters on the beta and the new formula doesn't seem to achieve the intended results. A cube of thrusters split into two layers now receives a 50% bonus in thrust. A cube split into 4 layers gets a 100% bonus, 10 layers 150% and so on.

 

The new formula would work fine if thrusters could not function while obstructed, but as long as they can, awarding bonuses based on shape guarantees that layering will continue to be an exploit. The only ways I can see of making this work is to either make thrusters purely volume based, or to change your position on obstructing them. Or perhaps as a compromise make it so that obstructed thrusters continue to function unless they are obstructed by another thruster.

[Subquake]

Or perhaps as a compromise make it so that obstructed thrusters continue to function unless they are obstructed by another thruster.

That can't work, because you could put a small layer of normal hull to bypass that.

[AstroOwl]

I've been playing around with thrusters on the beta and the new formula doesn't seem to achieve the intended results. A cube of thrusters split into two layers now receives a 50% bonus in thrust. A cube split into 4 layers gets a 100% bonus, 10 layers 150% and so on.

 

That is because thrusters built in such a way are redirecting force from other directions. Their power is volume based, but distribution of power can be governed by shape.

 

I mean, when you split cube into 4 thrusters facing forward-backwards axis, you get more brake thrust from such array than from a cube, but at the expense of thrust in other directions.

 

If you split thrusters into very small layers, they would provide almost no thrust in other directions, unlike the old system. So there is a reason not to make ultra-thin slices: you will achieve redirecting of all thruster power to one direction, but you will not get additional power.

[gblnk]

That can't work, because you could put a small layer of normal hull to bypass that.

 

True. Though, you would get diminishing returns doing that as the layers got thinner. At some point you might even start to lose thrust. You'd be better off just leaving an empty space between each layer. Which kind of underlines the difficulty with making thruster obstruction work. The calculation would have to be based on all of the mass between the face of the thruster and the outer edge of the ship, whether actually contacting the thruster or not.

 

It does seem like going with purely volume based thrust is a much simpler solution.

 

If you split thrusters into very small layers, they would provide almost no thrust in other directions, unlike the old system. So there is a reason not to make ultra-thin slices.

 

I do get that. But the average ship's need for lateral thrust is so much lower than the need for braking power that it really isn't much of a sacrifice.

[BlackWyvern]

It's me again. This time with math and empirical data.

 

So I made a 20x20x20 cube of titanium generator on the stable branch.

 

 

 

I slapped 20 10x10x0.05 thrusters on each side. We can see a brake thrust of 55m/s. We SHOULD be able to assume that this same power and velocity will apply to all the faces of this cube.

 

 

We can already see an issue here though. If thrusters are supposed to give equal thrust across their vectors.. Why is pitch abysmal?

 

I then merged the thrusters into 10x10x1 blocks.

 

 

You can see our rotations are down from 2 to 0.14 and brake force down to 3.3. Again, pitch is boned.

 

So then I took the same approach on the Beta.

I made a cube, and slapped a thruster array on each side.

 

 

Our thrust is down to 15ms, and our rotation is sitting on 0.48. I guess at this point, pitch can be called a wash.

 

I then merged the thrusters together as per last time.

 

 

Our thrust is down to 12, which is a significant increase from merged stable, and rotation to 0.52

 

---

 

Lets look at the numbers empirically.

 

THRUST ARRAY

STABLE -> BETA || % CHANGE

Brake 55.3 -> 15 || -72.87%

Yaw 2 -> 0.48 || -76%

Pitch 0.02 -> 0.01 || -50%

Roll 2 -> 0.48 || -76%

 

SOLID THRUSTER

STABLE -> BETA || % CHANGE

Brake 3.3 -> 12.2 || +369%

Yaw 0.14 -> 0.52 || +371%

Pitch 0.02 -> 0.08 || +400% (But still broken)

Roll 0.14 -> 0.52 || +371%

 

ARRAY TO SOLID MERGING - STABLE

ARRAY -> MERGED || % CHANGE

Brake 55.3 -> 3.3 || -94.03%

Yaw 2 -> 0.14 || -93%

Pitch 0.02 -> 0.02 || No Change

Roll 2 -> 0.14 || -93%

 

ARRAY TO SOLID MERGING - BETA

ARRAY -> MERGED || % CHANGE

Brake 15 -> 12.2 || -18.6%

Yaw 0.48 -> 0.52 || +7.69%

Pitch 0.01 -> 0.08 || +800% (But still again broken)

Roll 0.48 -> 0.52 || +7.69%

 

---

 

So what do we have here..

1. The changes from Stable to Beta hits thrust arrays with a baseball bat, shot from an air cannon at five feet.

2. Merging thrust arrays on Stable hits them with a locomotive going full tilt on a straight piece of track.

3. The changes from Stable to Beta give what appears to be a 400% BUFF to solid thrusters.

4. Merging thrust arrays on Beta doesn't seem to do a whole lot, either way.

5. Pitch is completely FUBAR.

 

What should we take from this:

1. STABLE - Thrust arrays give disproportionate amounts of rotational force when they shouldn't.

2. STABLE - Merging thrusters is absolutely not recommended for any reason.

3. BETA - Fixes the disproportionate rotational force, which is nice.

4. BETA - Gives a HUGE buff to solid thrusters, far beyond that which was initially suggested.

5. BETA - Even with this huge buff, thruster arrays lose 77.9% of their vectoring power on all faces when merged and compared to Stable.

6. BOTH - Pitch is broken. It needs help. Maybe a snickers bar.

 

In closing:

Anyone using thrust arrays is completely hosed, with the 77% loss in power coming PERFECTLY in line with all other reports.

[cy414]

i know some say its  rude to post ideas like this, but from what i am reading here things seem to be going pear shaped.

please check these ideas i had about thrusters and engines.

http://www.avorion.net/forum/index.php/topic,1826.0.html

http://www.avorion.net/forum/index.php/topic,1828.0.html

i just hope they help.

[DerPavlov]

I think thrusters are for maneuvering only and should not be used for deceleration at all.

 

I believe the main problem at the moment are the engines, because they are just dumb one direction fuel burners. Is there a way to make engines smarter with "thrust vectoring" and "thrust reversal"? The engine should be enough to steer your spaceship, while thrusters are only needed for rolling and moving sideways during docking maneuvers. This would lower the amount of needed thrusters on a ship. When modern aircraft can do something like this, why not a spaceship with a hyperspace jumpdrive?

 

References:

https://en.wikipedia.org/wiki/Thrust_vectoring

https://en.wikipedia.org/wiki/Thrust_reversal

[koonschi]

Copy-Pasta from the other thread:

 

Most of all, I'd like you to know that I won't be doing the new thrusters without proper compensation. To give you some numbers: With the latest changes I'm getting brake thrust of ~60 m/s² without too much effort and without making the ship an ugly mess. But if you relied on thruster pancakes then you'll have to rebuild your ship. This is not debatable, since thruster pancake stacks make no sense. Same goes for solar panels, but I'm looking into a different solution on these.

 

I've been looking over the code and have found some old code that was meant to ease controls of ships which is now getting in the way. I've removed/reworked all that stuff and overall I think I've found a pretty good solution. I'll give you some more details on all this soon.

[Dinkelsen]

First of all, it IS possible to stop the ship by turning around and boost thrusting, I do it all the time. The velocity bar gets smaller until it disappears. The catch is that you have to have your ship pointing EXACTLY in the opposite direction of where you are going or you will get a sidewards motion, thats the way physics work.

 

Thrusters are one of the systems that cannopt be enhanced with a module, which is very sad. It might help if we had a chance to enhance them.

 

Just cleaned my battleship (10 module slots) it can go 1500 m/s but has brake thrust of 6 m/s²... This will take some serious tweaking until its right.

 

Edit: Was on wrong brnach... On beta branch it has 15 m/s² brake thrust, I can live with that.

[Tsunamik]

To be honest, I kinda like new thrusters... the way it was previously, it was just cheap cheat, when even most heavy ships would turn like made from paper...

 

I currently have over half a million ton heavy ship, and I made its spine by 5 huge thruster blocks going along main axis. And guess what, they can´t be seen, ship is beautiful, brakes with 90m/s and roll/pitch are about 0.8... just great :)

[gblnk]

I also don't mind thrusters being weaker. I think it adds an element of strategy in ship construction that's generally lacking at this point.

 

But I really dislike the surface area biased thrust distribution formula in the Beta. It forces people to build them in certain shapes and sizes to achieve specific results for no particularly good reason.

 

Purely volume based monodirectional thrusters would be far less confusing and would give people a lot more aesthetic freedom.

[Falkenherz]

Braking thrusters with 60m/s??? That´s huge! Usually, my nimble ships have thrust around 25m/s and half of it as braking thrust. I never needed more to dance around NPC ships and gleefully dissect them with my lasers...

[thedamnatus]

Here's something I've always wanted to talk about. It's hard to stop a large ship, that's true. But there's something else that's weird.

Have you ever noticed how fast do any ships brake from boost to regular forward speed?

It's pretty much instant. But then, braking from regular forward speed to 0 takes ages. I think there's something to be done about this.

[Kane Hart]

I sometimes feel half the gameplay is breaking :P