Enjay

I would prefer if they left out the baffle / secondary aperture to artificially stop down the lens at the short end. The whole point is that this property is added in addition, the glass for the brighter wide end is already there. Here is the secondary aperture in action: https://www.lensrentals.com/blog/2020/01/finally-the-nikon-z-24-70mm-f2-8-s-lens-teardown/ "There’s a cam that inserts into the zoom barrel, changing the size of this secondary aperture as you zoom." And without it, that lens would not be 24mm f/2.8 but something like 24mm f/2.0, without any other changes to its construction. If you like not having that extra stop, that's fine with me though.

This is oversimplified of course - the effective entrance pupil, depending on the exact construction of the zoom (elements in front and back of the aperture), does change as you change focal length. That is what I called "magnification" of the aperture, as seen through the front lens. But not that much, which is why all these zooms have values like 3.5-5.6. A standard pro 24-70/2.8 zoom would definitely be something like a 2-2.8 if it wasn't for the said mechanism or additional light baffle, depending on the exact model. The f/2 would be soft and have strong vignetting, but it would at least be available.

Yes, but the aperture does not change its size by itself when you change focal length! Such a lens would be 50/50=f/1 at 50mm and 100/50=f/2 at 100mm. To keep that lens at f/2 so that it can be marketed as 50-100/2, the aperture has to close down to 25mm at the short end. This is done with a special double aperture mechanism that 1. closes down as you zoom out and 2. closes down more depending on what you set in the camera, firmware controlled. The first part is purely mechanical. You can see this on some strip-downs lensrentals.com have done. Detach the lens from the camera, zoom out, and watch the aperture close, without any power supplied to the lens. Constant f-stop is a complete marketing fabrication and not a "natural" property of any lens. As focal length decreases, the aperture gets larger in relation. Zooms are always brighter at the short end than the long end, unless the manufacturer actively prevents it.

The brightness is equal, but only because there is a special aperture that closes down at the short end because otherwise the physical aperture size does not change while the focal length varies. With a 4x zoom, the same physical aperture is (relatively) 4x wider at the short end than the long end. Some of this is for technical reasons. A 4x zoom with let's say 1.4-5.6 would be very soft at the short end, and I'm not sure modern zoom designs have the same effective aperture at all focal lengths (the "aperture image" is magnified depending on the focal length). But the f-number definitely does not stay constant unless the manufacturer artifically creates this effect. It's marketing because people associate high-end lenses with constant f-number. Decades ago, when cameras could not easily compensate for changing f-number, pros really did need the constant aperture value, which I guess is the source of all this.

Have you guys ever considered that constant aperture zooms are also seriously crippled? They are really brighter at the short end, I would say by about 1 stop. The diaphragm starts to close as you zoom out. This is easily visible e.g. with the current RF 24-105/4L and also with my old Sigma 24-60/2.8. Speaking of which, I just compared the two lenses at 35mm f/5.6. The Sigma's performance, which came out in 2004 and was always much cheaper, is almost identical.

But the R6 can shoot 4k?!

Says who? What from I read, you can take photos AND use the 1080 & cropped / non-HQ 4K modes (that are still way better than compressed to death YT videos) all day. Anything else especially concerning photos would be a first, and a huge failure. 4K+ streaming is still a joke anyway, all we are seeing is the higher bitrate services reserve. I have an interface traffic graph running at all times on my PC acting as a router, the bitrates are laughable. You could upconvert by at least one resolution step and probably no one would notice.. I guess the mistake Canon made was giving the samples out primarily to a vocal minority of videographers. There are real issues like a fixed viewfinder eyepiece that cannot be replaced by the magnifying version I was using for more than 11 years. Or the fact that the ergonomics are limited for left-eyed shooters (face obstructs access to some controls on these smaller camera bodies). Or baked RAW files below ISO800 with baseline noise reduction applied. Or the huge issue that DPAF cannot sense horizontal detail at all. I generally do not watch video "reviews" because reading is 3x faster and I cannot stand listening to someone reading his text at "single speed", but I bet not a single one of these YTers has addressed anything of this.

Funny how this issue gets so much attention, but literally nobody complains about Canon charging 10% higher netto prices in the EUR/GBP zone compared to Yen/US$. They started that with the 5D 4. THAT is what kept me from buying the R5. Third party lens manufacturers like Sigma or Tamron actually have lower netto prices for many lenses here, it cannot be a general thing.

The Canon I used to know since I bought my 350D in 2005 would not have included the "overheating" video modes or the H+ 12 FPS mode at all. Especially H+ is clearly a prototype style feature they would have omitted because it only works under very specific conditions.

Well, this is what happens if you stop using the "cripple hammer" people often complained about, and simply offer what the hardware can do. 12 FPS H+ mode with mechanical shutter is also severely limited, check out page 452 of the manual. In most instances, the camera will deliver between 7 and 10 FPS. From a developer's point of view, Canon has included features once only available with hacked firmware like Magic Lantern. This is one of the factors that allowed the leap forward in performance. Personally I like this, because there are no arbitrary limits, but it is different from what people are used to from this manufacturer.

The comparison with the car is funny. A lot of foreign (i.e. non-German) cars are known not to be what we call vollgasfest, "full-throttle proof", and will actually have their engine overheat after a while at maximum speed. Even some of our models theoretically have that problem, e.g. VW group models with the 6.0 W12 engine, but it is normally masked by the 155 mph speed governor.

Where does it say that 4k/24p also overheats? The table specifically lists 4k/60p. 24p is limited to 29:59 recording time, sure, to avoid import taxes for video cameras. The camera could technically record longer.