Dark Orbit - finally a "real world" application for math

Discussion in 'General Archive' started by El_Burro, May 19, 2014.

Dear forum reader,

if you’d like to actively participate on the forum by joining discussions or starting your own threads or topics, please log into the game first. If you do not have a game account, you will need to register for one. We look forward to your next visit! CLICK HERE
  1. so in simple words please what is best configuration in for lets say one has to be speed one slow, and assuming both are fighting on this way, because i still didnt get if you calculated upgrades etc
     
  2. El_Burro

    El_Burro User

    From all the information scattered in this thread ;)
    For example, I determined the average cost for an epsilon gate here.
    With a 25% chance for a LF4 as a reward you need on average 4 Epsilon gates per Lf4...I think you get the picture.

    In simple words: there is no such thing as "the best" configuration. As it turns out, there is quite a large range with good configurations. I created this post mainly because I fought lots of players who had no idea where to start searching for a configuration that is competitive. For more information on the assumptions I made to get meaningful results please read the introduction on this topic.
     
    Last edited: Nov 7, 2014
  3. I am reading the first page and just want to exclaim. I'm building a Beta Gate. Can you explain to me how if I'm activating a multiplier EVERY Time! at X2 I can not hit a Beta Part multiplied for 3000spins? I reached 45beta parts 3000+spins ago and haven't hit a multiplier since. I have gotten 2 single parts the last of which was 1200+spins ago. My Beta build currently stands at 4,700spins and counting!! I completed a Gamma build when I had 45beta parts and began a new one that is already at 75 parts at 2,875spins which would sit likely on average to the mean pace. But really I feel like I'm setting some kind of precedent here by building a Gamma from scratch before receiving my last 3Beta parts.

    My previous 4 Beta gates btw went at a 2169spins. Fairly close to your mean number.
    Also Btw, My previous Beta gate build I reached 44parts and within 100spins I managed to complete for the final 4parts. The one prior I did in 89spins. The one before that 500spins for final 4parts. and the first one 340spins for final 4 parts.

    Just saying again, I'm now at 3,000spins w/o a beta part when activating the multiplier!
     
  4. El_Burro

    El_Burro User

    First of all: There are sill quite a few people playing Darkorbit, lets assume 10'000 people for simplicity.
    If each of these 10'000 players builds 100 gates on average that are one million gates.
    The confidence intervals I gave for the number of turns are 99% intervals. So 0.5% of all gates built MUST have taken a higher number of turns than in the rightmost column of the table you are referring to. That are 5000 gates in total in our example.
    So I am not surprised that people come here to complain that their gate took a much higher number of turns than any of the values I gave in the table. Blame statistics, not me ;)

    But we can definitely say that you were extremely unlucky. Even in my simulations of 10^11 turns, only 0.007% of all Beta gates took more than 4700 spins with optimal strategy. Yet unless things like that happen very often, there is no reason to assume that something went wrong.
     
    Last edited: Nov 11, 2014
  5. Well add another 300spins w/o hitting a Beta part when activating a Multiplier. That's the question I'm posing you. If it's 33% Beta part and based on a 13% part spin rate how many spins should I have expected to hit the Beta Multiplier once? I'm now at 3,300 spins since hitting on it 5000 spins overall on this gate build.
    If I do the math in my mind we're about 1 in 7 spins after a multiplier should net a gate part. 1/3 of that occasion is a Beta so wouldn't it be 21 multipliers activated I should net a Beta part? In my latest 300spins I got 45part spins(nice) 3 Alpha gate multipliers for 6parts 7 single Alpha parts and 1 single Gamma part so that means I had 35multipliers if I'm correct? I should have activated the Beta gate almost twice if I'm correct? Now keep in mind this has been going on for 3,300 spins not reaching the Beta gate part in an activator? oh year 3/35multipliers=8.57% in a super small sample but I feel like completing gate build lately has gotten out of hand. If indeed I should achieve 13% rate of hitting an activator and I'm 2parts short or less, one should hit that multiplier almost twice in 100spins. Yet I've documented in a Zeta going 0-29 in activators missing 29 straight before some magical mercy appeared and I hit 3 of the next 4. But it's strings like that where I think oh I'm 300spins away from completing the gate if not less and then proceed to watch 300+ under 600spins go by before getting the final gate parts.
    I appreciate the numbers I've tried explaining this to my clan how to lower their overall spins taken to complete a gate. They were the ones trying to tell me use the X6 or leave an Alpha/Beta gate completed to get the Gamma parts and I told them how that's a bad idea wasting multipliers that would be the first dozen or so parts to a new Alpha/Beta.

    Im just super frustrated because I know that I'm approaching what should amount to a 3rd Beta Gate built in the amount of spins it's taking to finish just this 1 gate. I'm missing out on what? 53,000 uri+40,000 X4 ammo almost twice over now. And I'm pretty sure I'm not going to magically build multiple Beta Gates for 1600-1700 spins multiple times to make up for this occurence. It's also frustrating that if Im struggling to get Beta parts on activation, well then I should be seeing a rise on Alpha/Gamma activators and with it you'd assume a lower cost spin to the mean for building those two gates. But my Alpha builds since this Beta began have taken 1,831/1,735/1,641/1,619 continually less at least but still more than your Mean avg. The Gamma gate I completed which began before this beta build was 3,835spins and I'm now at 3,085 with 6parts still needed to complete. At least appears I should beat your mean number for a Gamma but it wont be by much. I am just sick of watching all this uri+UCB-100 I should be making disappear in to nothing!
     
  6. El_Burro

    El_Burro User

    It seems like you need a hug more than an explanation. But since I rarely hug, here are my two cents:

    from 3'300 spins with every multiplier activated instantly you should get on average 3'300*0.13=429 gate parts.
    Lets assume that from these 429 gate parts about 1/2 were multipliers, that would be 215 multipliers.
    Using every multiplier instantly, you hit 215*0.13=28 gate parts with activated multiplier.
    Now the probability of NOT hitting a single Beta part with activated multiplier is (2/3)^28=0.0012%.

    I already said that this is some serious bad luck, but still possible.
    But look at it this way: In your entire DO career, you will build quite a few Beta gates. I am at 96 now.
    If I built 95 of them at the avarage cost of 2261 turns and one of them took 6000 turns my average is still 2300.
    You simply cant expect to build EVERY SINGLE GATE with the average cost. Some of them will be much more expensive, some of them will be cheaper. And the cost will sometimes even be outside the confidence intervals I gave in the table. There is no point on worrying about the cost of one gate.
    Maybe this helps: If you didnt use the best strategy but x6 multipliers instead, you would encounter Beta gates with more than 5000 or even more than 6000 turns much more frequently. If any of the Beta gates you build from here on until the servers are shut down takes more than 5000 turns again although you use the best strategy, you can report back and I will give you that hug.