Do you actually have live stats for the various coins (not delayed/cached for multiple minutes at a time), and do you use the *actual* reward instead of the average reward for "random" coins?  If not, it's yet another site with bad information like Coinwarz.

Multipools aren't killing coins (directly).  They're killing the profitability of all the shitcoins that exist.  Nobody is bothering to pump & dump most coins anymore because any attempt to do so is met by a massive amount of mining power shifting to that coin and immediately dumping 100% of the newly mined coins.  Pump & Dumps are only profitable if you can fool enough idiots to bring the average bid price higher than your average price per coin bought during the pump up period.  With multipools, that just doesn't happen because they will dump coins to the highest bidder within an hour of the pump.

As a result, multipools are going to be fading away pretty fast.  The profitability is shrinking rapidly in comparison to straight DOGE mining (and if that fails, straight LTC mining).  They are a dying breed because pretty soon there won't be enough coins left for multipools to bother with.

KGW/DigiShield/whatever does nothing to prevent multipools.  It actually *helps* multipools because it means those coins will drop in difficulty quickly, returning them back into their coin rotation.  They prevent the coins from literally *dying* due to multipools, but their profitability is gone regardless.

FRY was never convertible, which is why it was never part of auto switching.  MOON is not convertible anymore because Cryptsy removed the MOON/BTC market, which is why it was removed from auto switching.  The only balances people should have of those (outside of the ones who manually picked it) are extremely minor from the minute or two wasted on those due to the bug during pool restarts mentioned above.

FRY, KDC, and NOBL were never on auto switch, except for a bug when the mining process was restarted and it accidentally mined those for a minute or two.  MOON was removed from auto switching about a week ago, a few days before it was removed from Cryptsy (MOON/BTC was removed from Cryptsy).  MOON was similarly mined for a minute or two at a few points during server restarts.  That behavior has since been patched to the pool always defaults to DOGE or LTC when it restarts.

The less I have to touch Cryptsy the better.

Eh, most of them did adopt stratum (only one I can think of that didn't that still exists is Deepbit...and they hardly qualify as existing anymore).

Stratum was a needed improvement.  Getwork was just too damn inefficient for mining remotely, and ASICs sealed the deal.  I can't imagine the horror of my bandwidth bills (or the bandwidth usage for a large ASIC farm) if we still had to use getwork today.  I've actually been debating the idea of a new mining protocol over UDP, but there isn't really a "need" for it, other than the fun of developing something *slightly* more efficient [though a pain in the ass for people to implement probably since UDP would require more work for error handling/lost packets vs TCP].  Would also allow for some interesting DDoS mitigation systems (anycast + multiple destination servers?).

This is a fundamentally flawed argument.  You will never, nor will any pool, ever prevent a 51% attack again.  We're very close to private entities [ASIC manufacturer farms, giant ASIC warehouses, and "cloud" mining services] controlling > 51% of the hash rate already.  It is NOT possible for a public pool to hit 51% ever again unless every other pool vanished off the internet simultaneously.


EDIT:  Exception here possibly being ghash.io.  Though they are majority private speed from their cloud mining service, so I have a hard time calling them a public pool.

https://www.btcguild.com/new_protocol.php

This includes a breakdown of all the major components of stratum, with detail on what each parameter means for each method.


Obviously you're not going to get something with a guide showing you how to build a TCP socket server.  But the above page includes all the methods you need to support in order to implement a stratum server, and a description of what each piece is.  It's a purely plain-text TCP system.


EDIT:  Forgot slush copied most of my new_protocol.php page and added it to his stratum page.



Slush doesn't even post on his own pool thread anymore.

Because they aren't using a pool in most cases.  Then again the 'Unknown' number is also wrong, since it's using the blockchain.info 'Unknown', which is NOT unknown (at least the majority of it).

That transaction has over 70 confirmations.  You missed part of the ID (cb99a0ee184ced5f0433805b52d7150de0d3956b8777413d4cc18d4aad0194a6).  It had *many* confirmations as of the time of your posting, so make sure your wallet is sync'd.  If you use a web wallet (WHYYYYYYYY!?!?!), contact their support.

Just a request:  When you quote somebody...please don't quote a giant half-page (or full page on low res monitors) image.  Edit that out!

I'm not sure if any exchanges have an active MOON/BTC market.  I know on Cryptsy you can trade it for LTC (then trade that for BTC).  Before you do, make sure you have enough for the minimum trades/withdrawals though.

Basically this.  Multipools have become major holders of the scrypt hash rate, and as a result, any coin that doesn't have some kind of community is basically being nuked.  Scorched earth mining.  Pump & Dumps aren't even happening like they used to, which is what kept most coins from falling into oblivion in the first place.  It's pretty damn hard to pull a successful pump & dump when multipools will quickly mine a bunch of <insert crapcoin here> and dump it at whatever the highest bid is.

Nope, the server runs quite well, better than expected honestly, considering how many of these junk coins are using extremely old & inefficient bitcoind/litecoind code.  The biggest problem right now is going to be database size, since it's running on SSDs.  The two largest tables are PPLNS shift details and the BTC conversion history.  The PPLNS table is getting an update on 4/1 when I remove those coins mentioned, and the BTC conversion table is probably going to have some code added to trim older transactions out and store them in a more space-efficient way.  The current DB schema is designed for speed rather than storage efficiency.

Past luck has no impact on future results.  Every block is a completely independent chance.

I replied to an email about luck earlier today, thought this is a good illustration:


Let's say you have a miner that mines at 1 GH/s.  That means every second, your miner is generating 1,000,000,000 (1 billion) hashes per second.  Now, as a miner, your target on BTC Guild (default) is a difficulty 2 share.  That happens *roughly* every 8,400,000,000 hashes, or 8.4 seconds at 1 GH/s.

To make it easy, think of that work you're doing as a billion little scratch-off lottery tickets, where each slot has only two possibilities: Keep Going or Lose.

To get a difficulty 2 share, you need to get 'Keep Going' 33 times in a row.  Once you get that share, you hand your ticket off to the pool as proof that you found one with at least 33 in a row, which happens 1 in 2^33 (~8.4b) times.

Now, your ticket had at least 33, but it may have even more.  If you get enough of them, you "win" the lottery (block).  The pool only cares about you getting 33 in a row, that's what your proving you found, and that's how your paid.  The pool then checks how many more 'Keep Going' entries you have.  Just like a perfect coin flip, each one is a 50% chance.

As of today, the odds are (roughly) 1 in 2^64 to solve a block with a given ticket.  That might not sound so bad, afterall your 1 GH/s is doing a billion attempts per second!  With the pool at 6.2 PH/s, the pool is doing 6.2 quadrillion attempts per second, it surely can't be that hard.

For those who don't know just how fast powers of 2 increase.  While 2^32 is only 4,200,000,000 (a couple seconds at 1 GH/s), 2^64 is 1 in 18,446,744,073,709,551,616.

Pool size is mostly irrelevant.  Smaller pool means you get a larger share of the blocks, but they're further apart.  Smaller pools have more variance, so higher highs when you're lucky and lower lows when you're unlucky.  So your speed really doesn't play a huge factor in picking a pool (although I'd probably avoid pools so small that you're larger than the entire pool).

From what I've been told, the US server is better for connections from Asia.

There's been a lack in updates this week as I've been going over the pool to decide the proper path going forward.  Many scrypt coins are being nuked into oblivion by multipools at the moment, and profitability has dropped sharply over the last few weeks as a result.  Since the 1.6 update to DOGE, it has been 80%+ of our mining time, and about 10% WDC, and the remaining 10% split to various coins for very brief bursts (mostly MEOW/LOT due to their random block rewards occasionally spiking the expected value).


1)  I'm going to be shrinking down the number of coins the pool currently works on, by eliminating coins which are not auto-convertible.  I'll also be removing the entries for the PoW+PoS coins (FTC, NVC, TAG) until I have time to update the pool backend to differentiate between the PoW and PoW+PoS coins.  I'm giving 11 days notice (Dashboard message, so unmissable unless you completely ignore the pool) for those coins that people may have balances for.  Frycoin and Klondikecoin may or may not be fully available due to KDC having a fork with very little warning causing miners to get paid a lot more than they should have (some corrections were made but not enough), and Frycoin is currently broken beyond use for Linux as far as I can tell (permanent "Safe mode" warnings).

2)  Very minor change in reward calculations to reduce the overhead in tracking shifts.  Should be unnoticeable.

3)  Likely adding 1-2 coins to replace the coins being removed, though being very careful about which ones are added.




Depending on how the next month plays out, ScryptGuild may be changed to a more traditional pool for DOGE+LTC, with profit switching between those two (and maybe 2-3 other coins, like WDC, which have shown consistency).  Having more coins is helpful for profitability, but if that profit is too marginal, I'd much rather have a stable pool that allows seamless switching between known good coins and more consistent returns.

100% luck is significantly better than the old PPS.  For the old PPS to be better, luck had to be approximately 95.5% or less, assuming 0 transaction fees and 0 namecoins.  I think that's one thing a few people failed to understand.  Luck wasn't measured against the 7.5% PPS, it was/is measured against 3% PPS (AKA:  Perfect neutral luck minus 3% pool fee).  If the pool finds *exactly* as many blocks as expected for the shares submitted, you'd have "100% luck".