Topic: [XMR] Monero - A secure, private, untraceable cryptocurrency - page 1907. (Read 4670606 times)

This coin (and technology) is either gonna be a Big Hit or a Big Bust...
Not the kind of marginal alt where your short term logic might apply.

Sure, that's possible, but it's speculation either way. Previously, so-called "gpu unfriendly" algos have generally shown order of magnitude increase in hashrate/watt when a GPU miner comes out. Even if the GPU miner is as low as, let's say, 1.5x the efficiency compared to a CPU miner, you've still got your influx of the GPU miner community who are less eager to dump at whatever the going price, meaning the effect of botnets is diminished, even with conservative assumptions.

I agree in part -- yes, the history so far is short, and an over-valuation doesn't affect fundamentals. But on the other hand:

My interest is long term, and that is the only reason why I give a damn about day to day price. That might sound contradictory, but consider that the success or failure of any coin depends on adoption, and for better or worse, most of the money is crypto, particularly in the early stages of a fledgling altcoin's life, is in the hands of speculators who only care about day to day price. Most people who put money in crypto have little to no idea how to analyse fundamentals. So, sure, put the work into the fundamentals, but don't ignore market dynamics to the point where speculators lose all faith in the coin. If it turns out that MRO is currently overvalued by (let's say) 5x from recent hype and eventually finds equilibrium around .0007, that may be beyond breaking point for investors & miners to abandon it. Meaning the need for a solution (like a GPU miner, as I'm suggesting) is urgent. On the other hand, if it turns out not to be so overvalued, the urgency is less. But afaik there's no reliable to estimate this. The way I see it, there's nothing to lose by kicking off a bounty for a GPU miner right now, hence the suggestion.

It depends on when they last found a block mainly (5hrs ago at the time of writing), the pool code assigns each miner a share of the reward according to hashpower contributed and then when the coins have matured it sends them out to the miners. On the bigger pools which find blocks every few minutes they may run a regular hourly payout job instead so miners arent receiving lots of small payments, but the smaller pools dont have to worry about that. Doesnt make any difference to what you get paid in the long run, on a smaller pool luck is more of a factor for the first few days (if the pool gets lucky and finds a couple of quick blocks you will receive more coins) but it averages out over time.

Network hashrate has wobbled by about a megahash in the past few days - that gives a rough idea. They're certainly the majority of the network.


The emission's contribution is independent of the dumping, though. Any curve will tend toward equilibrium. So the real issue is that there was too much hype and MRO reached a non-equilibrium price. It went up from by a factor of ~8x over just 24 hours, and only declined to ~3.5x of the initial price. But this would've happened regardless of what the emission curve was. This is why I'm confused by your partial focus on the emission.


Possibly, sorry. Much of the complaining about the emission curve stems from a demand to change it discontinuously, without a retroactive correction (read: instamine). I'm pro-GPU mining.

It won't make as much of a difference as you'd imagine - the whole point of the CryptoNight algorithm is to reduce the gap between GPUs and CPUs. You may find that even with a decent GPU miner, it is still more cost-effective to CPU mine at scale.

Either way, I think we've still got a very microcosmic view - we're looking at 5 weeks of history, and we're particularly interested in they last, what, 10 days? All the downward pressure in the world isn't going to affect our ability to innovate and continue to improve Monero by adding utility instead of adding hype. If you're interested in Monero on a long-term basis, it's probably best to ignore the market for 3-6 months whilst some sort of equilibrium is sought and whilst innovation continues to come from us. We aren't trying to drive things hard for a bit and then disappear, so sweating the small stuff that happens in the market on a daily basis isn't going to have any long-term benefit for us as the core team:)

[/quote]

thanks again, will need to look into this more but for now I'll hash away.

quick question do exchanges or pools offer a mixing number? and is ones IP address associated with blocks found when mining solo or in a registration free pool?
[/quote]

I dont think they do at least for now. And i also think that right now your IP can be associated easily if you dont use a proxy. But there is a partnership with i2p network that should deal with it soon.

Thank you my friend.

Thumb up 4 you 

cheers

What kind of numbers are we talking?

To me, botnets are an inevitability for any popular cpu coin, so my skepticism would need a lot more than rough estimations of top users' hashrates based on what people claim in IRC chats. By that I mean, I'd be skeptical if I were you and that's all I had to go by.

I agree that it's meaningless to refer to an unstated standard -- but I wasn't doing that. Any given reward structure is fine if buy pressure can at least keep up with sell pressure to maintain the price. It doesn't make sense to only consider the math of the block reward in isolation, ignoring other variables. For e.g., bitcoin could get away with a higher emission rate because of its market penetration / network effect. The point I was making was simply that the two factors together (dumping and emission rate) contribute to downward pressure on price. I could have phrased it both ways to say the same thing:

"dumping from botnets may mean the current price is unsustainable going forward"
"dumping from botnets, given the current emission rate, may outstrip demand"

So never mind the math of the emission curve, it's beside the point.

It seems you're conflating issues. For that matter, someone writing a GPU miner is not the same as playing with supply.

thanks again, will need to look into this more but for now I'll hash away.

quick question do exchanges or pools offer a mixing number? and is ones IP address associated with blocks found when mining solo or in a registration free pool?

You type this on the wallet: transfer 0 address amount  id

0 is the number of mixing you want to have. 0 is actually least anonymous. Bigger number is better but has more transaction fees.
id is optional. If you want to send to an exchange for now you have to add the id that is given at the exchange.

There may be a botnet. But I know the rough hashrates of some of these cloud miners and know that they constitute a large part of their network on their own. The botnet can only be the remainder after accounting for regular miners.



It's meaningless to talk of some unstated standard and declare emission is a problem. I specifically compared to the main two alternatives: (1) BTC/LTC-like and (2) other CryptoNote emission curves that are slower. Both are worse, if you do the math. Bitcoin is simple arithmetic, CryptoNote requires a loop calculation. Please specifically show your math if you disagree with the numbers, and post the code.

It seems altcoins have gotten cozy with instamines, post hoc supply changes, and other tricks. If you think a well-defined PoW that sticks to its word like BTC and LTC is no longer viable, that's a bit ironic. The point of cryptocurrencies was that there was no central entity playing with the supply.



The equilibrium is set by spot instance pricing.

In any case, you guys can be pretty sure a GPU miner will come out sooner or later. It's never not happened to date.

It's more about the numbers. Everyone has a CPU; only a fraction of people have a decently performing GPU. So for coins where high-end GPU rigs make up the majority of hashrate, the effect of botnets is reduced.

the point limlim is correctly making is that the price can never reach an equilibrium, or at best a very low one - this even gets more relevant when the price would go up massively.  

the problem is that money needs to have a store of value function - it would be sad if this really promising project fails due to that flaw.

I don't really follow the logic here. Yes, you can expect butt mining to be a large chunk of hashrate for any cpu coin (if it's profitable). But you can also expect a large chunk to be botnets. Unless you had some specific evidence or reason to believe botnets are not in operation? I'm skeptical -- botnets are generally the most profitable way to cpu mine.

I don't buy the logic here either. Emission rate is an issue if it's contributing to sell pressure that's kicking the ass of buy pressure. Coins can be abandoned & forgotten in weeks/months if they underperform, whereas the timescale for the emission rate to taper down is in the order of years.

thanks, that a great start, next to figure out how to transfer, how does one send coins?

Interesting observation and suggestions. Are GPU's not as easily bot net'd? I haven't heard of that problem on LTC, so wonder... I imagine many GPU miners are on Linux and also mining only rigs, so that might help lower the incidents of bots.

My one real fear with CPU coins is just what you mentioned btw.

I agree that a GPU miner would stabilize the economy better. But these two factors aren't correct. It's not a botnet: there are definitely multiple miners with a large number of EC2 instances mining away at MRO and they dump to cover their costs, which can be thousands per day. (I can even name who some of them are from IRC chatter.) This is an inevitable part of CPU mining. The same is true with GPU/ASIC mining because of farms, but probably worse in CPU mining because no prior investment in hardware is needed. You can whip up cloud mining instances at will if the coin is profitable (which MRO is).

The emissions rate isn't an issue, as I covered in this post: https://bitcointalksearch.org/topic/m.6986004

We're more inflationary than a coin that deliberately cuts reward as difficulty goes up (like DRK), but much less inflationary than a PoW coin with constant rewards like BTC or LTC (at the same age). And actually somewhat less inflationary than a slower CryptoNote emission curve like QCN. The latter point may seem paradoxical, but you can plot circulation or inflation v. time to confirm this. It's because the most inflationary part of CryptoNote emission is right at the start and slower emission curves actually spend more time in this regime.

1) Yes this long string is your address
2) Yes, you can mine without opening the wallet by typing on daemon     start_mining  yourwalletaddress numberofthreads. Thats for solo mining...
3) Yes there is only one address but you can make more wallets and i personally run at least two on the same daemon. I think there arent problems with that but i am not sure.

better others asnwer for 4 and 5

So I've been buying up some MRO at the current low, and I've been watching it closely from its beginnings. The market is doing some unusual things, and I have a tentative analysis. I believe the buy pressure is not being adequately represented in its price movements. I suspect this is because of a combination of two factors:

1. A large proportions of the hash being from botnets (where the operators immediately dump), and
2. The relatively high emission rate.

This creates a disproportionate downward pressure on price that differs from other coins, and which appears to be consistently outstripping buy pressure, despite strong interest in the coin. This is not a good thing, and it may drive the price into the ground. Without the expectation that the price will go up, fundamentals don't matter a bit. Miners and investors will lose confidence and move on.

There is a solution to this: write a GPU miner. In its current form as a CPU only coin, if the MRO market doesn't ramp up buy pressure to consistently match/exceed the expected constant dumps from botnets, the coin will fail. And it's easier to scale up botnets than it is to scale up the ordinary userbase, so this disparity may worsen. Your average GPU miner is more likely to hold than your average botnet operator who's looking for immediate turnover and who doesn't care if the coin lives or dies. Perhaps equally importantly, a GPU miner would massively increase adoption among the community, which is another thing currently holding MRO back.

I know people here may be currently enjoying nice profits from CPU mining, and may not appreciate the suggestion of a GPU miner. Personally I'm more concerned about the longer term future of the coin, and I think Monero has a vital role to play within the anonymity niche. Its fundamentals are good and I'd like to see that reflected in its market performance. I'm willing to put a good chunk of MRO towards a bounty for an open source GPU miner if there's enough support for it.

- on all these UNIX tuts on monero.cc, I would add "--no-check-certificate" parameter to all "wget" commands
- the debian script needs: "apt-get install python-software-properties" added, as the used "add-apt-repository" command is not installed default