Dirk Gently

Thanks indeed for your insights on this. I'm happy to stand corrected on Canon's Cine Raw Lite, not least since it expands the number of codecs in which I can be confident of shooting without causing any ripples in editing and grading.

This is apparently what Canon have done... After all the complaints concerning the 1.74x crop in 4k when the EOS R was released, Canon admitted that that was the best they could do at the time - I have copied below a section from a Canon UK Q&A that sought to explain away the huge crop. (They were limited to one-to-one pixel mapping, with a 4k 'cut out' from the middle of the sensor, as you rightly put it.) But it appears that Canon have now advanced their tech to do away with a crop and in the larger 1DX form factor, which allows for much greater heat dissipation, at least if the reported Canon France version of the 1DXIII is reliable ( https://www.mac4ever.com/actu/147834_avec-l-eos-1d-x-mark-iii-canon-passe-a-l-heif-et-a-la-4kp60-10-bit-raw-interne-sans-crop ). There is inevitably room for misunderstanding when details are being relayed via calls to groups of journalists in different countries, but I think it's fair to say that the 4k30 at least will have no or only a very modest crop (e.g. a crop that gives 1.5 times over-sampling from a 28MP sensor). "There's a lot of discussion online about the 'crop factor' when you shoot 4K video on EOS R. The lens's field of view is cropped by a factor of 1.74. So where the maximum 16:9 still image you can capture on the sensor is 6,720 x 3,776 pixels, the 4K video is a slice in the centre of that, 3,840 x 2,160 pixels. Let's be clear: this is 4K resolution, but the complaint is that this is not using the full field of view of the lens. "This is down to the technical limitations of the sensor and the image processor. We could have tried to pursue full-frame 4K video capture, but it would take a lot of processing, for example, to scale and resample the capture, and this would risk causing the camera to overheat and shut down prematurely during video recording. We didn't want to risk that sort of instability. With the hardware we had available or in the pipeline, this is the best balance we can achieve at this time to deliver 4K capture." source: https://www.canon.co.uk/pro/stories/eos-r-system-faqs/

Perhaps I was not sufficiently clear when t I referred to the 'friendliness' of the codec. Few have found Canon Cinema Raw Lite friendly to work with in the sense I intended, which is to say that it poses a heavy load on the processor and is not straightforward to grade (unlike the XFAVC codec produced by the C300 Mk.II, say, which most find very agreeable). Of course, ProResRaw is too new to feature in any but Apple's own fcpx to date. But it has been designed as a particularly friendly codec to work with. I'd be very surprised if Apple didn't deliver on this promise (it's the sort of thing their good at). Of course, this is no reflection on your own experience with Canon's Cinema raw lite - if you find it good n friendly then all well n good.

Main positives re S1H: 1. Near-bullet-proof AF (Canon and Sony's video AF are coalescing in being as good as each other, only in slightly different ways) 2. Internal Raw (S1H raw will be external to Atomos) Negatives re S1H: 1. No IBIS [though Canon are said to be working on IBIS for an EOSR (X) version - who knows when this will come - in combination with lens OIS and 'digital' stabilisation]. 1DXIII will have no IBIS; I only make this footnote in respect of the likely EOS R equivalent of this DSLR. 2. Canon's Cinema raw lite (as per C200) is not a NLE-friendly codec; the ProRes Raw that will come from an S1H feeding a Ninja V (post the planned S1H ProRes raw upgrade) is likely to be much more pleasant to work with. A couple of posters have asked about the crop in 4k. Given that the 4k crop of the 1DX Mk.II is 1.3, we can be certain that the 1DX Mk.IV will have a crop of 1.3 or less - it would be so wantonly alien to Canon's modus operandi to do something so disruptive on their 'most pro' body like causing pros to have to buy a whole new set of lenses by switching from 1.3x in an existing model to (say the crazy) 1.74x of the EOS R/5DIV in the next version of that same model. For me, the fascinating comparison - in terms of positioning - is with Sony. The Panasonic S1H is a video-centric beast of wonder - very much the one-off space alien at the current time (and plaudits to Panasonic for producing such a model). But where is Sony, the disruptor-in-chief or yore? In the last 6 months they have produced new versions of their A9 and A7RIV cameras, both of which could have featured 10 bit (at least to a Ninja V) and more in terms of upgrades to the video specs. Even if the compact dimensions of the A7 bodies meant that heat dissipation precluded internal 10 bit 4:2:2 at a decent bitrate, they could have provided this externally. At the same time, the previously PR-savvy Sony let rumors run wild about the new cameras, such that the (real) improvements in other areas were overshadowed by the failure to provide even half of the expected video bit-depth, sub-sampling and bitrate improvements.

Very clear explanation - thanks a lot!

A different topic from the recent ones, but still very much within the domain of the thread overall... The S1H has a dual gain structure, with two native ISOs, of 640 and 4,000. Probably my question stems from a misapprehension on my part but, in asking it, it is likely that someone with much more technical insight will be able to put me right! Does the S1H's lower native ISO of 640 mean that anything taken at an ISO lower than 640 (whether stills or video) will be inferior compared to its having been taken at ISO640 (and with an ND filter used, say)? In effect, I am asking whether the 'lowest native ISO' of 640 means that anything taken at lower than ISO 640 will have been subjected to digital gain reduction from an 'actual' ISO640 (e.g. you set to ISO320 but the S1H is actually shooting at ISO640 then reducing the exposure by one stop digitally). The latter would seem so odd (in terms of quality of shots at under ISO640), yet I can't understand on the face of it how a dual gain such as this can function otherwise...Unless there is additionally a true 'base ISO' of 100 (not mentioned)... Any thoughts welcome!

Thank you; this is a lovely film. The end credits detail each lens and setting used. Since there is no further qualification of the settings used, may we conclude that this was shot internally in 8 bit 4k (not in N-log 10 bit externally)?

I think there are many users in a similar position...in many ways the Z6 is a good fit (particularly for those of us with lots of F mount glass) but Nikon's effort seems to have been disjoint...the battery grip is only about to be released (having been announced as 'coming soon' at the time of the Z6/Z7 launch in summer 2018), the ProRes Raw is delayed, it has transpired that AF and IBIS when piping 10 bit externally are much less responsive... I agree with you that, overall, the Z6 was a fine first effort in FF mirrorless from Nikon. Yet the gaps and inconsistencies make you wonder about the degree of commitment from the very top of the company (not least since Nikon is the one maker without a high-margin cinema line of video cameras to shield). Such a 'rookie error' as omitting any metering gauge from Live View/video makes you question their internal processes, too...doesn't someone notice such a bizarre omission at some point during the new product commissioning process??! And yet...I like very much the quality of the 10 bit N-log images that I've seen. The 8 bit internal 4:2:0 looks to have quite harsh highlight rolloff and limited DR (even by 8 bit standards vs other FF cameras) but there are occasions when it's sufficient, and low light is very good indeed.

Reviews of the Z6 when feeding 10 bit N-log to a Ninja V (or similar external recorder) suggest that AF slows down a *lot* when used in this way; IBIS performance also suffers (versus internal 8 bit 4k recording, as if the processor has its hands too darned busy with the 10 bit N-log to have much capacity for managing AF and IBIS). However, the Atomos CEO recently 'leaked' news that a near-future firmware update will see improvements to the Z6's AF (see, for example, https://***URL removed***/forums/thread/4427659 ). Is it likely that the impending firmware to which Atomos is referring will see AF when exporting 10 bit N-log improved to a level comparable to internal 8 bit recording?