207300+ entries in 0.474s
mircea_popescu:
http://btcbase.org/log/2017-03-18#1629015 << odds are it couldn't be re-done in a current fab,
the vast majority of current fab constraints go
towards avoiding echos and other self-crits
the design may impose upon itself. as
the engineers at
the
time didn't labour under
the same constraints, and
they're cutting out way more
than 99% of
the
theoretical design space,
the odds
their work product would work in 14nm process are ze
☝︎ mircea_popescu: that said, it could of course be re-done, by people who understand completely what it did. but a lot more work
than "print me
these old family films on modern paper".
phf: aah, yeah, i'll
test it in xl
phf: macivory is an ivory on a board with all
the i/o handled over a bus by a mac
phf: xl1201 is an ivory (ivory being a cpu inside
the box, merlin being
the architecture)
phf: i don't have an alpha
though, same reason i don't have a macivory, i don't want
to have
to study/rely on all
these other architectures
phf: both approaches are viable, you can use emulator as reference, you can also use real hardware as reference, with
the solution meeting somewhere in
the middle.
phf: my goal with
that
thing was
to give it
to zeptobars, but i'm not going
to moscows for at least another 4 months
phf: i'm not sure if it's working, it's from a "possibly broken" box, but i can lend you
that one after i
test it in my ivory, and if you have a viable project of some sort, i can probably get you more
phf: i suspect
there's a handful of pals
there?
a111: Logged on 2017-03-13 00:13 asciilifeform: !~later
tell phf i dug out my crate of dead bolix boards,
turns out i'm a chump,
the pcb in question was a dead xl1200 ~i/o~ board.
the only cpu on it is what looks like a z80! 0 smbx silicon.
phf: also by signal interposer you mean attaching a logic analyzer, or an actual gadget
to sit on bus and do
things?
a111: Logged on 2017-03-13 16:49 asciilifeform: !~later
tell phf i was
thinking' of buying
that 'ivory' and making a signal interposer . any idea whether
the magic cpu was soldered down, or LIF socketed..?
a111: Logged on 2017-03-18 14:19 asciilifeform:
http://btcbase.org/log/2017-03-18#1628912 << whoever it was, he has problems paying, apparently, i bid on
the 'macivory' and ended up getting '2nd chance offer, high bidder didn't pay' from seller
shinohai: Nice Japanese pron you found
there ben_vulpes
mircea_popescu bows out of
this discussion lest he fucks up
the laboratory glass.
ben_vulpes: the methodology i recall was 'for each
tx,
take
top 32 bits, use as index into array, increment byte at offset"
ben_vulpes: when you say 'colliding
txids', you mean 'colliding
top 32 bits', right?
deedbot: diana_coman rated lobbes 2 << Made a very useful bot for #eulora. He's very modest,
too.
diana_coman: !!rate lobbes 2 Made a very useful bot for #eulora. He's very modest,
too.
mircea_popescu: not how it works irl on any models
that don't include "we need
to be seen as doing something please god" as a hard constraint.
mircea_popescu: more like,
the usg is under
the pressure
to appear
to be doing something, has settled on "chipping", ie, o noes if an eth splits off bitcoin
that'll be bad for bitcoin!!!1
mircea_popescu: i know
that out of
the past 1mn i
traded, 0% was on an exchange.
mircea_popescu: eh, "nearly 20 exchanges", wut. "exchange" idiots have managed
to
trade
the 2012 situation, where
they had gone from 0
to almost 50% of bitcoin
trading for a slow and agonizing death. i'd be surprised if all of
them (which within 90% is
the first
two) actually cover 10% of bitcoin for fiat
trade.
mircea_popescu: plus
they got into an argument over
the carpeting color in 1956 and split
the congregation.
mircea_popescu: absolutely no conceivable reason outside of some jolene or other needing
to feel special.
mircea_popescu: asciilifeform you don't see what
the point for deeds is ?!
mircea_popescu: anywya,
the reason i linked it is because obviously
the ~whole space readily reduces
to "what a number of people understood and remembered of what
they at some point read on #trilema"
mircea_popescu: l0 is verified by
the bitcoin network as part of its functioning. l1 can be verified by anyone with a full blockchain.
a111: Logged on 2017-03-18 11:03 Framedragger:
http://btcbase.org/log/2017-03-10#1624858 << re rdtsc etc, i'm curious, does `grep -E '^model name|^cpu MHz' /proc/cpuinfo` consistently report
to you mhz which are (+/-1 a mhz or so) equal
to model frequency? (i know you can
turn off linux powersaving etc features)
mircea_popescu: not
to mention it keeps on growing, like any blockchain.
mircea_popescu: add
to
this
that
there's a long list of such must-know (o hey, bought new house in so and so zone ? does it have a basement ? here's your geiger counter. what do you mean you don't understand ? atten-hut! nobody asked you what you understand, and it'll be 49.95
mircea_popescu: so it stands entirely unrevisable : either you understand it or not ; whether you want it or not : here's your shot.
tyranny.
mircea_popescu: tyranny, exactly. "do
this or die."
that's
the definition. and it's not yielding : if
the wife/husband have
the intellectual wherewithal
to understand
the whys and wherefores, fine, and bully for
them. but if
they do not -- it isn't
thereby going away, lol.
mircea_popescu: this is, of course, also
the natural basis for
tyranny. it's nice and good
to have independent agents, but
there must be some sort of authority which
tells
them "hey, you're rh-, your husband rh+, you're seven months in, here's your rh antigen shot."
this authority may be
the wife, may be
the husband, but if it's neither of
them it'll have
to be something outside, and if it is what do you call it ?
mircea_popescu: (hydrops fetalis / rhesus disease crumbs
the rh
thing)
mircea_popescu: but what
this means, obviously,
taking
the converse... is
that
there's a lot
there
than meets
the eye. which is
the point.
mircea_popescu: which is WHY
the "dun fuck your sister"
thing is such a huge
taboo : a) it will kill you and b) you won't notice.
that's
the problem with
these boiled frogs scenarios, from carbon monoxide poisoning
to sibling conception
to etc : you got no warning bells connected
to
THIS failure mode.
mircea_popescu: the problem, however, is
that under
the skin-deep "kid is fine", his passive genetics have seriously weakened. if he lives as part of a larger population
that does
this, a few generations down
the road genetic decay manifests and
there's literally no hope --
the population will either move
to exogamy or die out.
☟︎ mircea_popescu: oimmunity against either current fetus or, even more perverse, against
the NEXT. girl has kid fine, but also prepares herself
to murder next horribly. nature is FULL of such
traps.)
a111: Logged on 2017-03-17 14:21
trinque: hm, if so
the
thing falls apart
mircea_popescu:
http://btcbase.org/log/2017-03-17#1628783 <<
to briefly revisit : contrary
to what general public might expect (tm), if brother and sister conceive
together
the result is going
to be a perfectly fine baby by any standards. yes it's possible
they spawn a monster, but not likely, and moreover
this is possible anyway. (for instance : rh negative women (but not positive) impregnated by rh positive man can and often do develop aut
☝︎ Framedragger: (also due
to modern out-of-order execution in cpu pipelines, may need
to force all commands before
the rdtsc call
to be 'flushed', i.e.
to have completed before
the counter instruction is called; just noting for self)
Framedragger: (okay apparently if your cpu supports 'constant_tsc' (as seen in cpuinfo)
then
this
timestamp counter actually counts
time and not processor
ticks which is a *good
thing* given freq scaling etc.;
this is available in all new intel processors; what a rabbit hole, man)
Framedragger: it doesn't do
that for me (but i'm yet
to double check
that some 'additional' powersaving / freq scaling features are not enabled)
a111: Logged on 2017-03-10 23:41 asciilifeform: if it dun run at a constant rate: a) adjust your bios b) if fails,
Throw it out!!
Framedragger:
http://btcbase.org/log/2017-03-10#1624858 << re rdtsc etc, i'm curious, does `grep -E '^model name|^cpu MHz' /proc/cpuinfo` consistently report
to you mhz which are (+/-1 a mhz or so) equal
to model frequency? (i know you can
turn off linux powersaving etc features)
☝︎☟︎ a111: Logged on 2017-03-15 16:27 Framedragger: did you already have a chance
to check if
there are any
tx collisions in
the 32 bit address scheme? just curious