Topic: [ANN] Woodcoin [LOG] Pure Skein POW, Logarithmic Release, No Premine nor ICO - page 11. (Read 126093 times)

is that the oldest one that support solo mining? or i can choose a version which is more recent? i see that all the old version need compiling, do you have this version compiled please, i can't compile myself, thanks

oh i see it's complicated i thought you can just send two time the coins you mine with your own fork but you actually need to purchase something with a big amount, at this point i think the exchange can see that you ar emoving a large sum and block the deposit or prevent the withdrawal right?

I still use v1.0.0.0-unk-geo without issues.

I always only solomine coins and I strongly think that removing solomining (while it makes sense for bitcoin's wallet) is awful for decentralization and node count.

maybe it could help to have the possibility to solo mine here, there is a version that still work of woodcoin which is based on another client that is not bitcore like the old version that allow solo mining?

Being a GPU miner I'm partial to GPU friendly algos. 

But at the end of the day it doesn't matter much I think.


Come to think of it, with multiple algos we could essentially chop different types of wood. 

Hey thanks bathrobehero   I am going to read the code on the coins you mention and do some more research.

I don't think having multi-algos eliminates the threat of a 51% attack or double spend, but it could make it more difficult in some circumstances.  If there is a liquid market of hashpower, see e.g. nicehash, then one can always buy the hashpower one needs to make an attack.  Anyway I am going to take a closer look at the consensus mechanism. 

If we were to shift to a multi-algo system, what other hash functions do you think would be appropriate, other than double-skein? 

The speed of the retarget (slow or fast) really only depends on the network's hashrate compared to lone miners joining and leaving the coin. And the reason I called it slow was due to - as I said - any small miners can affect the difficulty. This was heavily increased if there were ASICs.

I personally dislike merge mining. It's a great idea on paper but most of the time the only thing that it achieves is more multipools; people mining multiple coins due to increased profits (which translates into lower coin prices) without even knowing the names of the coins they mine. But then again, I don't know any other coins using this variant of skein.

Coins like Joincoin, Digibyte, Myriad, Auroracoin are great, even considering not all of them have a "dynamic" dificulty retarget that changes the diff of stuck algos (only changes them on an individual pow basis) but it still allows for a fluid blockchain.
And - as far as I know at least - it almost completely eliminates 51% attacks (as you would need to either flashmine blocks on a single algo or dominate multiple algos, but you could make it so that a single algo can only be solved consecutively X times before a block with another algo have to be found (e.g. myriad)).

Unfortunately I don't know any disadvantages of having multiple algos but I'm sure there are some.

Ineteresting, yes I like the myriadcoin approach.  I think the difficulties of each algo there retarget independently from each other. 

It's interesting that 30 blocks to retarget (at 1 block per 2 minutes) appears slow; compared to LTC or BTC this difficulty adjust is lightning fast.

At various points over the last couple years we've talked about changing the difficulty adjust and/or POW algorithm of LOG.  Anything is possible if this is too annoying to deal with.  A change to a myriad-like network is attractive, what do you think?  Another possibility offering even more stability is moving to a merge-mined system.  However it seems that our fluctuating difficulty is not really such a headache.  Hop miners are still adding their hashpower to secure the network.  As we grow in size these effects will be relatively smaller. 

     

Step 1)  Start mining your own Foocoin chain without publishing it.  Include other transactions you hear on the network if you like, or don't.  Up to you.  What's important is that that you are solving blocks on this private fork with more hashpower than the rest of the network all put together. 

Step 2)  Make a large purchase using a Foocoin transaction.  Publish this transaction on the public chain, but NOT the one you are mining on privately.  On your private chain, make a transaction moving those Foocoin to another address you control (this is the double spend). 

Step 3)  Once the transaction is complete and you have walked away with your goods, publish the private chain you have been working on.  Because it has 51% of the network hashpower, it is now the longest chain and nodes will reorganize their block chains to reflect this "new" history.

And presto, the block chain no longer shows the transactions in step 2 - the foocoins are back in your control - and yet you still have the goods.       

An old article from now defunkt blog on the topic if you are interested to hear more : 
https://web-beta.archive.org/web/20160827095420/http://frass.woodcoin.org/godzilla-vs-the-51-attack/

Block #   date                           difficulty
503759   2017-04-29 16:05:15      77.014
503760   2017-04-29 16:08:58      275.054
...
503789   2017-04-29 19:33:11      275.054
503790   2017-04-29 19:35:02      79.368
...
503819   2017-04-29 19:48:27      79.368
503820   2017-04-29 19:48:29      311.931

and so on.

I said this before, the difficulty retarget is not just too slow but since the network's hashrate is small, virtually every small miner can make a dent in the difficulty.

Every single multipool (nicehash, yiimp, etc) will cause this effect. It just simply means the difficulty retargeting algorithm is just utterly inefficient. I wouldn't call that an attack though. Block withholding or 51% exploits are both real attacks though as they both can leave users with lost funds (unlike whatever difficulty is set by the retarget mechanism).


This is why having multiple algorithms in parallel is beneficial (if done well); optimally, when an algo is stuck at high difficulty, the retarget can reduce that algo's difficulty every time a block is found on another algo for a much smoother retarget. With one algo, the difficulty retarget can only ever change the difficulty once a block is found. Which is where there are constant overshoots both ways.


Regarding 51% attacks, I don't know much about them but as far as I remember, the blockchain accepts the double spend and so you can see in the blockchain (block explorer) if it were to happen. I'm not 100% sure though.

how can you do a double spend with 51%? i keep hearing about this but i'm curious to now the procedure, don't worry i don't have the gpu to do it lol

I think a way to fix hop mining is being able to predict the next difficulty, and try to push for the next difficulty when current profitability sucks, so there would be less extreme hashrate swings.
It would be nice if wallet makers could provide an rpc for expected next difficulty expecteddiff(x), assuming a block is solved in x seconds.

Thanks for asking Ayers  Rather astonishing swings of hash rate last week!!  I've never seen anything like it on any other coin.  In addition to price changes, and pools going off and online, the drive-by woodcutter is back in full effect.  

Yes the network has been undergoing this stress-test off and on for almost a year now.  It does appear that there is a single entity which controls enough hashpower to perform a 51% attack.  However they only leave it on for short bursts of time, and there has never been a report of a doublespent transaction, in private or from any exchange.  Lets put it this way: on no exchanges is woodcoin the lowest hashrate coin (or most 51%able) to choose from.    

My agents tell me that this party probably has somewhere from 20 to 80 top of the line nvidia GPUs they are throwing at our network.  However just like bitcoin, there could also be efficiency increases in performing the hash algorithm.  The question of why is an interesting one, it doesn't appear the the technique does much for our drive-by wood cutter apart from an upward pressure on the price and a some extra time to mine another coin with the rig during the high difficulty times.  It could be an exploratory technique.   

At any rate, the concern is a real one.  I believe exchanges that accept small-cap currencies without any know-your-customer do so carefully, not just by requiring multiple confirms but also by delaying any large withdraws from accounts that have recently input small-cap deposits, with random delays, and with maximum withdraws per day.  I believe we'll see more double-spend attacks on small-cap coin exchanges in the future, so your warning is a good one.  

with this very low hashrate this coin is at danger of an attack, how do we know if this coin is not being 51%'ed and the attacker is already double spending on the exchange? the explorer say something about it? i ask because i see huge hash appearing and disappearing

Heya pokey!  

Try this one out and see if it does anything.  I built myself but haven't tested on windows or with ccminer yet, so I would put the chances of it working out of the box somewhere in the slim to middlin' range.  

https://github.com/woodcoin-core/woodcoin-windows/tree/master/RC1

sha256sum *.exe
869faaf0645dc445e2ccb1e4cdf9f454bd9edf28bbe9dd10eac5e040328b9250  woodcoin-cli.exe
42d8d49c7291e96218624752f8e60488f14f890e306e43e2fc2e6b8518018681  woodcoind.exe
e0b82e9298fbcb64e6f39e775ba92bd2a3f6161b48a73aa5bf7e5ff82be84b5f  woodcoin-qt.exe
29299257a0671dd15005b79817f6b9cbc6a19fc7e57f9e63a771081509852c48  woodcoin-tx.exe

This update kicking your arse?  Any progress on getwork for windows wallet?  Thanks - pokeytex

It's possible if they are mining the same chain off-line. Then when the big miner goes off, they publish their longer chain. I didn't know this was still going on, and I defer to funkenstein_ to talk to the now rogue chain rewriter.

Yes well.. i dont see the problem, one single card can do so much these days. Are you editing the blockchain by hand ? how is that even possible to lose blocks after a lot of confirmations (and new blocks)

I think i will have to implement some extra security feature on yiimp to handle that (to virtually double or triple the confirmations required)

I think you just recently got an abnormal amount of orphans because someone just hit the network with 6GH, none the less if the pool is suffering then.....lets please fix it
Epsylon3 Rocks!! Thanks for the pool btw!

im sorry but i cant let this pool online as it is

i get orphans after the blocks are confirmed, and some send tx are cancelled too.

imo, you should increase the amount of confirmations for new blocks