v0.200.334 Bonus Stabilization Mechanics

[WalkerGain]

Does anybody really understand the way these new dimensional bonuses work?

I've been playing around stabilizing my ship, and I'm confused as hell. The update notes provided on Steam and nowhere else state: "To calculate the bonus of 1 of the 6 sides, we take the stabilizer group with the most efficient of that side and compare it with the most efficient group on the entire ship. If they’re equal sized, then you get the full bonus, else only a fraction."

So I tried starting simple, adding a group of 9 stabilizers left and right of the reactor, but pretty close due to the ship's dimensions. Then when I added a group of 9 to the front dimension, the multiplier for "equally big groups" decreased by 0.3, down to 0.7 total.

I'm really lost. So I tried building large stabilizers on the sides. Now when put the front stabilizer group on, each additional block added stability, until suddenly the next blocks start to subtract. But the front group is nowhere near approaching the size of the left and right groups. I thought I would increase stability by making them similar sized, but it didn't look like that was happening at all.

Sorry if I'm just being obtuse, but the calculations seem crazy and impossible to divine. I finally got myself to 100% stability just by adding and subtracting randomly from this and that group.

Is anyone else having difficulty in the practical implementation of this system, or am I just being obtuse and not understanding simple instructions?

 

[NTIMESc]

The update notes are always posted on the news page, which is also displayed in the launcher: StarMade News - StarMade v0.200.332 - Reactor System Enhancement and Fixes

With multiple groups, I noticed that the real-time info was a bit less useful while placing stabilizers, as the numbers fluctuate with the bonuses making it more difficult to predict the outcome until the groups are all the same. It did seem to work out fine, at least for the three-group system I was experimenting with, though.
[doublepost=1518064662,1518064402][/doublepost]

Sorry if I'm just being obtuse, but the calculations seem crazy and impossible to divine.

Yeah, a more detailed explanation of the fractional bonus for uneven sizes would be nice.

 

[Antikristianos]

it is not based on the block ccount of the groups.
its based on their efficiency. so if you place a groupe of the same size farther away from the reaqctor as the other groups you already have the new one is more efficient and becomes the one with the 100% bonus.

the others will get a decrased bonus related to the efficiency difference to the now most efficient group.

 

[YamiHikari]

The bonus system pretty much works like this:

Assume stabilizer groups are present in 4 dimensions. The stabilizer groups contribute 5%, 5%, 2.5% and 1% of the required stability for max stability, respectively. The group with the maximum stability contributes 5% stability, or 500 stabilization points (this value varies depending upon reactor group size). So this 500 sp (stabilization point) value is taken and compared to all other dominant stabilizer groups. I say dominant because each dimension only considers the group contained within it contributing the largest amount of stabilization. So by comparing the sp, you divide the max present sp from the dominant sp value in all other dimensions, getting 1.0, 0.5, and 0.2, respectively. These values are then added together to obtain the final bonus determinant of 1.7. This value is then multiplied by a config value (which I don't know the location of) to obtain the final bonus amount that is added to the overall stabilization.

This is a rough example of how the bonus calculation works (which I pretty much took from Lancake's explanation of it (you will need to ask him for a more thorough explanation)), but it should be able to give you a general idea.

 

[NTIMESc]

Thank you for the information. The main part that was missing was that it was the efficiency of the groups, not their size.
This clears up why the numbers fluctuate when changing groups, though it doesn't really make the process much more intuitive.
Any thoughts on a better way to present the info to the player so it's more clear what's going on?

<addendum>
The release notes do actually say it's the efficiency, but I think the highlighted part is what confused me, referring to the "size" of the efficency:

To calculate the bonus of 1 of the 6 sides, we take the stabilizer group with the most efficient of that side and compare it with the most efficient group on the entire ship. If they’re equal sized, then you get the full bonus, else only a fraction.

 

[Captain Fortius]

This system is getting more and more convoluted and unnecessary.

I checked the latest patch out, put stabilizers on four different sides of the reactor, and the info panel still said I'm only using one side.

Can we just please delete this whole stabilizer mechanic?

 

[YamiHikari]

This system is getting more and more convoluted and unnecessary.

I checked the latest patch out, put stabilizers on four different sides of the reactor, and the info panel still said I'm only using one side.

Can we just please delete this whole stabilizer mechanic?

I'm guessing you put the stabilizers on 45 degree angles from the reactor instead of 90 degree. To solve this problem you only need to rotate the reactor axis by 45 degrees, which should be an option in the ship stats panel on the left side of the build mode screen (unless it is in the advanced build mode interface).

 

[WalkerGain]

Can we just please delete this whole stabilizer mechanic?

I wouldn't mind it as much if it were well-documented and well-explained.

<addendum>
The release notes do actually say it's the efficiency, but I think the highlighted part is what confused me, referring to the "size" of the efficency:

Yeah, I think that was confusing me too. So it should read, "If they have equal efficiency...", correct?

Still, I don't understand why if the two left and right sides were much bigger, I would reach a point where adding blocks to my front group actually decreases the overall stability.

 

[YamiHikari]

if it were well-documented and well-explained

*Smashes phantom agree button* (in attempt to help facilitate the formation of a new inside joke)
[doublepost=1518146655,1518145969][/doublepost]

Still, I don't understand why if the two left and right sides were much bigger, I would reach a point where adding blocks to my front group actually decreases the overall stability.

Yes and no. The bonus is calculated based off of a combination of all dominant stabilizer groups compared to the largest-contributing stabilizer group in the form of a ratio (ratio of each other group to the largest group added together) and that stabilizer group's efficiency. Assuming a different group is the largest-contributing group, if you add more stabilizers or distance to a different stabilizer group, increasing its overall contribution, it will eventually surpass the stabilization of the current largest-contributing group. Now you have a new largest-contributed stabilization from a single group, which is then used to calculate the total bonus stability points added. The reason it isn't quite as simple is that as the new largest-contributing group's contribution increases, the ratio of the other groups' contributions to its contribution decreases, mitigating any increases in bonus that come out of the added stability of the largest-contributing group (it sounds a bit muddled, but that's actually how it works). The best way to demonstrate this is by playing with sets of four numbers. Start with the following set of numbers: 2, 5, 3, 4. The current largest-contributing group has a value of 5. You can then add the ratios of all of the other numbers to 5 together to get the bonus multiplier. Multiply this multiplier by the current largest-contributing group's stabilization (5) to get a rough idea (I'm not sure if it's exactly what it will be unless I crunch some actual numbers in game, but I know it's generally around that) of the bonus stability points added. Now, to demonstrate different scenarios, add and subtract different values from each of the numbers in this set, or even add or remove numbers from the set entirely. This is how I figured out a lot of how this system works.

 

[NTIMESc]

...and just to stabilize all this explanation, the bonus percentages given in the notes are derived from some other percentages in the config

Namely, 1.0, 1.0, 0.8, 0.6, and 0.5, for 2, 3, 4, 5, and 6 sides.

This works out, when rounding up: 100/sides + sides - (sides*bonus)

 

[Scypio]

So doing excel sheets to plan your ship ahead with previous defensive effect was tedious and needed to get away for the sake of building.

But i need a 10 full page documentation with graphs to understand the new stabs mechanics.

I will never understand anyone saying that the new system is simpler than previous one.

 

[kikaha]

So doing excel sheets to plan your ship ahead with previous defensive effect was tedious and needed to get away for the sake of building.

But i need a 10 full page documentation with graphs to understand the new stabs mechanics.

I will never understand anyone saying that the new system is simpler than previous one.

Just play a bit with the game and you'll find that perfect balance.
And by the way, 45° is not necessarily the best for reactor axis.
I got this salvager with 4 groups of stabs at the front of the ship. I angled one plane of the reactor axis to 21° and i got 200 % + buffer from my stabs. So i was able to remove quite a lot of them to come down to just above 90 % buffer AND at the same time i reduce my instability from - 186 to -16.

You've got to try things, excel sheets is way way horrible to do when playing. Fiddling is better.

 

[YamiHikari]

So doing excel sheets to plan your ship ahead with previous defensive effect was tedious and needed to get away for the sake of building.

But i need a 10 full page documentation with graphs to understand the new stabs mechanics.

I will never understand anyone saying that the new system is simpler than previous one.

Actually, all it took me was the chance to play with it and a decent visual example with some numbers to go with it. If we were to get an actual demonstration of it (I'm looking at you, Saber ) I'm sure everyone would have a much more complete understanding of the mechanic. I also expect the wiki to be updated with explanations of the new stuff, too. All you really need to do is understand how the system works. Past that, it isn't too complicated.

 

[Captain Fortius]

I'm guessing you put the stabilizers on 45 degree angles from the reactor instead of 90 degree.

No. I built the simplest example possible. 90 degrees. One big brick for the reactor, one big brick of stabilizers of exactly the same dimension directly above, below, left and right. Doesn't get any 90'er or simpler than that.

 

[YamiHikari]

No. I built the simplest example possible. 90 degrees. One big brick for the reactor, one big brick of stabilizers of exactly the same dimension directly above, below, left and right. Doesn't get any 90'er or simpler than that.

Hm. I feel like I've been hearing that a lot in the past few days. It seems there was a bug that was missed that messes with the inclusion of stabilizers. The same issue came up in a discussion with another player last night, as well as in one of Gmodism's recent streams. But don't worry, the devs have been made aware of the issue.

 

[Nosajimiki]

Just play a bit with the game and you'll find that perfect balance.
And by the way, 45° is not necessarily the best for reactor axis.
I got this salvager with 4 groups of stabs at the front of the ship. I angled one plane of the reactor axis to 21° and i got 200 % + buffer from my stabs. So i was able to remove quite a lot of them to come down to just above 90 % buffer AND at the same time i reduce my instability from - 186 to -16.

You've got to try things, excel sheets is way way horrible to do when playing. Fiddling is better.

I think you are missing the point that people will reverse engineer any mechanics this game has down into formulas to optimise their building. While finding exact mechanics in game via experimentation is fun, once you have them, they should model well. People don't just want to be able to use formulas to engineer better ships, it prevents tons of wasted time on over or under built systems that can force major redesigns of almost finished ships. Most older players have learned the hard way that 10 minutes of math can save you 10 hours of building. Power is the first step to EVERY system in the game. It is the most important system to model well.

While novice builders can make a functional ship just fine through over engineering things, making the math more complex does not force everyone to build that way. It just alienates the "middle-class" builder who likes to optimise, but is not an IRL programer/engineer/accountant/etc. Old power, once you understood what it was doing, it was easy to plan out. I could figure out the exact block counts needed for a turret in my head while taking my lunch break. It was complex enough to be fun, but not so much that it was frustrating.

In short, the complexity of such a basic system makes the game less accessible to a general population.

 

[kikaha]

I think you are missing the point that people will reverse engineer any mechanics this game has down into formulas to optimise their building. While finding exact mechanics in game via experimentation is fun, once you have them, they should model well. People don't just want to be able to use formulas to engineer better ships, it prevents tons of wasted time on over or under built systems that can force major redesigns of almost finished ships. Most older players have learned the hard way that 10 minutes of math can save you 10 hours of building. Power is the first step to EVERY system in the game. It is the most important system to model well.

While novice builders can make a functional ship just fine through over engineering things, making the math more complex does not force everyone to build that way. It just alienates the "middle-class" builder who likes to optimise, but is not an IRL programer/engineer/accountant/etc. Old power, once you understood what it was doing, it was easy to plan out. I could figure out the exact block counts needed for a turret in my head while taking my lunch break. It was complex enough to be fun, but not so much that it was frustrating.

In short, the complexity of such a basic system makes the game less accessible to a general population.

Well i'd say it depends on how you do your builds.
I "plan" my ships or station in my head.
They get a shape, a function then because of that function they'll get some changes to their shapes and so on. All the while i'm taking into account that my power should go there (in a "logical" kind of way) my systems need to be around here cargo, living spaces, work spaces etc have to be here to be coherent with the build.

I do all of that at the same time. the most i've done before starting a build is a drawing and looking at pictures of inspirationnal ships/stations.

So maths for me... never use them. Yet my builds are fairly decent when it comes to combat. far from being the most feared of fighter but good enough.

 

[Nosajimiki]

I've built my fair share of ships that way, even won a few fights with them; so, I understand your opinion. But I've then gone back and re-systemed those ships later once I understood the math only to find out that I could make the same ships pack 3-4 times as much firepower with the same sized power system just by being smarter about how I handled my load-balancing. More importantly, I could predictit with very simple math how much space and mass I would need to set aside to make a power stable ship; so, many of my latter ships were not only stronger, but took way less time to make work correctly.

Don't get me wrong, endpoint systems like shield regen and weapon damage can get away with being a bit less transparent because if a single subsystem comes out a bit wrong, you still tend to have a usable ship that might just need some fine tuning at some point, but power is the first step to everything else. If you make it easy to mess up, you make everything easy to mess up, and the need for total ship overhauls goes through the roof.

FYI: I'm not saying that you have to build your ships systems first, but you do need to know if the systems you need will fit into the hull you create.