Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.forth > #13439

Re: The Map Is Not The Territory

Show all headers | View raw

On Jul 1, 10:41 pm, rickman <gnu...@gmail.com> wrote:
> On Jul 1, 2:40 pm, Bernd Paysan <bernd.pay...@gmx.de> wrote:
>
>
>
>
>
> > rickman wrote:
> > > I was
> > > reading about his effort to use the GA144 as a video controller and
> > > when I read the part about the video being generated on the fly rather
> > > than read from memory, I realized that he is indeed following a path
> > > less traveled.  I have no idea how he can do anything other than
> > > produce a text display (like they described in "TV Typewriter
> > > Cookbook" some 30 years ago) and I expect he doesn't either.  I get
> > > the impression he is playing, not working.
>
> > Our student in the local Forth group has created his own GA144 board,
> > and did some video output a few months ago.  Well, he's a student, and
> > of course he's playing, and we encourage him to do so - programming
> > Forth with the restrictions of the GA144 is fun.  He got so far to have
> > a synchronized pixel clock (which works even on hot days, something you
> > can't say about Chuck's video driver), and he can display hex numbers -
> > the font "rendering" table takes a complete core, and it only works by
> > restricting the fonts to 6x3 pixels (which fit into the 18 bits).
>
> > The path here is well traveled, it is how games like Pong worked - they
> > all generated the image on the fly, because memory was expensive back
> > then.  Games like Asteroids used even vector graphics and special
> > monitors.
>
> > Now, we don't have 1970 anymore, we have 2012.  Memory is cheap, dirt
> > cheap.  We don't generate images on the fly, because we have frame
> > buffers.  Well, we actually do generate new images frame by frame, but
> > we use double or even tripple buffering to make it smooth (one image is
> > currently displayed, one image is ready-to-display, and one image is
> > rendered to).  We do that because memory is cheap.
>
> > The only place where memory isn't cheap is power consumption.  The main
> > reason why smartphones can't and won't have standby times in weeks as
> > dumb phones, is DRAM.  DRAM requires constant refreshing, and by doing
> > so drains a lot more current than is leaking through the capacitors.
> > Maybe some memresistor device will change that, but not yet.
>
> Are you sure the DRAM is the cause of the short battery life of the
> smart phones?  I was just looking at a data sheet for SDRAM and it
> lists a 1 mA current for the refresh function and 300 uA standby
> current.  I think most of the power goes to the 1 Watt CPU that uses
> the DRAM, not so much the DRAM itself.
>
> > In so far, as the map is not the territory, I think Chuck used an
> > ancient path and wandered off what makes sense today.
>
> I think the F18 makes some sense, but it needs a LOT of work and in
> particular it has to allow some deviations from the "simplicity"
> syndrome to make it practical.  If a 700 MIPS CPU can't implement a
> high speed phy in software, then maybe a custom interface is needed.
> If the 64 word instruction/data memory is not large enough to
> implement a protocol, then maybe a few nodes need to have more
> memory.  At some point concessions need to be made to the app, not the
> other way around.
>
> I haven't given up just yet.  We'll see if I can tame this beast!
>
> Rick- Hide quoted text -
>
> - Show quoted text -

I'm with you on that. I'd like to see the cores have more memory.

As a compromise, how about the inner cores having more memory than the
outer-perimeter cores? Then, by convention, one would design such that
the outer cores simply send/receive data from the outside world, and
pipe it to inner cores that have more memory to do the serious
crunching.

I don't want to have to split my MODBUS or IP implementation across 10
different cores just to get it to run - that is adding complexity at
the software layer in order to compensate for the simplicity at the
hardware level. Adding software complexity is not Forth. Chuck himself
being the biggest proponent of that paradigm.

Back to comp.lang.forth | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

The Map Is Not The Territory Jason Damisch <jasondamisch@yahoo.com> - 2012-06-27 17:44 -0700
  Re: The Map Is Not The Territory dambere@web.de - 2012-06-28 02:32 -0700
    Re: The Map Is Not The Territory Mark Wills <markrobertwills@yahoo.co.uk> - 2012-06-28 02:47 -0700
      Re: The Map Is Not The Territory John Passaniti <john.passaniti@gmail.com> - 2012-06-28 13:19 -0700
        Re: The Map Is Not The Territory dambere@web.de - 2012-06-29 14:29 -0700
  Re: The Map Is Not The Territory quiet_lad <gavcomedy@gmail.com> - 2012-06-29 14:44 -0700
    Re: The Map Is Not The Territory "Elizabeth D. Rather" <erather@forth.com> - 2012-06-29 12:17 -1000
    Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-06-29 16:15 -0700
  Re: The Map Is Not The Territory quiet_lad <gavcomedy@gmail.com> - 2012-06-29 14:43 -0700
    Re: The Map Is Not The Territory Jason Damisch <jasondamisch@yahoo.com> - 2012-06-30 10:46 -0700
      Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-06-30 19:54 -0700
        Re: The Map Is Not The Territory Howerd <howerdo@yahoo.co.uk> - 2012-07-01 09:14 -0700
          Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-01 14:12 -0700
        Re: The Map Is Not The Territory Bernd Paysan <bernd.paysan@gmx.de> - 2012-07-01 20:40 +0200
          Re: The Map Is Not The Territory Howerd <howerdo@yahoo.co.uk> - 2012-07-01 12:15 -0700
            Re: The Map Is Not The Territory Bernd Paysan <bernd.paysan@gmx.de> - 2012-07-01 23:52 +0200
          Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-01 14:41 -0700
            Re: The Map Is Not The Territory Bernd Paysan <bernd.paysan@gmx.de> - 2012-07-02 00:23 +0200
              Re: The Map Is Not The Territory Paul Rubin <no.email@nospam.invalid> - 2012-07-05 00:06 -0700
                Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-05 10:49 -0700
              Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-05 10:38 -0700
                Re: The Map Is Not The Territory Bernd Paysan <bernd.paysan@gmx.de> - 2012-07-06 02:22 +0200
                Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-05 18:57 -0700
                Re: The Map Is Not The Territory Mark Wills <markrobertwills@yahoo.co.uk> - 2012-07-06 02:13 -0700
                Re: The Map Is Not The Territory Bernd Paysan <bernd.paysan@gmx.de> - 2012-07-07 01:11 +0200
                Re: The Map Is Not The Territory Albert van der Horst <albert@spenarnc.xs4all.nl> - 2012-07-06 10:58 +0000
                Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-06 06:00 -0700
                Re: The Map Is Not The Territory Paul Rubin <no.email@nospam.invalid> - 2012-07-06 20:41 -0700
                Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-07 06:56 -0700
                Re: The Map Is Not The Territory Paul Rubin <no.email@nospam.invalid> - 2012-07-07 21:45 -0700
                Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-08 15:53 -0700
                Re: The Map Is Not The Territory Paul Rubin <no.email@nospam.invalid> - 2012-07-08 21:08 -0700
                Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-09 16:32 -0700
              Re: The Map Is Not The Territory Brad Eckert <hwfwguy@gmail.com> - 2012-07-05 13:22 -0700
            Re: The Map Is Not The Territory Mark Wills <markrobertwills@yahoo.co.uk> - 2012-07-02 01:26 -0700
              Re: The Map Is Not The Territory stephenXXX@mpeforth.com (Stephen Pelc) - 2012-07-02 10:15 +0000
            Re: The Map Is Not The Territory "Paul E. Bennett" <Paul_E.Bennett@topmail.co.uk> - 2012-07-02 19:28 +0100
  Re: The Map Is Not The Territory Albert van der Horst <albert@spenarnc.xs4all.nl> - 2012-07-01 11:35 +0000
    Re: The Map Is Not The Territory Howerd <howerdo@yahoo.co.uk> - 2012-07-01 09:31 -0700
      Re: The Map Is Not The Territory Paul Rubin <no.email@nospam.invalid> - 2012-07-01 14:03 -0700
        Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-01 14:57 -0700
    Re: The Map Is Not The Territory rickman <gnuarm@gmail.com> - 2012-07-01 14:50 -0700

csiph-web