Andrew Reid

Download now for just $19.99 ($29.99 usual price) The EOSHD C-LOG and Film Profiles Pack is now available and works in both video and stills mode - on all Canon DSLRs. Crafted using my Canon 1D C as a development camera, the EOSHD Film Profiles pack installs "Canon LOG" to cameras previously without it, plus a range of film simulation modes to DSLRs such as the 5D Mark IV, 5D Mark III, 1D X Mark II and T2i, etc. Read the full article

All hail the democratising C700! Read the full article

That is correct. However the main point of C-LOG for me is the LUT workflow. If you can get a custom LOG-style profile that is as close to the real LOG curve as possible it gives you that. Very big advantage and would make me trade my 1D C for the 1D X Mark II. I really want the Dual Pixel AF.

I am working on one of these as well. The 1D C is a great test bed as it as the real thing, Canon LOG The dynamic range won't be as good as real C-LOG but the main reason I like LOG is for the colour grading and LUT possibilities it gives you in post, whereas standard pic profiles are baked in. If you're going to have a faux LOG profile it has to give good colour and work with most LUTs already out there for Canon LOG - even if that means going less flat and sacrificing some dynamic range.

Yes I agree it's interesting to look at the signal to noise ratio improvements. The problem I have is when these are hyped as making a 'dramatic' creative difference to images, which is what the DPReview piece implies all the way through.

Pixel peeping is sometimes relevant, sometimes not. But important creative differences can be very subtle, look at Shane Hurlbut's comparison of the Leica Summilux vs Cooke cinema lenses. They do add up to a big difference creatively. How dramatically the low light advantage over the 5D Mark III translates into video mode remains to be seen. A 1:1 pixel crop in 4K does not leave any room for the clever noise reduction or oversampling that the A7R II Super 35mm benefits from. I do hope for a nice 2K image out of the 5D MK IV but only if Magic Lantern can give us raw, otherwise I'd rather stick to the Mk III even if it is a tiny bit softer. Again this applies to stills. No Canon LOG or RAW on the 5D MK IV as it stands so don't expect much improvement to video DR. Sigma 18-35mm is a beautiful lens, the crop is still a huge shame though as I love full frame and am heavily invested in FF lenses. Sigma 50-100mm F1.8 would be great on the 5D Mk IV for punched in shots - the crop won't matter then, it would even be an advantage for telephoto shots. The UK price is even worse - £3600? May as well get a 1D X Mark II! Only if Magic Lantern bring raw to the 5D Mk IV will I consider it, and even then it is highly unlikely to be 4K raw with the lack of CFast 2.0 slots. But I agree with you overall... Magic Lantern 14bit full frame raw in 1080p with dual pixel AF? I'll take 10!

No you wouldn't do it with any camera so why are we even considering the DPReview studio test as a true dynamic range test? It's more a technical measure of the sensor than a useful real-world test.

Yes looks to be, plus the recent XC10 firmware update to speed up AF a little bit more. Pathetic and lazy.

Who in the real world pushes their shadows 6ev anyway? The whole image just ends up looking like puke.

There is no way the C700 has a compelling feature that will make people suddenly drop their Arri and REDs And the XC10 already had Canon LOG! XC15 just adds 2 pro features and calls it a day.

It's not a "DRAMATIC" (their words) or MASSIVE (your word) improvement though! It's tiny! The 5D Mark III has 5 year old sensor tech, is it a dramatic surprise they have managed to squeeze out half a stop more dynamic range in 2016 with the mark 4!? And the 70D has a fundamental sensor architecture / image quality that you can trace all the way back to 2009 with the original 7D and T2i So DPR hyping the 80D and 5D Mark IV dynamic range improvement is just that... HYPE

Should be interesting but prepare to be pissed off... XC15 is probably merely an audio port C700 is probably $50k And a 1D C Mark II is probably AWOL But hey, I await to be pleasantly surprised

In the studio test at least, the skin tones have gone to hell on the 5D Mark IV shot in the shadows too. Which is why - for example - exposing for a window behind your actress and then pulling her out of 6 stops of dark shadows is a really really bad idea Anyone interested in designing a REAL world DR test for EOSHD? I can start my own comparison catalogue

Wow... Unbelievable.

"Dramatic dynamic range improvement" "Back when we were briefed by Canon about its update to one of the most popular line of digital SLRs in existence, Canon told us that dynamic range was one of the top concerns of current 5D-series owners." "A major leap forward" https://***URL removed***/news/3229755227/canon-5d-mark-iv-brings-dramatic-dynamic-range-improvements-to-the-5d-line ... So let's take a look at the reality behind the hype... Oh hey it's their own test scene... +6EV shadow recovery on a shot exposed for the highlights Samsung NX1 has it for breakfast The A7R II turns up and has it with milk and tea And the D750... 2 years old... Mentioned NOWHERE in the DPReview article, is an order of magnitude less noisy at +6EV recovery All in all meaning the 5D Mark IV has less dynamic range than the Samsung NX1. I double checked... Yeah it does. And that my friends is why in a nutshell I stopped being a writer for DPReview.

Indeed because it could not output the full 18MP in any mode. You couldn't make use of it. But on the GH5 it can and we need to make a distinction between the actual total resolution of the sensor and what the different aspect ratio windows of that give you.

However you look at the maths, unless the sensor is letterbox shaped, the 43rumors spec of 20MP total resolution for the CMOS sensor do not allow you any kind of 6K mode, be it stills or video. So the total resolution must be higher than 20MP, preferably in the region of 24MP.

New info I am told they are indeed using a multi-aspect sensor. It measures 18.9mm x 13mm in approx. 3:2 aspect ratio 20MP stills but the total resolution of the chip is 24MP 24MP 3:2 is approximately 6000 x 4000 (just like the Sony A6300 in 3:2 stills mode). Then the 6K photo mode is 3:2 with full sensor readout up to 30fps burst. The normal stills mode is a 20MP 4:3 crop of the sensor. This is why 43rumours say 20MP sensor. The 4K video is a 16:9 crop of the sensor using the full 6K width but downsampling on the chip. Absolutely take this is a rumour until confirmed. But it makes total sense to me... There is *no other way* of getting 6K photos from a 20MP 4:3 sensor like that featured on the GX8.

6K is always measured horizontally so the sensor would have to be at least 6000 pixels wide to do 6K resolution. So to have the 20MP sensor appear alongside a 6K stills feature STILL does not make any sense at all. A 3:2 sensor would have to be 24MP A 4:3 sensor would have to be 28MP Very unlikely for Panasonic to use a video or cinema aspect ratio sensor in a stills camera! So the mutli-aspect sensor is the only way they can do it, but still the rumoured specs just don't seem to quite add up to 6K!

Coming to a cinema near you... But probably NOT destined for a cheap camera near us! http://www.canonrumors.com/canon-announces-cmos-sensor-with-global-shutter/ I am pretty sure they will use this in high-end industrial machine vision rigs at first, before a slow roll out to broadcast cameras begins. But the pixel structure, design, patents, general tech looks very exciting, it seems they have solved the global shutter issues with dynamic range and noise completely, even turning the situation on its head so that the new pixel structure results in increased full well capacity for better DR!

Here's how I think it would work (scroll to half way) http://www.eoshd.com/2016/08/exciting-true-panasonic-gh5-sounds-like-bomb-including-6k-20mp-sensor-something-not-add/

So Micro Four Thirds has a 4:3 aspect ratio for stills But in the past Panasonic have had multi-aspect sensors Maybe this is another multi-aspect sensor That would explain 20MP stills in 4:3 But in 16:9 or DCI 6K mode it could use a wider area close to 6K wide?

That is IF the specs are true. The 20 megapixel Panasonic GX8 sensor was 5184 pixels wide How are they getting 6K from a 5K sensor? Correct! Maybe their sensor is a new 16:9 aspect ratio chip, 5760 pixels wide but I can't ever see that in a 4:3 Micro Four Thirds cam

Sony RX100 IV sensor has been doing 20MP 6K output for a while. If Panasonic can take their new 20MP sensor and back it onto some very good image processing I can believe the GH5 shoots 6K. which will make it absolutely unique for the price and in the form factor it is in. VERY exciting specs if true.

I see where you're coming from but... That is the same as saying H.264 wasn't designed as an acquisition format, yet now it's the basis for Sony XAVC and Canon XF codecs. As long as H.265 has the hardware encoder / decoder chips to support it there's no reason why it can't be an acquisition format and an edit-ready codec.