Nanodata vs Microdata
 |
I have a hard copy of the second edition of Microdata's manual; those less fortunate can find it on Al Kossow's web site.
I recently bought, cheap, second-hand, an old issue of a British astronomy magazine. It had reviews of the Meade RCX400 telescope, and the Cape...
But my memory being not what it should be, when I recently watched a DVD of Koyaanisqatsi, and saw the moon pbutt behind the Microdata building, I thought that my brochure was from Nanodata. I had buttumed that these two companies, with similar sounding names, made basically similar computers; minicomputer-sized boxes with vertical microcode.
While that is what Microdata made, suitable researches and Al Kossow's site allowed me to learn how mistaken I was.
The Nanodata QM-1 sported a mainframe-sized price tag of $250,000 according to one web site, and, from the manuals at Al Kossow's site, it had *horizontal* microcode, with a 360-bit width.
Not only that, it had a second level of microcode which invoked the nanocode. This implementation technique was said to have been used internally on the Motorola 68000 chip as well.
Given that the *first* Pentiums were like unto the hardwired System 360-195, I have no doubt that the technology exists today to put something like a QM-1 on a chip. Given that current chips are not microcoded, but instead hardwired, the moment has indeed pbutted by, as such a chip would no longer provide compebreastive performance. A software method, "Just In Time" code translation, on a high-performance conventional microprocessor, would likely do better in emulating almost any conceivable architecture.
This has been discussed here before. While NASA certainly bought a decent number of early IC's, it was ICBM R&D that was really driving electronics to be smaller...
John Savard Usenet Zone Free Binaries Usenet Server More than 140,000 groups Unlimited download