Title: Uncovered 1985 Intel 80286 Engineering Samples (ES) with authentic "Pass" handwritten test bench note!
Posted by pizber@reddit | vintagecomputing | View on Reddit | 10 comments
Hi everyone,
I just uncovered a fascinating small treasure from an old engineering collection. Found this tray of Intel CG80286-10 ES (Engineering Sample) chips.
The most incredible detail is the handwritten note "Pass" (and on another chip, "Runs Hot") written directly on the purple ceramic package in silver ink. It looks like an engineer’s test result directly from the workbench during development.
Here’s what I’ve found:
Front Side: Marked "INTEL '85" and "CG80286-10 ES", identifying them as 10 MHz Engineering Samples from 1985. The "Pass" marking is clearly visible.
Underside: Stamped "MALAY 8536" (36th week of 1985) and has beautifully gold-plated pins.
Context: They are in a plastic tray and were found along with a 1985 business card from a Silicon Valley engineering firm in San Jose.
I’m curious:
-How rare are these ES chips with authentic handwritten test bench notes?
-What kind of historical or collector value do items with this level of provenance hold?
Vinylmaster3000@reddit
I had no idea they put 286 CPUs in pga sockets, I always thought that was something they started with in the 386 era
Tokimemofan@reddit
Actually it started with some 80186 processors however those weren’t found much in PCs as a main processor
mallardtheduck@reddit
The 80186 was introduced at the same time as the 80286 (well, Wikipedia lists the introduction dates as a month apart, but without sourcing). Unlike other 80x86 CPUs, the numbering does not indicate a difference in "generation"; they were designed at the same time for different purposes/makets.
The 80186 was an early "system on a chip" type design aimed at "embedded" systems. The 80286 was the improved replacement for the 8086/8088 designed for microcomputers. The additional hardware included in the 80186 (timers, interrupt controller, DMA controller) was not compatible with the seperate chips used in the IBM PC, so it wasn't possible to use an Intel 80186 in a PC-compatible computer. Non-Intel 80186-compatible CPUs without the extra hardware, e.g. NEC's V20/V30 chips were sometimes used though.
Linuxmonger@reddit
Fun side note; I worked at Kings Island Amusement Park in Cincinnati in the early 90s, we had a ride called Phantom Theatre that was completely powered by pnuematic cylinders moving things around. There were about 4 thousand cylinders, all controlled from 5 42U rack cabinets.
The entire thing was controlled by a single 80186 processor and ran code from a set of 4 27C32 PROM chips. It had a pile of 8255 PIO controllers for IO and 4000+ socketed 24 volt opto-isolators.
Fun times, you could debug issues with a logic probe in no time at all.
Tokimemofan@reddit
This is correct. Ironically I have seen quite a few of them and in early PCs, most often as part of a SCSI controller
Tall-Introduction414@reddit
Every once in a while I would run across a 186 machine back in the day (80s and 90s). Pretty rare. Maybe 2 or 3 times.
tes_kitty@reddit
The 80286 came in at least 3 different packages: PGA, LCC and PLCC.
AgentOrange96@reddit
I really appreciate this! So for context, I am a system level testing engineer at AMD. I literally carry around a Sharpie when I'm in the lab so I can write this kind of stuff on the lids of units I'm testing during the bring-up process. I've absolutely written "Pass" or sometimes just "P" on engineering x86 processor lids. (And many other notes) Seeing the "tradition" going so far back is just super endearing to me! (:
aardvarkjedi@reddit
That one was made in the 36th week of 1985.
marx2k@reddit
👍