asciilifeform: aa

asciilifeform: mircea_popescu: i see eggog

asciilifeform: 'small pieces' is pretty good b00k

asciilifeform: neato spyked . keep in mind that you gotta use the ada subset displayed in ffa.

asciilifeform: aite

asciilifeform: mircea_popescu: http://btcbase.org/log/2017-10-08#1722595 << esr ? ☝︎

asciilifeform: ads on these trains, ftr, not cheap. ( and certainly not 'to allcomers' )

asciilifeform: in other lullies, http://www.loper-os.org/pub/nsawagenhoneypot.jpg << found on washington metro train today ☟︎☟︎

asciilifeform: i have deeply nfi

asciilifeform: http://www.loper-os.org/?p=568&cpage=4#comment-18272 << in other strange. ☟︎

asciilifeform: ( yes you set the low bit to 1 )

asciilifeform: primes >2 are odd, noose at 11

asciilifeform: worxgreat

asciilifeform: ( where first suggested , ftr : http://btcbase.org/log/2017-08-14#1697598 ) ☝︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722415 looked like a 'who needs miller-rabin' ☝︎

asciilifeform: http://btcbase.org/log/2017-10-08#1722470 << is why i suggested it to begin with, zaps items with factors up to 16bit or so quickly ☝︎

asciilifeform: ( try it )

asciilifeform: http://btcbase.org/log/2017-10-08#1722468 << quite acceptable, 1 in few thou ☝︎

asciilifeform bbl

asciilifeform: ( orig., iirc, thread : http://btcbase.org/log/2017-07-02#1678490 ) ☝︎

asciilifeform: ( see phf's http://btcbase.org/log/2017-07-03#1678660 , or even naggum's, re why ) ☝︎

asciilifeform: because if it did, it is meaningless

asciilifeform: http://btcbase.org/log/2017-10-07#1722367 << i gotta ask if this figure included sbcl load time !? ☝︎

asciilifeform: tldr : initial py snippet i had lying around was braindamaged.

asciilifeform: ( mainly, i suspect, by recognizing masses of 0 in karatsuba and returning 0 when they get mul'd )

asciilifeform: http://wotpaste.cascadianhacker.com/pastes/saynG/?raw=true << all1s. 0.028s. tho i do suspect it shortcuts internally.

asciilifeform: 0.018s on this box.

asciilifeform: ( and skips unused hammings... )

asciilifeform: http://wotpaste.cascadianhacker.com/pastes/0A6fb/?raw=true << python 'fair fight' ver

asciilifeform: oh ffs, here goes,

asciilifeform: if somebody wants to make the physically possible version of this, to see what happens on max hammingweight...

asciilifeform: for the obvious reason.

asciilifeform: OverflowError: Python int too large to convert to C long

asciilifeform: with eggog:

asciilifeform: phf, mod6 : funnily enough i went and tried the 'fair fight' max(4096b) a^b mod c in python, http://wotpaste.cascadianhacker.com/pastes/GHATB/?raw=true , but it... bombs

asciilifeform: i proposed primorial strictly as an initial winnowing to replace the idiot trial divisions koch et al used.

asciilifeform: ( certainly not even for as large a number as 64bit... much less 4096 )

asciilifeform: so no, nobody is replacing miller-rabin with gcd(primorial, x). ☟︎

asciilifeform: but certainly not 129.

asciilifeform: http://btcbase.org/log/2017-10-07#1722415 << if you have a comp the size of jupiter, you could ~maybe~ have such a thing as a 128bit primorial. ☝︎

asciilifeform: ( where there is no assurance of not consing and not branching )

asciilifeform: but 2 ) the python example is of course not closed form, and it is imho meaningless to even attempt to write the closed form item in a language like python or cl

asciilifeform: http://btcbase.org/log/2017-10-07#1722411 << 1 ) ffa is closed form. i.e. it CAN be written as a number of nand gates, with a 'funnel' at the top, to which you present a,b,c, e.g. 4096bit, numbers, and at the bottom in a little cup you get a^b mod c , and with NO UPWARDS FEEDBACK FLOW of information , i.e. answer comes after same interval of time always, and with strictly downwards signals. ☝︎☟︎☟︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722408 << you might consider reading the code ? it has all been posted. ☝︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722405 << in no case can the 'cheap initial primality test' primorial exceed the size of current ffa width. thinkaboutit. ☝︎

asciilifeform: all other methods leak info via timing , amperage, rf noise.

asciilifeform: the ONLY correct method of generating cryptoprimes, is to 1) get N bits from FUCKGOATS 2) determine, in fixed spacetime every single time, whether that string of bits constitutes a usable prime. ☟︎☟︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722402 << this is a fundamentally wrong way to generate cryptographic primes. we had a thread about it, http://btcbase.org/log/2017-08-14#1697562 ☝︎☝︎☟︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722400 << bernstein's gcd method is neither here nor there, i certainly don't need anything of the kind in ffa, and quite likely it fundamentally does not ffaize ☝︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722397 << i don't see anything that only wants ~lower~ half... whatcha talking about ☝︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722395 << compute and then what ? gotta multiply ☝︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722394 << this looks very , very painful to prove correctness of. i'ma come back to it. ☝︎

asciilifeform finally eaten log...

asciilifeform: http://btcbase.org/log/2017-10-07#1722376 << modmuls, not exps ☝︎

asciilifeform: because gotta get upper word somehow

asciilifeform: so a word mul is actually five MULs

asciilifeform: 0 asm

asciilifeform: and incidentally my base cases are ultra-slow, in theory

asciilifeform: 'старый и злой -- чем не жених!' (tm)(r)

asciilifeform: slow and broadcasting seekritz for miles around, whatsnottilike!!111

asciilifeform: i also suspect that they are in fact slower for maxhammingweight case of exponentiation and modulus, vs ffa. ☟︎

asciilifeform: the interesting imho discovery is that heathen bignumtrons don't win much (or even any!) speed by normalizing the ints being added/subtracted ☟︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722358 << point was exactly to compare like items. i.e. heathendom does NOT get to 'win' by 'oh hey the hamming weight of exponent is only 2, not 4096, so we only do 4 modexps and not 8192' ☝︎☟︎☟︎☟︎

asciilifeform: http://btcbase.org/log/2017-10-07#1722372 << in fact we have closed form. ☝︎☟︎

asciilifeform bbl : meat.

asciilifeform: on heathentron

asciilifeform: betcha it won't be 1.2s nomoar

asciilifeform: phf: now try same width exponent !

asciilifeform: out of curiosity, how long the py item takes on mod6's box ?

asciilifeform: ( nor will anything else. )

asciilifeform: in ffa, unlike in the python example, elongating the 0x10001 to full ffawidth will not change the required time.

asciilifeform: phun phakt, this calculation is taken from the gpg autopsies last summer, when asciilifeform was chasing imaginary rng boojum after somebody found a real one

asciilifeform: this is even though python uses a c bignumatron internally.

asciilifeform: ( interestingly, it takes 3.8 sec on my box )

asciilifeform: you can run it, get same answer.

asciilifeform: mod6: http://wotpaste.cascadianhacker.com/pastes/H4UGn/?raw=true << for comparison, py script computing same arithm problem

asciilifeform: ^ the 'two second' item, modexp

asciilifeform: mod6: http://wotpaste.cascadianhacker.com/pastes/0k78K/?raw=true << example from asciilifeform's torture room, of what his test looks like

asciilifeform: it is tempting, because currently i suspect that ~nobody is actually running my pastes

asciilifeform: currently i generate them with a pyturd

asciilifeform: mod6: unit tests will work as pcode known-good in/out pairs

asciilifeform: *changes

asciilifeform: i've been holding off on releasing the p-interpreter because there are several quite broad changed in the way that it worx, in the pipeline, and i'd rather folx not get used to the old form.

asciilifeform: ( it will need a small adjustment in re http://btcbase.org/log/2017-10-02#1719728 but otherwise oughta work ) ☝︎

asciilifeform: mod6: you should have one already, the factorial thing

asciilifeform: mod6: i've been using (unreleased) 'p' as the tester.

asciilifeform: hanbot: wanna try yer hand at ^ puzzler ?

asciilifeform: heya hanbot

asciilifeform: ... could even live with this, if i had a hard proof that it's never moar than 3.

asciilifeform: ( if it isn't obvious from where the error comes : observe the 3 Karatsuba_Term additions. in ordinary K., they walk over the upper half of XYLo ( lower half of result.) but in TopOnly K. we lose XYLo, so that carryolade is lost. )

asciilifeform: now! this procrusted-karatsuba is only used for the barrettron, so theoretically could compensate for that 3 with 3 additional subtractor-muxes. and still win ~4x speedup vs last night's . but this is mega-ugly.

asciilifeform: ( Karatsuba_Term is same for both )

asciilifeform: http://wotpaste.cascadianhacker.com/pastes/TgRkm/?raw=true << ordinary karatsuba, for convenient comparison.

asciilifeform: ^ this 'upper half only' karatsuba works, but the answer is always off by 0 to 3, because the carries from the bottom halves are ( recursively ) lost. somehow gotta be finessed.

asciilifeform: in other puzzlers, http://wotpaste.cascadianhacker.com/pastes/6l4uH/?raw=true << mod6 et al ☟︎

asciilifeform: '...I think a formal apology should be handed out, and the teacher involved should be reprimanded,” he added. '

asciilifeform: lol!

asciilifeform: '“the year is 1935 and you have been tasked with creating a mascot to represent the Nazi party at its political rallies.” “Think about all of the information you have learned about Hitler and the Nazi party,” the assignment directed. “You will create a COLORFUL illustration of the mascot. Give the mascot a NAME. You will also write an explanation as to why the mascot was chosen to represent the Nazi party.”'

asciilifeform: phantom limb, i tellya.

asciilifeform was about to upload crc book'o'crypto and then remembered that dulap is gone...