good morning and thanks for your answer.
> Your blog page is the real outcome based on luck for anything but true PPS.
no. its not based on luck, but luck is a part to consider. of course luck plays a big role when i measure real outcome for 14 days, but lateron i will try another approach which i use longer time range and try to eliminate as much luck as possible. so you can always say pah it luck, it nothing worth. or you can try to get as close to the expected value as you can. so lets go.
sigh, okay i try to explain the difference between luck and expected value and what i want to tell with my review :
there is a good explanation of expected value at wikipedia:
"In probability theory, the expected value of a random variable is intuitively the long-run average value of repetitions of the experiment it represents. For example, the expected value of a die roll is 3.5 because, roughly speaking, the average of an extremely large number of dice rolls is practically always nearly equal to 3.5. Less roughly, the law of large numbers guarantees that the arithmetic mean of the values almost surely converges to the expected value as the number of repetitions goes to infinity. The expected value is also known as the expectation, mathematical expectation, EV, mean, or first moment."
so i stated at least 2 times in my review, that of course 14 days does not really fit the law of large numbers. thats why i asked you which date range you would suggest for being a large number where in my results the expected value would be more exact. that the expected value is always influenced by luck is true per definition, thats why expected value converges to a value, and cannot be exactly measured in this case.
so you cant say slush is better or antpool is better.
thats why i wrote that the trend is interesting. and the trend is, that slush and antpool have good earning values for bitcoin miners, and that ghash with a high probability is some pool which has hidden fees or some other things decreasing your output. i also sent my hashing power to GHASH when it was lucky and found 28 blocks (11th december) and when it was unlucky and only found 20 blocks (10th december).
so here is my approach for eliminating luck as much as possible, maybe you can add a suggestion for eliminating luck more, let me know and i will change the setup again.
since i only have 1 Antminer S4 i cant do this experiment with it, but i have 2 of my Antminer S3 with the same setup, kernel versions and an average almost identical hasrate of 454 GH/s so my experiment will be based on comparing 2 pools - if i had more identical machines i could add more pools, but by now we will compare the 2 pools GHash and Slush.
i now will use my continued numbers of average outcome per block after i sent 1 S3 for at least 4 hours to slush and when i sent the other S3 for at least 4 hours to GHASH at the same time.
experiment starts now, 18th december 11:54 am CET, as of 18th december 3:54 pm CET (when the full share power in both pools are build up) i will count all blocks, norm them to 25 BTC and state the normed outcome of Slush and GHASH and calculate the average earnings per day with these networks. you can see now, about 1 day after starting the experiment, the variation and luck factor is VERY high, but i think after 2 weeks or 2 month we will get a nice converging expectation value of the real outcome of both pools. i will also add a chart the next days that you can see the converging values better evolve with the law of large numbers.
GHASH has -because of his size - a fast PPLNS system where your power is fully build up in 4 hours, and slushs system is also building up your full power in 4 hours (maybe 3 hours would fit also, but i want to keep it as "unlucky" as possible, so 4 hours.)
wait! we also eliminate the additional transaction fees by setting each block value for this comparision only to 25 BTC, thats what i call "block value norm" and "my BTC norm" and "sum BTC norm". we also keep in mind, that slush has a 2% pool fee, and GHASH has 0% pool fee so we expect that the expected earning value of slush must be in the long run 2% below the expected earning value of GHASH.
so here is the live review:
https://bitcoinsfaq.wordpress.com/2014/12/19/bitcoin-mining-2-longtime-slush-and-ghash-mining-pool-race-started/i will update my post at least once a day, so you can watch this experiment life.
regards,
ronny
