NEC 8001

As previously mentioned,  I lived and worked in Japan from  1983 to 1985.  Once I was selected for the job, I made an initial “get acquainted” trip in October of 1983.  During that time I found an apartment.  I had decided not to take my Franklin computer to Japan because it wasn’t clear whether it would work properly on Japanese electricity (slightly lower voltage, 100 v., than U.S. voltage, and slightly lower frequency, 50 hz.)  So I decided to shop for a Japanese computer.  The idea was that I would buy it in October, bring it home, transfer my files from my Franklin, and then take it back in December, when I went there to live. 

 

The IBM PC had come out by now but was too expensive for my budget.  The IBM PC ran an operating system that they had jointly developed with Microsoft, which IBM called PC-DOS and Microsoft called MS-DOS.  Microsoft had made the deal with IBM that Microsoft retained the right to distribute the OS on other non-IBM computers that used the new Intel 8086 chip.  I knew or discovered that there were a number of Japanese computers running MS-DOS.  So I set out to Akihabara, the electronics district of Tokyo, to find a computer.  I wanted one that ran MS-DOS and had English documentation.

 

What I discovered was that there were plenty of MS-DOS computers, but none with English documentation.  Therefore, I decided to buy a Japanese CP/M computer instead, on the theory that I already knew CP/M and could get along without documentation. 

 

I bought an NEC 8001 Mk III.

 

 

The keyboard could be switched between Kana (Japanese syllabic script) and Roman characters.  The ROM Monitor was in Japanese but CP/M was (mostly) in English. 

 

The documentation was in Japanese, but there was a source listing of the BIOS, which had comments written in English.  This was enough to teach me the low-level programming interfaces to the hardware. 

 

I took the computer back to my hotel room and set it up.  I discovered that even though I knew CP/M I didn’t know how to use it as well as I thought.  For one thing, I had never learned the free CP/M line editor, ED, because I had bought a full-screen editor.  And there were still hardware operational questions that I didn’t know the answers to. 

 

While I was in Tokyo, I went to a PC trade show at Harumi, the main convention center in Tokyo.  NEC had a booth, and I asked them if they had English documentation for the 8001.  They referred me to a retail shop in Ginza.  I went to the shop by taxi;  I remember it was hard for the driver to find—addresses in Tokyo are assigned in the order buildings were built, which makes them effectively random.  Eventually the driver stopped and called them on a pay phone.  Once in the shop I found that in fact they did not have English documentation, only Japanese.  I looked at it to see if maybe there was enough information I could glean, but decided not to buy it.  I already had a basic set of Japanese documentation that came with the computer.

 

When the airline found out I was bringing a computer with me, they made me buy additional insurance.  But the computer arrived in California safely. 

 

Once back home, I called NEC America, and they referred me to some consultant in Los Altos who sold me a set of English manuals.  It was not a full set, however, so for some things I had to take the Japanese manuals to work and ask a Japanese co-worker to translate for me.  One even offered to translate the entire manual, but that seemed too much to ask.

 

While my NEC computer was in the U.S. I set about the task of copying data from my Apple-compatible computer to the NEC.  I focused on the Apple CP/M system since it contained most of my “real” data—most of my Apple OS programs were games.

 

This was in the days before the Internet.  In order to download software, I had to use a modem and dial into some BBS (bulletin-board system).  These were typically run by individuals on personal computers.  If the system was not local, I had to pay the long-distance phone charges. 

 

I downloaded a program called zmodem, which was an improvement over the then widely-used program, xmodem.  xmodem was essentially a peer-to-peer file transfer program.  One had to establish a terminal-emulator session using some other software and then start xmodem on the remote system.  (The remote system was generally also a personal computer running CP/M.)  Then one would start xmodem running on one’s local computer and it would establish a connection with the remote program.  There were no IP addresses or anything like that.  It was strictly point to point:  the local copy of xmodem would simply talk to whatever it found on the serial port.

 

In fact, in my case, there was no modem, the two computers were sitting physically next to each other.  I used a “null-modem” cable that simply crossed wires so that the “transmit” wire at one end was connected to the “receive” wire at the other end of the cable.  Several control wires were similarly crossed. 

 

The major difference between zmodem (the program I used) and xmodem was that zmodem had a way of transferring multiple files at once (for example, all the files in a directory) whereas xmodem had to be started on both ends for each single file to be transferred.

 

Zmodem was written in 8080 assembly language.  I think it came with source code.  In any rate the first 256 bytes of the program was reserved for system specific interfaces to the serial port.  There needed to be a unique interface for each type of system, since each system had its own native interfaces to the serial port.

 

I expected it would be easy to find an interface for the Apple system, and more difficult to find the interface for the Japanese NEC system.  In fact the opposite turned out to be the case:  I was able to download the interface for the NEC system, and ended up writing my own interface for the Apple CP/M system. 

 

I then used zmodem to transfer all my source files, executable files, and data files to the NEC computer’s CP/M system.  When the time came, the computer was packed along with other furniture and personal belongings and shipped to Japan.

 

We arrived in Tokyo in early December of 1983.  We stayed in the Okura Hotel, ranked by business travelers as the second most luxurious hotel in the world.  My boss allowed us to rent a suite because our daughter was with us. 

 

We were to move into an apartment being vacated by a previous manager from my company.  For complicated tax reasons he had to leave on New Years Eve.  So we ended up moving in shortly after the start of 1984.

 

The apartment was furnished, so we were able to simply move in with out suitcases of clothes.  Our household goods (including my computer) were coming by ship and didn’t arrive until much later—I don’t remember exactly when they did arrive, but it was probably in March. 

 

My job in Japan was that of a technical liaison between my company in California, and its Japanese joint venture in Tokyo.  This was not a programming job, and I did no coding on the job.  Therefore my Japanese PC was a welcome outlet for my desire to write code.