Name that Ware, January 2023

The Ware for January 2023 is shown below.

Thanks to cpresser for contributing this wonderfully photographed circuit board as this month’s entry.

10 Responses to “Name that Ware, January 2023”

  1. AZeta says:

    A front-end readout board of the muon detector system KASCADE (KArlsruhe Shower Core and Array DEtector): https://publikationen.bibliothek.kit.edu/270048462/3814005

  2. José Araújo says:

    SGS IP32 – 32 Channel Shift Register Card for a “Streamer Chamber muon tracking detector”.
    Found an AON reference, has the 32 channels (32 identical circuits around the edges), “BUS IN” “BUS OUT” suggesting a chain connection, and the IC date codes (1996 e 1997) are around the same time as the papers I’ve found referencing it (1999): “A large Streamer Chamber muon tracking detector
    in a high-flux fixed-target application” and 2000: “Large Muon Tracking Detector in the air shower experiment KASCADE”
    Just found now second paper has a picture of the board, “ST Muon Tracking Detector”, page 20.

  3. Keith says:

    Isn’t the entire point of these simply to exercise our brains, look at the components and other features on the board in order to guess/speculate what it might do? Versus using search engines that is.

  4. KE5FX says:

    Yeah, doing a reverse image search kinda breaks the game. Not sure how it could be effectively policed, though.

    • Keith says:

      Just figured it needed to be said. It’s not like the prize is a car or anything. It’s a shame some people tend to just ruin things or feel the need to post “first”.

      • bunnie says:

        It’s hard to judge how some of the fast responses came out — more often than not, it just so happens that some reader is a domain specialist in XYZ and they’ve seen the board before, and they can just name it. I think that’s a legit win, even though we’re not examining the substance of the board to deduce its function. This is also part of the reason it’s hard to do ntw around say, vintage arcade PCBs or pinball machine components. They’re really beautiful boards but there’s a big enough enthusiast community there that no amount of redaction could make it a competition that lasted longer than 5 seconds.

        Some of the fastest/repeat winners I’ve had are just really good at figuring out what the board does, and they do use a search engine to corroborate and refine their initial intuition, which I think is perfectly fair game. That’s how I’d do it, at least.

        Simple reverse image search with no subject knowledge would be disqualifying, but I don’t think there’s a slam dunk case for that here. I do reserve the right not to give the prize to the “first but terse” response, and instead give it to someone with a more thoughtful analysis (this has been done in the past).

        But I think in this case, even though the board is obscure to most, I suspect it’s been included as an image in enough doctoral theses and physics journals that this was guessed simply because it’s known within a particular community. Now that I know this, next time I get a “big physics” board I’ll probably start with just a subsection of the board to make it a little more challenging.

        • AZeta says:

          @bunnie
          I apologize to have unwittingly violated the spirit of the competition. It won’t happen again.
          Feel free to consider my answer null.

          @all
          I’m a CS engineer by law, FW engineer by trade, with limited knowledge in electronics. I try to identify main submodules and functions as long as I can, then I turn to other clues: component markings, board layout, etc.

          My guess was an industrial board, part of some kind of signal detector: 32 identical SMD inputs connected to 8 MAX908 comparators, themselves connected to the bus by the discrete logic nearby (7402 NOR and 7420 NAND gates, 74166 shift registers). Given component latencies, sampling and bus frequency is at a few 10s of MHz. As José Araújo mentioned, BUS IN and BUS OUT suggest a daisy chain connection with other boards.

          The other parts were tricky to address: the board is industrial but the only identifier is the handwritten W196; from its dimension and side connectors I supposed an Eurocard w/ latched VME connectors but couldn’t find any reference. Markings on side connectors (Temeco or Semeco) led nowhere; markings on the blue component mid board are unintelligible, save for VDE logo. ANOR and DIGOR are two RF signals on C connectors, maybe clocks for analog and digital domains but wasn’t sure; marking on the IC near ANOR led nowhere, meaning of the -OR suffix was unknown. Searching the two terms led to the paper.

          I didn’t attempt to reverse image search because that would be 1) cheating 2) useless anyway, as I expected the image to be present only on this blog. I submitted the answer after a cursory read of the paper, finding a match between the board and frontend description; after posting I continued reading and was surprised to find the board pictured later on page 20.

          I’m sorry for not substantiating my response and ruining the game for everyone.

          • bunnie says:

            @AZeta — don’t feel bad! I didn’t think that you cheated. A lot of people give quick one-liner responses, in part because it takes a long time to write up a long answer. I want the competition to be fun and casual, so there is a fairly low bar for making an entry. I do think European contestants have an edge because I tend to raise the post at midnight in my local time, which is the middle of the day CET and most of silicon valley is still fast asleep then.

            A fast win may be a bit frustrating to others who wanted to have a go at answering it, but, I’ll take the blame for that. Sometimes the ware is less obscure than I had thought and it’s easier to guess. In this case, the board is in enough physics papers that a definitive match could be found with a bit of legwork, and I did not anticipate that.

            I’ll also note to others reading this comment that many times (but not this time) the first guess is maybe only 80% correct, and quite frequently the “absolutely nailed it” guess only comes out a couple weeks later, after someone takes a more critical look at the ware. So I usually let the competition run to the end of the month even if facially it looks like we could have a winner.

            Thanks for sharing your process, it’s 100% legit, and of course, 100% correct.

            • bunnie says:

              I’ll also add that some readers, if they are too busy to write up the long-form answer, will just post a hash of their one-liner, and then later on reveal the text that created the hash along with their longer form answer. I’m sure most readers know this trick already, but for completeness the command line looks something like this:

              echo “My guess for name that ware” | sha256sum
              a077fe12bb565348dc2102ac90230cb7d010881d15ee18b7e0384f723ed7ceab –

              And then the hash only (a077fe12bb565348dc2102ac90230cb7d010881d15ee18b7e0384f723ed7ceab) is posted.

      • cpresser says:

        As the person that phograhped this board I had hoped for more guesswork/deduction based on the function. Like in other #ntw challenges.

        I enjoy reading those comments because I can learn a lot from them. People explaining how things work, and why they are designed as they are a big driver of own engineering work.

        Even though my work is in the field of particel-physics-detectors, I just recently learned about the specifics of the KASCADE experiment when extracting this board. Typing in ‘ANOR’ and ‘DIGOR’ as written on the PCB will get you to one of the referenced papers on the 2nd page or search results. TBH, I only checked the first page and decided that none of the hits are related to electronics, so It should be good. I did not anticipate it to be this easy. Sorry about that.