log☇︎
27200+ entries in 0.015s
asciilifeform: i can't even think of ~one~ post-'09 title i liked, other than 'portal'
mp_en_viaje: however, if you sit down to play gothic, or kb or any fucking one thing it'll be time spent playing, full screen, immersed ☟︎
mp_en_viaje: the entire collection of "video games" output 2009-2019 can command HALF the play hours off your time that ANY ONE TITLE of 2000-2009 can command
mp_en_viaje: i didn't even start with it ; the actual history as it unfurled is fucking emblematic. 1 "oh, steamos ?" 2. nowait, no os, idiots ; 3. ahahaha, THEY HAVE NO GAMES, how can they survive 4. omfg bbq NOBODY DOES HAVE ANY
mp_en_viaje: then alf is like "but mp... you don't need accelerator card for those" and mp is like... ikr.
asciilifeform: i may have mentioned, built exactly similar 'motorcycle' 2y ago, and to same result.
mp_en_viaje: this is not merely true, but fractally fucking true. consider the actual, lived experience of yours truly, who ended up buying that motorcycle of a 5lb vid card and that mega screen etc to... in the end ... play heroes3 on it ?
mp_en_viaje: but surely all metaconsiderations & heuristics point that way
mp_en_viaje: possibly so ; tho we yet lack the experimental buildup to say.
asciilifeform: mp_en_viaje: i suspect this hypothesis extends just the same to iron.
mp_en_viaje: 1. most of the "need" and "use" you perceive is entirely imaginary, resulting from the intersection of your WSOD and their marketing efforts and 2. most of the actual utility comes from older stuff in the first place, and will be delivered better by resurected older stuff than by the newer stuff.
asciilifeform: the current kernel is <2MB. and prolly trimmable further. boot to shell in 3sec or so. ☟︎
mp_en_viaje: kinda what all this experience seems to be converging towards.
a111: Logged on 2019-06-04 00:51 asciilifeform: incidentally, i generate these by machine, and it takes about 3sec per. would have put it as a net-connected hopper thing aeons ago, BUT it of course uses a heathen render (there are no 'demonstrably electrically correct' pdf eaters, and i dun expect one to exist) and suffers from the obvious problem
mp_en_viaje: basically, a flat rejection of the userland past 10 years or so is the ~exact equivalent of not talking to indians in call centers on the phone. "i'll just talk to the manager, get lost paki."
mp_en_viaje: "but mp, if you don't get latest chromed piss you can't connect to appsites like fetlife". like hell i can't! i wouldn't fucking use their idiotic interface to interact with their own backend/userbase/wife-and-daugther if they fucking paid me anyways.
a111: Logged on 2019-05-16 21:23 asciilifeform: http://btcbase.org/log/2019-05-16#1914135 << bahahahahaha! and for extra lulz, phuctor was 'replaced' in <2hrs -- whereas this took, wat was it, 6mo ?
mp_en_viaje: e advances" in gaming and the web is STILL better seen through lynx than through firecrash/chromeburn. what, i lose out on loading github ? linkedin ? lordy. ☝︎
mp_en_viaje: then again, i can not name any program published after... uh i dunno, 2005 or so that i actually fucking use. a large part of the advantage in dealing with these idiots is still ye same old : nobody needs aything they "did" for anything. much like i have no use for "all the advances" in bitcoin (what, segwit ? bwahahaha) i also don't have any use for "all th
a111: Logged on 2019-07-17 12:16 asciilifeform: problem comes if you want to run non-contemporary proggies on it ( musl, gnat 4.9.x, etc . ) linus permitted abi to change.
mp_en_viaje: http://btcbase.org/log/2019-07-17#1922951 << yeah, that was a fucking problem. ☝︎
a111: Logged on 2018-10-26 02:14 asciilifeform: meanwhile, in gnat bugs : apparently ( and this is documented or mentioned nowhere ) : it is impossible to have a Ada.Finalization.Limited_Controlled type ANYWHERE inside a static library, unless it is generic all the way down (i.e. if the lib package is generic, any sub-packages must also be instantiated as generics )
asciilifeform actually has the with-what engine, and it's blocked on that gnat finalization bug ☝︎
mp_en_viaje: it'll be so fucking nice when we finally manage to ditch that dependency tho
asciilifeform: mp_en_viaje: doesn't require hand-hexing, since trb actually stores all blox, simply needs re-walk trigger
mp_en_viaje: of course, this also requires familiarity with bdb, which is a lot like familiarity with baud modems.
a111: Logged on 2019-07-17 01:21 asciilifeform: http://btcbase.org/log/2019-07-16#1922900 << i dunthink i've ever witnessed the corrupted db (tho i believe the folx who have.) possibly because i have never run node off anything other than always-on sinusoidal inverter ups, tho
mp_en_viaje: http://btcbase.org/log/2019-07-17#1922922 << fortunately bdb will actually recover itself from ~most "omfg error" states it loudly reports upon unexpected shutdown. not 100%, but in general it can be hacked back into coherence without much trouble. ☝︎
asciilifeform did write some experimental patches, and they were marked with jolly roger, and certainly aint about to recommend'em to anyone who does not consciously know wtf he's doing, much less hand-hold to install..
asciilifeform: esp. for safety-critical component that is wallet.
mp_en_viaje: besides the "i utterly hate the republic and wish to fuck it over", there's no incentive for the stance available.
mp_en_viaje: more to the point : why ~the fuck~ would you want to spend your time "hand-helping" someone on "how to install an improper patch" in preference of, spending the same time making it a proper fucking patch, which has the side benefit that now you don't have to hold their hand.
asciilifeform: mod6 when you wake up -- wtf was this ? ☟︎
asciilifeform: i dun even have that patch on my hdd, turns out
mp_en_viaje: get the fucking shit aligned or stop talking about it altogether. it's either abandoned, in which case, it is not mentioned ; or it is not abandoned, in which case it is fixed. ☟︎
a111: Logged on 2019-07-17 01:18 asciilifeform: http://btcbase.org/log/2019-07-16#1922895 << recall that mp_en_viaje prescribed that the wallet oughta be sawed off into own proggy. imho this is entirely Right Thing. but no one has yet found the free hands with which to do, afaik
mp_en_viaje: http://btcbase.org/log/2019-07-17#1922917 << for what it's worth im well irritated by all this out-and-out saboteur work of subversion, "oh, there's been this improper patch for years now that we keep deliberately maintaining improper and keep mentioning" nonsense. ☝︎
a111: Logged on 2019-07-17 01:01 asciilifeform: ... and mips in general. ( how do you suppose the ball-of-shit arm arch became dominant. to this day chinese pay tribute to britain for 'permission' to make arm. why ? at one time mips had patent, but expired in 2009 or so. and in march of '19 orig. mips verilog entirely opensoresd and made public... yet 0 fast mips on the market. guess why. )
mp_en_viaje: http://btcbase.org/log/2019-07-17#1922915 << i dunno, ima take a stab in the dark here and guess ? ☝︎
asciilifeform: mp_en_viaje: unrelated, i found 1 of your belarus tomb folx, comment in mod queue.
mp_en_viaje: i dun think i actually moved to 3 branch.
asciilifeform: mp_en_viaje: rk and dulap 4.x , but heavily hand-cut. ( and i suspect can be cut further still . ) i dun operate kernels other than hand-cut.
deedbot: 2019/07/14 22:26:25 <trinque> I'm sure it does feel like a bit of gruntwork, and yet, later you will always have that curl, and it will always build.
diana_coman: !!seen trinque
asciilifeform: kernels, however, no hard limit found yet, erry 3.x seems to work (with my patch)
a111: Logged on 2019-07-15 10:05 diana_coman: trinque: from what I see though the genesis.vpatch is a snapshot of /cuntoo/portage dir *only* which means that the actual tarballs with the code are not included anyway - so basically it will still fail to find them as soon as whatever URI in the ebuild doesn't host them anymore, what am I missing?
diana_coman: http://btcbase.org/log/2019-07-15#1922470 - trinque, any chance of getting an answer on this? ☝︎
asciilifeform: and 4.8.1 appears to be the last gcc w/ properly-behaving mips1 backend.
asciilifeform: ftr the last version of buildroot that does 100% Right Thing (i.e. built working kernel + userland) is 2013.08-rc1 .
asciilifeform: ( how ? e.g. 'b _end_cycle if $r9==0xdeadf00d' will stop when 'program counter' reg of mips is equal to 0xdeadf00d', then single-step to see what instr dispatched... then 'i r' and see regs. etc )
asciilifeform: ( pretty gnarly, incidentally, had to debug crashed kernels using ~pc host~'s gdb, as there is no dedicated debuggism in the 12kB emu thing, it'd massively slow down execution just by existing)
asciilifeform: mp_en_viaje: i expect it'll be somewhat easier than the piece already done.
a111: Logged on 2019-07-17 00:57 asciilifeform: near as i can guess, this was done in order to help kill loongson sinomips.
mp_en_viaje: http://btcbase.org/log/2019-07-17#1922914 << the schmucks are pretty shameless, it is true, but perhaps not many notches above trivial to reconstruct musl atop a proper (ie, without Peter Korsgaard &rest of retards) ☝︎
asciilifeform: ( in the 3 'byte r/w' instructions, nowhere else )
asciilifeform: if switch to 'little', you will get a roughly 0.01% speedup, all it'll do it get rid of the xor addr, 0x3 instr.
asciilifeform: ( fwiw the posted prototype has selectable endianism, albeit via an 'ifdefism' . thus far only tested with 'bigendian' kernel + userland tho. )
asciilifeform: so apparently orig cmips d00d was not to blame for this oddity.
asciilifeform: ts ), and so dun matter how the instrs are represented, from user's pov
a111: Logged on 2019-06-22 16:43 asciilifeform: re: loose ends: fwiw asciilifeform did in fact add the final missing piece to mipstron. BUT! can't test with the dummkopf's orig linux image as he... guess wat, his system had little-endian word accesses but big-endian byte read/write ! so his image in fact will boot on NO existing mips, nor any afaik other emulator.
asciilifeform: btw for thread-completeness -- the 'weird-endian' puzzle had simple answer : turns out the mips arch uses 'little' representation for the instructions per se, even on 'big' machine. the rationale for this, was that the instr. length is fixed, and jumps into unaligned addresses are forbidden (i.e. iron will not load an instruction from an addr that doesn't end with 2 (on 64bit -- 3) zero bi ☝︎
asciilifeform: and you can plug'em into virtually any box, from 'vax' to 'pogo', all you need is a serial port of whatever description.
asciilifeform: single-phosphour crt, w/out shadow mask. hence razor-sharp ( and actually why bolix made the lispm console a b&w tube. was only way to get 1280x1024 in 1980s tube, with sharp text) . for this reason asciilifeform for many years had actual glass tty, when first started unixing, was 1 of the best junkyard finds ever imho
a111: Logged on 2019-07-17 11:35 mp_en_viaje: yellow on black too, cga 4 eva
asciilifeform: http://btcbase.org/log/2019-07-17#1922942 << cga, by the standards of that world, is 'newfangled'. the glass tty's typically emulate ibm's 3270 -- a 1971 product... (or vt100, dec circa 1978!) ☝︎
a111: Logged on 2019-07-17 11:35 mp_en_viaje: in more uplifiting news, travel agent (this twentysomething "rocker" kid, kinda cute at that) used honest to got TERMINAL. none of those java clucker interfaces.
asciilifeform: http://btcbase.org/log/2019-07-17#1922941 << afaik all the airlines still are using the ~physically same~ ibm mainframes as in 1990. ( simply, in 'civilized' world, the agents run 'terminal' proggy under microshit. whereas in orclands, ye olde ibm glass terminals. i saw one in e.g. argentina, in coupla places ) ☝︎☟︎
asciilifeform: it dun even ~have~ to be mips, i picked it cuz it is the 'smallest' , moving-parts-wise, arch, for which there was existing gcc back-end.
asciilifeform: when you have a ~defined~ iron, can lose 99% of the driverism crapolade.
asciilifeform: the incident where i glued together rk pilot plant, to only ~then~ find out that nobody knows when the fuck proper gnat will actually build arm binaries w/ working threading, was instructive. ☟︎
a111: Logged on 2019-07-08 17:28 asciilifeform: meanwhile, asciilifeform while cooking war plan went and thought 'oughta clean up the june item for pre-pub'. and so, for the pleasure of the audience : http://www.loper-os.org/pub/tmsrmips-demo/tmsrmips.asm
asciilifeform: re 'lunapark cars' -- the rationale for the mips experiment is only half 'make honest vps for piz', other half is 'make reproducible linuxable iron' ☝︎
a111: Logged on 2019-07-17 11:33 mp_en_viaje: http://btcbase.org/log/2019-07-16#1922886 << this is such fucking recurrent bs... linker WILL emit same crap n omatter what sjlj thing you say to it ; and so on in this manner, "computing" in the sense of conway island wheel "cars", will spin regardless of how little kids turn and twist the "steering wheel"
a111: Logged on 2019-07-17 10:59 mp_en_viaje: http://btcbase.org/log/2019-07-16#1922882 << reminds me of zimmerman fucktard, wrote to me "oh, lost pw to my orig key"
asciilifeform: http://btcbase.org/log/2019-07-17#1922928 << ftr i had to disasm his demo kernel to find how the fuck he made it go with the simplistic mips timer. (why ? cuz linux porting docs -- lie !) ☝︎☝︎
asciilifeform: problem comes if you want to run non-contemporary proggies on it ( musl, gnat 4.9.x, etc . ) linus permitted abi to change. ☟︎
asciilifeform: the beauty of synthetic 'iron' is that you can actually use just about any kernel, because there is no driver gnarl, you're already baking the drivers with own hands
asciilifeform: it was ~2.4~ that some folx are keeping alive, cuz it would fit, with trimming, into <1M. but if yer using 2.4.x you will need a contemporary userland.
asciilifeform: never found any reason to keep alive 2.6.x, hand-cut later kernel can typically sit down in same footprint even in small irons
asciilifeform: mp_en_viaje: i don't have anything running a 2.x, and haven't for some years ( ye olde 'wrt54g' was i think the last piece of iron i had which ran one )
spyked is using different kernel versions (but mostly 3.x) depending on the hardware.
spyked: iirc the rk (and everything on arm64) is on 4.x
a111: Logged on 2019-07-16 21:52 asciilifeform: re that kernel -- someone gotta genesis a kernel. ( if no one has any constructive input re ~which~ one, then i will, and it will be somewhat arbitrary. ) atm that patch is a bad-old-style patch, rather than vpatch.
mp_en_viaje: http://btcbase.org/log/2019-07-16#1922889 << i thought we were mostly using 2.6 - 2.9 sorta thing ? is everyone really on 3 branch ? ☝︎
mp_en_viaje: yellow on black too, cga 4 eva ☟︎
mp_en_viaje: in more uplifiting news, travel agent (this twentysomething "rocker" kid, kinda cute at that) used honest to got TERMINAL. none of those java clucker interfaces. ☟︎
mp_en_viaje: http://btcbase.org/log/2019-07-16#1922886 << this is such fucking recurrent bs... linker WILL emit same crap n omatter what sjlj thing you say to it ; and so on in this manner, "computing" in the sense of conway island wheel "cars", will spin regardless of how little kids turn and twist the "steering wheel" ☝︎☟︎
diana_coman: heh, at least there are pretty pictures!
a111: Logged on 2019-07-17 10:19 feedbot: http://ossasepia.com/2019/07/17/the-mirror-land-notes-on-graphics-for-eulora-iii/ << Ossa Sepia -- The Mirror Land (Notes on Graphics for Eulora, III)
mp_en_viaje: http://btcbase.org/log/2019-07-17#1922927 << this stuff's pretty fun to read ☝︎
a111: Logged on 2019-07-16 21:39 asciilifeform: also in the process made sad discovery : previously asciilifeform always pronounced 'gcc 4.9 is usable' but! turns out some time b/w 4.8.1 and 4.9 , someone silently sabotaged support for classical mips1
mp_en_viaje: i guess by now we all hold the key usage longivity records.
mp_en_viaje: i'm like "fucker... why the fuck am i talking to you. are you even in show business ?!"
a111: Logged on 2019-07-16 21:38 asciilifeform: orig. author of 'cmips' btw was of 0 use, i wrote to him a while back and not long ago he answerd 'i lost hdd, lost errything'
mp_en_viaje: http://btcbase.org/log/2019-07-16#1922882 << reminds me of zimmerman fucktard, wrote to me "oh, lost pw to my orig key" ☝︎☟︎
feedbot: http://ossasepia.com/2019/07/17/the-mirror-land-notes-on-graphics-for-eulora-iii/ << Ossa Sepia -- The Mirror Land (Notes on Graphics for Eulora, III) ☟︎
BingoBoingo: <asciilifeform> aaand gcc broke support around same time. << And the hard opensourpush was made to push a crippled subset of SPARC as MIPS alternative around that time as well iirc
asciilifeform: mod6: re backups -- if one absolutely must make a hand-cranked copy of a node, the correct method is 'dumpblock' (on donor end) and 'eatblock' (on recipient) , dun require taking down either.
asciilifeform: for the smaller boxen (various arm, 'pcengine', similar) you can actually put a 12v lead-acid in line with the dc connector. and never suffer 'oops, mains flicker' db corrupt.
a111: Logged on 2019-07-16 22:25 mod6: Couple of other things that I wanted to mention quick, girlattorney: Just be sure to make frequent backups of your entire blockchain. Be aware also that TRB does not handle power-outtages very nicely as BDB can get corrupted; UPS and the like can help to mitigate this.