SmallChange [research-only] [Litecoin based] [15 seconds blocks] [*update now*]

[alex_fun]

Sounds interesting.

It's not so simple because there are many other constants which calculated from maxblocksize. We need ~ 64k for maxblocksize at such blocks rate. But it will affect max transaction size, because it's calculated as maxblocksize/5. This should be changed also.

[lightenup]

I think that we need to perform the following steps:

1) contest for the real name
2) contest for the best logo
3) send petition to btc-e

 

lol

Well, I know the momentum is good right now and of course it would be nice to recycle (parts of) this coin and create a new more thought through one (in fact: anyone can just fork and do so right now).
At least I wouldn't do this, until we have some more novel changes that make the coin solve problems better than other currencies.. also I really like to have some models first that give us hints how the coin could behave in the future or during attack scenarios, .. .

But block size and transaction volume limits should be adjusted before.

working on that.

[Otaci]

Please include more detailed build instructions for Linux.

Ubuntu server 12.10 - extract from my sysprep script (run by admin)

Code:

# dependencies
sudo apt-get -y install build-essential libssl-dev libdb++-dev libboost-all-dev libminiupnpc-dev

# enable in firewall
if ! sudo ufw status | fgrep --silent 9030
then
  sudo ufw allow 9030
fi

Extract from installation script (run under user account which is going to run the daemon)

Code:

#!/bin/bash

# smallchange installation
# intended to be run as the user who will run the service

# get software
if [ ! -d ~/smallchange ]
then
  cd ~
  git clone https://github.com/bfroemel/smallchange.git
  cd smallchange/src
  make -f makefile.unix
fi

[Otaci]

Is there another pool going right now?

[Otaci]

Stat data for the last while

timestamp (Central European Time),difficulty,networkhashps,blockcount

Code:

2013-04-22T20:20:10.287287,0.00358563,446273,8952
2013-04-22T20:24:02.725794,0.00358563,436780,8956
2013-04-22T20:30:01.859238,0.00358563,423760,8964
2013-04-22T21:00:01.929899,0.00358563,353620,8998
2013-04-22T21:30:02.363443,0.00358563,1690777,9139
2013-04-22T22:00:01.436636,0.00358563,2019693,9379
2013-04-22T22:30:01.794579,0.00358563,1375014,9571
2013-04-22T23:00:01.865779,0.00358563,827224,9669
2013-04-22T23:30:01.955984,0.00358563,769046,9760
2013-04-23T00:00:02.027385,0.00358563,603336,9821
2013-04-23T00:30:01.116974,0.00358563,579135,9890
2013-04-23T01:00:01.187934,0.00358563,611724,9962
2013-04-23T01:30:01.277549,0.00358563,607901,10031
2013-04-23T02:00:01.348973,0.00238493,349995,10084
2013-04-23T02:30:01.437857,0.00238493,378793,10158
2013-04-23T03:00:01.511540,0.00238493,463318,10239
2013-04-23T03:30:01.600787,0.00238493,492856,10322
2013-04-23T04:00:01.672620,0.00238493,376819,10371
2013-04-23T04:30:01.763360,0.00238493,376357,10447
2013-04-23T05:00:01.834324,0.00238493,426948,10522
2013-04-23T05:30:01.930203,0.00238493,432354,10602
2013-04-23T06:00:02.001589,0.00238493,478095,10685
2013-04-23T06:30:01.873047,0.00238493,727758,10811
2013-04-23T07:00:01.945017,0.00238493,629704,10924
2013-04-23T07:30:02.034515,0.00238493,489324,11004
2013-04-23T08:00:01.105271,0.00238493,408095,11071
2013-04-23T08:30:01.197656,0.00238493,427542,11150
2013-04-23T09:00:01.269012,0.00238493,473673,11235
2013-04-23T09:30:01.358329,0.00238493,488546,11318
2013-04-23T10:00:01.431617,0.00238493,546304,11417
2013-04-23T10:30:01.551037,0.00238493,460886,11493
2013-04-23T11:00:01.625378,0.00238493,505628,11579
2013-04-23T11:30:01.723748,0.00238493,467192,11654
2013-04-23T12:00:01.795047,0.00238493,467548,11736
2013-04-23T12:30:01.904825,0.00238493,394348,11802
2013-04-23T13:00:01.979512,0.00238493,452072,11885
2013-04-23T13:30:01.071837,0.00238493,425176,11955
2013-04-23T14:00:01.142614,0.00238493,442152,12042
2013-04-23T14:30:01.231736,0.00160842,309318,12119
2013-04-23T15:00:01.302790,0.00160842,411401,12224
2013-04-23T15:30:01.396977,0.00160842,370077,12319
2013-04-23T16:00:01.470698,0.00160842,404180,12426
2013-04-23T16:30:01.559716,0.00160842,427526,12541
2013-04-23T17:00:01.630961,0.00160842,385031,12639
2013-04-23T17:30:01.723401,0.00160842,393437,12742
2013-04-23T18:00:01.793916,0.00160842,401439,12850
2013-04-23T18:30:01.885580,0.00160842,403589,12957
2013-04-23T19:00:01.956461,0.00160842,397589,13062
2013-04-23T19:30:02.044965,0.00160842,1695241,13369
2013-04-23T20:00:01.115978,0.00160842,626112,13617
2013-04-23T20:30:01.204680,0.00160842,339603,13704
2013-04-23T21:00:01.284260,0.00160842,662119,13850
2013-04-23T21:30:01.389782,0.00160842,470472,13973
2013-04-23T22:00:01.460658,0.00160842,352904,14064
2013-04-23T22:30:01.552932,0.00172858,419644,14167
2013-04-23T23:00:01.623853,0.00172858,657979,14328
2013-04-23T23:30:01.713310,0.00172858,582290,14477
2013-04-24T00:00:01.784005,0.00172858,1538695,14776
2013-04-24T00:30:01.879285,0.00172858,1693734,15171
2013-04-24T01:00:01.953688,0.00172858,1554807,15564
2013-04-24T01:30:02.048584,0.00172858,616542,15768
2013-04-24T02:00:01.120638,0.00172858,770012,15948
2013-04-24T02:30:01.211224,0.00172858,631399,16097
2013-04-24T03:00:01.282491,0.00155466,416025,16207
2013-04-24T03:30:01.371976,0.00155466,968880,16432
2013-04-24T04:00:01.497975,0.00155466,2049269,16796
2013-04-24T04:30:01.587557,0.00155466,3761804,17652
2013-04-24T05:00:01.774160,0.00194696,5253695,18673
2013-04-24T05:30:01.863612,0.00194696,5145927,19842
2013-04-24T06:00:01.934426,0.00428524,8593763,20776
2013-04-24T06:30:02.023743,0.00428524,8593763,21630
2013-04-24T07:00:01.629016,0.01655359,13694467,22247
2013-04-24T07:30:01.718538,0.01655359,7285784,22496
2013-04-24T08:00:01.789313,0.01655359,8980687,22706
2013-04-24T08:30:01.880990,0.01655359,7777258,22905
2013-04-24T09:00:01.951783,0.01655359,9458595,23144
2013-04-24T09:30:02.041752,0.01655359,10341397,23400
2013-04-24T10:00:01.112909,0.01655359,11865998,23664
2013-04-24T10:30:01.208693,0.01655359,11623505,23937
2013-04-24T11:00:01.284000,0.06622092,37178649,24202
2013-04-24T11:30:01.377144,0.06622092,17082082,24302
2013-04-24T12:00:01.448927,0.06622092,19425156,24428
2013-04-24T12:30:01.537948,0.06622092,22468729,24573
2013-04-24T13:00:01.655400,0.06622092,21317926,24702
2013-04-24T13:05:01.792797,0.06622092,21976819,24725
2013-04-24T13:10:01.915288,0.06622092,22647644,24754

[Eb0la]

update on my experience with the coin so far.  I left my miner going over night and when I woke up I had 8200 confirmed coins.  I thought, okay great.. So I had to do something on my machine and restart it.  I turn my wallet back on and it starts to resync to the network... Now I have a total of 12 coins mined.. and the rest show as not being credited to me.  Why would it show as confirmed and then now take them all away?  I think 15sec blocks might be too quick for the network if you have 1000's of people hitting it with their machines.. Or am I just sounding like I am crazy?

I had scan time set to 1.  Queue 0.  and expiry 1

[c4n10]

Hmmm, so, I've been mining just fine for the past few days and now I suddenly have "0" connections, stuck at block: 20, 668


EDIT: Fixed by opening my listening port in iptables. Not sure why it was working before then suddenly stopped working, but whatever, back to mining   

[lightenup]

Thanks Otaci!

update on my experience with the coin so far.  I left my miner going over night and when I woke up I had 8200 confirmed coins.  I thought, okay great.. So I had to do something on my machine and restart it.  I turn my wallet back on and it starts to resync to the network... Now I have a total of 12 coins mined.. and the rest show as not being credited to me.  Why would it show as confirmed and then now take them all away?  I think 15sec blocks might be too quick for the network if you have 1000's of people hitting it with their machines.. Or am I just sounding like I am crazy?

I had scan time set to 1.  Queue 0.  and expiry 1

Not at all.
Seems to me you have been split from the (main)network and mined alone or in a smaller separated sub network. Of course you would confirm your own blocks, but as the hashing power of the separated sub network has been smaller than the main network, your block chain was in the end smaller than the block chain of the main network. Hence by design you lost everything after the split when you resynced.

Splits could have been prevented by a more dense network (i.e., more connections); and some main hubs (well connected nodes announced by dns, instead of the IRC channel).

I think 15sec blocks might be too quick for the network if you have 1000's of people hitting it with their machines..

Up until block 24 180 the block generation rate was much faster than 15 seconds.. at the beginning of the block reward (block 17280) we even sometimes had 2 blocks per second.. only about now we have reached approx. the target block generation rate.
So 15 seconds block rate might work, while I agree that a block rate of only several seconds or even subseconds can't work well with the current design.

[Eb0la]

So I forked myself

Okay, makes sense.  I will do a complete resync to the main network and put my miner back on it and see what happens.

[TheSwede75]

So I forked myself

Okay, makes sense.  I will do a complete resync to the main network and put my miner back on it and see what happens.

How do you perform a resync? (noob question). I suffer from the same. TONS of solved and 24 coins in 7 hours.

[mill0601]

what happened to the pool? I would really like to get my coins out.

[TheSwede75]

what happened to the pool? I would really like to get my coins out.

You COULD go ahead and read the previous reply's to the thread. It was DDOS'd @ 4.5GB/Sek. Wallets/DB are safe and owner is working on getting it up and disbursing coins.

[lightenup]

So I forked myself

Okay, makes sense.  I will do a complete resync to the main network and put my miner back on it and see what happens.

How do you perform a resync? (noob question). I suffer from the same. TONS of solved and 24 coins in 7 hours.

For the non-GUI client:
Optionally, delete ~/.smallchange/peers.dat (contains all previously found peers, but since you're split, they don't help you much) and restart client. Look at ~/.smallchange/debug.log and make sure, that there is something like that:

Code:

IRC :pelican.heliacal.net NOTICE AUTH :*** Looking up your hostname...
IRC :pelican.heliacal.net NOTICE AUTH :*** Couldn't look up your hostname
...
IRC :pelican.heliacal.net 001 u5XG5jmFiYFpABt :Welcome to the LFNet Internet Relay Chat Network u5XG5jmFiYFpABt
IRC :pelican.heliacal.net 002 u5XG5jmFiYFpABt :Your host is pelican.heliacal.net[173.246.103.92/6667], running version hybrid-7.2.3
IRC :pelican.heliacal.net 003 u5XG5jmFiYFpABt :This server was created Jun 28 2011 at 14:26:11
IRC :pelican.heliacal.net 004 u5XG5jmFiYFpABt pelican.heliacal.net hybrid-7.2.3 CDGabcdfgiklnorsuwxyz biklmnopstveI bkloveI
.
.
.
IRC SENDING: JOIN #smallchange00^M
IRC SENDING: WHO #smallchange00^M
IRC got join
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
IRC got who
.
.
.

Unfortunately, irc.lfnet.org seems to point sometimes to a dead IRC server, hence you might need to try a second time.

To verify your connection status:

Code:

# ./smallchange getinfo
{
    "version" : 60300,
    "protocolversion" : 60001,
    "walletversion" : 60000,
    "balance" : 75.40000000,
    "blocks" : 25617,
    "connections" : 7,
    "proxy" : "",
    "difficulty" : 0.06622092,
    "testnet" : false,
    "keypoololdest" : 1366762258,
    "keypoolsize" : 101,
    "paytxfee" : 0.00000000,
    "mininput" : 0.00010000,
    "errors" : ""
}

The field 'connections' should be > 0. In case the coin survives the first 2 weeks, I'll setup a domain name entry and a well connected node. Then such issues should not happen anymore.

[Cheshyr]

So it sounds like, the faster the target block time, the faster we need to retarget the difficulty.  I also like reward delay and graduate reward increase idea.

These parameters bounce of each other in interesting ways, but it doesn't seem  like a terribly complicated system to model.  The hard part would be finding the boundaries of the variables.  Things like minimum time required to request and start new work.  minimum latency requirements.  maximum data throughput to a casual users wallet.  That would give us limits on the model, which we could then fine tune to give the performance we want.

...

brain ran off the deep end here for a minute.  this may be completely unrelated, but i'll dump it here anyway, since we have people who actually think in this thread.  bare with me; i'm shooting from the hip here, and the coffee is only starting to kick in.

dynamic difficulty retarget seems a given.  what about dynamic rewards?  it sounds like a terrible idea...  i believe liquidcoin made an attempt at it, but hear me out a sec.  dynamic rewards plus dynamic difficulty would allow you to tune for a specific number of coins a day, regardless of network has rate.  that seems like a reasonable way to control inflation. 

... i had an idea here regarding arbitrage, but I just talked myself out of it.  either way, inflation is one of the critical variables that needs to stay well damped in a fast moving currency.

[alex_fun]

I think the coin with some changes can really be working fine. Maybe we can form a developers group and make it work nicely this week. I know where to change some parameters like difficulty, etc. Maybe someone more experienced can do more. And yes make maybe new name and logo as balhazar suggested

I think that we need to perform the following steps:

1) contest for the real name
2) contest for the best logo
3) send petition to btc-e

 

lol

Well, I know the momentum is good right now and of course it would be nice to recycle (parts of) this coin and create a new more thought through one (in fact: anyone can just fork and do so right now).
At least I wouldn't do this, until we have some more novel changes that make the coin solve problems better than other currencies.. also I really like to have some models first that give us hints how the coin could behave in the future or during attack scenarios, .. .

But block size and transaction volume limits should be adjusted before.

working on that.

[alex_fun]

Lol when i send coins from 1 address to another address between 2 wallets, the receiving wallet balance goes down by amount of coins sent then back up to its previous state? wtf lol

[wmikrut]

Lol when i send coins from 1 address to another address between 2 wallets, the receiving wallet balance goes down by amount of coins sent then back up to its previous state? wtf lol

lol... double spend built right into the wallet!
Now that's innovative!!   

[lightenup]

Lol when i send coins from 1 address to another address between 2 wallets, the receiving wallet balance goes down by amount of coins sent then back up to its previous state? wtf lol

txid ?

[alex_fun]

I got a bit pissed off and closed putty terminal with txt id Win wallet shows 2 active connection. Coins sent to addresses generated by win wallet are yet to show Well perhaps soon it all gets fixed

[alex_fun]

Send more coins txt 2d8e697b040e310aad5bf00d708433fe7a9ebbcc54c5234c5d6c638f8471775e coins yet to appear