March 2012 Update

Loper's I2C controller is working.  The SPD ROM on the DDR2 RAM stick attached to the Xilinx ML-501 board is read correctly.  The video controller is working (though not feature-complete.)  The DDR2 SDRAM controller is still under testing, as is the cache SRAM controller.  The gigabit Ethernet controller is not yet complete.

I should probably say a few words about the (presently experimental) Loper machine architecture.  The paradigm of choice is dataflow computation.  The goal is to create something which can one day be seamlessly converted to a fully-asynchronous design, implemented using Muller C-gates. The latter is extremely difficult (though not entirely impossible) on available FPGAs, so the "adult" form of the design would have to be fabbed in actual silicon (fat chance of this, at least in my lifetime.)

Loper has no "cores", no instruction set (in the usual sense,) no interrupts or DMA.  These concepts are, for the most part, artifacts of the Von Neumann paradigm.  That is to say, epicycles.  And the destiny of epicycles is -- to go away: "Beauty is the first test: there is no permanent place in this world for ugly mathematics." (G. H. Hardy)

Eminent people and well-funded organizations would like to resurrect Moore's Law. The most straightforward way to do this is to move to a fully-asynchronous dataflow architecture, where you can buy computational capacity "by the pound," so long as your bottlenecks consist of something at least theoretically parallelizable. But they will not succeed, because they are unwilling to let go of the epicycles.  Their investment in existing idiocies - not only monetary, but psychological - is far too great.

Also note that the motives of the would-be Moore's Law resurrectors are far from pure.  Scarcely anyone wishes to sell you a genuinely-fast computer.  What is really happening is that shitware peddlers fear that their days of selling ever-more bloated bloatware written by progressively-deskilled slave laborers might be coming to an end.  The introduction of practical dataflow machines could save these wretches, but only in the sense of "destroying the village in order to save it."

This entry was written by Stanislav , posted on Tuesday March 13 2012 , filed under Cold Air, Computation, Hardware, Hot Air, LoperOS, Progress . Bookmark the permalink . Post a comment below or leave a trackback: Trackback URL.

12 Responses to “March 2012 Update”

  • Aaron says:

    Well put. I look forward to further progress on Loper OS. Exciting times, these!

  • Chris Smith says:

    Please keep up the good work.

    A small rant regarding shitware peddlers: The only way I see these going away is if someone goes and smashes the system they reside in. The "system" (I will not refer to it as capitalism as it's too broken even for those ideals) supports products of epic complexity simply to provide labour opportunities and to keep everyone marching along. People chase growth but never simplicity or elegance.

    • Stanislav says:

      Dear Chris Smith,

      The good news is that the "system" is indeed collapsing. The bad news is that it is, well, collapsing.

      I am acutely aware that my work is likely to be buried in the sand. But I like to think that it might be remembered by some, and perhaps inspire future inventors - like the Aeolipile.

      Barring a complete reversion to barbarism, an energy-poor future might still contain computers of some kind. But the Computer may very well go back to being an installation - a place, like a large church organ or a particle accelerator. Like the calculating machines of the 1950s. So let's leave our descendants with plans for a computer worth their while to build: one which is a true mind-amplifier, rather than a comically-inefficient video teletype.

      Yours,
      -Stanislav

  • Michael Olney says:

    At a time where the cancer of complexity is reaching the irreversible stages it's good to see something other than excuses not to start taking action to purge the disease. I live in the hope of seeing simpler, smaller systems where the most or all of each stack is backed by personal investment and responsibility. This is totally alien in a culture where the sole value is producing more software, even if most of it is either spinning its wheels wastefully in the background or serving "needs" manufactured by the industry itself.

  • MG says:

    Have you come across GreenArrays? Seems to be a dataflow-like architecture that might be easier to work with than FPGAs. Perhaps underpowered for a modern operating system however, and from reading a few blogs it seems like implementation at this point is very chip-specific.

    http://www.greenarraychips.com/

  • Milton says:

    Have you looked at Achronix FPGAs? They are asynchronous and it appears that they allow the (easier)implementation of Muller gates. In fact, it seems they use them internally but, I'm not sure since not many details are available about their architecture.

    • Stanislav says:

      Dear Milton,

      I have looked at it. Achronix is "not for people." I have never seen a development board of theirs for sale anywhere. One might be able to get it from one of the listed "industry partners," but at what cost? In my experience, any manufacturer whose site contains no prices is a "if you have to ask, you can't afford it" sort of business. Also: it isn't clear what kind of software tools (free? Linux-based? or shit?) they offer. Theirs appears to be one of those products that doesn't "really exist" for the consumer.

      Yours,
      -Stanislav

      • Milton says:

        The tools run on linux[1]. It is indeed a problem. They provide very little information. It is unfortunate because the FPGA technology seems very interesting and I heard that the price per board is approximately the same as Xilinx but, maybe they only sell in bulk.

        [1] - http://www.achronix.com/docs/ACE_Product_Brief_PB002.pdf

        • Stanislav says:

          Dear Milton,

          I get my Xilinx boards on Ebay. (Always for a fraction of their original price.) Should I ever see an Achronix board there, I might very well buy it. Assuming that I could find even elementary specs for it.

          Achronix is, at present, "not for people." This means that it sells exclusively to corporate entities willing and able to sign legal agreements and form industrial partnerships.

          Yours,
          -Stanislav

  • geo says:

    Hi Stanislav,

    I'm not sure maybe this can help? Or maybe you already read this one:

    http://www.ics.forth.gr/carv/async/demo/

  • akater says:

    Your metaphors are almost always charming. “Epicycle”, however, is an undeservedly pejorative term. Ptolemy described the motion with harmonic analysis, which is even more modern than Newton's tools. For now, it serves certain purposes worse and others better but in no way it's gone. It's been developed and improved since XIX century (Fourier).

    After writing this I noticed that your Wikipedia link contains all this at the moment but given the ever-floating nature of Wikipedia, I'd still post.

    Given Hardy's quote, “ugliness” is a point of view. One could argue that Hardy's advances in analytical number theory belong to the uglier pieces of mathematics.

    What you are doing is admirable, of course. Good luck and good health!

Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong> <pre lang="" line="" escaped="" highlight="">


MANDATORY: Please prove that you are human:

82 xor 39 = ?

What is the serial baud rate of the FG device ?


Answer the riddle correctly before clicking "Submit", or comment will NOT appear! Not in moderation queue, NOWHERE!