The 19 july statistic is flowed
ypool fixed it block finding few days ago.. now you can see its finding 5-6 block per hour...

Also mine i5 haswell find a block per few days so not so bad, and i think diff started to drop a bit

you have to have backward compatibility for blockchain integrity checks but its not too hard to implement.
trivial case is to add something like
if (block.nheight < switchoverheight) {
  block.checkblockOld(hash);
} else {
  block.checkblockNew(hash);
}

It is probably possible, like mustyoshi said.

However, if that is to be done the difficulty has to be taken into account, since different hashing algorithms have different speeds. Moving from, say, SHA-256 to Keccak, would entail a significant hashrate alteration (some searching around gives me 12.5 cycles/byte for Keccak and 14.3 cycles/byte for SHA-256, a difference of ~15%), resulting in probable difficulty plunges or spikes. A good difficulty adjustment algorithm would be able to handle this properly.

*sigh*

Just because it crashes doesn't mean the process ends.

Megacoin, for example, has a tendency to freeze under certain conditions. The process is still running, and it will even accept RPC requests, but it never responds. A watchdog would have a tough time realising that the daemon needs to be restarted.

Similarly, how would a watchdog know that the local chain is a fork and not the correct blockchain? These issues require manual intervention.

Finally, no sysadmin in their right mind would "write a cron script" (whatever that is, you're mixing technical terms that have nothing to do with each other) to poll ps and restart a process when all of the major init systems already support this functionality. The vast majority of major Linux distros nowadays use Upstart as their init system, and Upstart bakes in auto restart functionality, respawn limits, etc. A sysadmin may write a custom watchdog to poll RPC calls, or he may choose to write this functionality into an Upstart script.

You're right - ignorance *is* a two-way street.

I'm announcing middlecoin.com

It's a new mining pool with several features.

- Switches to the most profitable coin to mine, like multipool.in, but with several more coins. The switching code is accurate and fast. At the moment, it's monitoring 12 coins, with several more suspended.
- Automatically exchanges coins for bitcoins, and pays miners entirely in bitcoins.
- Predicts the value of the next block, for coins with variable block rewards.

There is no registration. Simply connect your scrypt miner to middlecoin.com:3333 with your bitcoin address as your username, and any password. Payouts are automatic, and occur once a day, if your balance is greater than 10 mBTC.

See the FAQ at http://middlecoin.com/faq.html

1Dbj9BZsuKvqEjXFGNDAsXb7ZUQ1fW2sCz

thank you!

1a4Bwyheka94fPWqFt9sJUCT73TvNqnPk

thanks!

I think I saw an estimate that difficulty 8 was 30x harder to find chains for than difficulty 7. (But verifying just means checking one more prime, so it's only 1/7 harder in that case.)

I'm curious of the efects of the network once difficulty gets really high and prime numbers to be found are insanely difficult to compute let alone verify by nodes to be efficient in any meaningful way.

There is some use to the data gathered for pure mathematicians (disproving conjectures, possibly discovering new patterns). In the same way you get programs online that allow people to search for higher and higher prime numbers (often Mersenne primes). It's usefulness is in a similar vein to those. The coin in itself will also strongly motivate people to write efficient algorithms that crunch through prime numbers, so a by product of the coin could be a breakthrough in that direction. A lot of 'pure maths' doesn't (yet) have any real world applications. Over the last hundred years though time and again pure maths that was thought to not be useful was later found to be ideal for solving a problem.

At the end of the day it's still primarily a currency, with 1 minute confirmation times. That's it's main purpose. But as a bonus, at least the proof of work contributes to society in some way rather than meaningless forgotten hashes.

Once the URL target changes, the block can no longer be validated.

How about create a P2P (decentralized) cloud. which can be used to store data or computation.
This cloud can be used to host websites, host personal files, do distributed computation. It has a bitcoin-like block-chain to store the payments/transactions. Something similar to Namecoin but more dedicated.
I think it will be much useful.

Another paper with a nice table, gives you some idea on how things will go.

http://trone.di.fc.ul.pt/images/e/e2/ASAP11-paper.pdf

GTS8800 [17] GTX8800 [10] GTX260 (This paper) GTX580 [39] Intel W3565 [46] AMD Phenom II 1090T [46]
Cores 112 128 192 512 4 6
Frequency (MHz) 1188 1350 1294 1544 3200 3200
Price (USD) 250 173 100 500 300 200
TDP (W) 150 155 202 244 130 125
GFLOPS 399 518 715 1581 102 153
Modexp/s 6504 11074 41426 149464 32608 77002
Modexp/s (scaled) 13052 15282 41426 46973 N/A N/A
Modexp/s/W 43 71 205 612 250 616
Modexp/s/USD 26 64 414 298 131 385
Table I
COMPARISONS OF MODULAR EXPONENTIATION PERFORMANCES ON VARIOUS CPU AND GPU IMPLEMENTATIONS.

When the masses come in next year they're gonna buy into ideas that either sound good or ideas they agree with.  Some will buy a coin just cause it sounds cool or its associated with someone famous or has a good rep.  Hardly anyone is gonna ask well, does this one waste less energy than this one and are the transactions on this one faster than on bitcoin.

Also, to be honest, what development work is there on Noirbits besides changing some Litecoin parameters? I feel like I'm the only one not drinking the clonecoin kool-aid around here.

i am starting to believe in the hard coded thing. I have been considering methdos of reserving then taking addresses.

Example, i create a new coin. I have hardcoded 10 addresses to each receive 0.3 %

the first one is mine, the next two will be for the two best pools over a period of 1 month. The next two are for The first two exchanges to take the coin.
One will be for a faucet, another for manual bounties and another for giveaway. the last two will be reserves in case the dev team needs a hot wallet.