log☇︎
412300+ entries in 0.267s
asciilifeform: http://log.bitcoin-assets.com/?date=29-08-2015#1256109 << what exactly is meant by this ?? ☝︎
mircea_popescu: asciilifeform also im pretty sure power didn't actually fail. derpy kids doing things to fix things etc. ☟︎
mircea_popescu: but in no small part (to answer punkman's q) this experiment is interesting to me specifically to see how well things do in cramped spaces. ☟︎
mircea_popescu: i dunno what the proper item to measure is.
asciilifeform: hence, from 75% to 0
asciilifeform: mircea_popescu: strictly cpu cycles? the answer depends on whether anyone is querying the db
asciilifeform: it changes ENTIRELY every single time a modulus is processed !
asciilifeform: understand, the running-product is an integer, presently ~1GByte in size.
trinque: yeah, you want a ton of fast RAM, I'd assume, and not want to be writing to disk constantly
assbot: Logged on 29-08-2015 02:49:45; asciilifeform: not only does it not, but it absolutely impossible to do anything re: the running product having to be recalculated on warmup without 1) increasing complexity 25x 2) slowing down processing palpably
assbot: Logged on 29-08-2015 02:48:54; mircea_popescu: dulap may be the latest piece of software in actual use that doesn't actually benefit from periodic rebooting
asciilifeform: and the budget is approx. $0.
asciilifeform: all of this could be had, but FOR A PRICE
trinque: ^ that point about the kids these days and rebooting, bleh. no kidding.
assbot: Logged on 29-08-2015 02:47:55; mircea_popescu: i think the latter kids don't really grok this "don't reboot machine " thing
asciilifeform: in fact, phuctor is presently NOT getting enough oxygen to live, as per http://log.bitcoin-assets.com/?date=29-08-2015#1255136 ☝︎
asciilifeform: experimentally demonstrated that neither runs ADEQUATELY (much less with any degree of trust!) without this condition being met.
asciilifeform: punkman: since you asked, it also doesn't hurt to point out that both phuctor and the node MUST HAVE reasonably-cpu-and-ram-rich PHYSICAL machine.
asciilifeform: (not all spare time is created equal. i'm both free from paying work AND awake enough to program, for perhaps a few hours a month!)
asciilifeform: (contrary to what some may think.)
asciilifeform: i, on the other hand, donated my spare time, of which i don't actually have very much.
asciilifeform: if mircea_popescu had not donated the cpu cycles, the thing would still be sitting on aws and being essentially useless
asciilifeform: can't exactly charge the ships, for the light.
asciilifeform: there is no way in heaven or hell that either of those things could ever pay for itself.
asciilifeform: this is YET AGAIN the 'can't charge for sunlight' thing !!
asciilifeform: for no especially good reason other than that i felt like it, and mircea_popescu saw it fit to supply a machine (via subcontractor) out of his own (for now) pocket
asciilifeform: phuctor and the node are run as a public service.
asciilifeform: and neither phuctor nor therealbitcoin bring in any revenue
asciilifeform: ;;later tell mircea_popescu dulap is a quad-core machine. current 'werker' (phuctor) uses strictly a single core, at the maximum. but what is happening appears to be bus cycle starvation (shared cores share a cache; and at any rate all processes share a single ram bus)
assbot: Logged on 29-08-2015 19:48:13; asciilifeform: ;;later tell mircea_popescu potentially interesting observation: dulap has been ~50 blocks behind semi-permanently since reboot. in as far as i can tell, it is not mainly from 'blackholing' but due to cpu starvation! on account of phuctor re-grinding the product...!
assbot: Logged on 29-08-2015 02:22:08; asciilifeform: mircea_popescu et al: dulap rebooted today for no reason known to me.
assbot: Logged on 29-08-2015 21:09:21; mod6: this should do the trick: `LC_ALL=C ./bitcoind -myip=127.0.0.1 -addnode=195.211.154.159 &`
asciilifeform: recall that we don't exactly have a deterministic system here.
asciilifeform: http://log.bitcoin-assets.com/?date=29-08-2015#1256010 << if you specify what exactly ought to be measured, i might be able to do it ☝︎
mod6: ah, gravy, thanks.
punkman: also thinking of making a fork that doesn't mine/wallet
punkman: I might work on part2 this week
mod6: punkman: naw, i haevn't had a chance yet. i'll get to it though, it's on my list.
danielpbarron: gentoo made with trinque's script
punkman: anyone try the final debug_sanity_part1 patch?
danielpbarron: and it is up to height=281399
mod6: this should do the trick: `LC_ALL=C ./bitcoind -myip=127.0.0.1 -addnode=195.211.154.159 &` ☟︎
danielpbarron: Bitcoin version 0.5.4-beta << I'm currently running this, made with the rotor instructions
mod6: just see if you can get it to start up and sync some blox
danielpbarron: what does it mean to rotate?
mod6: yup, that's required on every system that i've encountered when built with the `rotor'.
trinque: with the LC_ALL=C
mod6: danielpbarron: still have it? try to fire it up see if it it rotates.
mod6: i'll put that in there. "Linux Mint, 13-14ish." :P
mod6: ;;later tell danielpbarron You did test out rotor w/TEST2 or stator this month right? Or did I mis-remember that?
trinque: I could run over all previous records at some long-ish interval and see whether that occurred, update db accordingly.
trinque: punkman: I have not handled that case. As I understand it, transactions would then go back into the mempool for inclusion in a subsequent block.
mod6: chettty & mircea_popescu: btw, eulora seems to be functioning well since changes. although, i don't use any automation in there. just good ole hard rock minin' by hand. with a mule^H^H^H^Himproved pickaxe.
mod6: i guess i do remember this now.
mod6: punkman: hey! glad i asked. thanks :]
punkman: mod6: I tested stator on Mint. didn't leave it running for full sync though
mod6: ah. ok. np. thanks!
jurov: no i was too busy and then on afk vacation
mod6: ;;later tell jurov You did test out rotor w/TEST2 or stator this month right? Or did I mis-remember that?
mod6: ;;later tell punkman You did test out rotor w/TEST2 or stator this month right? Or did I mis-remember that?
mod6: ;;later tell davout You did test out rotor w/TEST2 or stator this month right? Or did I mis-remember that? ☟︎
assbot: You rated user gernika on 04-Jul-2015, with a rating of 1, and supplied these additional notes: New blood..
asciilifeform: since we're on the subject, i do not regard it as a proven fact that schwartz tied the rope by himself
phf: no atheists in foxholes and all that
asciilifeform: it is just something people ~do~, like the boyars impaled by ivan iv famously screaming 'god save the tsar' as their arses stretched
asciilifeform: it is why the fact that (in american courts) the plea never really ~works~ does not affect its popularity
phf: right, also same point at which aaronsw chose to off himself
asciilifeform: rather than any kind of rational undertaking
asciilifeform: i see the ulbricht plea, and schrodinger's, etc. as similar
asciilifeform: there is a long history of tales re: folks having conversations with (personified) Death, and asking her 'not now, why me, ...' etc
phf: well, in the case ulbricht his plea was dumb, the point which you guys consistently drive in other cases
asciilifeform: phf: iirc this began as a thread about r. ulbricht's shameful plea for mercy to the judge
phf: bakunin has equally shameful confessions written to nicholas i
assbot: Logged on 26-05-2015 17:45:33; ascii_field: ;;later tell mircea_popescu http://wiki.mises.org/wiki/Miscellany:Confession_to_the_F%C3%BChrer << schrödinger's shameful letter. english.
asciilifeform: http://log.bitcoin-assets.com//?date=26-05-2015#1145721 << thread ☝︎
asciilifeform: erwin schrodinger, famously, did his physics almost as just a thing to fill the time between random fucks. but he was not enough 'non serviam' nietzschean superman to avoid sucking hitler's cock (unsuccessfully !!) and having to run for his life
asciilifeform: considering how the latter is almost inescapably a less-rewarding affair than the former
asciilifeform: i suspect that folks who are child prodigies at programming people (esp. programming the female of the species) do not usually end up having the patience to seriously deal with machines
asciilifeform: or, more appropriately to the case at hand, 'program pc or don't program'
asciilifeform: ben_vulpes: the correct setting for the gedankenexperiment is 'serve usg or take the cyanide'
ben_vulpes: i do not get the 'non serviam' routine. we all serve the khan.
asciilifeform: but instead other things.
asciilifeform: for instance, iirc mircea_popescu was also pretty handy at mathematics as a young man. but notice that we do not come here to learn mathematics from him.
asciilifeform: and it also turns out that for some reason the folks who are very excellent at non serviam - don't program. (napoleon's work in geometry notwithstanding...)
asciilifeform: at least, at individual steps of the way.
mod6: anyway, not that it need be the same. just curious.
phf: sure, korolev was tested by that limit, and i'm not in his position, so i can't say "сука буду i've offed myself", but when the choice is between doing csv and sexp for a customer..
mod6: lol, the css 'solid green', appears slightly different tone than 0x00FF00 on thebitcoin.foundation, or is it just my eldery eyes?
asciilifeform: phf: mircea_popescu seems to suggest that if korolev, etc. were real men, they would have killed themselves instead of serving the hated master.
phf: <asciilifeform> ben_vulpes: take me to this planet of 'healthy relationships.' << i am equally confused by this position. non serviam doesn't necessarily need to translate to its extreme expression. if you're that figured out how to push the lever, i think the last option that you can possibly have is to keep pushing the lever for others. once you accept it to its logical conclusion, are you not free to do as you please?
ben_vulpes: call it 3K then.
asciilifeform: ben_vulpes: actually don't be so sure. it is almost certainly possible to shorten it.
mod6 ribs trinque a bit :D
punkman: trinque: would deedbot notice a bundle tx that got reorg'd out of main chain?
asciilifeform: http://log.bitcoin-assets.com/?date=29-08-2015#1255945 << lol, how big did ben_vulpes think it to be ? ☝︎
mod6: this thing has a cool 0x00FF00 && 0x000000 website
trinque: it's rather dumb, doesn't watch for blocks in the log, though that'd be nice
mod6: trinque: fantastic. thanks.
asciilifeform: and they won't even be able to eat my corpse, wrong-chirality proteins
trinque: mod6: shall pop out at the next hour