Loading ...
Sorry, an error occurred while loading the content.

Football Game?

Expand Messages
  • Brooke Clarke
    Hi: HP brought their automated 8410 Network Analyzer to Aertech in Sunnyvale, CA for a demo. http://www.prc68.com/I/Aertech.shtml They had a small van custom
    Message 1 of 37 , Feb 12, 2012
    View Source
    • 0 Attachment
      Hi:

      HP brought their automated 8410 Network Analyzer to Aertech in Sunnyvale, CA for a demo.
      http://www.prc68.com/I/Aertech.shtml
      They had a small van custom fitter with a tilting rack. The 19" rack panel could be rolled up to the van, attached to
      the tilt mechanism then rotated to horizontal and slid inside.
      http://www.prc68.com/I/Rack.shtml

      While the higher up people went to lunch they started a football game. It allowed the player for Stanford and the
      player for Cal to call different plays, maybe something like:
      1 - short run
      2- short pass
      3 - long run
      4- long pass
      5 - kick
      and the odds of completion were based on the Cal-Stanford history.

      Does anyone know about it?

      Are there any 2114, 2115, 2116 or 2100 computers for sale?

      --
      Have Fun,

      Brooke Clarke
      http://www.PRC68.com
      http://www.end2partygovernment.com/Brooke4Congress.html
    • Quentin North
      Hi Mike The simh throttling issue is still a bit of an issue. Im now running on Simh v3.8.2rc2 which has some new throttling code that allows you to sample the
      Message 37 of 37 , Feb 15, 2012
      View Source
      • 0 Attachment
        Hi Mike

        The simh throttling issue is still a bit of an issue. Im now running on Simh v3.8.2rc2 which has some new throttling code that allows you to sample the performance of the processor and set an absolute throttle rather than a relative throttle. Unfortunately, because of the need to start up the two processors on Access, the main starting before the iop, simh still does not determine the correct host clock cycles per emulated cpu cycle until Access is fully up. Therefore, it is not possible to script the throttle before starting access, rather I still have to manually throttle it after Access is available. Having said that a throttle of 40% on main and iop seems to work fairly fine. Anything less than that causes the interconnect to fail as I think the timing just gets too slow. Sadly though, because of this limitation I cant yet make it turnkey and therefore cannot start producing HP Access Plugs that just plug in and work.

        On performance, I think the principle cap is the single core ARM 5 processor which limits the performance as both the main and iop are sharing the same cpu cycles. I think a dual core would solve it, but Ive not yet found a plug computer that comes with dual cores that is in the sub $100 market. I think that using two plug computers, one as main, one as IOP, may well work as they should just communicate using tcp/ip for the interconnect, but its not something I have tried. One thing I might try when the Raspberry Pi comes out (a $35 computer, www.raspberrypi.org) is using two of those, one as main and one as IOP.

        One thing for the purists is that the ARM5 SheevaPlug computer that I am using has a serial UART for the console, so you could in theory attach a genuine serial device. However, as shipped the UART is presented through a USB converter on board and so you end up using a laptop and a terminal program as the console anyway as Im yet to come across an ASR33 with a USB interface.

        For background, the SheevaPlug computer is a 1.2Ghz ARM 5 cpu with 512Mb of RAM and 512Mb of flash which is supplied pre-loaded with Linux, Debian in my case as this boots quicker but Ubuntu is available. After initial setup of the plug computer on the network and installing a few additional packages on the supplied debian (mainly gcc), I compiled hp2100 from source, no problems. Then it was a simple matter of using the clean_access image from the group repository to startup an initial Access system, and off it went. To overcome the need to manually adjust the throttle on the main and iop to stop the mux being so bursty and slowing down significantly under load, I loaded main and iop using the Linux terminal session tool "screen", which allows multiple sessions on a single telnet or console session. This is ideal for the serial console as you can start the main, put it in the background, start the iop, then switch back to main, bring up access, throttle main, switch to iop, throttle iop, and then switch back to main where the console is running.

        Below I give you a terminal log of starting up Access on the SheevaPlug and I'll put a photo up of the plug on the group in case it is of interest:

        root@hpaccess:/home/hp2000/access# screen -d -S HPConsole -m ./hp2100 asmain
        root@hpaccess:/home/hp2000/access# screen -d -S HPiop -m ./hp2100 asiop
        root@hpaccess:/home/hp2000/access# screen -r HPConsole

        HP 2100 simulator V3.8-2
        Listening on port 4020
        Listening on port 4021
        Throttling disabled
        minimum sleep resolution = 2 ms

        MAG TAPE SELECT CODE? 16
        2754? N
        LOAD WHICH MODULE? 2883
        SYSTEM GENERATION? N
        MAG TAPE RELOAD? N
        LOAD OR DUMP COMMANDS?
        DATE? 046/12
        TIME? 2344
        HP22687A-1812
        ROS
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....

        and the system is now listening for telnet connections to the mux on the IOP. To ensure that performance is optimal I then set the throttle on both main and iop using screen commands (ctrl-A and d) to switch between the two on the same serial terminal:

        ^E
        Simulation stopped, P: 70140 (STB 70024)
        asmain> run 2000
        Simulation stopped
        sim> set throttle 40%
        sim> c
        ROS
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....

        ^Ad
        [detached from 18240.HPConsole]
        root@hpaccess:/home/hp2000/access# screen -r HPiop
        HP 2100 simulator V3.8-2
        Listening on port 23 (socket 4)
        Connecting to IP address 127.0.0.1, port 4021
        Connection established
        Connecting to IP address 127.0.0.1, port 4020
        Connection established
        Throttling disabled
        minimum sleep resolution = 2 ms

        Simulation stopped, P: 20563 (ADA 20572)
        asiop> run 2
        Simulation stopped
        sim> set throttle 40%
        sim> c
        ^Ad
        [detached from 18244.HPiop]
        root@hpaccess:/home/hp2000/access# screen -r HPConsole
        ROS
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....
        .... .... .... .... .... .... .... ....

        ... and then try a terminal session (notice support for lowercase on the mux, this allows Tektronix 4010 graphics!):

        Connected to hp-access.dyndns.org.
        Escape character is '^]'.


        Connected to the HP 2100 simulator MUX device, line 0

        ^M^J
        PLEASE LOG IN
        hel-t001,t001

        BRIGHTON POLYTECHNIC HP ACCESS SYSTEM

        THURSDAY 16 FEBRUARY 2012 AT 0:07:54 AM

        TYPE EXEC-$NEWS FOR THE LATEST NEWS

        cat
        NAME LENGTH RECORD NAME LENGTH RECORD NAME LENGTH RECORD



        Cheers,

        Quentin



        On 15 Feb 2012, at 20:05, mgemeny wrote:

        > Quentin,
        >
        > That's very cool. Did you ever get SIMH to throttle properly?
        >
        > I wonder of your sluggishness on mult-user is the result of the flash being slow to write. If this is the case I would wonder if we could get disk drive 0 onto some sort of RAM disk. We would have MLOCK most of drive 0 to prevent user programs and files from being stored into the RAM disk, as it will be lost in the event of a power failure. The biggest question is if we can get the system to allocate the ID tracks on a drive other than drive 0. If so then we could end up with a system where the only things stored on drive 0 where static things like the OS libraries, and volatile things like the user area swap tracks. That would allow a fresh copy of the drive to be copied onto the ram drive from flash at startup.
        >
        > Would you care to share with us what you had to do to get it running?
        >
        > I also was wondering how hard it would be to use a plug computer as a single port RS232 terminal server. That could allow a classic terminal to be used with a telnet connection to a simulated system.
        >
        > Once I was able to get Access running across a network with the IOP SIMH running on a different host from the MAIN SIMH. I wonder if that would improve performance on the plug computers?
        >
        > I guess I just went from 1 plug computer to 3 or more, but at $100 each, and 5 watts, I guess that's still not too bad.
        >
        > Still very cool, I hope you can share more about it.
        >
        > Thanks again,
        > Mike
        >
        >
        > --- In hp2000family@yahoogroups.com, Quentin North <quentinnuk@...> wrote:
        >>
        >> Hi
        >>
        >> I have a public Access system at hp-access.dyndns.org also if anyone wants to use it, drop me an email. It is running on a $100 plug computer with a USB flash drive for disk. It's a bit slow when you get more that 3 or 4 concurrent users, but otherwise is fine.
        >>
        >> Sent from my iPad
        >>
        >> On 15 Feb 2012, at 00:32, Bob Brown <bbrown314@...> wrote:
        >>
        >>> Any way my public Hp em system on Mickey.ath.cx could be leveraged or helpful in this effort?
        >>>
        >>> -Bob
        >>> Sent from my iPhone
        >>>
        >>> On Feb 14, 2012, at 6:11 PM, Doug Humphrey <doug@...> wrote:
        >>>
        >>>>
        >>>> Re: my last message
        >>>>
        >>>> in fact, and I am serious about this, since I suddenly do NOT have
        >>>> any spare time, but I want to do this project, who would be interested
        >>>> in building a PC with proper emulation and setting up TSB/2000 Access
        >>>> for me?
        >>>>
        >>>> I can see two things that I could do for the person doing this. One
        >>>> possibility is that I buy two sets of hardware (PCs I presume are the
        >>>> right thing - maybe laptops would be best? or maybe not....) so that
        >>>> the builder keeps one and I get the other, and of course Two I would
        >>>> be willing to pay some amount (in the hundreds? ) for the effort.....
        >>>>
        >>>> I know, almost not sporting to do it this way, but if I wait until I have
        >>>> the time to do this, well, it will freakin' never happen.... and what I
        >>>> want to do is to be able to sit down at the 33 and hammer away on
        >>>> the old machine, and to show it to my kids as a working example
        >>>> of "how it was" since "hello world" or whatever we did way back then
        >>>> is NEVER as impressive as when it gets hammered out by a teletype :-)
        >>>>
        >>>> I have the teletype, so the project is just to get it to an RS232 port I guess,
        >>>> and then I can try to handle the current loop translation local - I have a
        >>>> few VT100's around I am pretty sure too....
        >>>>
        >>>> thoughts? I am very open to suggestions ;-)
        >>>>
        >>>> thanks
        >>>> doug
        >>>>
        >>>>
        >>>> On Feb 14, 2012, at 6:58 PM, Doug Humphrey wrote:
        >>>>
        >>>>>
        >>>>> On Feb 14, 2012, at 6:43 PM, Ralph Smole wrote:
        >>>>>
        >>>>>>
        >>>>>> I miss programming in Time Share Basic. Back during high school,I wrote a weather prediction program that won me 1st place in the Buffalo Grove HS science fair. I also had a more dubious achievement when I wrote a program that almost got me expelled. It was a program that sampled all the ports on the HP2000 system and printed on the teletype whatever people in other schools were doing,unfortunately it also crashed the system. I was called into the principals office one day which was filled with school district officials AND a representative from HP. Fortunately,my little program exposed a hole in the system. I escaped with a stern warning AND a thank you from HP. Guess I was an early day hacker :)
        >>>>>
        >>>>> you are a member of a select but measurable
        >>>>> club of people :-) the 2000 was the perfect machine
        >>>>> for the adventurous programmer, with enough nice
        >>>>> fruit within just the right difficulty of reach :-)
        >>>>>
        >>>>> It rewarded those who put in the time to figure it
        >>>>> out with A000 and other goodies.... and started
        >>>>> a whole generation of high school hackers.
        >>>>>
        >>>>> Bless it!
        >>>>>
        >>>>> It is funny - I am thinking of bringing TSB up on an
        >>>>> emulator just do that I have something to talk to with
        >>>>> my 33 teletype - unix is no fun on an uppercase only
        >>>>> machine - so its either the 2000 or a PDP-11
        >>>>>
        >>>>> doug
        >>>>>
        >>>>>
        >>>>
        >>>>
        >>>>
        >>>> ------------------------------------
        >>>>
        >>>> Yahoo! Groups Links
        >>>>
        >>>>
        >>>>
        >>>
        >>>
        >>> ------------------------------------
        >>>
        >>> Yahoo! Groups Links
        >>>
        >>>
        >>>
        >>
        >
        >
        >
        >
        > ------------------------------------
        >
        > Yahoo! Groups Links
        >
        >
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.