Set> Hmm it never even occurred to me that some pools do let you specify the difficulty of shares you want.

I always just ifgured the pools are adjusting it based on how fast you get solutions to them, so as to slow you down or speed you up as they think best?

I was mining mmpool, not sure if it even has a way I can tell it a specific difficulty I'd like.

-MarkM-

EDIT: Okay apparently it is cgminer that I tell my desired difficulty to, I thought it used to be some add-on code you put in the pool URL or something but apparently it is a commandline argument to cgminer. What kind of diffuclty would be better? I am have been seeing pretty good hashing speeds. Also that killed remark seems to hint it is the O/S or the web scripts or something external to cgminer that is killing me not some error inside cgminer...

-MarkM-

When ckolivas provided a link to a version of cgminer that supposedly might fix the problem of your miner just stopping, I ssh'd in and wgot it and ran it and all seemed well for a while.

But just now it stopped. The lasting thing my ssh session saw was:


It seems to be saying that something actually killed it rather than that is ran into some kind of fatal error?

If so, what would have killed it, and why?

/var/log doesn't look too helpful in figuring out what happened...

-MarkM-

The big problem with blockchains is they are insanely expensive to secure or they are insanely insecure.

Look at all the crapcoins that came out the last bunch of months, most of them have huge problems trying to secure their chain, and the more of them there are the less hashing power gets devoted each, on average. Plus the miners jump around, in effect attacking the chains instead of really helping them.

That is a large part of why all the specific-arts specific-crafts etc coins the devcoin project plans to implement are not going to be done using blockchains; we still have not even been able to make the devcoin blockchain itself as secure as bitcoin's blockchain, it would be insane to increase our expenses by trying to budget hashing to protect even more blockchains.

Thus all the little specific-purpose currencies are most likely going to be implemented using Open Transactions.

-MarkM-

So this is the same blockchain she relaunched?

The commandline setting to tell it where to put its data directory doesn't work, I told it to use ~/.cubits but it ignored that and created ~/.Cubits2 instead.

-MarkM-

How do you do an eleven second retarget when blocks are only every 33 seconds?

Plus I thought the original dev re-launched this with 3.3 minute blocks because people complained that 33 seconds was too fast?

Also, how many months worth of blocks have you already blown through with your premine/instamine?

-MarkM-

Well diamonds are kind of iffy, y'know. All that bull about "investment diamonds", the totally artificial scarcity, the subjective "quality" and "clarity" and "cut" stuff, bitcoins are nice simple objective mathematics, much better...

-MarkM-

Rather than more blockchains, which are afterall a ridiculously expensive format due to the cost of protection from 51% attack and so on, the plan is to use Open Transactions to branch out the DeVCoin system into more artistic and creative fields. Various domains have been acquired toward those goals but maybe some of the delay might have been due to problems trying to get all the required domains. So a lot has already been moving behind the scenes, it was not considered a good idea to talk a lot about those efforts before securing all the relevant domains otherwise we'd likely just find domain squatters grabbing all the domains they could predict from our discussions that we would be wanting.

Maybe if all the domains have at last been secured we could finally talk about those plans...

-MarkM-

It does seem to indicate that if you are going to invest electricity into cryptocoins the scrypt coins might not be the best ones to invest your electricity into...

-MarkM-

I grabbed RoadTrain's version from github, it seems to be working fine for me so far...

-MarkM-

Still it could be a nice feature of some types of client or clients for some types of people to have the client watch such things and point them out to you.

-MarkM-

Make a Windows GUI front end for p2pool, then maybe computer-illiterate Windows users will use it, which will be good for everyone.

Maybe you could even sell it, since Windows users are often actually used to paying for software and maybe even prefer stuff they paid for to free stuff.

-MarkM-

p2pool is a distributed pool and extremely easy to set up, you don't even need to set up a database.

Plus it even allows the pool owner to merged mine. (That is, run your own copy of p2pool and you get all the coins you choose to merge as well as your share of the bitcoins.)

So basically no matter whether you are a large miner or a small miner you can very simply run p2pool for yourself and thus get the variable-amelioration of a pool while still using a decentralised solution instead of a centralising pool.

-MarkM-

There is some startup processor that does about twenty shitcoins already, maybe Ixcoin doesn't happen to be one of them right now but that could change if Ixcoin does better than one or more of the 20 or so they already are using. Plus they plan to add more anyway once they have the infrastructure to let them run even more daemons.

-MarkM-

I doubt individual merchants are all that important to any coin, as presumably most merchants will use payment processing services and thus not care what the heck that service chooses to accept since the merchant still gets whatever type of currency the merchant wants.

So basically any coin can get lots of merchants any time simply by getting added to the growing list of currencies some payments processor supports.

-MarkM-

People can post such news, at least the English translation of the Chinese news, in their own words of course not simple thheft-by-translation of other people's articles, on Devtome to earn Devcoins...

-MarkM-

See TAGcoin.

-MarkM-

They don't have to invest in bitcoin, just accept it. Companies that mostly accept fiat mostly pay wages in fiat, and furthermore mostly in the specific flavour of fiat that they mostly accept.

Just go ahead and start a successful business and choose to accept some kind of cryptocoin.

You will probably find the hard part is starting a successful business, if you can do that the choice of which currencies to accept is relatively the easy part.

Part of the idea of http://www.devtome.com/doku.php?id=cpu_mining is it in effect creates jobs, by providing a huge number of products that people can produce. The fact that the products are virtual is what makes it relatively easy/affordable to actually do it; if we tried to provide real quarries and real goldmines and real ironmines and so on for people to mine real metal and stone in, and real forests for people to chop real wood in and so on it would cost a heck of a lot more to set up than the virtual mines forests and so on this method lets us set up to provide "productive" employment as in employment that creates "products".

-MarKM-

Well any business that sells its goods or services in cryptocoins could offer its employees their pay in such coins. Maybe as more crypto millionaires arise there will be more lawn-mowing window-washing and so on jobs available paid in cryptocoins?


Seems likely. Presumably the idea is they will scrape up a few millibits or microbits shining shoes or mowing lawns or whatever then hold them a few years whereupon they will be rich.

-MarkM-

Coins like Ixcoin, I0coin, Devcoin, Groupcoin, Fairbrix, Tenebrix, Coiledcoin, GeistGeld, BBQcoin and probably others have had some success using the approach outlined in http://www.devtome.com/doku.php?id=cpu_mining as a kind of "CPU mining" type of thing that does not require super hardware and which is not all that great for botnets either...

-MarkM-

If you like fast, check out GeistGeld. Like IXCoin it is merged mined, unlike IXCoin it recently got a slight fixup for the RAM usage problems all merged mined coins need to fix and keeps on mining coins forever. It has like 30 or 40 second blocks I thinl, something like that. Basically so fast that some people were/are concerned that the need to keep getting work for it so often would risk slowing down the getting of work for all the other coins you merge. Until recently it needed 16+ gigs of RAM but the quick and dirty fix, which is the same quick and dirty fix that was applied to I0Coin (and by the same programmer) before a proper fix was then made for I0Coin, has it now using less than a gig, close to half a gig maybe.

So if you were putting off merging GeistGeld because of its RAM needs, grap rsnel's gitup repo with the quick and dirty fix and use that...

...There will be a couple of days lead-time though as there are a LOT of blocks and you basically need to get all the blocks from scratch as it uses a different blockchain format on disk from the old RAM-hogging version.

Then if you like it, join the campaign to try to bribe rsnel into doing the full and proper fix, the one he did to I0Coin, to all the merged mined coins. (Namecoin, Devcoin, Groupcoin, Ixcoin, Coiledcoin and Geistgeld.)

-MarkM-