log☇︎
2000+ entries in 0.136s
asciilifeform: mircea_popescu: you might be thinking of old flamefest ( possibly with diana_coman & mircea_popescu ) where asciilifeform was asked 'why dontcha quit yer dayjob' and asciilifeform answered 'would consider it, if , like the 2 of you, had enuff coin to permanently leave slave market , with x% engineering margin' or sumthing of the kind
asciilifeform: while it boils : btw log readers can ignore the earlier arithm, it is entirely off ( asciilifeform computed r * 2^s = n instead of r * 2^s = n - 1 ... )
asciilifeform: to redo the earlier munged example: bin(12) == 1100. s = 2. r = 3. (n - 1)^r mod n == (12 - 1)^3 mod 12 == 11^3 mod 12 == 11, i.e. n - 1. witness.
a111: Logged on 2019-01-26 08:58 diana_coman: http://btcbase.org/log/2019-01-25#1889973 -> didn't rummage as too late yest but woke up this morning with this: when a= n-1 step 2.2 aka y=a^r mod n becomes y= (n-1)^r mod n; using for instance binomial expansion you have (n-1)^r = n^r + rC1*n^(r-1)*(-1)+rC2*n^(r-2)*(-1)^2+....+rC(r-1)*n*(-1)^(r-1)+(-1)^r; since the whole thing was "mod n", of all those terms you are in fact left with only the last one, aka y = (-1)^r mod n ; r is odd so
diana_coman: 2 is odd?
asciilifeform: diana_coman: bin(12) == 1100 . then r = 2. then, (n - 1)^r mod n == 11^2 mod 12 == 1. 12 is pronounced composite.
asciilifeform: diana_coman: correct. in implementations where a is forced to <= n - 2, n - 1 is excluded
diana_coman: that's (-1) mod n so y= n-1 and then there's nothing to do at step 2.3
diana_coman: http://btcbase.org/log/2019-01-25#1889973 -> didn't rummage as too late yest but woke up this morning with this: when a= n-1 step 2.2 aka y=a^r mod n becomes y= (n-1)^r mod n; using for instance binomial expansion you have (n-1)^r = n^r + rC1*n^(r-1)*(-1)+rC2*n^(r-2)*(-1)^2+....+rC(r-1)*n*(-1)^(r-1)+(-1)^r; since the whole thing was "mod n", of all those terms you are in fact left with only the last one, aka y = (-1)^r mod n ; r is odd so ☝︎☟︎
BingoBoingo: Trump gives Pantsuit voters paychecks through 2-15-2019, after that they deliver his wall or he declares and emergency and heads roll to be impaled on artistically designed steel border slats
asciilifeform: so algo is then obv.: 1) obtain pies 2) treat the kraut 3) win the 1200 ...
asciilifeform: 'choose a random integer a , 2 <= a <= n - 2'
diana_coman: asciilifeform, can you point me exactly to what you mean re n-2?
asciilifeform: seems like ~50% of the various lit blindly copies the n - 2 thing , also
asciilifeform: mostly unrelatedly -- while diana_coman is awake -- diana_coman can you think of why vanstone et al give n - 2 upper bound for m-r witness ? cuz by theorem, it's good through n - 1
feedbot: http://qntra.net/2019/01/foreign-language-course-offerings-dropped-5-3-percent-in-3-years-at-us-universities-while-enrollment-fell-9-2-percent/ << Qntra -- Foreign Language Course Offerings Dropped 5.3 Percent In 3 Years At US Universities While Enrollment Fell 9.2 Percent
mircea_popescu: much like in the case of usual gcd -- checking a random number for divisibility with 2 provides a much larger benefit (eliminates 50% of candiates) than checking a random number for divisibility with 15,485,863, which only eliminates 0.000% of candidates.
mircea_popescu: specifically because density is uniform, if you ~only~ check say 64 bit numbers as witnesses (ie, nothing smaller than 2^63) there's a set of composites < certain limit that'll pass ; whereas if you only check 1-63 bit numbers (so the same ~number~ of numbers) there's a set of composites that'll pass, but they're LARGER than the same limit.
asciilifeform: one oughta be able to feed any valid, per the theorem, witness, up to n - 2 , and get the expected output.
asciilifeform inclined to reject koch's optimization ( which diana_coman retained ) where witness consists of rng(bitness_of_n - 2) , and actually make witness equal to rng(width) mod (n - 2) for full range
a111: Logged on 2019-01-11 17:30 asciilifeform: btw per asciilifeform's chalkboard, the physical cost of constanttime m-r is ~equal to that of (2 modexps of the given width) x (number of witnesses) .
asciilifeform: meanwhile , measurements : 2.533s is cost of 1 shot of 4096bit m-r (on standard tester iron) ☟︎
mircea_popescu: so the dood's businessplan is, 1. get a list of websites ; 2. while $list do curl -e "esif22.zcodesys.hop.clickbank.net"; done.
deedbot: billymg paid BingoBoingo invoice 2
billymg: !!pay-invoice BingoBoingo 2
asciilifeform: mircea_popescu: re crapple, it's become a rich generator of lulz : their current strategy seems to consist of, per iteration of product : 1) cut an essential component (e.g. headphone jack) , and declare 'fashionable' 2) double price
mircea_popescu: sometimes i sit here and regret the days i was wriging 2-3-5 articles/day on trilema. then i read one of those, and it's always, "o cute, 1/10 of an article".
a111: Logged on 2018-01-03 16:05 asciilifeform: but in very unrelated lulz, https://archive.is/bON1p >> 'It’s a bit absurd that a modern gaming machine running at 4,000x the speed of an apple 2, with a CPU that has 500,000x as many transistors (with a GPU that has 2,000,000x as many transistors) can maybe manage the same latency as an apple 2 in very carefully coded applications if we have a monitor with nearly 3x the refresh rate.'
asciilifeform: mircea_popescu: 16bit ( barring various 'banking' trickeries ) addresses 2^16 == 64k.
a111: 2 results for "\"hc-8\"", http://btcbase.org/log-search?q=%22hc-8%22
mircea_popescu: consider the item i was discussing with the girls last nigyht over coffees : there ~are~ rules for divisibility of integers for ~any~ integer, not just 2 or 3 or 5 or 9.
mircea_popescu: yea, here's how they resemble : 1. "why are you wearing that old rag ?" "it's pretty" "maybe it was" "same difference" "mkay" vs 2. "why are you using that old math ?" "itworks" "maybe it did" "also". 1===2~!!!1111
a111: Logged on 2019-01-21 00:07 asciilifeform: in asciilifeform's o(1) tx indexer ( will be welded to an experimental bdb once i get the mmap thing resolved ) there's a 2-level storage -- a 'write-once' o(1) index for blox of age N ( N can be 100-500 in practice ), and a much smaller rewritable one kept strictly in ram ( for 'recent' blox, where the longest chain is potentially movable )
asciilifeform: when i have a free hand or 2, i'ma apply diana_coman's published method to the thing and see whether cured.
a111: Logged on 2019-01-22 06:57 BingoBoingo: <asciilifeform> BingoBoingo: how does it cache the e.g. tB of rng i happen to generate in racked box in jp and dload via ssh ? << I will say that this does suggest .jp s a candidate for Pzarro rack 2 when that time comes
BingoBoingo: <asciilifeform> BingoBoingo: how does it cache the e.g. tB of rng i happen to generate in racked box in jp and dload via ssh ? << I will say that this does suggest .jp s a candidate for Pzarro rack 2 when that time comes ☟︎
BingoBoingo: And it turns out that Airstrip 2 (Formosa) is indeed large enough to facilitate dying alone in the elements https://www.news.com.au/travel/travel-updates/incidents/bikini-climber-freezes-to-death-after-20m-fall-from-mountain/news-story/5248010cf59fe8983422969ca3498e77
BingoBoingo: If your endpoints hits Netflix for 5gbps they'll let you leverage 1.2 gbps every half day for 20gbps+
BingoBoingo: <asciilifeform> assertion is, the cost of a processing job oughta 1) be calculable in advance of performing said job 2) depend only on ~said job~, rather than what else the machine happened to have done recently or may be doing concurrently with remaining cycles << This is a violin. We gotta get a lumber yard stocking something other than Spruce/pine/fir or we hit the sill plate problem.
asciilifeform: assertion is, the cost of a processing job oughta 1) be calculable in advance of performing said job 2) depend only on ~said job~, rather than what else the machine happened to have done recently or may be doing concurrently with remaining cycles
asciilifeform: in other lulz, 'numpy' ( 1 of the few useful libs for that lang, does various numericalmethods ) Officially turns tard, proclaims 'this is last ver to support python 2.7, from nao on 3'
mircea_popescu: and even then -- nothing under 2-3k or so worth the mention.
asciilifeform: runs off 2 physically separate 220v ps.
asciilifeform: in asciilifeform's o(1) tx indexer ( will be welded to an experimental bdb once i get the mmap thing resolved ) there's a 2-level storage -- a 'write-once' o(1) index for blox of age N ( N can be 100-500 in practice ), and a much smaller rewritable one kept strictly in ram ( for 'recent' blox, where the longest chain is potentially movable ) ☟︎
asciilifeform: the inner loop or 2, tho, definitely can.
asciilifeform: nao, it isn't as if the current ffa, with 2.7sec 4096-bit modexp, is immediately usable to eat packets at line rate. but that part at least theoretically parallelizes ( i.e. a rack fulla multicore boxen running ffa, can theoretically eat packets at line rate... )
asciilifeform: i prolly oughta add to the http://btcbase.org/log/2019-01-20#1888508 thing : 1 of the items which seemed like a speedup, but in actual practice sucked, was the use of (constant-time) 2 (ditto 4) -bit windows for modexp ( iirc apeloyee suggested ) ☝︎
asciilifeform: 'locality' effect is actually why e.g. comba wins over ordinary 'kindergarten' o(n^2) mult
mircea_popescu: i suppose you might say "it shouldn't be x2, my expectation would be it's x2^1/2", but w/e.
a111: 2 results for "jekem", http://btcbase.org/log-search?q=jekem
asciilifeform: 40 / 2 sounds moar like current-day ameri-trailerpark
asciilifeform: 2 jacks.
mircea_popescu: now, if indeed per 2 enemy can perma-trash your boxes, then 3 becomes moot.
mircea_popescu: well, i'm not sure what your decision tree is like, exactly. but proceeding logically : 1) it can not be the case cpu is sad on rk ; 2) it is not shown that deffect is organic rather than nsatronic ; 3) the selling point of rk is cheapness/disposability. you can afford to throw these away, unlike an adult server board+deckings.
asciilifeform: i.e. it gets through 1) mask rom 2) load of kernel from sd
asciilifeform: http://nosuchlabs.com/pub/dead-rk/run_1.txt http://nosuchlabs.com/pub/dead-rk/run_2.txt http://nosuchlabs.com/pub/dead-rk/run_3.txt http://nosuchlabs.com/pub/dead-rk/run_4.txt
asciilifeform: there's ~2 ~indistinguishable 'alternatives' to errything.
BingoBoingo: <asciilifeform> lol 30 m^2 << Smaller
asciilifeform: lol 30 m^2
mircea_popescu: so in this sense, because the t-34 cost 18 units and provided 6.5 to 30 or so sovheads while tiger cost 77 units and provided .1 to 24 naziheads, we can resolve the matter : 18 for 195 vs 77 for 2.4
mircea_popescu: in any case, the problem in terms of "1 tiger is said to have been match for 2-3 t-34" is already misstated. the point is : marginal war gain per german soldier from tiger was like .1, marginal war gain per soviet soldier from t-34 was like 6.5
mircea_popescu: this then illustrates a principal line that constructs my http://btcbase.org/log/2018-12-19#1881924 ; to go with other main http://btcbase.org/log/2018-12-20#1882318 and i think there;s maybe 2 or 3 more and that's it. ☝︎☝︎
asciilifeform: 1 tiger is said to have been match for 2-3 34. problem for germans is that they brought 5,6 per.
asciilifeform: err, 2
asciilifeform: BingoBoingo: i'ma include 2 spare raid controllers in the cargo, iirc our spare pile of these was depleted
asciilifeform: ( not even to mention the fact that -- rare, for a kochism -- the problem dun even exist in stock gpg, which defaulted to 2 subkeys, 1 for sigs and 1 for encipherments )
asciilifeform: ox as well as it did for esr tho, his www already for year or 2 contains a begtron
mircea_popescu: " this gives 2 1u slots remaining, they can be occupied by colo passengers, will carry rk plants otherwise"
mircea_popescu: amusingly, the 2 seems the better deal.
asciilifeform: 2
asciilifeform: it is also time to speak of the next crate. and the customary four cargo slots. 1 is to contain a replacement for uy1 ( and ideally will run cuntoo . ) 1 will contain a 1u that holds rk's. this gives 2 1u slots remaining, they can be occupied by colo passengers, if these stand up and wish to ride , or pizarro irons, at BingoBoingo's option. ☟︎
asciilifeform: incidentally manifest actually includes 2 FG units ( pizarro-owned ) . i had to fly'em back in april , if anyone recalls, cuz of ben_vulpes's misadventure where they were pawed by orcs for whole night
asciilifeform: mircea_popescu: what do you typically do with yours ? throw'em out after 1 peculiar reset ? 2 ?
asciilifeform: mircea_popescu: how wouldja distinguish via 'flow of air from rack' 2 cpus at 50c ea. vs 1 at 0 and 1 at 100.
asciilifeform: hanbot: we dun know yet wtf reset the box ( and it happened no fewer than 3 times, in 2 day span )
asciilifeform: even if i turn up nuffin in next day or 2, i'ma have at least 1 moar planned reset, see whether ben_vulpes remembered to enable eccism when he plugged the thing in
asciilifeform: (in span of 2 day)
asciilifeform: iirc there's a sequence that nails all composites < 2^64 . so could be handy for primorial generators ( vs. eratosthenes etc ) .
mircea_popescu: re http://pizarroisp.net/2019/01/ : leaving aside how annoying it is i can't select, and how it's the case trying to comment timesout : it seems the case pizarro ~made~ about 2 bitcoin during november, on the strength of its fiat-denominated holdings, taking it from being worth ~14 to being worth ~17 bitcoin ?
diana_coman: myeah, that "into the black in two months" seems to me weird here; the trouble is not even "in two months" but at all; because it has been way more than 2 months and if it keeps going as is, I doubt 10 months would make a difference
asciilifeform: take for example diana_coman's system , where 16 witnesses are used. ( i'd use moar, but let's go with the example. ) so if we're generating 2048b primes (for 4096b rsa mod), per ch.14b timings on asciilifeform's iron this costs ~2.9s per modexp, and thereby ~93sec per m-r procedure.
asciilifeform: btw per asciilifeform's chalkboard, the physical cost of constanttime m-r is ~equal to that of (2 modexps of the given width) x (number of witnesses) . ☟︎
diana_coman: I suppose only in some silly situation such as "let's take 2 numbers from user; compute gcd and then proceed to divide and print results"
asciilifeform let it go on 2 diff opterons and 2 intels
asciilifeform: 'Jam runs on UNIX, VMS, NT, OS/2, Mac OS X, and Mac MPW' << if troo, notbad at least in re portability
mircea_popescu: IF 4 takes too little to bother, 2-3 aren't worth doing ; otherwise, otherwise.
asciilifeform: step 2 takes ~1msec (on asciilifeform's aged box, and less elsewhere.)
asciilifeform: let's spell out the algo. 1. reg := read_from_fg(2048bits). 2. l = gcd(reg, primorial) 3. if l != 1, goto 1 4. mr = m-r(reg, shots) 5. if mr != true, goto 1. 6. reg is probabilistically-prime .
deedbot: BingoBoingo paid lobbes invoice 2
BingoBoingo: Anyways, it seems the reason he didn't pile up 2 dozen stiffs over the years is that same poverty
lobbesbot: lobbes: Sent 8 hours and 2 minutes ago: <BingoBoingo> http://p.bvulpes.com/pastes/r6rks/?raw=true
asciilifeform: 'In September 2016, it was announced that Renesas was to acquire Intersil for $3.2 billion'
mircea_popescu: imagine this wonder, "we're winning world war 2, llc", manufacturer of extremely effectual shove handles. THEY NEVER SPLINTER!
mircea_popescu: in the very strict terminology of the turing test : if you get a mystery box you can't explain whether it contains type 1 moron or type 2 moron.
asciilifeform: m mircea_popescu's olympus they are indistinguishable, the way untrained human cannot distinguish 2 types of wasp necessarily, will think 'wasp'
asciilifeform: item contemplated upstack was no sort of 'berth of freedom from inca'. d00d designing turbine parts in broom-closet-sized office, seeing no living faces other than in lift, is a thing, seen with own eyes. for the privilege he pays 2-3x tax ( and can , yes, be 'defrocked' at the whim of whatever gosplan commissions the workpieces, if rocks boat. ) however it is quite diff from life as inmate inside cube hell, subjectively. granted fro
asciilifeform: garrett with 2-3 d00ds, no 'hr'.
mircea_popescu: there, intel kicks in : julia sowells, lives with mom @3036 Powell Rd, Blossvale, NY 13308, a single family home worth about 100k market / 60k on tax basis. 2 bedrooms, 1 bath, 1ksqft.
asciilifeform: but the 2 hypotheses aint distinguishable from my perch
mircea_popescu: scraps, and shaped into curls.[1] The cigarette is ignited at one end, causing it to smolder and allowing smoke to be inhaled from the other end, which is held in or to the mouth. Most modern cigarettes are filtered, although this does not make them safer. Cigarette manufacturers have described cigarettes as a drug administration system for the delivery of nicotine in acceptable and attractive form.[2][3][4][5] Cigarettes ar