Solutons to discourage gigantism?

[Keptick]

Remember that post?



With that phrase in bold you started to disrespect those who do not think like you, maybe was my lack of english striking again, then i must ask sorry to missunderstand you, and i don´t understood your point but it's curious how learning to weave energy blocks in the three axes is called "know how to build an eficient ship", and I'm not talking about to build only lines in the three axes.

A large ship = Around 200 meters and less 250.000 blocks, forget the mass, mass don´t lag, the block numbers does.

Fat asses = below 200 meters and more than 250.000 blocks.

mass is the number of blocks divided by 10 (for 99% of blocks).

 

[Leanson]

Yup

 

[Leanson]

Thx pal:

Lol

 

[Keptick]

Thx pal, now i can Fix the thing i said.



But returning to the treath, what´s the best way to implement a heat system to starmade?

I forgot to mention that most blocks have a 0.1 mass and that a few exceptions have slightly less. That's what I meant by 99%, as in 99% of blocks have 0.1 mass.

 

[NeonSturm]

But returning to the treath, what´s the best way to implement a heat system to starmade?

Mass ^ (2/3).

Or with (a = largest axis; b = second largest, c = smallest axis) do (surface = 2* a*b + 2* a*c + 2* b*c) and (natural heat dissipation = surface)

That would limit a ships growth in active parts to something closer to (length^2 + fixed average thickness) instead of (length^3).


Personally, I like ships >200m <500m.
But I dislike how hard they are to build and that their existence limits smaller ships to 1m thick blocks rather than 0.5m which would allow 4x details.


In my opinion:

1 . Ships should be smaller or equal to earth's moon's radius (which has 1737.10km radius according to wikipedia) (moon maps)

2 . If they are too big (>300 .. 500m), they should require too much time - rotating toward a gate - or charging their jump drives
- to actually be play-able as anything but a carrier-base or moon-destroyer on it's way to an enemy faction outpost
(for gates I mean because of the size, for jump because of the mass)

3 . Ships should be encouraged to contain at least 70% vacuum (compared to it's box size) and 5..10% air (interior) which is 33..67% of the ship volume (=box -vacuum).
Everything else should be discouraged to make balancing RP ships and carriers easier (most peoples want to build for defaults).

4 . NPC factions should have common 24..32m regular 48..64m and rare 96..128m gates and share it with the player.
Everything above should be annoying to setup as a single player on faction-centered servers and thus take days to actually travel somewhere or have hard times catching something smaller.​

 

[Mariux]

My personal problem with big ships is that they're not fun to play with.

Well, that's the real problem. We need to make big ships fun, not just wipe then out of the game... If a lighbulb stops working, you change the lightbulb, not the house.

 

[Qweesdy]

The problem isn´t the server or crapy players computers pal, it´s the game engine.

"Lag" is several completely different things: server CPU load, networking bandwidth, and client graphics.

For server CPU load, a small number of huge things is lot better than huge number of small things. This is mostly because of collision detection (both for ships/turrets and for projectiles). E.g. with 3 huge objects there are 3*2=6 possible collisions that need to be tested, and with 1000 small objects there's 1000*999 = 999000 possible collisions that need to be tested (in other words, overhead grows exponentially with the number of objects, and doesn't grow linearly). There's also other factors (e.g. lots of bobby AI rather than a small amount of bobby AI).

For networking bandwidth, a small number of huge things is better than a huge number of small things. This is because each thing has meta-data (e.g. location, orientation, velocity, shield, etc). 2 huge ships might be a total of 20 million blocks (that don't change often and can be cached effectively by client) plus 2 lots of meta-data (that changes often and can't be cached well). 2000 small ships might be the same 20 million blocks (that don't change often and can be cached effectively by client) plus 2000 lots of meta-data (that changes often and can't be cached well, and flood the networking with a large number of "object meta-data updates"). Of course this includes projectiles (beams, missiles, etc) - e.g. a single massive turret that creates one very powerful projectile is better for network bandwidth than 1000 small ships/turrets that create 1000 weak projectiles.

For client graphics, it mostly depends on how many blocks and projectiles the game thinks might be visible (which is different to the number that actually are visible). For a huge solid ship it's easy for the game to figure out that most of the blocks can't be visible, and for many small ships this isn't the case; so (e.g.) "10 times as many ships that are 10% of the size" means a lot more graphics overhead (even though it's the same number of blocks).


Basically; for the same number of blocks, a smaller number of larger ships/turrets/projectiles is better than a larger number of smaller ships/turret/projectiles; for all of the different types of lag.

For different numbers of blocks, less blocks is better for lag (e.g. a single smaller ship is better than a single larger ship and a lot better than multiple smaller ships). This is why people think small ships are better for lag when they aren't - they're not doing a fair comparison, and are actually comparing "less resources" to "more resources" and finding that less resources is less lag.

Of course if you limit people to "less resources", then you can expect the majority of players to give a small amount of effort for their small reward, and then get bored and find a different game to play after a small amount of time.


If you want to limit lag but don't want to limit "reward for effort"; then the most sensible approach is to encourage players to spread their ("unlimited") reward over an unlimited number of sectors. For example, if a player or faction has 100 million blocks, then it'd be better for them to have 100 sectors where each sector has a 500 thousand block ship and a 500 thousand block station; rather than having a 100 million block ship/station in one sector, or having 1 million little (100 block) ships in one sector.

With all of this in mind; the solution to both gigantism and lag is to encourage people/factions to try to own the most territory. This could be done by (e.g.) having various resources in different areas of space, and making people need those resources. For example, one sector might have a gas giant (that the faction uses for cheap fuel), one sector might have a planet that's idea for growing "hamburger cows", one sector might have an asteroid field with a lot of one type of ore, another sector might have a planet that has a lot of another type of ore, etc. A faction/player might set up mining operations in each of these sectors to get a good supply of each of these resources (and have to defend their mining operations); and end up spreading their resources across many sectors. Of course people are greedy and/or cunning (e.g. a large faction might try to take control of every sector that has a gas giant, so that their enemies don't have a cheap source of fuel); so this could also add a lot of strategy to the game (where war is more than just a single"biggest faction wins" battle where the "winner" is the loser that lost the least; and where war becomes many battles for control of important resources).

 

[NeonSturm]

With all of this in mind; the solution to both gigantism and lag is to encourage people/factions to try to own the most territory. This could be done by (e.g.) having various resources in different areas of space, and making people need those resources. For example, one sector might have a gas giant (that the faction uses for cheap fuel), one sector might have a planet that's idea for growing "hamburger cows", one sector might have an asteroid field with a lot of one type of ore, another sector might have a planet that has a lot of another type of ore, etc. A faction/player might set up mining operations in each of these sectors to get a good supply of each of these resources (and have to defend their mining operations); and end up spreading their resources across many sectors. Of course people are greedy and/or cunning (e.g. a large faction might try to take control of every sector that has a gas giant, so that their enemies don't have a cheap source of fuel); so this could also add a lot of strategy to the game (where war is more than just a single"biggest faction wins" battle where the "winner" is the loser that lost the least; and where war becomes many battles for control of important resources).

Agree with this paragraph, less with some others.

I think building 1000m long ships and then have only 5% of it's length with nice interior is a waste of CPU-time compared to what you get for it.

If peoples are not encouraged to build interior and have it useful (for trading latest blueprints at least or NPCs) you have a boring game made of 95% block spam.

Also support for ships >1000m is a burden.
Personally I would really like a way to not require 200m^3 for 64 bit ram in 1 second write, 2 seconds read (which is worse than a 1990 computer with the same size in meters!!!)

 

[SomethingSlimy]

3 . Ships should be encouraged to contain at least 70% vacuum (compared to it's box size) and 5..10% air (interior) which is 33..67% of the ship volume (=box -vacuum).​

​

70% is ridiculous. Ships need actual guts too.

And how is the game going to identify "interior?" And why would a ship need 70% vacuum anyways if it already has 5..10% interior?

 

[starhaiden]

The only problem with large ships is the turrets that protect them

 

[jayman38]

​

70% is ridiculous. Ships need actual guts too.

And how is the game going to identify "interior?" And why would a ship need 70% vacuum anyways if it already has 5..10% interior?

@NeonSturm may be referring to the volume in relation to the ship's overall bounding box dimensions, not the actual ship. For instance, even a fairly boxy-looking, non-deathcube ship will only occupy an average of 50% of its bounding box volume. Neon, can you confirm?

 

[Qweesdy]

@NeonSturm may be referring to the volume in relation to the ship's overall bounding box dimensions, not the actual ship.

Ah,. so like a 700*700*700 cube with an additional 1*1*300 stick of rod lights slapped on the back so that it only consumes 70% of the bounding box and therefore isn't a deathcube?

 

[jayman38]

Ah,. so like a 700*700*700 cube with an additional 1*1*300 stick of rod lights slapped on the back so that it only consumes 70% of the bounding box and therefore isn't a deathcube?

Correct. It's a deathtadpole. *ducks*

 

[starhaiden]

instead of a solution to discourage gigantism there needs to be a solution to discourage doom cubes.

 

[FunnyBunny14]

There is already one (sort of). You know about the setting in the server config for dimension rotating speed, right? If you set to to zero, you should no longer have rotation speeds influenced by boxdims.

I'd like to see more ways though.

 

[Comr4de]

There is already one (sort of). You know about the setting in the server config for dimension rotating speed, right? If you set to to zero, you should no longer have rotation speeds influenced by boxdims.

I'd like to see more ways though.

No, that's just the rotational multiplier.

 

[Daffor]

instead of a solution to discourage gigantism there needs to be a solution to discourage doom cubes.

I think general scorn at their ugliness and people who fly them can be applied to this effect. Well designed and good looking bringers of galactic death should be welcome.

I think large ships should be viable and hopefully with the development of the game economy we will see them being more expensive to field.
Turning on the option of buying blueprints with blocks is one solution, but it is hardly viable as it is, since to buy a large detailed ship you often need more blocks of different types than what fits in your inventory. The factory revamp is a step in the right direction in regards to manufacturing.

I for one want to be able to build my big dumb objects in the future and not getting them blown up by a few fighters. Starmade is after all my go to game for building planetary salvagers, system destroying nukeboats, death stars and other semi-plausibly ridiculous concepts.
A single titan should not be able to just wipe the floor with a fleet of enemies, but neither should it be disabled or crippled by a few fighters poking at it.
I am also worried about the upcoming thruster changes leading to a point where a large ship won't be able to have weaponry if it wants to move due to power constraints.

 

[SomethingSlimy]

@NeonSturm may be referring to the volume in relation to the ship's overall bounding box dimensions, not the actual ship. For instance, even a fairly boxy-looking, non-deathcube ship will only occupy an average of 50% of its bounding box volume. Neon, can you confirm?

Thanks for clearing that up. I'm still not comfortable with the idea though, what if I wanted to build a Borg cube? It limits the freedom of what you can build, it seems to me.

 

[jayman38]

Thanks for clearing that up. I'm still not comfortable with the idea though, what if I wanted to build a Borg cube? It limits the freedom of what you can build, it seems to me.

I guess you'll need to have lots of internal space in your borg cube. Nevertheless, the 70% is just an idea and kind of runs counter to Schema's desire to keep the game as open to creativity as possible. Maybe if a fill percentage is ever implemented, maybe it could be server-configurable with a high default value.

 

[NeonSturm]

With "Encouraged to have 70% vacuum" I meant that we remove the advantage of Borg-cubes over ships with detailed shapes.

Few peoples actually want Borg-cubes, but want the most efficient ship, thus build them as no other ship can rotate as fast and still be powerful enough to fill that role.

Also if you build a cube, you are an almost perfect target.

Build a spider-web and you are very hard to hit.​

If a ship contains 5-10% of it's boundary box in blocks it will be ok too.
How vacuum is determined? I made a post somewhere about it related to how to distinguish exterior from interior hull.