Edited for Brevity...
Secondly- Just to touch on what Jero said- he is correct. With one exception. Digishield was not unforeseen. Back in late June, the suggestion of Digishield was made. I believe by another member of the CripToe team, 24Kilo. There was some chatter about it before, but Kilo was very clear in his warnings about needing to add Digishield. I'm not saying this as an "I told you so" moment, but rather that this has been a discussion in the past. It wasn't until now that there was irrefutable proof that it was an absolute necessity. So while it probably should have been on the roadmap a little sooner, it's on the roadmap now and that's all that matters.
-Fuse
Since Fuse referred to my early concerns regarding diff swings and KGW... I will weigh in on the subject.
I no longer necessarily support the implementation of DigiShield(this will be news to Fuse as I have not spoken to him about this yet) because I don't think that it is the best solution. Let me explain...
I make a policy to not cross post regarding other coins that I hold an interest in... but I am going to make exception as I believe it would behoove the dev team to spend some time reading and researching the diff algo of Bitmark(BTM) -
https://bitcointalk.org/index.php?topic=660544.0BTM launched as scrypt-clone using the latest Bitcoin code base by a very, very talented and modest coder with some very clever ideas. And one of those ideas was a very interesting take on difficulty retargeting algos.
The short version is this - BTM has a 120 second block time with a retarget of 720 blocks(so approx. every 24 hours) but the twist is that the block maturity is 720 blocks. This effectively removes multipools with their mine and dump business plan from mining BTM since it takes a complete difficulty cycle before mined coins can be redeemed and put on the market.
So let's make this practical... say BTM has a diff of 150... a multipool moves in with 4x the hash-rate ramping up the diff to 600... then leaves with the coin and the nethash returns to pre-multipool level... this means that it will take 4x as long to mine through this 600 diff set of 720 blocks... which means that the multipool may have to wait for 4 days before the coins are in the multipool wallet and able to be put on the exchange... because no coins will be confirmed until all the 600 diff blocks are mined and the diff retargets for the next 720 blocks... this means that multipool cannot capitalise on the exchange rate that triggered the multipool to mine the coin in the first place... so multipools are essentially unable to mine BTM profitably and leave the coin alone.
Here is a quote from Mark regarding his reason for using this retarget algo opposed to using KGW, Digishield or other retarget algo that change the diff with every block...
Good, our friend intelle is back, they have jumped on and off the network since day three I think.
Difficulty, revisitedSatoshi told us, and we have quoted before:
In the absence of a market to establish the price, NewLibertyStandard's estimate based on production cost is a good guess and a helpful service (thanks). The price of any commodity tends to gravitate toward the production cost. If the price is below cost, then production slows down. If the price is above cost, profit can be made by generating and selling more. At the same time, the increased production would increase the difficulty, pushing the cost of generating towards the price.
now what has happened with other currencies, is that they have changed the special formula from "If the price is below cost, then production slows down."
through short diff changes and things like kgw/dgw they have changed it to "If the price is below cost, then production cost falls"
which has killed them, or made it hard for them to survive
whereas with us we have the special formula from satoshi, on sunday I wrote, in slack:
so maybe what follows is only a temporary alignment for a day, or two max, followed by a slow/fast cycle like before going from 100-400 on diff, and double the price range of the last on market
for maybe 2 weeks
through which time volume on both could intensify, and close the large spread
so at diff 400, the network may quarter, slowing production down, giving the market a time to adjust, and making our btm more scarce, and thus more valuable
so the adjustment goes up, not down, on market - as long as we earn it
and with that, our network changes difficulty to just over 400, and we enter this phase.
Let this be a lesson to trust those before, and reinforce our decision to not change how our difficulty works.
So the short version is that KGW,etc cause the cost of production of a coin to fall when the difficulty rises... thus making coins cheaper to produce... when what we really want is not for the cost of production of coins to drop, but for the production of coins to slow.
Allowing the cost of coin production to fall introduces cheap coins to the market pushing down the value of the coin by increasing supply at lower price.
Slowing production of coins makes the coin more scare, increasing demand by reducing supply.
I was extremely skeptical about this retargeting implementaion, but a brief look at the BTM pricing will show this works well, if you spend some time researching and analyzing the data... you will see that the price of BTM has risen every time the diff has jumped sharply because the cost of production has fallen.
While they are huge miners that roll into BTM when the diff falls and chew up the 720 blocks in no time... BTM has remained usable with a maximum block time of about 20 minutes at worst.
Here is a good summary of BTM after the last big diff change...
Difficulty made a very weird jump. It went from 642.97400619 to 160.something (estimate was about 360) to exactly 642.97400619 again.
Every 720 blocks the difficulty changes. So it will be most profitable to mine the 720 easy blocks, but your payout will be slow as it takes 720 blocks to mature.
So the miners will need to put some hash to it also at the high diff blocks to get their payout. This ensures that miners wont be dumping, but selling of slowly (if they sell, i think alot just hold).
Just another genious thing about this coin.
When it is slow there is less currency mined too, sometimes 5x less.
BTM Supply: 792240 BTM, and we had planned: 957700 BTM - there is 165460 less BTM in the world
So while the hashrate is slow you find there is no currency released that was just mined, and no new currency to speak of. 2-3000 BTM per day when the trading volume is 10-20x that is not much.
So the net effect is that BTM is actually being emitted slower than scheduled... again increasing demand because of reduced supply.
My apologies to the community for expounding on another coin, BTM is not a competitor of NLG, BTM and NLG are both very unique coins focused on two very different goals, but have a common thread of being recently launched scrypt coins wanting to embrace ASIC's and the strength and security that ASIC's bring to a network.
A side benefit of the BTM algo is that being based on the BTC/LTC code, it is essentially exploit proof and time-proven the most secure retarget difficulty algo to date.
I am hoping that JohnDec2 will weigh in here as well on this subject.
Thanks for reading,
24Kilo
Edit - I have been mining both NLG and BTM since launch and have a large stake in both. As for NLG... I am one of the 1700...
