log☇︎
740800+ entries in 0.478s
MisterE: yea I see a lot of support at 500 TestingUnoDosTre
MisterE: or nukes in the back yard?
tg2: My thoughts are, my power costs me .05c
MisterE: heh yea think of that scale jump
MisterE: so if they can amortize it spending 2k on a BTC is acceptable I guess.
MisterE: it goes out in pieces to the electric company
asciilifeform: tg2: that's what i was wondering. perhaps it isn't always accidental
asciilifeform: tg2: many 'home' miners end up doing just that.
MisterE: I heard they were being sued the first time a couple months ago but that also people were lining up to file
MisterE: I hope they all do
tg2: I'm glad they decided to do it, they were on the fence about it for a few days
asciilifeform: from this crackpot theory, mining doesn't need to have much +EV so long as it isn't too far in the minus.
MisterE: BFL could not deliver the product my buddy ordered and instead of refunding sent a lesser spec
mircea_popescu: anyway, i'm off. take it easy all.
tg2: I hope this Canadian class action against gox bears some fruit
MisterE: plus I thing it was stamped on the box
asciilifeform: think 'silk road' and 'why didn't they just go to the dark alley to buy the coke'
mircea_popescu: unless they're taking a microscope to the box, who's to say what's in there ?
tg2: I honestly think they started out with the goal of scamming and then btc blew up so fast lol
mircea_popescu: asciilifeform someone they'll still be trusting.
mircea_popescu: ahh, remember the days when am was derping about "having 30% of all hashing for a year"
asciilifeform: some people will spend 2K usd to get 1 btc, if it means they don't have to bend over for 'aml' crapola or trust someone
mircea_popescu: asciilifeform not even. more builders went out of business than stayed in.
tg2: Even the "next gen" 600gh is garbage
ozbot: Meet the manic miner who wants to mint 10% of all new bitcoins | Ars Technica
mircea_popescu: MisterE it's a theory. it's not outrageously or massively wrong, but im not too persuaded.
asciilifeform: mircea_popescu: that'd be because the only way to have reliable access to latest xxx is to... build it
MisterE: yea tg2 does a good job explaining that depreciation, expecially of 2nd gen sha-256 asics
mircea_popescu: "large concerns" that thought they had access to the latest widgets expected to come out... november 2012 are mostly bankrupt today
asciilifeform: but there's a standpoint from which it makes sense to buy a junk miner
MisterE: but medium - small guys are throwing money away?
asciilifeform: the poor schmucks buying thirdhand asic, not so much
asciilifeform: MisterE: large concerns with ready access to the latest widgets, sure.
MisterE: there was just the article in Ars abotu the guy who controls ~10% of hash rate
mircea_popescu: asciilifeform those people trade otc.
MisterE: asciilifeform: the mining centers I've read abotu who are doing it right have paid back their investments within months
asciilifeform: not everyone has an 'otc' vendor that they're willing to trust
MisterE: smart ones supply the miners and speculators
asciilifeform: think of it this way - some people are prepared to pay double, triple, the 'usual' cost for btc, so long as they aren't using an exchange
MisterE: BTC still has a gold rush mentality that personally I dont think is doing speculators any favors
mircea_popescu: rational behaviour isn't exactly the rule.
MisterE: and generate transactions that help keep the network healthy
mircea_popescu: asciilifeform how about the people who "invest" despite negative roi ?
tg2: Yeah I think that's a way to look at it
asciilifeform: i've always wondered if folks interested in mining despite the negative ROI are really just looking to convert fiat to btc by whatever means possible.
tg2: They assemble them out of house
MisterE: they also mine so there is a conflict of interest iirc?
tg2: They are probably one of the best right now
mircea_popescu: MisterE not as bad as all that.
MisterE: I heard knc was not reliable and they have had a lot of delivery misses int he past?
mircea_popescu: the reason, of course, is that bitcoin is worth three to four degrees of magnitude more.
tg2: Although knc sure is trying to prove me wrong lol
mircea_popescu: nearly as prolific as bitcoin miners they will never be
tg2: I stand by my analysis though that it won't be near as prolific as sha256
asciilifeform: hence 'lifecoin'. you're stuck with the RAM
tg2: They went for the "less ram faster latency" approach
asciilifeform: why brute force. it's like nobody gives a rat's arse enough to even try
mircea_popescu: tg2 you realise people wouldn't build the ram as part of the asic do you ?
asciilifeform: 'Current/Next gen bitcoin/sha256 asics are already at the forefront of manufacturing process, they cannot get more power efficient without waiting for a new manufacturing process to be available (14nm)' << nah
tg2: Yeah knc updated their numbers, I should update the post
gribble: SAT solving - An alternative to brute force bitcoin mining: <http://jheusser.github.io/2013/02/03/satcoin.html>; Proof of concept code for SAT-based bitcoin mining: <http://jheusser.github.io/2013/12/30/satcoin-code.html>; jheusser/satcoin · GitHub: <https://github.com/jheusser/satcoin>
MisterE: oh it's on litecointalk thats why didn't find it right away... [2014-03-27T12:19:20+0700] <tg2> https://litecointalk.org/index.php?topic=17948.0 // Discourse welcome!
asciilifeform: the ideal botnet-optimized (i.e. what scrypt wants to be) algo is described above.
MisterE: asciilifeform: probably not for you it's about the state of and medium-term future of hardware and mfgr. proceses
mircea_popescu: who had that link ?
mircea_popescu: stompysteve read the log will you.
stompysteve: anyone got any info on neo and whats going on past they stopped trading due to sus activity
asciilifeform: any surprises therein for folks who understand the algos?
kakobrekla: but mostly trash
MisterE: yea from what I can tell its mostly trash
asciilifeform: except when linked here - and even then, every 50th time or so
MisterE: asciilifeform: did you read tg2's analysis of sha-356 asics vs scrypt asics on bitcointalk?
asciilifeform: anyone who has the apparently common distaste for the asic folks should give this some thought.
moiety: cazalla: yeah definately, i got the live version :D he is a very good sensei
cazalla: moiety: that gpg guide is good, this was much easier than i thought it would be
asciilifeform: likewise, scaling the difficulty in time and space becomes trivial.
asciilifeform: and people are stuck having to compute the state transitions for the whole field, as many times as specified
asciilifeform: so long as you use an automaton that has been proven turing-complete, and use a reasonably entropic starting field state, you're pretty much safe from shortcuts
ozbot: White House regulate cow flatulence climate agenda | The Daily Caller
asciilifeform: (yet another 'discovery' by people long ago that herr W claimed for himself)
asciilifeform: most possible cellular automata degenerate to a steady state rather than give the kind of 'soup' one normally looks for.
mircea_popescu: yeah, i think the lowest would be 3 ? or 5 ?
asciilifeform: although it would need to be called something else
mircea_popescu: how many numbers are the product of two others ?
asciilifeform: now all you need is a reason to actually do this.
asciilifeform: that'd be how it has to be.
mircea_popescu: but the size of the field to compute.
mircea_popescu: you can have difficulty scale as... the size of the field.
asciilifeform: alternatively, run 'life' on the entire field, for M moves, and output is sha(field)
asciilifeform: corresponding to 'bridge is being built'
mircea_popescu: is it there then !?
mircea_popescu: ;;later tell benkay "One cannot issue ratings until one is rated oneself. This is a useful feature that prevents a lot of misuse of the system. If scammers cannot stuff the WoT database with their sockpuppets, they cannot artificially inflate their ratings. Granted, the mechanism is weak and easily gamed, but the simple presence of a barrier reduces much malfeasance." this made no sense. you know it made no sense. why
asciilifeform: then run langton's ant, starting from the middle of the field
asciilifeform: fill bits, in sequence, with sha2(xxx), then sha2(sha2(xxx)).... until field is full
asciilifeform: proposed algo, for those who insist: ☟︎
gribble: Stephen Wolfram, A New Kind of Science: <http://www.cscs.umich.edu/~crshalizi/reviews/wolfram/>; Ten Years of Monster Raving Egomania and Utter Batshit Insanity: <http://vserver1.cscs.lsa.umich.edu/~crshalizi/weblog/915.html>; Collection of Reviews of Wolfram's A New Kind of Science, Etc...: <http://www.math.usf.edu/~eclark/ANKOS_reviews.html>
asciilifeform: let him steal this one!
tg2: Not even a heavy site to lift
samO_: also, is there a chat log history of this channel somewhere? thx!
asciilifeform: (why anyone should wish this is another question.)