Originally posted by richandfleur The video behaviour is a bit weird, like you don't get focus peaking when it starts recording, the image still crops when you turn off the software SR, the image is generally soft, the low bitrate can't handle busy scenes, limited framerates etc.
I'm pretty sure all that is due to Ricoh saving money on the Fujitsu chip they are using.
The model they are using is most likely not capable of doing all these things at the same time.
It seems like Ricoh are prioritising a competitive price for a stills camera over having great video capabilities.
Personally, I don't fault them.
Ideally, they'd offer a stills photographer only camera (without a mic and headphone jack and the corresponding electronics) and a video-capable model as an alternative. Then the people interested in video could sponsor all the investment that is necessary to support good video. Note that Panasonic are charging for the firmware upgrade that makes the S1 more capable in terms of video specs, so it's not like anyone hasn't thought of such a two-tiered approach before.
Just because other manufacturers let everyone pay for video specs that some don't use at all, shouldn't mean that Ricoh has to play the same game.
BTW, I guess that limited processing power is also the reason why SR isn't available during video anymore.
I think the explanation given by Ricoh that the movement is audible is pretty terrible. Serious video users employ external mics anyhow and at worst one could just dub the footage with a soundtrack and only use the SR for such shots.
Since SR appears to be working fine during live view, it probably isn't a matter of SR working well only for "one time shot actuation", hence my assumption that there isn't enough processing power to handle video recording and SR at the same time.
Originally posted by richandfleur ... so it would be nice to see it get some imporved AF tracking. Must be some reason to add a thumbstick control?
The joystick is probably there to support navigation to a larger number of AF areas. There is a wider "intake" for the AF module which suggests that AF coverage of the frame will be much larger than usual (there is no reason it couldn't be as large as the AF coverage of the K-1 in APS-C mode). It would make sense to accompany such an upgrade with an increase of AF areas.
This does not necessarily imply better AF tracking (AF-C) performance. However, one would hope that they were able to make advances in that area.
Originally posted by richandfleur My only issue with it would be the number of times I've taken group photos and found the hedge in the background is in focus and the people I aimed at are not.
Have you tried using the "Spot AF" mode? It uses a smaller AF area than the "Single AF point" mode (although it doesn't look like that in the viewfinder). This may help with precisely aiming at a person.
Originally posted by richandfleur Just not sure how you bring these mirrorless/mobile phone type smarts to the DSLR focussing system approach?
The Nikon D500 already supports face detect while one uses the optical viewfinder. They are employing a metering chip that has a higher resolution than what is needed for metering alone. The current 86K metering chip in recent Pentax cameras won't do the trick but the D500 chip only has ~1.5 more resolution and there are cameras with even better resolving metering chips, so it seems at least possible from a technical point of view for Ricoh to support face detection (even when not using live view).