1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

[Q]: Is SUN going to become x86'ed ??

Discussion in 'Sun Hardware' started by Kosta Xonis, Nov 27, 2008.

  1. Kosta Xonis

    krw Guest

    I know it's been done. Yes, that one in particular was done by
    design. That feature helped the Vietcong out greatly.
    krw, Jan 17, 2009
    1. Advertisements

  2. And possibly not even then.
    Look in the tanks (or use a dipstick) before you start, then use
    the consumption figures that you already know. Mind you, this
    still won't save you if you switch to one tank and forget about
    it - but a sputtering engine makes a wonderful reminder of that.
    Charlie Gibbs, Jan 18, 2009
    1. Advertisements

  3. True enough. Just ask Microsoft about file formats.
    Charlie Gibbs, Jan 18, 2009
  4. Kosta Xonis

    jmfbahciv Guest

    Not knowing about how to do dimensional anaylsis is the lesson in this
    thread. A lot of computer types have no idea that meters and feet
    don't mix.

    jmfbahciv, Jan 18, 2009
  5. Kosta Xonis

    jmfbahciv Guest

    Heaven forbid that he bother to learn anything which be
    very useful in his work.

    jmfbahciv, Jan 18, 2009
  6. Kosta Xonis

    jmfbahciv Guest

    Even when the "new" these days looks cheap and won't hold together
    for more than a couple of years? My Southboro house had shelves
    and cupboards built using old wooden packing crates. Those boards
    never sagged, unlike today's cabinetry. The only problem was
    the guy didn't know about sandpaper before he built them.

    jmfbahciv, Jan 18, 2009
  7. Kosta Xonis

    jmfbahciv Guest

    And looking at the wood gives enormous pleasure, especially when
    you've done all the work. I liked working with wood when I made
    my bookshelving.

    jmfbahciv, Jan 18, 2009
  8. Kosta Xonis

    Joe Morris Guest

    And the totalizer is at best only a measure of how much fuel has been fed to
    the engine. Unless you know reliably how much fuel was in the tanks to
    start with, you're still unable to say how much you've got left. The only
    really reliable states are either when there's no fuel in any tank (not
    recommended at takeoff time - although I've seen a C-172 run out of fuel on
    takeoff roll) and when all tanks simultaneously have fuel up to the "full"
    tab (which still could be a bit off if the bladders are wrinkled).

    Of course, "how much fuel was in the tanks to start with" might not be
    reliable if you've got a leak or a loose fuel cap that's siphoning fuel out
    of the tank in flight.

    The issue of available fuel *is* legitimately on-topic for afc. How many
    shops (or homes for that matter) have been oh-so-proud of their backup
    generator that's expected to carry them through an outage that's beyond the
    capability of their battery UPS, only to discover that they don't have as
    much fuel as they thought?

    Joe Morris
    Joe Morris, Jan 18, 2009
  9. Kosta Xonis

    jmfbahciv Guest

    At this point, I'd be happy with people knowing that single point
    failures exist. The last month's posts have been getting worrisome.

    jmfbahciv, Jan 18, 2009
  10. Pah! Whippersnapper! In 1976 I was hacking the RSTS terminal driver at
    a place called EPT (Electric Power Transmission) (so they could drive a
    Calcomp 960 roller bed plotter) They were doing FE analysis to design
    towers, driving CNC machines remotely to drill and cut angle girders,
    and (on-topic) trying to solve the hamper packing problem to make the
    most efficient cuts in their stocks of steel angle. All on a pretty
    modest PDP11-45.

    There was an air of getting it done and making money with it before you
    found out it was impossible.

    The plotter's drawings were used on site, the first time humans saw
    what the tower was going to look like.

    (as an aside, this was my first contract after leaving DEC, it was just
    before Christmas, and I supplied my friends and family with wrapping
    paper illustrated by variants of a*sin(bx +c) +d*sin(ey+f)
    Well, there *was* an interesting hardware bug in the RTS/CTS interrupts
    that needed extensive testing of the workaround <grin>)
    Elliott Roper, Jan 18, 2009
  11. Kosta Xonis

    Mensanator Guest

    Mensanator, Jan 18, 2009
  12. Even when the "new" these days looks cheap and won't hold together
    for more than a couple of years? My Southboro house had shelves
    and cupboards built using old wooden packing crates. Those boards
    never sagged, unlike today's cabinetry. The only problem was
    the guy didn't know about sandpaper before he built them.


    Most construction is cheap, we romanticize the past because the offal
    stuff has been destroyed and only the best remains.
    Walter Bushell, Jan 18, 2009
  13. Kosta Xonis

    kkt Guest

    Design as good as that (for the weapons users) doesn't happen by accident.

    -- Patrick
    kkt, Jan 18, 2009
  14. Kosta Xonis

    Paul Guest

    Agreed, tho' I used up my personal quota long before I got the 'old'
    Paul, Jan 18, 2009
  15. When one looks at the reason for the incompatability one has to wonder
    design or accident.

    Bill Gunshannon, Jan 18, 2009
  16. this is reference to somebody modified CP67 at MIT USL ... to drive what
    I believe was an "ascii" plotter over at harvard. I had done the changes
    to add TTY/ascii terminal support to cp67 as undergraduate in the 60s.
    In the changes I played some games with one byte arithmatic for buffer
    calculations. My memory was that the (MIT/USL) modifications increased
    the max tty length from 80 to something like 1200 (for the plotter
    device) ... but didn't catch the hack with one byte operations.

    now as part of the original work (adding tty/ascii terminal support to
    the 2741 & 1052 support) ... i had tried to make the 2702 (mainframe)
    terminal controller that it couldn't quite do. this somewhat was behind
    the motivation for the univ. to start a clone controller project
    .... reverse engineering the mainframe channel interface to build a
    channel board for an (initially) Interdata/3 ... and programming the
    Interdata/3 to emulate 2702 (plus what I wanted it to do). Some past
    posts ... including reference to some article blaming four of us for the
    mainframe clone controller project

    note that clone/pcm controllers were then blamed for the motivation
    behind the future system project ... some past posts

    recent specific quote ... in thread about clones:
    http://www.garlic.com/~lynn/2008s.html#17 IBM PC competitors

    and then the distraction of the future system (which was going to
    replace all 360/370) suspended activity on new 360/370 activity
    .... which contributed to allowing 370 processor cloans to gain a
    foothold ... another old reference:
    http://www.garlic.com/~lynn/2001f.html#33 IBM's "VM for the PC" c.1984??

    and as mentioned in the "IBM PC competitors" post ... after future
    system project was killed, there was a mad rush to get new software and
    hardware items back into the 370 product pipeline.
    Anne & Lynn Wheeler, Jan 18, 2009
  17. The person I asked about them lived here (US) for decades before moving
    "home" again and said they're the same as in the US. He was an old timer
    that makes his own tools, including kitchen knives and axes. Of course,
    the axes are made of two types of metal welded together and the kitchen
    knives are zone tempered.

    Aside from being inch, do BSP part mate with NPT ones at all?
    Cydrome Leader, Jan 18, 2009
  18. I think that this is like the Traveling Salesman problem: you can
    get fairly good approximations, but you can *not* ever be sure that
    you can get the *optimum* solution. There is an algorithm used for
    this sort of thing: simulated annealing.


    It is also represented in _Numerical Recipes in <some language>_.

    Simulated annealing can give you a *good* answer in a reasonable time.
    Like General George S. Patton said: "I'd rather have a good plan
    today, than a perfect plan next week."
    Charles Richmond, Jan 19, 2009
  19. Kosta Xonis

    jmfbahciv Guest

    And remember to ask your officemate about the dimensions used in
    the code.

    jmfbahciv, Jan 19, 2009
  20. Kosta Xonis

    Joe Pfeiffer Guest

    Well.... it isn't that you can't get an optimal result for TSP, it's
    computationally infeasible to do it -- unlike the halting problem,
    where it really is impossible.

    I'm not sure whether this problem, as stated, becomes an instance of
    binary knapsack (which is infeasible like TSP... but the number of
    alternatives is small enough it might well be solved through brute
    force) or general knapsack (which can be solved with dynamic
    Joe Pfeiffer, Jan 19, 2009
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.