I'm referring to "Why is the Intel 8086 CPU called a 16-bit CPU?", which is being on HNQ now,
Well, a question being in HNQ or not is not really OT for us, as that mechanic is not controlled by any moderator (AFAIK).
just one month since "Why was the Sega Genesis marketed as a 16-bit console?",
This just means the question hit a nerv for users browsing for Information
which I see as essentially the same question.
At first it may look alike, as it goes around the same topic, but it takes a different angle. This is especially relevant, as the topic itself is nothing that can get a definitive answer.
The Sega question was about the general usage of 16 vs. 32 bit happening for one and the same CPU over time. In general introduced due the difference between ISA (32 bit) and implementation (16 bit), enhanced by marketing trying to position it in the best fiting market range (originally 16 bit, later on, when 32 bitness was a thing, as 32 bit) to generate sales.
In contrast, the 8086 question asks purely along implementation detail and here even more the otherwise not even mentioned relation of address bus size vs. bitness.
Yet the second question is not being closed but happily answered, including by people who answered the first one. Is there something I don't see which makes it worth discussing the bitness of each CPU separately?
To me it was the different focus set by the OP(s). And here especially I tried to go a step back and show up the various ways claimed bitness can originate. Eventually to help selecting the one to take, as well as to help to understand why other stances are taken or have been taken at various points in time.
If the community considers such questions valuable (and considering the outcome of this discussion), it seems that I should go ahead and ask a few more such questions, but I figured I ask here first.
Well, I do definitely encourage random questions with just different CPUs inserted but asking the same. But if you bring a different and genuine new angle to view this complex issue, then I'd love to think about.
Now, while Wizzwizz4's answer does have some valid points, the approach may be misleading. None of the listed points is a on it's own or in combination a reason for not closing the question - and have it point to one already tackling that topic. This includes the last (although true her) as it's redundant. Being a different question is always a reason for not being a dup.
But he's quite right in his remark that "Just because the answers are roughly similar, doesn't mean the questions are duplicates". Questions may circle around the same core topic and be well answered by (mostly) the same answers, but they need to ask may come from a quite different angle - like here, bases on address size instead of marketing claim.
This duplicity is, in my view, even an important feature here, as information is not just searched by the answers given, but also questions asked. (RC.)SE is not wikipedia where the goal is to write a single all answering entry for a keyword, but a network of n:m questions:answers, which takes much of it's power from not being normalized to a keyword, but context. Thus this additional question inherently adds to the site, even if the answers are (mostly) the same.
So bottom line: No, we do not want dupes just for the sake of it, but we do love to have as many valid views onto a topic as there are.
Now, in practical terms, it might be a good way to add a link to the other question(s) within the question text. If not done by the OP, then via a distinct edit (tried to do this already for the question in question :))