If you're picking out a CPU or graphics card for a custom-built personal computer, ignore basically every number the manufacturer provides to quantify its performance, and go look at some benchmarks. Not because the numbers the manufacturers provide are inaccurate, but because there are so many factors that go into how good hardware is besides the those numbers, that you will never get as accurate an estimate from them as with direct measurement of the performance.
Also, make sure they are compatible with your motherboard.
This is an extension of a comment I made that I can't find and also a request for examples. It seems plausible that, when giving advice, many people optimize for deepness or punchiness of the advice rather than for actual practical value. There may be good reasons to do this - e.g. advice that sounds deep or punchy might be more likely to be listened to - but as a corollary, there could be valuable advice that people generally don't give because it doesn't sound deep or punchy. Let's call this boring advice.
An example that's been discussed on LW several times is "make checklists." Checklists are great. We should totally make checklists. But "make checklists" is not a deep or punchy thing to say. Other examples include "google things" and "exercise."
I would like people to use this thread to post other examples of boring advice. If you can, provide evidence and/or a plausible argument that your boring advice actually is useful, but I would prefer that you err on the side of boring but not necessarily useful in the name of more thoroughly searching a plausibly under-searched part of advicespace.
Upvotes on advice posted in this thread should be based on your estimate of the usefulness of the advice; in particular, please do not vote up advice just because it sounds deep or punchy.