The guy is not trying to solve the problem, nor does he suggest it as being a solution. All he is trying to provide is the possible proof, that there might be an artificial timer used, instead of the thermal read-out.
You are too bold to mark all internet being just armchair fools. Many ppl might have some technical backgroud. E.g. I am an IT guy for 30 years, we have built Astro CCD camera, doing the HW, firmware, software and all that.
While I don't feel competent enough to judge the current situation, I would not downplay other's experience. There is definitely something fishy here, like sitting in menu for one hour, doing few pics and not getting even a 1sec of a 4K HQ video, measuring 43C temperature, whereas getting to 63C during the 8K video and getting 20 minutes of the shooting time from the cold start.
The problem might not be the limitation of the lenght of particular video modes, those got advertised, but weird cool down times, most probably not being based upon a temperature reading or not a temperature reading alone (the timer stuff). Simply put - noone wants to use camera for stills in a lightly manner for cca half to one hour, and get zeroed on the 4K HQ video capability.
I give credit when due even on the interwebs, and all I have seen so far are whiners and armchair engineers making idle speculation or outright controversy out of nothing. Speculation taken too far makes people fools. No offense, but IT doesn't make you an expert in the Canon R5 thermals. Neither do people pointing a FLIR at the R5 suddenly know all about the situation. None of these tests equate to die temperatures for ASICs and other internals, and you have no idea what temperature threshold is the protection limit for these parts. Some parts really don't like to be hot over their lifetime and so for reliability are kept at lower temperatures.
I have specific experience with high speed circuit design and thermal management and will leave it at that, but I would never go around speculating I understand a design unless I did the thermal analysis or pulled out schematics or looked at the code for ASICs, FPGAs or uControllers myself. To pretend I know, even with my experience, is a lie unless I have done adequate testing and reverse engineering. On a product like this, insanely compact and complicated, that isn't going to happen from some youtube test. To sit around doing inane amateur tests and then to make biased assumptions about artificial timers and blah blah blah is people stoking their ego, NOT honest intellectual discourse. Sometimes saying "I don't know" is the best course, but too many people don't like the feeling of admitting it, even when the subject matter is way out of their area of knowledge.
Calling the time limits artificial or fishy in the first place is an assumption and an emotional label. Engineers have to make a choice on how to put in stops around various physical realities of the components and their interaction as a system. Even if the record times WERE arbitrarily chosen, SO WHAT?! Canon told us about it including the part about other camera activity reducing those times. So for the millionth time I suggest people go buy another camera if you don't like it instead of falsely acting like some detective! Since people are not having an honest discussion about how electronics and product engineering actually work, I will continue to down play what these people are saying because the way they are going about this is amateurish, has an unintelligent and non-rigorous tone, and the vast majority of them have an agenda of some sort: to get clicks or prove the camera is flawed. It isn't; Canon told us exactly how it behaves before shipping started.
There are plenty of reasons the camera internals might heat up just in the menu or shooting stills. In my own tests, just using the EVF for long periods of time generates heat. IBIS also seems to generate heat. Try a 1 hour timelapse at 5 second intervals and just disabling those two is the difference between a warm and cold body at the end of the hour. So I could draw conclusions from my own amateur test that something about the heat and cooling is real, not artificial, but why bother? The camera worsk for me as designed. I'm sure everyone has heard that removing the CFE card helps. SERDES for the CFE might always runs idle characters even without data flowing. We could speculate all day. I could do some pretty fancy tests if I really wanted to, or I could buy the right tool for the job (which I did, and it's a pretty great camera with a few flaws like every product must have).
If Canon has some errors to fix or optimizations for the firmware that make this better, great. Not uncommon for new product. If they actually made a hardware mistake and fix it, great, but I find that very unlikely. However, until I hear well researched logic that doesn't fall apart trivially I will keep calling out all these hokey tests and assumptions for what they are: hot air (see what I did there)?
The only testing I've seen that actually benefits anyone is from those actually working with the camera instead of trying to make it fail, or providing workarounds like using external recorders- those people are accepting the camera for what it is at least instead of continuing on this path of labeling things nefarious or fishy or artificial or whatever.
here is what I said:
