[HN Gopher] The Pentium as a Navajo Weaving
___________________________________________________________________
The Pentium as a Navajo Weaving
Author : GavinAnderegg
Score : 444 points
Date : 2024-09-01 16:46 UTC (1 days ago)
(HTM) web link (www.righto.com)
(TXT) w3m dump (www.righto.com)
| kens wrote:
| Author here for your obscure Pentium questions...
| hindsightbias wrote:
| Well done.
| kens wrote:
| Thanks! Curiously, the article doesn't show up on HN. Maybe
| my domain got banned or something?
| Joe_Cool wrote:
| I just saw it on the frontpage and very much enjoyed it.
| throwaway290 wrote:
| Check if you accidentally clicked "hide"... (hidden posts
| in your profile)
| teruakohatu wrote:
| That was a very interesting and well researched article. Thank
| you.
| omoikane wrote:
| I laughed at the bit where the gallery picked the wrong side of
| the rug to display, but this got me thinking -- presumably all
| the die photos we see are from the etched side of the wafer,
| what does it look like on the other side? Is it just all flat?
| kens wrote:
| The other side is plain gray silicon; I've looked at the back
| side of dies by mistake many times. (Intel is starting to do
| power through the back side, which would make the back more
| interesting.)
| Tuna-Fish wrote:
| Note that it's entirely transparent to some wavelengths of
| light, which is why a lot of die shots are taken from the
| backside, as it allows doing them non-destructively. You
| just cannot use visible light.
| kens wrote:
| Unfortunately it's not that easy. I was working with John
| McMaster to get backside die shots of the 386 using
| infrared but it didn't work out.
| OldGuyInTheClub wrote:
| "Such hearts, such brains, would be unable to comprehend that
| one's attachment to a masterpiece may be utterly
| overwhelming, especially when it is the underside of the
| weave that entrances the beholder and only begetter, whose
| own past intercoils there with the fate of the innocent
| author." --- Nabokov, "Pale Fire"
| n8henrie wrote:
| Just moved back to ABQ from Shiprock last month, after 8 years.
| Not every day you see Shiprock featured in an HN post!
| benatkin wrote:
| Not every day I see a place on the Colorado Plateau in an HN
| post! I live in the SF Bay Area but am from Flagstaff!
| chiph wrote:
| I had a dual Pentium II board (made by Tyan, with an Adaptec
| Raid Option slot) but I don't recall ever seeing a dual P54C
| board. Did a mainstream motherboard maker ever release one, to
| your knowledge?
| h2odragon wrote:
| there was tyan, supermicro, and MSI? at least
|
| i thin my board with the dual pentium 60mhz chips was msi.
| xanathar wrote:
| Pentium 60MHz was P5 though, not P54C ... if I'm not
| mistaken. IIRC the slowest P54C was the 75MHz one as they
| all had 1.5x multiplier or greater? Again, I wouldn't rule
| out my memory being wrong.
| TheAmazingRace wrote:
| That sounds correct to me. Also, the P5 was a 5V design!
| Intel went to 3.3V soon after the very first Pentium
| CPUs.
| gattilorenz wrote:
| You can find multiple of them here, with pictures too: https:
| //theretroweb.com/motherboards/?page=1&itemsPerPage=24...
| toast0 wrote:
| Not a pentium question, but..
|
| > Due to the end of the Cold War, Hughes acquired General
| Dynamics' missile business in 1991 before being acquired by
| General Motors in 1985 and sold to Raytheon in 1997.
|
| I'm guessing that's supposed to be 91 -> 95 -> 97?
| kens wrote:
| Trying to sort out aerospace companies is always a mess. I
| got the dates mostly right but reversed the timeline. GM
| acquired Hughes Aircraft in 1985, forming the Hughes
| Electronics unit which consisted of Hughes Aircraft and Delco
| Electronics. Hughes Aircraft acquired General Dynamics'
| missile business in 1992 to form the Hughes Missile System
| Company. In 1994, Hughes Electronics combined three
| operations into the Hughes Aerospace and Electronics Company
| (still a unit of GM). In 1997, Raytheon acquired the military
| business of Hughes Electronics Corporation from General
| Motors for $9.5 billion. At least I think that's what
| happened :-)
| MandieD wrote:
| As a hobby hand-spinner and occasional weaver who's been in IT
| my entire working life, and interested in learning more about
| what my home country has done to the nations that were already
| there, this article is excellent in all regards.
| __alexander wrote:
| Excellent article. I know Shiprock but I didn't know it's
| history with manufacturing processors. Thank you for writing
| it.
| wolf550e wrote:
| I get the feeling Intel needed to "bribe" the Navajo by
| ordering this work. Why did they need to do that? Was it part
| of an agreement with tribe leadership?
| kens wrote:
| I think it's more likely that someone at Intel thought, "You
| know what would be cool? A rug that looked like the Pentium
| die." Note that unlike Fairchild, Intel's plant is not on
| Navajo land so there are no tribal agreements.
| klyrs wrote:
| It's actually quite common for large companies to commission
| artworks inspired by their products; American companies
| frequently reach out to tribal artists for the uniqueness of
| their artistic style. Your comment is pretty depressing --
| despite the clear mastery demonstrated in this art, you're
| insinuating a nefarious plot involving a long-scapegoated
| population.
| h2odragon wrote:
| Fairly rare to be able to connect sheep with semiconductors.
|
| I wonder if anyone is working on ovine AI opportunities yet.
| eszed wrote:
| No. Singularity risk has them cowed.
| OogieM wrote:
| As a programmer and Shepherdess I can tell you that sheep have
| enough intelligence as is. Contrary to popular belief they are
| not stupid but certainly have a very different world view.
| NelsonMinar wrote:
| That story about Fairchild manufacturing at Shiprock is
| fascinating and heartbreaking. Glad to see it so thoughtfully
| researched and presented.
| tarellel wrote:
| And as sad it is, they cycle as has repeated itself and its
| still a sad situation around Shiprock. I'm someone who lives
| within the general area and poverty is still saturating the
| Navajo community. To make things worse the oilfield in the area
| has shifted to Texas. And within the last year the Four Corners
| Power Plant (PNM) and Navajo Mine (BHP) have shut down. These
| have been 2 extremely large employers of the area for the
| Navajo people since the 70's. Lots of businesses in neighboring
| towns like Farmington and Gallup have shuddered. And a large
| amount of people (who could afford to) have moved to places
| like Phoenix and Denver so they don't get stuck being part of
| the situation.
| BoingBoomTschak wrote:
| You know what they've said for millennia: "Vae victis"!
| niccl wrote:
| A fun full-circle. The Jacquard loom was influential for input
| for early computers, and now the computer is influencing the
| output of looms
| xiande04 wrote:
| I thought the same thing reading this! I'm currently reading
| Prisoner's Dilemma by William Poundstone, which details the
| life of John von Neumann. According to his brother Nicholas,
| von Neumann likely got the idea to use punch cards for
| computers from the Jacquard loom. One of his father's (Maxwell
| von Neumann) clients at the turn of the century was a Jacquard
| loom manufacturer, and Max would evidently discuss his work
| with his kids everyday over supper.
| kens wrote:
| That doesn't make sense, since computers were using punch
| cards before von Neumann got involved.
| xiande04 wrote:
| Ah, you're right. Babbage's Analytic Engine was designed to
| use punch cards, and I totally forgot about the Hollerith
| Machine that I read about just last year. I wonder what
| other false information Poundstone's book has. Thanks for
| the heads up!
| aoanla wrote:
| Not only did Babbage's Analytical Engine use punched
| cards, it was specifically inspired by the Jacquard Loom,
| as Babbage was a massive fan of it (and owned a portrait
| made on one!)
| swayvil wrote:
| Speaking as an artist and a programmer, has it ever struck you
| how utterly _low_ it is to sit in a room making things? You are
| interacting with little speck of dead stuff. Staring, unmoving,
| practically dead yourself. Playing with a little dream. There 's
| something deeply wrong with that. Spiritually wrong even.
| Sometimes I reflect and feel shame at my wastefulness.
| groestl wrote:
| Dead stuff is stuff without structure. I tend to think about my
| interactions with the universe as extending the structure that
| is me to dead stuff. Stuff is lifted to structuredness, and
| becomes part of us as a whole. I don't feel that's low at all.
| turtledragonfly wrote:
| Hello, fellow artist+programmer (:
|
| What do you think of the quote: "Whatever you do will be
| utterly insignificant, but it is very important that you do it"
| ?
|
| Personally, I think playing with little dreams is beautiful,
| but I do hope you find something that satisfies your soul.
|
| In my experience, artistry and a bit of depression often go
| hand-in-hand. If you spend a lot of time paying attention to
| things, you're bound to notice the Abyss.
| ahazred8ta wrote:
| "Gaze not into the abyss, lest you become recognized as an
| abyss domain expert, and they expect you keep gazing into the
| damn thing." --@nickm_tor@mastodon.lol
| robocat wrote:
| Soeaking as an engineer, has it ever struck you how utterly low
| it is to sit in a room making art? Sculptural materials and
| paint is dead. Words are but faint echoes of human life.
|
| Your comments are more a reflection upon your own worldview.
|
| To improve things you merely need to change your own
| perceptions.
| underbooter wrote:
| That's a remarkable question.
|
| Have you, reader of this comment, ever created something that
| will subsequently never interact with a living being?
|
| I don't think I have... except maybe when I've blinked Morse
| into the sky with a flashlight or laser.
|
| There's always someone or something on the other end. I try to
| remember that.
| actionfromafar wrote:
| The classic excercise is: write a poem, tear the paper to
| pieces and throw it away.
| Cthulhu_ wrote:
| Sounds like you're going through a bit of a nihilism phase, the
| futility of it all. It's cool.
| cubancigar11 wrote:
| Low and high are but you imposing your value system onto
| something that exists irrespective of it. "Low" could very well
| be the activity of breathing if one argues enough to show it.
| In reality actions don't and shouldn't exist to serve a higher
| purpose. Attachment with result is the cause of suffering.
| Attach yourself to the action.
| renewiltord wrote:
| Are there automatic looms the way there are CNC machines so that
| you can get a custom design on?
| Stratoscope wrote:
| Since 1804!
|
| https://en.wikipedia.org/wiki/Jacquard_machine
| renewiltord wrote:
| So _that's_ what a Jacquard loom is. I've seen it mentioned
| often. Thank you.
| kragen wrote:
| dobbies automated looms before that
| OogieM wrote:
| dobby looms are an automated version of draw looms. Some
| use a draw boy, small child in the loom itself to draw up
| the warps as required for the patter. Later shortened to
| dobby.
| ahazred8ta wrote:
| There are some old Jacquard card stacks which are literally
| the oldest machine readable datasets in existence, a
| century before Hollerith.
| cmiller1 wrote:
| The oldest datasets produced by humans intended to be
| read by machines. With our advances in digital image
| processing one could argue a cross section of a tree
| trunk is a "machine readable dataset"
| tithe wrote:
| That's some serious...multithreading!
|
| (I applaud others for resisting such cringe, but I just couldn't
| help myself.)
| koolba wrote:
| The entirety would have complete faster if they only used green
| threads.
| lelanthran wrote:
| Maybe if they switched all code to Java, they can run it on a
| Loom?
| DaoVeles wrote:
| Don't hold back, that level of puns is what I live for.
| layer8 wrote:
| Just wait for the Pentium 4 with hyper-threading.
| leed25d wrote:
| I always look askance at a blanket statement.
| drivers99 wrote:
| "The Analytical Engine weaves algebraic patterns, just as the
| Jacquard-loom weaves flowers and leaves."
|
| - Ada Lovelace, 1843
| Flop7331 wrote:
| Just couldn't weave it awone, could you?
| ein0p wrote:
| I've long considered "hard tech" (semiconductors, high end
| manufacturing, aerospace, etc) to be the embodiment of our
| culture, because it's not something you can just start doing, it
| takes generations of improvement to make anything worthwhile.
| DaoVeles wrote:
| That is how I generally see it. The technological suite needed
| to get to that stage is astounding.
|
| It is a great indicator of progress and decline, like how the
| quality of crockery dropped in the roman empire showed how
| their economics hampered their skills.
|
| Maybe computing is one of the big hurdles that shows the
| overall capabilities of economies and on a much more broad
| scale, species. Like wrap drives in Star Trek.
| Taniwha wrote:
| I'm building a stained glass window for our front entrance, as a
| sometimes chip designer it's going to be a half adder (so much
| much smaller than a pentium) - the big challenge is showing all
| the layers (si, implantation, poly, metal, vias etc) I'm only
| doing 1 level of metal. The result is going to be more of a thin
| layered sculpture than a traditional stained glass window
| bankcust08385 wrote:
| Neat. 2 RTL gates -> 16 CMOS transistors (10 for XOR and 6 for
| OR)
| Taniwha wrote:
| Actually I misspoke, it's a full adder and very similar to
| the one on page 5 here
|
| https://www.cerc.utexas.edu/~jaa/vlsi/lectures/8-1.pdf
| anthonix1 wrote:
| Don't bother with the rectilinear pakeha layouts, do your half
| adders in curvilinear patterns, Koru style
| Taniwha wrote:
| I am pakeha and tangata triti .... now I have an idea for the
| next window (I was thinking about a neuron ...)
| anthonix1 wrote:
| AHh gotcha.
|
| Well yeah I reckon you render a full custom 4004 w/ koru
| patterned transistors into about 4m x 4m stained glass
| panel. Would look good as the foyer panel for the CS dept
| at the University of Waikato
| Taniwha wrote:
| Probably best to tape it out to https://tinytapeout.com/
| first
| anthonix1 wrote:
| Do they support curvilinear cells?
| Taniwha wrote:
| The simple openlane flow just uses standard cells, but
| you have the ability to just tape out polygons so
| curvyish is possible
| Taniwha wrote:
| (I already have a 4-bit CPU there)
| kragen wrote:
| this sounds awesome! do you have drawings?
| LennyHenrysNuts wrote:
| Absolutely fascinating article. I wasn't aware of any of the
| Navajo history of Farichild Semiconductors.
|
| Thanks for sharing.
| 0xbadcafebee wrote:
| Wow. They have _got_ to turn this into a 99% Invisible podcast
| episode. (if you stopped at the weaving - which is incredible in
| its own right - you missed the much more amazing story)
| tug2024 wrote:
| Do sha nada da da !
| mlsu wrote:
| Wow, what a gem of a post. I knew about the Navajo codebreakers
| but didn't ever hear about the Shiprock connection.
|
| I also am admiring how deftly the author was able to weave in
| (heh) little bits of low-level computer knowhow. This is an
| article I can send to my non-tech friends!
| datavirtue wrote:
| This how I have always thought of the Giza plateau and other
| temple sites. They look like different digital components or
| circuits attached to each other. It was easy to convince myself
| that they were indeed designing some type of deep logic within
| the structures.
| squarefoot wrote:
| This triggered a memory about the Native American code talkers
| employed by the US military during WWII to encode communications
| using their own languages which were unknown to the Axis forces.
| First thought: "ha! Now they're doing firmware?":)
|
| https://en.wikipedia.org/wiki/Code_talker
| kens wrote:
| Interestingly, Peter MacDonald, the Navajo tribal leader at the
| time, had been a code talker during World War II.
| 1bent wrote:
| This sounds like the backstory for Alan Dean Foster's "Cyber
| Way". I enjoyed the novel, didn't realize how well rooted it was
| in actual history.
| elzbardico wrote:
| I wonder how much debate and how many PhD dissertations in
| archeology will result when this rug in found in a excavation
| some 2000 years from now.
___________________________________________________________________
(page generated 2024-09-02 23:02 UTC)