884600+ entries in 0.638s
jurov: now
teach
them
to factorize primes :D
kakobrekla: well raised sort of
thing. unlike me, slovenly unadjusted animal
kakobrekla: well yeah jurov
the polite
thing is
to
turn away when you see a scam i
think
jurov: anyway evoorhees, i'd like
to know if anything
tangible came out of
the conference, for s.dice or else
furuknap: 24 hours and roughly 25% of BFMines IPO sold. I'm not disappointed, even
though
that includes 5% for myself.
kakobrekla: eh
they have been doing it for years with
their broked
trading engine
evoorhees: the quality of
the entrepreneurs is improving...
the general
talent is getting better
evoorhees: I've been
to
three conferences now I
think, and
the level of professionalism gets better and better
evoorhees: yeah
there are good
things coming, and great connections were made
kakobrekla: so evoorhees, anything great comming out of
that meet?
evoorhees: we were on like 40th floor, looking down on
the JP Morgan
tower ;)
matthew_boyd: Damn, now you are making me jealous, I really wanted
to be
there, maybe next year...
evoorhees: london was more like business people gathering
to network
benkay: how'd it differ from
the US con?
matthew_boyd: Damn, I wanted
to be
there so badly, but I couldn't in such short notice, anything really good I missed?
evoorhees: last night I got back from
the london conference, it was fantastic
ozbot: Justin Carter,
Teen Jailed For Facebook Comment, Reportedly On Suicide Watch
Namworld: I recommend BTC-BOND. Just a flat 0.03% daily. 10.95% per year. I'd call
that good returns.
furuknap: Wouldn't know what
to do with
them :-)
furuknap: The miners I bought for BFMines is priced at 5% of what AM blades are
today (or were, until
they were pulled off
the market).
Namworld: We'll have
to see about
that, eheh
furuknap: Heck, Bitfury will be mass-producing his 55nm by
then, and at
that point, it will be a proven design which sells for a fraction of
the cost of AM hardware.
furuknap: If KnC comes out in September, do you really
think AM can sell 65nm in October?
furuknap: Which is safe and fine, it's just a strategy of caution rather
than risk. Great for investors looking for long-term profitability. Except for
the halving effect, of course.
Namworld: And once sold, does it matter? It's
the buyer's loss
furuknap: Friedcat said it would be 65nm, but I don't know if
they changed
their minds of course.
Namworld: They'll go right
to
the best option
ThickAsThieves: i predict you'll be wrong about all your predictions in
the future
furuknap: Yes,
they can design a 28nm chip. AM's next gen is 65nm.
furuknap: Sooner or later, I'll be right,
though, like every prediction made far enough into
the future.
furuknap: No,
they won't because
they don't have
the money. AM isn't near having $100M in money because
they pay out virtually everything
they make in divs.
Their
tech is always based on prevoius generation, which is cheaper but not nearly efficient enough.
furuknap: Now I have
to get back
to work.
furuknap: All of
this, BTW, is why I don't
think difficulty will keep growing for very long;
the risk
to large investors (ie
those
that need
to come along
to fund next-next gen chips sub 28nm) is simply
too great.
Namworld: Damn, why is it
that people put a 10-30% premium on calls/puts
that expire in 1-7 days?
That's way
too much for
too short periods.
furuknap: Unlike with clothes,
though,
there's no smaller quantity for order; it's either $100 million or nothing, and your clothes may not even work.
furuknap: And
that's
the risk everyone runs; Even Bitfury missed massively, delivering only 2GH when he expected 5GH.
furuknap: If
they don't,
they're infinitely less efficient :-)
furuknap: So either new
tech comes along, which is very expensive, or KnC or someone like
them must get much lower prices.
The rates
that brought AM
to insane heights are miles away from being efficient enough now, just a few months after it was hailed as
the cheapest possible chips on
the planet.
jurov: once several parties will have a working design, all it
takes is only making an order at
tsmc
furuknap: No,
the mass production isn't
there by a long shot. Like I said, KnC (which are
the most efficent miners as of now) must sell hardware for $200 million until FEbruary 2014
to maintain 10% network growth.
They're not even close
to
that.
jurov: haven't
they been made?
furuknap: Jurov; for
that
to happen,
the investments must be made. Would you invest $100 million in clothes
to possibly sell up
to $400 million in clothes, but very likely not sell anything?
jurov: and it can happen
this year,
this being of bigger concern
than halving
jurov: now i can only say
the hw prices will drop like a stone, cusing severe dent in existing players' revenues and hashrate piece of cake
Namworld: Well honestly, I don't
think
the halving is what makes
the biggest impact on
this.
The insanely high price of AM and lowering USD/BTC rate does.
furuknap: Just a monent ago, you argued we'd have 20nm. Of course
they, or someone, will succeed;
they have
to, otherwise Bitcoin mining is dead.
pankkake: how sure are you
that KNC will ship? will
their hardware be as good as
they promised?
jurov: yes. it knc and other can mass produce and ship,
thus depressing hw prices overall, *after*
that it will make sense projecting for halving
furuknap: The cost
to go from 28nm
to 20nm is very high and someone needs
to risk a huge amount of money
to possibly succeed in reaping a 1:4 ratio on investment. I don't
think you'll find many investors willing
to risk
that.
furuknap: Thus, it makes no sense
to invest
that much because
the gains
that can be had from
that is not nearly enough
to warrant
the risk.
jurov: when we will have
trillions of asics on 20nm process and pace of development slows
to moore's law,
then it's possible on look forward
to reward halvings
jurov: irl
the % of global hashrate is very hard
to maintain, for asicminer or anyone
furuknap: For example, if _only_ buying KnC Jupiters from now until February, we need
to invest $200million
to keep a 10% per change up.
furuknap: That is because of
the halving effect; investments now have much shorter
time before halving, so
they can't be as high as
they used
to be.
jurov: the article silently assumes when you buy PMB-like stuff, you are buying some % of hashrate... but
that couldn't be farther from
the
truth
furuknap: Quite
the contrary, I expect it
to flatten out, probably much sooner
than we realize.
furuknap: Namworld:
That what is going up?
Namworld: furuknap, you're assuming ASICMINER doesn't get smaller process/better ASIC out over
time, and
that old hardware never goes bust,
thus it's always going up.
furuknap: jurov; read
through
that article.
jurov: building any projection mostly oin halving seems foolish
to me
furuknap: We've seen an insane increase in difficulty past
three months and prices drop like rocks.
There's no causuality between diff and price, and even correlation is in question.
jurov: also, what if fees do increase due
to devs not increasing blocksize?
furuknap: Well, what if it goes down
to half?
ThickAsThieves: and what if
the halving causes increase in fiat value?
ozbot: What is
the Halving Effect in Bitcoin Mining Investments? | Furuknap's Cryptocoin Blog
furuknap: Again, sellig what? Miners will still need
to get ROI which will be
twice as difficult, so hardware sales will drop
to half
too.
ThickAsThieves: you do realize,
that by
then,
they will need
to primarily serve as a hardware sales compeny anyway
Namworld: Well
technically it would go down
to zero eventually without
TX fees.
furuknap: Because in *November 2016, AM income will go down by 50% unless
TX fees compensate.
furuknap: Actually,
take half of
today's price.
furuknap: Very simple
to factor in.
Take
today's price, divide by number of month until November 2016, multiply by
three.
furuknap: The issue, of course, is
that AM will probably never go above 35%, meaning
that no matter how little or much
the hashrate grows, only costs will grow for AM, not income.
furuknap: Because for AM
to be even close
to comparing
to
TAT.VM and BFMines, it would need
to exceed mining income by about 10x.
furuknap: Sure it is. What will
that be for
the next
three months,
then?
furuknap: jurov: No, because if AM has
the hash power
they claim on order,
their relative hashrate will remain steady.
ThickAsThieves: no one is pricing
the halving into anything right now, except you
jurov: furuknap, bunch of other asic companies likely shipping in next months have much bigger influence
than some halving
pankkake: I wonder why it had
to be cut in half brutally