Topic: [ANN] AEON [2019-09-27: Upgrade to version 0.13.0.0 ASAP HF@1146200 Oct 25] - page 219. (Read 625666 times)

Took a few hours of work, but it's done. Not used to Javascript.

My pool now estimates its share of the total hash (currently about 60%), and displays the effective fee (currently same as base). The hash rate estimation is done from the recent blocks found, so should not lag behind difficulty as it currently does.

I'll put that on github when it's switched on tomorrow and confirmed to be working well.

Yes minergate has about 40 KH/s as well. And I'm sure there are a bunch of solo miners out there too. I know there are people using the --donate option since the donation fund gets some blocks, although it is fairly small (which is fine, every little bit helps!).

BTW, I think a better estimate of pool share would be the fraction of blocks mined by the pool over some recent time period, say 1-2 days.

Note: this is one such time where the conditions are producing a poor instant estimation of nethash. Right now, if one only reads the stats on Moo's pool, one will think that it has 100% of the nethash, at about 150KH/s. In fact, there are 70KH/s at Arux pool at the moment as well, so the hashrate distribution right now is not particularly terrible.

Wonderful!

One thing to be careful about in implementing this is the disparity between the methods of measuring the pool hash rate (measured by counting shares over some time window unknown to me) and network hash rate (measured by difficulty, which is a trailing function of blocks over 720 blocks or about 2 days). This can produce very peculiar results such a one pool appearing to have >100% of the hash rate.

I updated the pool listing in OP

This is advance notice that in a bit more than 24 hours, my pool will switch to a variable fee system:

When the pool hash rate grows past 50%, the fee will start growing as 1% extra fee for every 5% of the network hash rate. So base fee being 2%, 60% of hash rate will cause the fee to jump to 4%. 100% hash rate will bump it to 12%. Both the pool fee and the donation get scaled the same. When the pool hash rate goes back down, the fee automatically goes back down with it. The above formula is not set in stone, I may need to tweak it depending on how it goes.

Hey I don't want to set expectations unrealistically high. I don't have a magic bullet to "fix mining", just something to hopefully keep the problem at bay here for a while.

I like some of the ideas from spreadcoin but there are major problems with it too. It comes down to, given what you call "pool culture" (and I like the term), still being able to pool if you can trust the pool. That means the biggest and most trusted pools will probably be the most used, potentially making the problem even worse.

Thanks for the diatribe. It is appreciated here.

Naw, I mentioned this in the IRC maelstrom that happened today. pool culture is just ubiquitous. I think its primarily because of ease of use and abject laziness.

Honestly, I don't know whats going to fix it, but I think its the number one problem in cryptocurrency right now. And yah know what, I'm gonna get myself another beer before I go on this diatribe. Because i feel it'll actually be appreciated here. There's been some response in the monero thread regarding this component, but not enough to really satisfy my tinglies.

I'm really hoping you have something slick up your sleeve smooth. I mean, what could be the strategy? I mean one of the primary problems is the fact that a malicous miner could spawn multiple instances of the mining function, and assuming this miner could always just hide themselves behind layers of abstraction...... I guess its just one of those things that I assumed was too difficult to solve, and hence it hasn't been solved by now. Because anyone who follows anything about fucking cryptocurrencies knows that decentralization is what gives it value. The lack of the ability of a single power to execute exclusive control is THE value of this technology. Otherwise you just end up in the same game to brinksmanship that exists now. Trustlessness comes from this decentralized architecture. You lose the decentralized architecture, you lose the trustlessness.

Of course, the alternative is that smart mining actually takes hold. But again, the problem could be pooling, because if Verisign decides to take on Monero payments, they're terminals could all well be pooled.

It's a fascinating problem - how do you actually isolate the fact that an individual is commiting resources as opposed to an orchestrated group of individuals? What is so unique about a given effort that can be parsed on the backend ( the protocol)?

Well, suffice it to say that the beer hasn't helped. If anything, its obvious things are more turbulent. Regardless, its still there. This feeling that it is actually possible. You can't do what spreadcoin did because even with spreadcoin trusted members were able to share their work.

Unless there was a way to prove that a hash hadn't been created by an alternative key. But then, one could simply share a key.

So, i solve the block, I provide the solution to you, but the blockchain doesn't accept the solution because it knows that somehow I created it, and you didn't.

Which is just cryptography, is it not? Keeping information contained?

wait a minute.... this is just an extrapolation of a ring signature I think. Except the one holding the key that is true is the blockchain. Is that a thing?

Ok i give up. I'll just wait for this solution of yours.

The interesting thing is there is not really anything about PoW's centralization pressure at work with this coin now, as far as I can tell.

As someone else pointed out, you an solo mine on one computer and get blocks every single day (and even getting blocks less often than that would still be fine for a small miner, and perhaps preferable to getting dust from a pool). There are virtually no orphans since we switched to four minutes, which was kind of the point, and I'm happy to say it worked. (I see maybe one per day on my nodes.) For that matter, Mooo's pool isn't even on the greatest connection, so if orphans were an issue, his pool wouldn't be a good choice anyway. But they're not.

If you look in the pool blocks on Arux's pool, plenty of blocks there, even with the current low hash rate, and again no orphans. So no reason not to use it if you don't want to solo mine, or are using GPUs and don't want to set up your own pool, can't afford the resources for a full node, etc.

What this tells me is that a large part of observed centralization of mining (speaking in general here, not just this coin) is something other than rational economic behavior. It is habit? Laziness? Brand loyalty? Comfort with that approach? I'm not even sure.


We're on same page. I support this, and stay tuned.

Unfortunately, most POW encourages the centralization of hashes so this is not a trivial problem to solve.
I don't have any ideas here.

While miners may not care for the long term prospect of the coin, I think pools need to do more than just ask people to move their hashes.
To achieve the same goal for a higher earning, pools can increase their fees to, say 10%, 15%, 20% temporarily. I think most in the community will understand this action.  To win a majority of the blocks, you don't need 100% of the network hash power.

There are many more innovations to be implemented for this coin, I hope miners are not so short sighted that they kill this coin before the features can be added.
 

What's your suggestion? Sincere question. I have some plans in the works but new ideas are always welcome especially if they can be implemented immediately.

Ultimately it comes down to how much the community members care about it, and it seems many don't. They just want to mine on one big pool and probably dump right away (I'm guessing here, of course, that if they wanted to hold longer term they'd care more about the legitimacy and such and not just mine on one big pool.)

Beside miners, pool operators also have the responsibility to ensure the well being of the coin. If no measure are taken to prevent more than 50% hash,  it will undermine all the great features and the legitimacy of this coin.

This is not good, we have a single pool approaching 100% of the network hash rate. 

I moved to Arux's pool after Moo's had the issue with the daemon and I haven't looked back.   In the long run, the payouts will theoretically be the same.

This is a great coin and community.  Really impressed.

+1

I've moved over myself

I second this, and add the suggestion of trying out solo mining if using a high'ish-end CPU.
I'm still solo'ing on the daemon with a single i7 4770 and I think there has not been a single day that I don't hit at least one block.
Some days I get two, so, double-yay! 

I really wish some of the folks on Moo's pool would move over to Arux's.  If you're concerned about the pool not paying out as much, don't be.  Instead of getting lots of 1 AEON payouts, you'll get less frequent, but much higher payouts.  It works out to be just about the same.

You explained it so plainly that even I do understand :-) Thank you very much.

Absolutely. The reason is that, the network hashrate is purely an estimation, based on the time that it took to find the last X blocks and their relative difficulty over that period.

In a similar sense, the pool hashrate is also just an estimation, since it is simply adding up the number of shares that it receives from miners over a certain period, and their relative share difficulty.

If you throw 100 KH/s at the pool right now, the pool stats will quickly update due to the massive influx of shares, while the network hashrate will take a good while longer to reflect the added hashrate, as the necessary sample data is much larger and the estimation takes much longer to settle after a drastic change.

Not sure I was able to come up with a half-decent description on this... Well, I tried 

http://aeonpool.mooo.com/ now:

Network
 Hash Rate: 125.87 KH/sec

Our Pool
 Hash Rate: 152.17 KH/sec

Can a pool have got a bigger hash rate than the network?

Bittrex wallet transactions are unfrozen.

This can't come soon enough...



all we need is another fork  


That's 80% of network hash!