[ANN] MemoryCoin

[stas]

it took few minutes , now both of my clients synchronized

[1714688865]

1714688865

[1714688865]

1714688865

[1714688865]

1714688865

[1714688865]

1714688865

[FreeTrade]

Sounds like you're on the right branch. Other updated nodes are -

addnode 24.6.21.198 add
addnode 180.183.154.90 add

[ajet]

im on block 1573 now. is this correct?

[stas]

I am on 1573 too

[gym520]

blocks" : 1573,
"currentblocksize" : 1000,
"currentblocktx" : 0,
"difficulty" : 0.00002835,
"errors" : "",
"generate" : true,
"genproclimit" : -1,
"hashespersec" : 15.03759398,
"pooledtx" : 0
"testnet" : fals

 how to know  all the  hashespersec ?

[FreeTrade]

im on block 1573 now. is this correct?

Yes. The blockviewer is up to date on the new branch -
http://memorycoin.21stcenturymoneytalk.org/

To be certain, you can check your debug log to ensure you have the 1573 block on the right chain - the log file should have the following hash near the bottom.

SetBestChain: new best=000009a170f40fb99e18407c9248e857dcec9980d9076d892ee503f99ca5c3ea  height=1573

[FreeTrade]

blocks" : 1573,
"currentblocksize" : 1000,
"currentblocktx" : 0,
"difficulty" : 0.00002835,
"errors" : "",
"generate" : true,
"genproclimit" : -1,
"hashespersec" : 15.03759398,
"pooledtx" : 0
"testnet" : fals

 how to know  all the  hashespersec ?

Looks like a good rate you have there . . six cores?

I anticipate difficultly should go down at block 1580 as the clients on the wrong branch are disconnected.

[stas]

looks like we are OK

[Stinky_Pete]

Clients on 2 machines updated and sync'd. (Windows XP & 7). Contents of wallets appear to have survived, including one block mined about 13 hours ago but not yet completely confirmed.

My IP 86.146.143.252

Now on block 1573 also

Is it safe to start mining again?

[FreeTrade]

Is it safe to start mining again?

How do you think we've been making these beautiful new blocks?

[SlyWax]

I think you are not flushing the debug.log every line !

My "tail -f debug.log" is one block behind the getmininginfo block number.

[Stinky_Pete]

{
"blocks" : 1581,
"currentblocksize" : 1000,
"currentblocktx" : 0,
"difficulty" : 0.00002014,
"errors" : "",
"generate" : true,
"genproclimit" : -1,
"hashespersec" : 18.16713767,
"pooledtx" : 0,
"testnet" : false
}

[B.T.Coin]

blocks" : 1573,
"currentblocksize" : 1000,
"currentblocktx" : 0,
"difficulty" : 0.00002835,
"errors" : "",
"generate" : true,
"genproclimit" : -1,
"hashespersec" : 15.03759398,
"pooledtx" : 0
"testnet" : fals

 how to know  all the  hashespersec ?

Looks like a good rate you have there . . six cores?

I anticipate difficultly should go down at block 1580 as the clients on the wrong branch are disconnected.

No, I think 4 cores.
This is 6 cores:

{
"blocks" : 1581,
"currentblocksize" : 1582,
"currentblocktx" : 1,
"difficulty" : 0.00002014,
"errors" : "",
"generate" : true,
"genproclimit" : -1,
"hashespersec" : 26.75097276,
"pooledtx" : 1,
"testnet" : false
}


Unfortunately the multicore bug is still not fixed so this machines only runs for a few blocks before hanging or crashing
Hope you can fix the bug very soon.

[YukonCoinelius]

ok, I squeaked onto the new code at block 1579, as soon as I saw it today. 8 hours from code release to hard fork time probably sets a new record.

FreeTrade, please consider my timeline comments above for any future forks/protocol changes. A conservative time frame from code release to hard fork time is about 4+ weeks into the future. Obviously you can get away with less with a smaller/new coin, but you really do not want to get in the thicket of multiple competing blockchains. Even when the main one goes forward intact, if there's outcry from miners who “lost their coins” on spurious chains, it detracts from the reputation/confidence.  

Looks good so far - nice work and support on sustaining the blockchain!

btw, I am also getting the crashes/hang on multi-core mining under Linux (it takes down the whole machine, very severe). (actually, to clarify, I was getting that on the old code, but have not yet reproduced a crash on this latest code - it usually takes 0.5 to 1+ days)

[B.T.Coin]

I'm pretty sure the hanging and crashing on multi-core machines only happens at the moment a block changes.
It doesn't happen while mining on a block, just at the exact moment when data has to be processed to move to the next block.
More data to process (at grant-blocks) means a bigger chance of the miner hanging or crashing.

I haven't looked at the code, but would it be possible to pause the mining on all cores very shortly every time data has to be processed for a block-change?
I'm pretty sure this would solve the crashes.

I think the crashes happen because one core processes the data to change to the next block, while the other cores keep on mining on data that is changed during that period.
This would also explain the more frequent crashes if you have more cores.

[FreeTrade]

ok, I squeaked onto the new code at block 1579, as soon as I saw it today. 8 hours from code release to hard fork time probably sets a new record.

FreeTrade, please consider my timeline comments above for any future forks/protocol changes. A conservative time frame from code release to hard fork time is about 4+ weeks into the future. Obviously you can get away with less with a smaller/new coin, but you really do not want to get in the thicket of multiple competing blockchains. Even when the main one goes forward intact, if there's outcry from miners who “lost their coins” on spurious chains, it detracts from the reputation/confidence.  

I agree that with a more established coin, a more conservative time frame is appropriate. As this one is so new, I thought the risk of losing momentum and the number and severity of the problems justified fast and decisive action. The still network looks in good order, although we've lost a few seeds. I hope we'll get those back in the next day or two. Overall I'd say it's gone very well, and we're in much better shape now.

Looks good so far - nice work and support on sustaining the blockchain!

btw, I am also getting the crashes/hang on multi-core mining under Linux (it takes down the whole machine, very severe). (actually, to clarify, I was getting that on the old code, but have not yet reproduced a crash on this latest code - it usually takes 0.5 to 1+ days)

Thanks, let me know how that goes. I think the voting code was leading to instability. It's still in there, but shouldn't be called except at startup. I'm going to prioritize stability over the next week - for the network and the clients.

[B.T.Coin]

(actually, to clarify, I was getting that on the old code, but have not yet reproduced a crash on this latest code - it usually takes 0.5 to 1+ days)

I have had my machines go down on the new code after block 1566 already (many times) so the bug is still there.

[FreeTrade]

Well, I've been mining with 2 quads uneventfully for 3-4 hours.

Is there any way to get an estimate of network hashrate?

I've made some calculations, but I think I must be out by a factor of 10 somewhere. Maybe someone smarter can review them and tell me where I've gone wrong!

A difficulty of 1 equates to 16^8 = 4,294,967,296 hashes per block
MC current difficulty is .00002191
 
4,294,967,296 * .00002191  = 94102

So you'd expect to mine one block per 94102 hashes.

Say you're doing a rate of 10 per second . . that's 864,000 per day . . so you'd expect to mine

864000/94102 each day . . . 9 blocks.

Which seems out by a factor of about 10 to me.

[YukonCoinelius]

btw, I am also getting the crashes/hang on multi-core mining under Linux (it takes down the whole machine, very severe). (actually, to clarify, I was getting that on the old code, but have not yet reproduced a crash on this latest code - it usually takes 0.5 to 1+ days)

Thanks, let me know how that goes. I think the voting code was leading to instability. It's still in there, but shouldn't be called except at startup. I'm going to prioritize stability over the next week - for the network and the clients.

Hung system again, after a few hours on 4 cores. 

[gym520]

This coin only CPU mining ？