Post
Topic
Board Announcements (Altcoins)
Re: [ANN] [QRK] Quark | cPoW | PC mining | Stability | Hashcows - QRK Payouts
by
BugSpirit
on 05/07/2014, 20:26:38 UTC
@Kolin
Quote
ha ha i'm sorry to laugh guys but 'm hearing a lot of "save Quark" ha ha

Yes Kolin, save Quark. I mean, Quark will most likely "survive" but I guess even to you it is not appealing to hold 247 millions of Quark and noone else wants to exchange it with you. As you may have noticed many people sold their Quarks and if you are going to laugh away problems that everyone else in the active community agrees exist then I can see more people turning their back on Quark. The hashrate issue is directly connected to price and trust in Quark, so yeah, we should and will do something about it.

@reRaise
Quote
Very interesting, i would be glad if you go into details.

Conditions:

A merge mining coin only makes sense to me if
a) there will be a prospect for a long-term value and
b) it is no direct competitor with Quark

This is why I am against a random roin that merge-mines. The coin should have a longer distribution scheme, let´s say 20 years and a different blockrate, say 3-6 minutes. This would make it distinguishable to Quark and avoid a fast boost with a short bust. When I said this could be an "experimental" coin I didn´t mean to think of it as Bitcoin Testnet but rather as a coin with a higher probability of hard forks when implementing new features.

To experiment on a different coin would sort of solve the discussion on whether we should change the Quark source code. As you know, some people wanted to push things forward and other were sceptical if that wouldn´t we a too high risk factor. To have a longer blockrate and distribution scheme would avoid the coin becoming a direct competitor with Quark as when it comes to adoption for many in-store trades 30 seconds would still be more attractive than 3-6 minutes and the inflation rate would be way higher as with Quark.


A copy from reddit. http://www.reddit.com/r/QuarkCoin/comments/29vtyt/a_nonstandart_merged_mining_hashrate_solution/

What if we create a merged-mining cPoW coin (let's call it SPARK Smiley ) with STANDARD money mass distribution. 30sec 50-coin block with halving every 4 years. Spark will allow GPU-miners to loot mine the coin on a trully asic-resistant algorythm.
Merged mining means that the task of solving Spark block equals solving Quark one and all the spark hashrate automatically adds to Quark one. MM pool will distribute 2880 QRK a day and 144k SRK. Thus Quark will gain additional hashrate without hard-forks, changing economical mechanisms (PoS) or inflating its money mass.

There are three mid-term scenarios.
-- SRK will be engaged in massive mining pump-and-dump schemes and all the other activities of our free market. On its way it gain much more popularity than Quark and maybe higher price (50M SRK in first year vs. 247M QRK). Large bagholders of Quark may dump all the coins and switch to it for some time diminishing quark status to Namecoin in BTC-NMC pair. But eventually as money mass of SRk inflated and PnD fade the capital will move back for in QRK where it will be safe from inflation.
-- Noone knows and care about Spark. We loose nothing.
-- For miners QRK is additional reward for mining Sparks, but much more scarse (50 to 1). Psycologically more scarse resource is more valuable. For this reason Quark will engage in PnDs and mining activity too if such exists.

In any case we loose nothing as on the side of qUark will be only a redistribution of a minor miners' rewars which in any case rewarded to someone every block. On the other hand we could gain more hashrate and security, use CPU- and GPU-miners as a resourse. They are now frustrated and lost in ASIC hysteria and very like asic-resistant algos even if the algos are not trully such (scrypt-N or x11/13/15). If we give them something they used to and show a quark logic and phylosophy as an unnesesary option, some of them will definitely understand and join us.
I could start a bounty let's say in 20k QRKs.