David Bowgett

The APS-C, 4k-capable camera closest to the G9 in terms of ergonomics would probably be the Nikon Z50 - and that doesn't have 10-bit video or IBIS (albeit it does have much better video autofocus). The A6400 and Canon M6 mk2 are also options in that price range, but again, neither of them has 10-bit video nor IBIS, both of them are pretty far-removed ergonomically from the G9, and the M6 mk2 has some of the absolute worst 4K you'll ever see.

From what I gather, the RF zooms are somewhat better than their EF counterparts. Not overwhelmingly so, though there's obviously a few which don't have EF equivalents such as the 28-70 f2. Primes are where the RF lenses really seem to shine; way better in terms of wide-open sharpness than their EF counterparts. But yeah, the 24-105 f4-7.1 really doesn't look too hot from the reviews and samples I've seen. A second-hand EF 24-105 f4L connected via the RF adapter would probably be a better option.

Between this and Olympus's announcement today, my first instinct was to wonder whether Panasonic was just throwing together a bunch of old parts into a new camera body so that they can use up their inventory ahead of abandoning the lower-end m4/3 market, if not m4/3 as a whole. But then you look at the number of clearly bespoke parts such as the EVF and microphone array, and I'm really not sure exactly what Panasonic's thought process was when they put this thing together.

Canon EOS 250D Nikon D7500

Even if it does turn out to have cropped video, on a 20MP sensor it would work out to the equivalent of an APS-C crop. Certainly not ideal by a long shot, but it'd be leagues better than the crops on the R and RP.

There were apparently some rumours that the R6 would use the same type of battery as the EOS RP. Turns out it actually uses the same type as the R5. https://www.canonrumors.com/the-canon-eos-r6-has-shown-up-for-certification/

The Panasonic S1 and S1H, and the Canon 1DX Mark II and III.

They never cut 24p from their 5D/ EOS R-tier cameras, so I don't think you've got anything to worry about on that count. Unless you're referring to true 24.00p instead of 23.976p, in which case I should think the R5 will have it, seeing how the 1DX Mark III has that feature (according to the spec sheet).

I wouldn't be surprised if it is - but if Canon have rediscovered their late 2000s/early 2010s mojo, then hopefully we'll be able to look forward to the same path that their VDSLR line took, from stupidly expensive (5Dii) to moderately expensive (7D) to actually pretty affordable (550D).

Canon's main weakness in that department has been that when it comes to flat image profiles, your choices have basically boiled down to C-LOG or nothing - or, on any camera that wasn't a 5Div, EOS R, 1DC or 1DXiii, just nothing (out of the box, anyway). I presume this thing's gonna come with C-LOG given the price tag, but it'd be nice if they had some Flat/Cinelike-D type profile as well, for when all you need to make are quick-and-simple grading adjustments.

To be fair, that seems to be fairly standard behaviour for focus peaking. Neither of my Panasonics display focus highlights in AF mode, and the Z6 I bought appeared to display the same behaviour in the roughly 45 minutes I used it before it exploded.

DPReview have just posted a follow-up video focusing on the A7iii. In short, the camera they tested wasn't defective; a second camera performed exactly the same. However... it also looks like the A7iii's audio performance can differ drastically based on the microphone that's being used. Some microphones end up sounding as bad as the initial test, but others end up in the S1H's ballpark. Not sure what's going on there - maybe Sony cheaped out on the camera's EM shielding or something? Also, based on what Jordan says at the end of the video, it sounds like they may have unwittingly tested the Z6 under optimal circumstances, and that with other microphones the results are much, much less rosy than their initial test indicated.

I was thinking of starting a PhD later this year, so I guess when I'm not editing I'll be doing groundwork and putting together a proper proposal on that. Fortunately, I managed to go into my former university's library and photocopy just about every textbook related to the subject area, which should keep me occupied for a good while. But if it drags on then I might have to, shock horror, actually buy some textbooks! If nothing else, I'm actually weirdly glad that the Z6 I purchased last month exploded less than an hour after I first turned it on, and Amazon refused to exchange it, offering me only a refund. Would have been annoying having a nice new mirrorless and being all but unable to actually make use of it. Plus now I'll be able to see if the Canon R6 might be a better buy after all.

I think absolute best case is that the R5 is priced similarly to the A7Riv, and the R6 somewhere just above the A7iii and Z6. And even the latter might end up being priced nearer the Z7, considering that pricing it much lower would pretty much instantly render the EOS R obsolete.

The DSLR version of the M6 Mark II is the 90D, which uses the same 32MP sensor. This uses the older 24MP sensor, which likely means that the 4K is cropped as on the M50 and 250D.

Yeah, it sounds like it's just the 250D/SL3 in a slightly more up-market body. If for whatever reason you want a Canon APS-C DSLR with any real concession to 4K video features, you really need to save up for the 90D.

Hopefully this section... ...means that the 8K video mode is actually going to be something usable, rather than something along the lines of the 1080p20 mode that Canon gave us with the EOS 500D way back when, or the 4k15 mode that appeared in the final few Nikon 1 cameras before the plug got pulled on that line.

DPReview have now put up some stills of the Z50's video modes: https://***URL removed***/reviews/nikon-z50-review/7 In a nutshell, overall 4k video quality looks to be about on-par with the GH5 and G9 (maybe even a touch sharper), which pretty much confirms that it's oversampling from a ~5.5k region of the sensor like those two cameras. Not quite as good as the 4k-capable A6x00 cameras or the X-T3(0), but still way, way better than the 4k video on the M6 Mark II. 1080p quality on the other hand actually seems a little worse than the M6 Mark II, GH5, and X-T3(0), albeit still better than the A6x00 line and Nikon's DSLR line.

Bear in mind though that a full-frame sensor has a 3:2 aspect ratio, so some of those vertical pixels will go unused when capturing 16:9 video. There probably is some slight oversampling going on, as the sensor's horizontal resolution is 4,240 pixels, but it'd only be in the region of a 10% oversample rather than 31%.

Weird how there seems to be this black hole when it comes to APS-C cameras and IBIS. Damn near every M4/3 and full-frame camera has it nowadays, but on the APS-C front there's only really the A6500 and A6600 that have it (plus some of Pentax's cameras).

Actually, if you're going to make claims like that, the onus falls on you to prove that's how Canon are doing it. Otherwise, the default assumption should be that they're running the sensor at 60Hz and using pull-down to get 24p (or technically 23.976p), because that's how damn near every video-capable ILC has gotten 24p in the past.

Canon almost never add new features to their cameras via firmware. The fact that they're doing this - and not offering up some lame excuse, such as bugs with the new DIGIC processors that needed ironing out, or that whole "licencing fee" thing - shows that they've finally realised they've made a screw-up of rare proportions.

Shame the USB-C port on these new models seems to be a data-only port. Being able to run one off a power bank would be pretty handy.

In fairness, how many sub-$1,000/£1,000 ILCs actually do offer 10-bit 4:2:2 HDMI output? The only one I can think of off the top of my head is the X-T30.

You would need a separate clock circuit for true 24.00p. However, the 23.976p that's more commonly used in consumer cameras is derived from a 60Hz base clock - in 60Hz mode my FZ2000 gives the option of "24p", 30p or 60p, but in 50Hz mode it's 25p or 50p only - so if a camera can do 30/60p, there's no fundamental reason it shouldn't be able to do 23.976p.