Dragorn of Kismet gave me a handful of identical mystery chips that might date back to the 1980s. They’re surprisingly large and covered with contacts:
There are no logos or identification anywhere on the chip. The back side is blank silicon.
The visible patterns don’t suggest anything obvious:
The metallization layers aren’t particularly intricate:
Surely there’s something tucked under the top metallization; I have neither the materials nor inclination to dissolve the thing one layer at a time.
I gave a sampling to our Larval Engineer, who says she may turn them into fancy jewelry. I’m sure the solder bumps contain lead, but …
The Smell of Molten Projects in the Morning 
well, there’s a diffraction patterm so there is probably a regular structure on the order of a micron in pitch that just doesn’t show at the magnification you’re using. It could be some sort of CCD image sensor—or a process test coupon chip that tests the steppers/implanters/metalization/etching.
Dang, I should’a snagged that electron microscope when I had the chance… [grin]
A “test coupon” sounds right: regular structures, simple electrical properties, easy to verify. I’d want a human-readable ID on it, though, because the first failure comes from testing the wrong coupon!
Some chips have an anti-tamper metallization layer on top to avoid probing attacks, and can detect if that layer is damaged and do a wipe. However, this doesn’t really look like any of those. I also thought it might be part of a printhead, but that doesn’t look right either.
Looks like a test pattern, presumably from a fairly advanced process, either production or development. Odd that there’s no ID–we always had stuff identified, even if the pattern would never go out the door. OTOH, there were a lot of academic labs doing advanced stuff in the ’80s and they wouldn’t always figure out the need to identify each element. Like you never have to tweak the test pattern. Yeah, right.
Assuming the timing is correct, it could be a development vehicle for solder bump processing….
We had a desk-top SEM at Signetics. Not a great tool, but it worked after a fashion. It was like the Kodak Brownie camera version of electron microscopes.