K-1, 14 months, 20000 photos (gets a bit nitpicky)

Not neccesary. There could be times you need 2s delay but manual control when the shot is done. It seems more that you need a remote or should use a smartphone to start your brackting sequence. And if this would no longer be possible someone else would be upset as they are using that feature.

In which instance do you want 2s delay between bracketed shots?!
I have the remote, but I think there can be little disagreement that it is generally best if we do not need extra equipment.
If there is REALLY someone who wants 2s between bracketed shots: Fine, add it to the bracketing drive mode.

It's not exact 2s between the shots that may be neede; bur longer delay. If you use bracketing to get the best exposure for the shot it may not matter if it is a long delay beween shots. Sometimes you just have to wait between shots for best conditions.

Bracketing is not only used for HDR.

I think we are talking a little bit past one another here.

While I can see that for some instances you do not want to do the shots as quick together as possible: If you want to space it out some time, you'd not use one-touch bracketing! Because then you'd take each shot of the series individually.

So I maintain: One touch bracketing with mirror-preflip should only flip the mirror up at the start and then keep it up during the whole series.

Yes you spotted the missing word. The old style PDAF had the advantage of working with all lenses. Old Sigma lensesit has showed that their reversed engineered frimware is not fully compatible with CDAF. If it is newer Sigma lenses they may have firmware updated that can fix the issue.

And it is not a Pentax specific problem users of other brand cameras has reported the same thing.

Interesting. This means that CDAF is not really a fully closed control loop, but has some open-loop or feed-forward characteristics.

I shall try and compare my screw drive Sigmas to my HSM 120-400. It's not an ideal comparison, because the 120-400 is much newer than my other long Sigmas, so it will not be clear whether a difference in performance is due to the other drive system or the improved reverse engineering.But it's the best I can do.

If you have links regarding this, I'd be interested in them

It's because they use two completely differrent designs, where the Sony sensor is fixed on rails while Pentax sensor is free floating. Most manufactures today use similar system to Pentax (including Sony).

Sure, that wasn't a criticism, just an observation

I wonder if that difference is due to the additional axes the new Sony and Pentax systems can compensate compared to the older Sony system? Not sure how it was for older Pentax systems.

The old Minolta/Sony IBIS could not rotate the sensor so it would have been limited to 4-axis, but at the time it had 2-axis when Pentax had 3-axis.

But I think the biggest disadvantage of if was that it was less responsive so it would not have reached the 5 stops of compensation that free floating systems allow. And it would be difficult to add all other fuctions that rely on SR. (pixel shift, AA simulator...)

Oh yes, things have improved for sure.

FWIW, I found the old Sony SSS works quite well during panning (Sony recommended to switch it off), but haven't done any detailed tests regarding the Pentax system during panning.

Cheers

Jens