tmsr - Search: i

60900+ entries in 0.516s

a111: Logged on 2016-12-24 01:02 asciilifeform: incidentally, since (to borrow the lament of turing's school headmaster) 'the room already stinks of mathematics', i'll share a tidbit that i promised folx some half year ago and promptly forgot:

asciilifeform: and while we're on subj, i'd still like to see the thing applied in for the purpose i originally suggested -- a 'enemy knows 0 structure ahead of time' bitstring transform

asciilifeform: ( i have not yet tried, hands -- very full atm )

asciilifeform: i don't see any theoretical basis for this statement

asciilifeform: i dun think i have quite 5k individual fpga in the parts chest, no

asciilifeform: i have fpga with >1MB ~internal~ sram right here.

mircea_popescu: i'm just hoping S is over 1mb for the 64kb/2kb item and yay.

mircea_popescu: this can't be a thing "i want a bunch of substantially different items that are all identical to the best performing one in any arbitrarily chosen criteria"

mircea_popescu: i confess i like the idea.

phf: i think maybe it's worthwhile as an auxiliary exploration. you can run it and it will either confirm what you already understood, but sometimes it'll correct your understanding. either way you have to understand how sampling profiler works (particularly when it comes to sampling freq), to use it effectively

phf: i think ascii already made that point, that if you're profiling lisp with the vm startup, then you should also profile c machine from boot time. at the very least the vm should be warmed up by loading all the dependencies into the core, doing save-lisp on it, and then making sure that your foo.lisp has an up to date fasl. inside lisp though to achieve the optimizations you run variants of your function inside (time ...) until you bring it within the ra ☟︎

phf: http://btcbase.org/log/2017-06-30#1677284 << i'll take a look. i've been periodically seeings parts of that lisp profiling and optimization thread ☝︎

mircea_popescu: and i think that should be it.

ben_vulpes: once i start collecting data in earnest, yes

mircea_popescu: i mean for the actual data

ben_vulpes: mircea_popescu: i'll letcha know after i rerun and capture that

mircea_popescu: yeah i think so.

lobbes: I gotta say, I'm experiencing the whole "WoT make decisions easier" thing in practice now. I.e. I know hanbot is good for it, and since I trust danielpbarron's rating of Birdman, I also trust that he, too, will be good for it

lobbes: mircea_popescu, hehe. I was thinking of implementing that once n00bs start using the thing. Make it Wot-based at some level; disabling auction service for people lower down in WoT once delinquent x amount of time

ben_vulpes: i will probably just implement the slice that i actually need

asciilifeform: ben_vulpes: i haven't 'updated' ql since 2011 and never again intend to !

a111: Logged on 2017-07-02 15:36 asciilifeform: btw ben_vulpes your mphash seems to use some shitlibrary that 1) i dun have 2) won't install via quicklisp

ben_vulpes: http://btcbase.org/log/2017-07-02#1678507 << "bit-smasher"? came in from ql just fine for me. i only use it as a crutch to beat strings into bit-vectors. ☝︎

a111: Logged on 2015-03-31 02:52 asciilifeform: i will now take the liberty of calling the hypothetical machine 'freyacopter'

asciilifeform: mircea_popescu: i can't imagine it wouldn't be well known. but i was digging for a 'schoolbook' description of it, which i thought existed.

mircea_popescu: asciilifeform it's well known, i had conversations re this in 2012. prolly can fish out quote.

asciilifeform: possibly i dreamed it.

asciilifeform: i have nfi where, if anywhere, it lives on paper.

asciilifeform: funnily enough, last time we had a related thread, i looked for it in schneier 1995 where i THOUGHT i learned it

mircea_popescu: this aside ; i'm kinda loath to mix rsa in both padding and encryption. for all anyone knows they resonate

asciilifeform: anyway i'll omit steps 4-6 and leave as exercise.

mircea_popescu: large part of usgization of normal processes. "oh, does your msg app support VIDEO ?" bitch, i don't want to watch overweight dudes with their cats braying into a mic.

asciilifeform: 1 of the things i've been curious re mphash is bits-of-message-touched vs message-length

mircea_popescu: if that'd covered, i'm happy.

mircea_popescu: asciilifeform i can't really visually saw that appart, but looks like it's a hlen ** b mlen or such.

mircea_popescu: kinda badly chosen cutoffs too, i don't specifically care re diff between 40 byte and 70 byte message. make it log on that side and do 16, 128, 1024, 8192 and 65536 byte messages, for 32, 256, 2048 bit hash lengths as a standard of testing.

asciilifeform: btw ben_vulpes your mphash seems to use some shitlibrary that 1) i dun have 2) won't install via quicklisp ☟︎

ben_vulpes: but if you'll excuse me, i'm going to go make a hash of breakfast

asciilifeform: http://btcbase.org/log/2017-07-02#1678477 << i have nfi why mircea_popescu went with ascii-010010010111.. for the output format ☝︎

asciilifeform: and to have folx suppose that i somehow want to perpetuate it.

asciilifeform: python2 was also garbage, infix pseudolisp with globalinterpreterlock, broken lambda, 1,001 eager idiocies. apparently i write a handful of throwaway rubbish proggies in it ( was less atrocious than perl, so i threw out perl) but now i'm condemned for life to hear about it

ben_vulpes: well if i time just the hash impl it does squeak in under the go implementation

mircea_popescu: i'd call it good

deedbot: http://www.contravex.com/2017/07/01/le-150ieme-partie-i/ << » Contravex: A blog by Pete Dushenski - Le 150ième – Partie I

ben_vulpes: in other optimizations: https://i.imgur.com/h7WcmU5.gifv

sina: I am out. hope all have a wonderful day

ben_vulpes: i'm well supplied with compute tho, thx

sina: ben_vulpes: I am about to head out the door so currently not the best time, but if it would help in anyway I can donate some compute to the effort

ben_vulpes: but i am definitely interested to see how performance plays out on large sets

ben_vulpes: well who knows, who'm i to make guesses like that

ben_vulpes: myeah 'tis what i'm thinking

sina: well, I guess lets see how the benchmarks play out over a larger dataset, maybe it evens out over a certain bitlength or bytesizer

ben_vulpes: sina: aye, i did

ben_vulpes: i think

sina: from reading https://stackoverflow.com/questions/12593768/how-is-lisp-dynamic-and-compiled it looks like GNU CLISP compiles down to bytecode, which I guess will be probably performance equivalent to pypy

a111: Logged on 2017-07-02 04:33 ben_vulpes: no i do intend to black box it

ben_vulpes: ah yeah i thought we'd put that thread to bed with http://btcbase.org/log/2017-07-02#1678389 ☝︎

sina: ben_vulpes: sorry. I am referring to pypy JIT vs python

ben_vulpes: buddy how do you think i've been comparing things?

ben_vulpes: but extending from what i know of java's jit (not much), no. the whole file (at least in the tests i'm running) is compiled.

ben_vulpes: i've no idea what you mean by that in this context

ben_vulpes: i wouldn't keep a lisp runtime hanging around just on the offchance i want to hash things

sina: well, I dunno too much about lisp, does it "JIT" for long running programs?

sina: if I make an mpfhf daemon, then no

ben_vulpes: no i do intend to black box it ☟︎

sina: but again, I haven't used that one

sina: systemtap is basically "dtrace for linux" I guess

ben_vulpes: sina: i wouldn't know, is why i'm asking

sina: I heard good things about systemtap but never used it myself

lobbes: I do!

asciilifeform: ( i still use bad ol' Pb cells, tho. )

mircea_popescu: i expect ya

mircea_popescu: asciilifeform i wasn't even rthinking getting the panels too. just get batteries, charge from network

mircea_popescu: i already have those

mircea_popescu: i could readily spare 6-10 m3. which...

asciilifeform: 'Encell ENC30012' 300Ah 12v ( i.e. 30 Ah 120v supposing ideal inverter... ) lasts, purportedly, 11,000 cycles, 48cm x 48cm x 61 cm.

mircea_popescu: that's what i was thinking, rather than get a diesel, get a very large iron cell set instead.

mircea_popescu: i suppose i could break down and put in... SOLAR PANELS. dun dun dun

mircea_popescu: right. that's what i mean.

asciilifeform: yeah these are terrible, i disrecommend

asciilifeform: ah hm often i forget the sheer expense of getting products of industry in bananilandia

mircea_popescu: sometimes i wonder what you're thinking, i'm connecting from a different dimension or what,.

asciilifeform: i also dun have a zoo!11

mircea_popescu: hey, i got a pool >D

asciilifeform: i can't picture civilized life in bananistan without a 2wk tank of diesel + at least 100kW of gen

mircea_popescu: asciilifeform those were in buenos aires. i didn't take along. all i got here is yhou know, inverters, batteries, stuff like that.

mircea_popescu: i suppose i should provision a whole generating plant again.

mircea_popescu: no power, no pumps (i live up on a hillside) and so i had the girls... fery water. from the pool, with ex-paper baskets repurposed as sacas, up the fucking stairs. ☟︎

mircea_popescu: i don't mean, here where i am. i mean, in costa rica period. and nicaragua. and panama. toda la centroamerica, mang!

mircea_popescu: i think that about covers it.

mircea_popescu: as far as i'm concerned, good party = no less than 60 females per 100, no loser/dork males present, good music not too loudly and some public fucking.

mircea_popescu: then again, i wasn't on the west coast ; nor do i have much patience for the flip flop and smoked out crowd.

mircea_popescu: honestly, i don't recall ever going to a good party states-side.

mircea_popescu: i dunno anyone who'd turn down her rsvp.

asciilifeform: point was rather ' i can't see diff b/w here and commissar, same faceless face '

mircea_popescu: point being : the "i'd give it to the soviet comissar rather than to this woman" preference ain't supported.

asciilifeform: i didn't see any naked ones. but then again did not dig.

asciilifeform: mircea_popescu: right, which is why i am not sitting here and smoking benjies

mircea_popescu: i suppose this'd be why you dun have a network then.