Social Networking

A project log for Reverse Engineering The Weather STAR 4000

The Weather STAR 4000: A Journey of reverse engineering the hardware to an iconic machine of the 1980s/90s.

techknighttechknight 03/08/2021 at 22:540 Comments

The next step in the process was getting a brief overview of the hardware that this thing is carrying, Dumping the ROMs to take a peek at the code, and then... potential networking with others in the TWC/Retro-computing hobby to see if anyone could have potentially saved the code when the feed was still online. So, I was reaching out to anyone I possibly could have.  

Sadly however, to my knowledge, nobody was able to capture the data from the feed to be able to replay later. the data frames were SDLC frames at 115.2Kbps, over an FSK subcarrier. So without the proper hardware, its just not easy to do. Maybe with a sampling logic analyzer with the ability to record, or using a SDR radio to demodulate the signal and just store it into a bitstream. Something. But alas, no. I connected with an individual who had run these machines in his house when they were still operating, and he tried various methods within his means to record the signal, but all attempts failed since you need the correct setup to do so. And I don't have a time-machine so I cannot go back and capture the signal. :-(

So the moral of the story is: The software is lost, and never to be recovered again. TWC likely got rid of the VAX machines that ran these, and wouldn't talk to me anyways, the original designers have long since retired and don't have any of the information either, being 30+ years ago, their memory of specifics is fading. 

There is also no documentation or schematics, or engineering diagrams for this machine. it literally was a mystery box of retro-computing hardware. So with that said, I was pretty much stuck to start from scratch. 

Luckily, the circle of friends I gained had crisp crystal clear reference footage I could go by. And, on top of that, he still retained the original capture machine he used to record the system while in service, as well as the system itself. This proved to be vastly helpful for reproducing the correct colors, since he could verify color output on the same machine as the reference footage, and same recording device. So I could quickly compare his footage against my software output and compare the differences. This allowed me to recalibrate all the color values. They should be accurate. But, more on that later. 

Anyways. Now it was time to start studying the hardware.