Now we see through a glass, darkly...
- Apr 5, 2016
As for the 24-105mm f/1.4L IS II, sometimes companies simply make mistakes. That's what happened in this case--it is a dog of a lens. Sure, they might have saved money per unit, but they are producing a lens with lower IQ than its predecessor.
If a lens that has basically the same performance as its predecessor can be considered a "dog", then I suppose it is a "dog". It is slightly sharper at some FLs and Avs, it is slightly less sharp at other FLs and Avs.
The "new version" of the 70-200mm f/2.8L IS is another indication that Canon has put further development of EF lenses in the rearview mirror. I don't know if a tear down would reveal a firmware change, some new chip maybe, but I think any internal tweak to it was done to make it work better with RF bodies. Maybe, maybe not, but otherwise it would be the single most useless "update" ever.
It may or may not be such an indication. It may just be that there wasn't much possibility to improve over the current optical formula without making the lens much more expensive (cough, cough... Nikon AF-S 70-200mm f/2.8E FL VR... cough).
There was apparently a need to upgrade the electronics for the upcoming EOS RF and it was well past time to add the superior anti-reflective coatings (ala the EF 35mm f/1.4 L II) that do significantly reduce lens flare in certain shooting scenarios. If you are a photographer that desires the better flare performance, this is no "minor" upgrade.
Just because something does not show up on a shot of a flat test chart under ideal lighting from a relatively short distance does not mean that it's not there.
Well, unless your only goal is to be able to brag that you own a lens with the best possible performance when shooting a flat test chart under ideal lighting from a relatively short distance.