log☇︎
79800+ entries in 0.015s
asciilifeform: ( if mircea_popescu meant a chip-that-checks-bounds. which existed at various points, e.g. symbolics lineup supported ada natively just as well as lisps . )
asciilifeform: what would that mean
asciilifeform: http://btcbase.org/log/2017-06-30#1677807 << see also thread... ☝︎
asciilifeform: ( and if i were producing $box by the million, i wouldn't deal with these derps and use their weird arch, but have my own asiced. )
asciilifeform: as in buy at digikey or other mainstream supply house .
asciilifeform: yes. but not on open market, is the point.
asciilifeform: but go and buy.
asciilifeform: unsurprise
asciilifeform: looks like golden toilet asicization exists (existed) ?
asciilifeform: ( that doesn't rely on closed xilinx 20GB turdchain to fill, doesn't contain a flash rom, doesn't double as a frying pan )
asciilifeform: what i was looking for, is actual physical off-the-shelf cpu.
asciilifeform: i have own, even.
asciilifeform: there's NO shortage of these.
asciilifeform: never existed as an actual physical chip.
asciilifeform: fpga only
asciilifeform: ?
asciilifeform: )
asciilifeform: ( other than by baking into fpga. which 1) introduces the 'which fpga, again' problem 2) if you have fpga, big enough for sparc, you can make up a SANE arch to go in it
asciilifeform: i mention sparc as a historic item, it is afaik no longer possible to source it.
asciilifeform: and yes it's ridiculous . but they forced themselves into it by having the window thing at all, is the idea.
asciilifeform: the thing being indexed, gets populated ( or not ) exactly like, e.g., ipv4, or the holes in your pci bus, etc
asciilifeform: the index into these is a register. having some immutable width.
asciilifeform: physical objects all contain immutable 'magic numbers'.
asciilifeform: the word is this-wide.
asciilifeform: mircea_popescu: you can't have infinitely elastic registers. there's N bits , physically.
asciilifeform not aficionado of sparc specifically. it's a complexityhog.
asciilifeform: mircea_popescu: it was a pretty ordinary thing in '80s
asciilifeform: ( largely because they historically were able to externalize the cost. )
asciilifeform: it apparently is a surprise to some people.
asciilifeform: one other gotcha: complexity has a cost.
asciilifeform: ( it can lower thermal output, so still makes sense sometimes to have empty. )
asciilifeform: ditto for a die made with particular process -- empty space doesn't get your money back
asciilifeform: correct. but the answers to these depend SEVERELY on the substrate. for instance, if your thing lives in a fpga, not filling it up doesn't make it cycle any faster. so you want to actually use the available gates, if they can be used productively
asciilifeform: ( or, on the extreme end of things, not having arithmetic other than add-one. as scheme chip did. )
asciilifeform: or a multer. etc
asciilifeform: or the cost of not having a barrel shifter.
asciilifeform: ( consider, elementarily, pipeline vs. none , for instance. )
asciilifeform: the basic tradeoff is, the fewer the moving parts in cpu, the more constrained is the horse by the switching elements' speed.
asciilifeform: there are several known schemes for this.
asciilifeform: ( as in, 'oisc' )
asciilifeform: upstack -- you can do a great deal with 1 instruction.
asciilifeform: and answer very very depends on what is required in terms of horse
asciilifeform: this is actually asciilifeform's primary interest re subj
asciilifeform: and the ones in cn also iirc.
asciilifeform: it's what the ru elbrus folx ripped off for elbrus.
asciilifeform tested
asciilifeform: at any rate the published opensparc fpga thing actually worx
asciilifeform: probably 9000 things
asciilifeform: lolwat
asciilifeform: major objective of the usg.oracle acquisition, was to nuke it.
asciilifeform: sparc fit all of the wish list conditions ( incl. being multi source ) when it was alive. today afaik dead.
asciilifeform: they still sell it, even.
asciilifeform: ibm
asciilifeform: which i'm quite confident leaves : the empty set.
asciilifeform: ( and no sole-source garbage period. )
asciilifeform: briefly revisiting the cpu thing, incidentally i forgot a 7) not ppc
asciilifeform: iirc they have the opposite
asciilifeform: i just dun care ~by whom~ documented.
asciilifeform: nm not different
asciilifeform: either documented or reversed.
asciilifeform: if anyone has any knowledge, even thirdhand, of such a thing existing -- plz to post.
asciilifeform: 5) not intel
asciilifeform: summary : asciilifeform is looking for a 1) 32 ( or greater ) bitness cpu 2) that is able to execute code WHOLLY from external bus ( a la 1980s ), incl from rom 3) not ARM 4) in production
asciilifeform: how else.
asciilifeform: ( see also http://btcbase.org/log/2017-01-26#1607160 observation ) ☝︎
asciilifeform: asciilifeform however habitually collects and referments the aborted foeti of items which nearly existed, to try an' grow item that exists. because no other materials available.
asciilifeform: leaving empty box.
asciilifeform: aha!!
asciilifeform: which in practice are needed to do anything particularly useful.
asciilifeform: for the debug extensions.
asciilifeform: can make board quickly. but nowhere does -- afaik! -- the full doc exist publicly.
asciilifeform: well yes, so asciilifeform picked up a classic devboard to investigate
asciilifeform: that was the contemplated item
asciilifeform: reporting today that result of this is still empty.
asciilifeform: for clarity : asciilifeform ain't inluvv with superH. but he ~does~ search for a cpu where 1) respectable (say, even trb-capable) horse 2) no internal flash rom 3) not ARM
asciilifeform: ( other than a pdf where 'flip the dip switches like-so, click in winblowz like-so' )
asciilifeform: the shitboard ain't.
asciilifeform: the ~arch~ is documented.
asciilifeform: it exists in everybody's 1990s game console, stereo, vcr, etc. and now also in that fpga item.
asciilifeform: ( dark ages proprietary pre-jtag debugger, it seems, implemented in on-board fpga )
asciilifeform: also typical.
asciilifeform: nowhere.
asciilifeform: anyway this isn't half the grief : if it were just a matter of flashing, i could flash via crocodiles, applied externally. but you need the debugger to singlestep.
asciilifeform: pretty common idiocy, sadly
asciilifeform: apparently they flashed it in from winblowz.
asciilifeform: mircea_popescu: pretty easily.
asciilifeform: this doesn't change the fact that the only other non-internalflashed serious-horsepower chips are ARMs.
asciilifeform: i can easily see why renesas is ~dead ( chip in production, but largely for existing designs )
asciilifeform: in very very other lulz, asciilifeform actually got hold of a superH ( http://btcbase.org/log/2017-07-31#1691143 see also ) dev board, sh7216. mega-disappoint : ZERO linux support for the on-board debugger. which one actually NEEDS to do anything whatsoever with the thing. ☝︎
asciilifeform: it was
asciilifeform: it was in the mass of us tv asciilifeform saw in '90s when he still had a tv, to suck in the language
asciilifeform nostalgiates
asciilifeform: sliders!
asciilifeform: so bad , counter underflows, and it's... good
asciilifeform: ^
asciilifeform: noshit.jpg aha
asciilifeform: who said pinkertons went out of style.
asciilifeform: ^ achtung pete_dushenski et al
asciilifeform: '...Lackman's belongings still haven't been returned. Nor can he access the TVAddons website or its social media accounts, which were also seized as part of the original order. That's because Bell, Rogers and Videotron have appealed the court decision and a Federal Court of Appeal judge has ruled that until the appeal can be heard, Lackman will get nothing back.'
asciilifeform: wo computer technicians, an independent counsel and a lawyer representing Bell, Rogers and Videotron.'