The recent price variation displays that magi runs the opposite to BTC, unlike those altcoins such as LTC following the same pattern. It is mostly the reason that one may cash out BTC to gain benefits. All altcoins, except those mainstream ones, may not avoid this end. It isn't all about price, but as far as price is concerned, altcoins are more likely pumped high and dumped low. When attention (speculation for example) is shifted off the coin, getting lower price is almost certain. This is awkward but it'll be weird if we talk about rational norms in this land filled with "coins" coming out everyday. A coin with number of coins generated in PoW each day is under the pressure of price lowering; however, diminishing this cause is yet to eliminate the essential reason towards low price. What's matter factoring in this process and stabilizing or pushing the price is wide adoption, by people, though merchants are one of the channels, they are the middle part reaching more people, and mining is the bridge too. In contrast to wide adoption, centralization is obviously the one which needs to be avoided. When cryptocoins are talked in nowadays, decentralisation, which, in certain conditions, is very much a dream, is a need; heading towards a dream directs what's need to be done.
To make magi to be towards a mainstream, legitimate approaches should be followed in my view. What/how those are legitimate? To be frank, I did have naive thinkings and concerns initially on the price; to think of legitmate, we might drop off price factors, or diminish the role of the factors which worry about the price too much. Though people care about the price, one must admit P&D isn't the way we're heading. We can push magi in a formal way. With that in mind, what's happening should not simply cure a symptom, or raise up additional unsettled issues. We may push out technical contributions which are useful in general and are very much needed to this land.
We may hold on a few basic features for magi without compromising for the purpose of curing the things, nor should we eliminate some of them as a remedy to the "brokens". Magi supposes 1) coin distribution among people, 2) low costs in maintaining the network; at the least these two make transformation positively towards the wide adoption. It would be very tough to make a purpose (or sort of) to please the miners. What's need of mining for profits is mostly contrast to the idea, and this isn't the purpose of the way that bitcoin is designed either. Instead how coins are distributed into miners is the crucial process. What's proposed by people in prior posts are not unlikely and doable but we'll need research & verification to make them sounds feasible firmly. In my limited knowledge by far, the following is a to-do list, mostly, which I had ideas coming up time-to-time, so they aren't minutes ideas at this writing. You may not expect perfect solutions which, without more depth research, are barely possible, neither the bitcoin is as well.
Listed are in order of preferences. Don't expect too soon; I am keen to carry on but would not be able to at this time. If one has patience enough, we may see these coming along at some point (I'm not rushing on the coin but more pleased to be with it (plus some personal stuffs as mentioned earlier)).
1) The algo and it'll need a look at possible change in the way how hash is found; however, if the FPGA is supposed to do anything like CPU, nothing special. We'll get the first-hand experience with FPGA or GPU etc for know-how, however. As I pointed out it is nearly impossible to deal with mining farms by any techs, and what needs to be dealt with are their capabilities.
2) Improvement on the difficulty adjustment. The difficulty issue is actually the center of the issues. This will benefit dozen of aspects and thus this improvement is almost a must. The block reward adjustment is based on difficulty; because of its insane variation, difficulty used in the block reward calculation must be averaged out by taking into account many past blocks. For this reason, there is a lag in what's happening in hashrate and what's generated in rewards. A proper adjustment in difficulty will benefit the blocks being generated in an equal time spacing by which we can avoid the 51% attack, and the network is quite secured. Obviously business merchants are pleased by this fact too. The more profound solutions in the difficulty may come out changes in the hash finding. Implementation of this is absolutely a great contribution to this land.
3) A smart adjustment of the reward. I pretty think this is crucial but we need a smart adjustment of the reward to be super responsive. With the improvement of 2), this won't be difficult. There are few situations:
a) big miners are for profits, they will be frustrated by the rewards, and they won't stay long;
b) big miners fire 51% attack, though I suppose the improved block time would resist this to happen, extra works need to be done;
c) big miners abuse the system; for no reasons, they stay. The PoW stops running by i) trivial / zero rewards, ii) shutting down PoW blocks. I need a comprehend thinking on this. Absolutely this would cause critical concerns since generic miners get nothings. These are the remedies:
i) I recalled a post about the idea, one runs a node and gets the rewards; I believe we could implement a thing like masternode or the like; under a (protection) mode, partial of coins turn to the network who run the nodes. I am actually not a fan of enforcing people to have a certain amount of coins to get the right to claim coins but it can be likely.
ii) We have been developing an idea of running cloud mining with the revenue used to buy XMG. I believed a portion of the coins can go to encouraging individual miners. This is yet to be in plan of implementation, and also not a solution of 3).
iii) This might be insane, so weird idea. The magi team build a super miner which comes along and compensates the hashrate because miners' leaving when coming the big miners. This is not a solution of 3).
In my view, the solution to the big miners' abusing is quite passive, and there are no real solutions to that by far to my knowledge. Shutting the PoW down and put the coin in a PoS mode is the way going. This solution is far better than bitcoin, though it is unlikely, but technically if one sets a super miner to control the network, that's the end; if this is not happening to bitcoin, but could happen to other altcoins.
It quite sounds weird by a fact that the big miners takes more than half of the hashrate. When we implement the above, we won't need to worry too much about that, as long as the network is secured. For the business side, it's none of their worries about who is mining, right?
4) This is pretty a plan I mentioned and am actually reluctant to talk it about. I am keen to implement a solar powered miner and make zero mining cost; this is one of the reasons we need PoW and would be great to reach more people. Mining should really go to the casual phase in my view, and should never be taken as a serious business (then I wouldn't order my KnCminer; giving them a chance to abuse or cheat the system), though it is likely a dream, it is a dream I'd like to advocate. Imagining solar powered miners running all over the world 24 hour a day and 7 days a week without interrupts by the costs, no matter how big miners are on or off. You are seeing that one may abuse the system, but I see a fact: coins go to majority hands than minority. Network security isn't an issue once we get improvement on the difficulty; surely security concern should be censored.
5) Any more?
Here's the socialized, partially centralized solution, are you ready comrades?
Is it possible to fix the PoW block reward at perhaps 5 xmg and pay it out to only one address in the network, no matter which miner finds it? Then using an advanced form of PoM rewards we can then pay out miners with equal amounts from funds in this address.
Each miner having at least 2000 XMG could register their address and use it for mining. PoM payments could automatically go out evenly each day to all miners with at least this minimum balance and a consistent 24 hour hash rate.
Having a certain amount of XMG (2000?) at an address that gets the PoM payments is needed so that people who are not supporting the coin through saving value can't register multiple miners for PoM payments.
I am thinking the reward reduction as fragilefungi mentioned, however it may not sound a right one way to go. Taking the BTC strategy (it may not sound a standard but it was an example), with 50 BTC/block initially, 7200 BTC in a day, which is little high XMG average amount of 5000 coins in a day. Provided rising in XMG's value over time, we will expect more miners and then reduction in the generated coins (plus yearly reduction see
https://bitcointalksearch.org/topic/m.9991269). Pushing maximum to 5 XMG may frustrates generic miners a lot, though we can hold the price and prevent big miners (they may also abuse the system with no reason), we may reach less people than expected. There exactly exists a balance point, when less coins are being generated in the long run. The mining revenue should be back to similar to level as what is now (suppose) that needs a reduction in the number of miners. If one is to abuse the system without caring about mining, then it's the same end.
Mining to a single address may sound like we make a big premine and distribute into the community as many coins done in the past.
In the long run I feel we need to gradually decrease the PoW block reward and go full PoS. (I thought this was the original plan) This will reduce potential issues. The way things stand right now can you imagine the problem that would happen if XMG was worth $500 a coin? A large hash rate miner can take almost the entire PoW block rewards for themselves - as a guess maybe over 400 or 500 xmg in a day? (The low PoS rate by comparison generates maybe one or two XMG per day for large holders.) That kind of potential would cause many investors and business some issues and I think it is best we do something about it before the coin gets more valuable. Kudos to people for pointing this out by the way, if there is a problem with the system we should address it, even though it is difficult.
The initial goal is like going towards the full PoS, but I've leaned that to make coins more energetic, we should be PoW/PoS. The original PoS is unsafe as people can fire up attack. Magi's PoS-II is a nice one but can't fully avoid "richer gets rich" which is one of the issues in PoS. For this reason, PoS interest should never go high. For a simple reason, I've seen many PoS coins go to dead end; this issue is very severe at the early stage of coin distribution (low coin value), and yes this will motivate people buying cheap coins, and can easily pile up. It is certain that price is pushed to a higher level, but we won't agree that coins held by few hands is a good sign, right? When few people hold most of coins, they are actually like the big miners; with the higher interest, the worse the situation. I'd not say this is a certain consequence but most likely. Coins going wide adoption which needs more and more people participating and own the coins; coins bough by a few people is really going the opposite. In addition, making people engaging buying as the primary channel to get coins is something. In my view, PoS is better to be an additive, which will take an important role once the PoW is done; while I am little concerning about the inflation by comparing with coins like bitcoin.
I forgot to mention the assumption of coin price. With price rising by 50000 times to be around $500, every one will gain 50000 of profits. Is this a case? No. Following is actually what happened; with the price rising, there will be more miners joining which reduces the amount of coins generated and individual shares as well, mostly ending up the same mining profit as you will get right now, or similar to mining other coins; This is a balance point, so called market.
By taking a deep looking at above, I believe this won't happen. Let me know any thoughts.
I agree with Lightsplasher and tittiecoiner here. In just the last few days we've seen nearly the entire block rewards vanish and or go to one large miner, at this point that large miner is single handedly controlling the coin rewards, that day has already come and we really should do something to fix it.
I guess I've covered the most: high interest, low PoW reward. Regarding the big miner, I'd say he is not controlling but prolong the PoW, of course we need an improvement on PoW reward as mentioned.
no, i meant it that way, that not the net hashrate is taken for calculation, but the hashrate of the acutal solver. is this a miner with high hashrate (as a pool for example), the reward should be lower. is it a solominer, the reward should be higher. with an overall cap by the whole nethashrate. that is my idea.
Actually the coin can't see the network, but only the time spent in finding blocks; so it's not likely to distinguish individuals; but I'll take a look at the possibility.
if you want large miner POW frustration, I'll give it to you:
you should kill pools, rework magi daemon to accept up to 250Kh/s - roughly three i7 CPUs
[...]
As some posts pointed out, people can split their hashrate, though we can imagine how difficult it is when speaking of hundreds of miners, I know people can run multiple daemons in one machine. People always have their potentials to do that. But I don't see this is unlikely and I'll take a look at this as well.
Just another example, let's say I would create my own coin using the same algo as Magi does (or another CPU only algo) and need miners. It would be easy t throw some big hashrate for a few days or weeks on Magi to make my own coin look more profitable.
I don't see this maters; they are many other coins out there probably with profits better than magi, and I failed to see really making mining profitable in either PoW or PoS is important though it is to some degree (that's the PoW reward adjustment), but we shouldn't make it as #1. Also I believe people mining XMG are not because mining is profitable.
Scrypt coins have similar problems with multipools. When they switch to a coin, the huge hashrate lets the difficulty go up considerably. The script checks all x minutes if a block has been found. If no block was found, the profit switch leaves the coin that remains with much lower total hashrate and a super high difficulty. In order to prevent this, some coins are using
DGW and similar techniques.
Maybe you can adapt something like this to solve the reward problem?
Exactly what I need to look at the difficulty we need to improve, and magi needs a more stable block time.