Canon L stuff is great - no doubts. I think I mentioned more than once that they are the kings in lens quality (although Sony has caught up lately).
Still, Sigma and Tamron are selling A LOT of lenses - at some stage during the DSLR era, Sigma was even the biggest lens manufacturer (not sure about today).
So, it's safe to state that most people WANT third-party lenses. And in the case of R-mount, that's simply a big fat negative bullet point.
Not everybody can afford L-glass. And regarding non-L glass, Canon isn't anything special, really. On the contrary.
As far as RAW vs JPEG testing goes - I hear you. In my own book, the struggle isn't so much about camera-produced JPGs. I, for one, would never use these simply because I pretty much always do shadow/highlight recovery in post (but your mileage may vary here). I struggle when seeing what something like DxO Photolab can do with the RAWs. The softness recovery capability is shocking at times. Lenses that I'd previously touched only with gloves suddenly produce results that are more than just decent - which, of course, raises the question of how far the classic "hard" criteria are still relevant (or at least THAT relevant). So, it may not be YouTube that is killing the review websites; it may be AI. I always suspected that AI would kill the need for "bokeh lenses" eventually, but I'm surprised that the first victim may be "lens sharpness".
While it is undoubtedly true that FF image quality is killing MFT at the moment, I do wonder whether FF sensors make any sense long term. The megapixel race has been silly for years anyway. Very few amateurs need more than 20 megapixels, really. AI can lift MFT images to a quality level where there's little to complain about already. The old "noisy image" argument is basically gone. However, Oly/Pana have to get their arses up and running to make AI-optimized bokeh a reality. If Apply/Google can do it, they should be able to do it as well. I hope that they can manage this before they've lost too much market share.
Still, Sigma and Tamron are selling A LOT of lenses - at some stage during the DSLR era, Sigma was even the biggest lens manufacturer (not sure about today).
So, it's safe to state that most people WANT third-party lenses. And in the case of R-mount, that's simply a big fat negative bullet point.
Not everybody can afford L-glass. And regarding non-L glass, Canon isn't anything special, really. On the contrary.
As far as RAW vs JPEG testing goes - I hear you. In my own book, the struggle isn't so much about camera-produced JPGs. I, for one, would never use these simply because I pretty much always do shadow/highlight recovery in post (but your mileage may vary here). I struggle when seeing what something like DxO Photolab can do with the RAWs. The softness recovery capability is shocking at times. Lenses that I'd previously touched only with gloves suddenly produce results that are more than just decent - which, of course, raises the question of how far the classic "hard" criteria are still relevant (or at least THAT relevant). So, it may not be YouTube that is killing the review websites; it may be AI. I always suspected that AI would kill the need for "bokeh lenses" eventually, but I'm surprised that the first victim may be "lens sharpness".
While it is undoubtedly true that FF image quality is killing MFT at the moment, I do wonder whether FF sensors make any sense long term. The megapixel race has been silly for years anyway. Very few amateurs need more than 20 megapixels, really. AI can lift MFT images to a quality level where there's little to complain about already. The old "noisy image" argument is basically gone. However, Oly/Pana have to get their arses up and running to make AI-optimized bokeh a reality. If Apply/Google can do it, they should be able to do it as well. I hope that they can manage this before they've lost too much market share.
Chief Editor - opticallimits.com
Doing all things Canon, MFT, Sony and Fuji
Doing all things Canon, MFT, Sony and Fuji