2014-Mai-21 23:28:16.33016 [P2P6][sock 94] Some problems at write: Broken pipe:32 That should be moved to log level 2 .. don't worry about it. good point! I've just fixed one bug that made this messages apeared frequently.
|
|
|
I see transaction with about 450BBR that stuck in memory pools(and two other transactons also).
The problem is in fee. Wallet use just default hardcoded fee(and here is problem actually). At the other hand daemon's transaction pool use "greedy" policy to collect transactions into block - and this transaction not fits to this conditions (it depends of fee/tx_size ratio).
To solve this problem we need to improve wallet to set fee up to current tx_pool state, to have fee/tx_size ratio enough to be included in block. It will be implemented soon, but for current version - i strongly recomend you to send big amount with few small transfers not bigger than 100BBR
Senders of those transactions can PM me, i'll try to help with this issue.
|
|
|
IMPORTANT NOTICEAliases now have restricted to this alowed characters: #define VALID_ALIAS_CHARS "0123456789abcdefghijklmnopqrstuvwxyz-." I'm very sorry to all who already assigned some extra symbols. This is made to protect your aliases from "pfishing" or confusing cheats. For example if there some well-known @goodpool alias, i want to make this name safe from same-looking names like @Goodpool or @goodpóól, or @goodpoo|.I'll be back when i finish current developent iteration, with short description of what is done. Good "luck". Your Zoidberg.
|
|
|
I'm pleased to announce that btc-mike kindly offered his help with this project. He started a thread for naming contest(yes, again) - please welcome.
|
|
|
Hello.
2all who wondered what is going on: don't panic. The problem is not in difficulty function. the function is fine. The problem is that someone have pretty big part of hashrate and periodically turn on monster mining.
Probably someone have done optimized or GPU miner. Our scratchpad have dynamically increasing size, but it's still to small to resist GPU, and it's theoretically possible to implement GPU miner.
Miner optimization is endless story and i'm gonna focus development on improving usability now.
As soon as someone will share optimized or GPU hash i'll immediately include it into project repo with copyright. But i'm not expect those miners will hurry up to share it, it's a business. So if some of devs want to support project with development - take care about hash optimization. Good "luck".
PS: sorry if i'm not answering to PM, have a lot of task to do, will answer to all... or almost all.
FYI: Difficulty function can'n be changed after currency launch without hardfork. It's for any PoW currency.
|
|
|
i think guys we have a GPU miners there.
if someone is ready to share his GPU miner, or ready to implement it - i will put sources to project repo with copyright and put respect here.
|
|
|
it took 44.5 minutes to find block 2341, and difficulty is going up
It looks like that the diff is not adjusted smoothly i guess until someone have a pretty big part of hash rate he could swing difficulty of any smoothness. This situation is the pest prove that even if optimized hash exists it is note produced by developer, since he would not be shake his own network.
|
|
|
Difficulty will go to sky.... What is the height a measure of? The difficulty doesn't seem to have changed too much. It changes with slight delay about 100-200 blocks. Difficulty calculated on last 720 block's timestamps(DIFFICULTY_WINDOW), but it begins to react already after the 60 blocks.
|
|
|
funny PoW 2014-May-19 17:03:07.381063 [miner 3]Found block for difficulty: 23421861201 2014-May-19 17:03:07.381563 [miner 3]Block with id: <80812e83e5601e8876a97e52e94e68e33d2d1856242c2fa2e4da38df4fe57bb6> have not enough proof of work: <1288ec1185b73220314913afd3e27a3a4079cc7afe8cea58df27707a4e1091c8> nexpected difficulty: 23421861201 2014-May-19 17:03:07.385325 [miner 3]ERROR /c/boolberry/src/currency_core/currency_core.cpp:370[virtual bool currency::core::handle_block_found(currency::block&)]mined block failed verification
Seems that block have absolutley wrong PoW hash. how often does this happen?
|
|
|
1. How much is the premine? 2. What is the formula for block reward?
1. No premine. 1% of money supply reserved for devs but emission of this reservation is smooth, the same curve as supply emission, and it controled votes in blocks (each block have a vote). 2. Formula: base_reward = (EMISSION_SUPPLY - already_generated_coins) >> EMISSION_CURVE_CHARACTER; But, reward is depends on block cumulative size also, take a look details on website: http://boolberry.com/index.html#reward
|
|
|
Difficulty will go to sky....
|
|
|
Sorry about this man, i'll help with pool protocol and thin miner to anyone who would have serious intentions to make pool.
I want to make a pool but the algo seems too complex for the pool to check the hash. Hash check is much easier than on bytecoin/monero. You just need to use another protocol with incremental delivering scratchpad to client.
|
|
|
Alias case sensitive.
Maybe this can lead to some confusion ! I've expected this thesis when you asked about case) May be, but it's to late to change it. The only thing i can do is to add soft rule in client software, to protect from unintended taking some case-related alias. But if someone is going to do it intentionally, he is able to do it. FOLK, i strongly suggest you to map aliases in lower case, since we going to make default wallet prefer low case aliases in transfers commands, to avoid confusion and cheats.
|
|
|
Good news!I''ve added HowTo section http://boolberry.com/howto.html and put detailed specification of the coin to website and announce. Feel free to PM me if you found some mismatch or mistakes(including english). And i want to ask seriously, where is the bug reports ? Please take a look, it is in our mutual interests to find a bugs now, until we went too far. Peace!
|
|
|
I found nothing in 30 hours of mining. This coin needs some pools soon, solomine is insanely hard.
Sorry about this man, i'll help with pool protocol and thin miner to anyone who would have serious intentions to make pool.
|
|
|
It would take about 3 sec to confirm so hosting a pool for 10 people would cost as much resources as mining directly.
Can you explain about 3 sec ?
|
|
|
Can you transfer an alias to another address ? How ? Are alias case sensitive ?
Yes you can, if you prove that you own this address (sign missage with address private key). Updating implemented in core, and in tests. But have not implemented in command line interface. I think we'll done this in UI. Alias case sensitive.
|
|
|
BoolBerry seems to compile fine with libboost 1.54 But does it work ? I'm asking since other cryptonight don't seem to like 1.54.
DON'T USE libboost 1.54 Thank you to point this, i've updated readme and cmakelists. Ok, anyway it seems the bug is only for windows so you could put this restriction on windows only... Done.
|
|
|
What should I use instead of simpleminer in case I have any number of miner nodes and want run only one copy of daemon?
I've cut off simpleminer since we have pretty different approach with PoW hash(blockchain-based pow), i'll support those who start developing pools with protocol implementation for thin miner.
|
|
|
BoolBerry seems to compile fine with libboost 1.54 But does it work ? I'm asking since other cryptonight don't seem to like 1.54.
DON'T USE libboost 1.54 Thank you to point this, i've updated readme and cmakelists.
|
|
|
|