Not really, when the R7 has a sensor with substantially faster readout. That means more AF calculations per second.
Where did you get this information, from my admittedly crude tests the readout speeds are the same. If there are more accurate/competent tests available I’d love to see.
It’s been tested by more than one person, and there’s a chart making the rounds. To be perfectly honest, I can’t be bothered to fish it out now
Nope, not when the denser sensor reads out faster.
I’m not so sure that’s actually true, doesn’t the whole sensor need to be read out?
Sensor readout speed depends far more on the silicon substrate technology than it does on photosite count. Naturally, if you have two chips, one with 8MP and one with 24MP, made using the same process, the less dense one will have a faster readout. But this doesn’t scale linearly, and different processes can have vastly different readout speeds, enough to offset any losses from extra photosites. Like it happens with the Canon 24MP and 32MP chips.
That's another matter altogether. Maybe Canon's current FW can't leverage the difference. Maybe it's only measurable in edge conditions. Maybe optimal technique can turn the difference into a negligible one. It's exceedingly hard to say. But, with everything else constant, a faster sensor readout should deliver an improvement in AF effectiveness.
I guess the question is, is the R7 sensor readout actually faster than the R10 sensor. Also, if you can’t actually detect the difference in actual use does it actually mean anything.
It does. Canon sensors today use similar strategies to Sony chips, meaning that they drop down to a much lower bitrate when in “EVF mode”, which is also how they find focus. Ten milliseconds don’t sound like much, but that means a couple of extra frames for the processor to scan during AF tracking each second. Over time, that definitely adds up.