Hi devs,
I have one account in my wallet with a balance of 620000 NEM and get an importance score of 0.03% to harvest a block. I have now splitted the amount to 4 accounts in the same wallet, one with 20000NEM one with 50000NEM one with 100000NEM and one with 450000NEM. After waiting a day it seems that i am now able to harvest with all 4 accounts. Every account now gets an importance rating of 0.02% to 0.03%. It seems i will get a ~4x better chance to harvest a block by splitting up the balance in the wallet to 4 accounts. I wonder what will happen if i would split the amount in the wallet to 62 accounts with 10000NEM each?
I would like to know if my notice is correct an if the behaviour is intended? Or is this just a temporary effect?
What are the addresses of those accounts?
In general, if you have a rather low balance and start splitting into more accounts you will gain importance because you let NEM flow from one account to another and the receiving accounts enter the calculation.
This is wanted and lets people with small balance boost their importance a bit.
But if would try that with a big balance (lets say 500M NEM) then you would realize that this effect is smaller, i.e. you cannot boost your importance by a factor 4 in that case. This behavior protects the network from big accounts taking over the network.
Hi BloodyRookie,
thanks for the explanation.
Origianal main account:
TB4OS4Y5CO2JWJ544I7FXLO2JAVTIGOWONUUYF6H ~450000NEM with current score:2.51
Newly created accounts to which i send the NEM:
TBWKKB2JAAWCSWME2IZGFJAWBD2PJHVN3TFMATXJ 100000NEM current score:2.69
TB2GUPYDAAZPPP3AL5G5LJ2O3ALYPI6YLKAOIFNI 50000NEM current score:1.98
TBBZNKLNYH4ANBIAGZPL6FTT2EQ2VP2AKQNJMCNE 20000NEM current score:1.91
Honestly i have no exact value for the score of the main account before i send out the NEM to the other three. But the harvest display in the client has always shown 0.03% for this account so i guess it stays more or less the same. And the factor 4 is as well only a rough calculation, i basically was interested if the effect is intended, and it seems it is.
Thanks.
What are the addresses of those accounts?
In general, if you have a rather low balance and start splitting into more accounts you will gain importance because you let NEM flow from one account to another and the receiving accounts enter the PR calculation.
This is wanted and lets people with small balance boost their importance a bit.
But if would try that with a big balance (lets say 500M NEM) then you would realize that this effect is smaller, i.e. you cannot boost your importance by a factor 4 in that case. This behavior protects the network from big accounts taking over the network.
Just curious here. What'll happen If I split my 150mil account into 15000 accounts that each have 10k NEM each :D. Now the ratio is 3.8%. How much would it be then roughly?
I would estimate the sum of all importances would be 6%.
I would estimate the sum of all importances would be 6%.
so its actuallly far more beneficial for larger accounts to split up their accounts into lots of little ones?
I would estimate the sum of all importances would be 6%.
so its actuallly far more beneficial for larger accounts to split up their accounts into lots of little ones?
Perhaps in the absence of transaction fees. This would only be feasible if someone had a lot of NEM in the first place. In a way, this rewards people who run nodes to support the network.
Hi, guys.
Where can I see the algorithm for the importance score calculation? And the "forging" algorithm itself?
Hi, guys.
Where can I see the algorithm for the importance score calculation? And the "forging" algorithm itself?
You can't yet. Whitepaper isn't published yet because the algo is still being tuned.
Hi, guys.
Where can I see the algorithm for the importance score calculation? And the "forging" algorithm itself?
Hey mthcl :)
You want to do some calculations for nem?
Yes, it would be interesting for me. But I really need to know the details 
Unfortunately the poi algorithm undergoes changes at this stage. So I guess you have to wait for the final version of the algorithm before you can start to investigate it.
OK, I can wait. But are you really sure it undergoes the right changes? 
Hopefully, mthcl. You noticed anything that contradicts this?
No, since I have no information at all. But, for example, do you use something similar to the NXT's BaseTarget adjustment algorithm?
Well, I would call it similar, yes. But there are differences in the actual algorithm how the difficulty is adjusted. You can see the result (block times) here:
[url=http://chain.nem.ninja/#stats]blockain stats
Use the slider to adjust how huge the part is you want to see.
Did you see this topic, by the way, https://nxtforum.org/proof-of-stake-algorithm/basetarget-adjustment-algorithm/ ?
Thx for pointing to that topic, mthcl
@mthcl I've sent you our retarget algo in PM
Thanks! I've answered it, but, for some reason, it doesn't appear in my "sent mail". Did you receive it?