My fascination with computers began when I took my first computer programming cours at the University of Hawaii in 1979. I soaked up this class like water and as final project I created a program in BASIC that did flight planning, and weight & balance for the Piper Warrior II I was flying back then.


My newest system was built using the following:

  • Gigabyte GA-MA790X-DS4
  • AMD Athlon 64 X2 5600+ 2.92GHz
  • Diamond Viper Radeon HD 3870 512MB PCI Express 2.0
  • (2) Kingston HyperX 2GB DDR2 800 - 4GB Total Memory
  • (2) Western Digital Caviar GP 500GB SATA 3.0 Drives - Mirrored 500GB
  • (2) Western Digital Caviar SE 250GB SATA Drives - Mirrored 250GB
  • LITE-ON Combo Black IDE DVD RW
  • Linksys WMP300N Wireless PIC Card
  • ANTEC P182 Super Mid-Tower
  • ANTEC EA650 650W PS
  • Microsoft Vista 64 Ultimate
  • The case was purchased via Buy.Com. The rest was purchased at I have had nothing but positive experiences with both of these companies and highly recommend them. I could have gone to one of the main computer manufacturers to get a system. However, I prefer building my own systems from scratch so that I know exactly what is in the system. I have yet to put a system together that did not have a little glitch. In this system one of the mother board connectors was 2 pins but the case connector expected 3 pins. Fortunately it was not a critical item so I could forgo it. I encourage anyone who has the interest to give it a shot. There is just something satisfying about the first power up, seeing the boot screen and loading the OS from scratch.


    The following is a list of the systems that I have had the opportunity to work on over the years. When I say work, I mean the operating system and application side.

  • HP 3000 - MPE (Probably)
  • Harris H800 - VULCAN
  • IBM 370/145 - OS/VS2
  • IBM 3033, 3081, ES/3090 - MVS
  • IBM Z-Series - z/VSE
  • IBM S36 - SSP
  • Periphonics - Unix
  • DEC VAX (?) - VMS
  • SUN (?) - Unix
  • Tandem NSII, TXP, VLX, CLX, K100x, K200x - Guardian
  • HP NonStop S-Series - Guardian

    Of these I spent the most time on the HP NonStop(Tandem) system. As a system administrator I upgraded the OS and recommended hardware and software. I did OS upgrades from Guardian A06 through G06. They were very interesting systems to work on and they were the most reliable hardware I ever worked on. While many facets of the system were archaic and difficult to master; the OS was solid and anyone who could program on it was in demand. Coding Non-Stop programs was difficult and in fact the examples in the manuals have yet to provide a working example that you could code from. I reported that but had never seen a manual updated with a working example. I miss working on these systems. If the opportunity presented itself to work on one remotely I would jump at the opportunity.


    My introduction to Unix came through the Periphonics Voice Response system BOH used in the mid-1980's. The initial system was Unix System III and the application was based on a finite state machine coded by guys at Periphonics. One of the guys there came out to help bring up the system and he was the most gifted programmer I had ever seen. I had a ball learning about Unix and C. I actually extended the application code substantially since the system was so well coded. I also got the opportunity to upgrade from Unix System III to Unix System V which was not a trivial exercise.

    The Periphonics also provided me with one of the most interesting problems and solutions I encountered as system administrator. The Voice Response System was actually composed of 2 separate computer systems that were connected via what was called an interprocessor cable. One system contained a proprietary OS that handled the phone lines and the voice. It had a cabinet full of cards that were memory to hold the voice recordings. The other system was the Unix system that ran the finite state machine application that communicated with the mainframe to get the customer data used. One of the memory boards in the proprietary system failed and the local vendor did not have a replacement available. After an hour or so I decided to check the listing showing how the memory was used and discovered that we were not using the last section of memory on the last board. Since each board consisted of 2 sections and we discovered that the board that failed only failed in the upper (last) section we could re-address the bad board to be the last card in the system. That way we could run until the replacement board arrived. That was a real exercise in using our knowledge to find away around a problem.