Some comments on my previous post

stellaris 2 - Some comments on my previous post

Previous post was A brute force statistical study of Interceptor-type corvettes.

This post rambles a bit. The TL;DR is at the bottom.

I worked on my code some over the weekend. I regret to say that barring a sudden increase in my computing power, I won't be able to expand the weapons options I simulated. In order to avoid too many possibilities, I divided the weapons and defenses into six tiers: one for each technology tier (0-5). If a weapon wasn't directly upgraded in the next tier, I kept it for one tier; this only really applied to tier 4->5 (no more upgrades, just a new weapon – Matter Disintegrator) and to the Energy Siphon and Null Void Beam. All others were replaced from one tier to the next (i.e. red lasers don't exist in Tier 1, only blue lasers). I'm still ignoring missiles, and only simulating small-slot weapons (so no kinetic artillery). I also reduced each fleet to a single ship: every battle is 1v1.

This worked fine for tier 0: there are only 16 possible ships, so I only had to simulate 240 match ups to cover the entire parameter space. At about 1/3 of a second per match up, this didn't much more than a minute. I saved the data for later visualization.

It got a bit muddier at tier 1, when we add the Energy Siphon. The number of possible match ups jumped to 1560. This took about 10 minutes to simulate.

Then we hit tier 2, where we add all the specialist weapons – plasma launchers, autocannons, disruptors, etc – as well as crystal infused plating. To cover tier 2 comprehensibly would require ~1.4 million match ups, which would take my home PC about 300 hours to compute. Needless to say I didn't run it.

I still have the comprehensive tier 0 and tier 1 data sitting on my Google drive. I plan to draw it up sometime this week and post it.

Now, onto the other bit: I put some thought into why the mixed builds did so much better than the non-mixed builds: the multipliers are not even. I'll draw a slightly-unrealistic situation for you to illustrate. Imagine we have two ships, and each of them has two 10-damage guns and two 60-point defense items. We won't worry about hull for now – lasers and mass drivers deal normal damage to hull, so once we get there they are equivalent. We'll also ignore accuracy and only count both-weapon hits for this. Also, for argument, we'll pretend one ship doesn't fire back.


There are only two weapon arrangements we need to worry about: double laser, and laser+cannon. (Double cannon's math works exactly the same as double laser.) Further there are three armor arrangements of interest: double armor, double shield, and armor+shield. So, six scenarios. Lets start with double laser.

Against double armor, the lasers deal 1.5x damage with every attack: so 10 becomes 15, and we are dealing 30 damage with each hit. This defeats the 120 points of armor in a mere 4 hits.

Against armor+shield, we first have to defeat a shield to deal with the armor. Since lasers deal 0.5x damage to shields, the 10 becomes 5 and we deal 20 damage with each hit. This takes six hits to beat the shields, plus another two to beat the armor – a total of 8 hits, twice the time required to beat the double armor.

Against double-shield, using the math above, we need 12 hits to beat the defenses. This is an eternity.

As you can see double lasers have a wide swing in their time to beat the defenses. If you see your opponent loading lasers, you should totally stack shields on them – or vice versa, if they're stacking armor, bring lasers. (Double-cannons have the same math, just backwards.)

Now lets look at a cannon-laser. It turns out that no matter what defense these are firing at, they will deal 20 damage per hit: one weapon is at disadvantage, one weapon at advantage, so, 10*0.5 + 10*1.5 = 5+15 = 20. So regardless of what your opponent is packing – double shield, double armor, or a mix – you beat it in 6 hits. Sure, if you are wearing double armor and your opponent is holding double laser (or shields vs cannons), they can get you in less time. But in every other circumstance your mix weapon setup wins.

If we turn this on it's head, and look at the defenses, we notice that twinned defenses will survive 4, 6, or 12 days, depending on the attacker's weapons. By comparison, we see that mixed defenses will survive either 6 or 8 days, again depending on the attacker's weapons.

If you mix weapons and mix defense, the only thing that stands a chance is another mixed weapons and defense ship… at least until we start dealing with more exotic weapons, then my math goes out the window.

TL;DR: Expect more simulation data in a more-readable form later this week, once I figure out how I want to display it. Also, mixed weapons and defenses are best if you don't know your enemy's setup, and usually even if you do. My simulation last week confirmed this.

Source: Original link

© Post "Some comments on my previous post" for game Stellaris.

Top 10 Most Anticipated Video Games of 2020

2020 will have something to satisfy classic and modern gamers alike. To be eligible for the list, the game must be confirmed for 2020, or there should be good reason to expect its release in that year. Therefore, upcoming games with a mere announcement and no discernible release date will not be included.

Top 15 NEW Games of 2020 [FIRST HALF]

2020 has a ton to look forward the video gaming world. Here are fifteen games we're looking forward to in the first half of 2020.

You Might Also Like

Leave a Reply

Your email address will not be published. Required fields are marked *