Robert Collins

Impressive specs.

I still don’t get the absence of a rear wheel. This seems to be pretty basic on any enthusiast camera. Am I missing something?

Yep.I will definitely take 2.7k/60p 10 bit over 4k/60p 8 bit any day.Hopefully the codec is much improved over the way that the Mavic pro handles sharpening/NR. I have done hyperlapse with the Mavic Pro before but it is somewhat limited by the flight time and the need to do a prerun in waypoint mode before. Also the app on my Crystalsky limits photos to once every 10secs as opposed to once every 2secs on the smartphone. At once every 2 secs you are looking at a 1 second hyperlapse for every one minute of flight time. The rumored price tag for the Mavic 2 Pro is US$1499 for the basic kit. That is a 50% price increase over the Mavic. While this sounds a lot I think it is pretty 'fair' based on the rumored specs.

These specs. Video af? from where?

Great specs. Although note that 60p is limited to 2.7k rather than 4k as some people expected. I am pretty sure that the 'photo timelapse' you refer to is 'hyperlapse'. I think we are going to get a 'hyperlapse' mode because there are two hyperlapses in the promo video. You can actually do hyperlapse with the existing Mavic Pro but it is a little complicated and very 'klunky'. Also excited by APAS and Activetracking 2.0 as well as a possible gimbal. My expectations for the Mavic Pro 2 are extremely high but DJI rarely disappoints. I expect the price to be higher but hopefully not insane.

The M43 sensor dates back further than M43. Olympus had the 43 system which rather encompassed the 'small sensor, big camera' notion which compared to 'big sensor, small camera notion sounds pretty silly although some manufacturers still try it. Their arguments included that the 4:3 ratio made more efficient use of the lens image circle (I have to give them that) and that light from the lens image circle would be more perpendicular from a short flange distance resulting in sharper and brighter corners in the image (again some sense here.) In my view they made two miscalculations. One rather obvious. 1) They probably didnt think that the cost of sensor silicon would fall as fast. It seems pretty clear though that sensor costs would fall (and there is only one in a camera) while lens costs (a variable) would remain pretty flat. 2) The other problem was near impossible to predict (I would have thought). That the emergence and dominance of even smaller sensor, smaller, cheaper (?) smartphone cameras would undercut them from below. So in effect M43 has been squeezed by FF at the top and smartphones at the bottom. The fact that M43 is regressed to a small sensor, big camera approach rather indicates exactly how badly they have been undermined by smartphones. If you buy a really small M43 camera and put a really small (but slightly slow) zoom lens on it (which in terms of size and cost has an enormous advantage over FF) what you find is the image equality isnt much better than a smartphone.

Seems to me Panasonic are treading increasingly precariously on shaky ground. A lot of their decision making seems to be dubious to me. One example is their tepid follow up to the LX100 (described by DPR has 'possibly the best compact camera ever made'). Another would be the GH5s which took away Panasonic's class leadiing ibis and didnt resolve their video af problems. Once you take away ibis and admit that Panasonic's cant really video af you might as well buy a BMPCC 4k with its fatter codecs and bitrates at a lower price. So Panasonic ends up being squeezed by Sony at the 'techy' end of video that the likes of Black Magic at the 'Cinema' end of video.

I suspect the 'HDMI' cable will be a mini HDMI to female fullsized HDMI cable.

One possibility here is that Sony is not planning to come out with a RX1riii (and are handing the line to Zeiss in return for some moolah). So essentially you get an RX1riii in Zeiss clothing. It makes sense to me in terms of branding...

They do have a large user base - a market share of around 23% of ILCs - which is pretty much the same as all the non-Canon competitors put together. I think the main goal with their first product is to get the existing user Nikon user base on-board with mirrorless. The only users they are hoping to 'switch' are those with a large Nikon lens collection that have switched 'for mirrorless'. The perquisites for this product to be very successful with the existing Nikon base I would think are: 1) An adapter for F mount that works far better than existing F mount to E mount adapters. A pretty low bar. Nikon should achieve this 2) IBIS as good as OIS for stills. Again I think they will do this. 3) AF for stills with Z mount lenses that pretty much matches that of their DSLRs I think that video specs are likely to be an afterthought for Nikon. I am sure the ospdaf and ibis will mean that video specs are better than on the DSLRs but it would be heroically optimistic to believe they are going to be class leading. But we will find out tomorrow.

I always think that when M43 users see the need for extremely fast, large, heavy and expensive lenses, they are missing the point. BTW, the Zuiko 35-100 f2 weighs 1.65kg which is more than the Sony 70-200 2.8 GM. You see what you are really asking for is a bigger sensor.... https://camerasize.com/compact/#482.460,777.639,ha,t

I think the A7iii v say GH5 is a tough comparison. The A7iii is a hybrid camera (which I would loosely define as a stills camera with the videocentric features thrown in for free) while the GH5 is a videocentric camera where you pay a considerable premium for its video features. It is pretty obvious that the biggest trend in mirrorless is the convergence of stills and video. Video is an important kicker for hybrid purchasers but isnt something that many want to pay a premium for. Which brings us to the A7siii which is clearly a videocentric camera line (especially as it will almost certainly add nothing for stills over the A7iii and be priced at, at least a US$1500 premium.) Personally I think that Sony should either rethink the A7s line (or possibly scrap it altogether.) By rethink - start with the implicit assumption they are selling a video camera. So make the body bigger (videoguys tend to like that or at least accept it.) That would make heat dissipation far better and leave room for better and more robust external connections (such as a fullsized HDMI.) It might also leave room/scope for a variable electronic ND which is something of a holy grail feature for this sort of product. Of course that would cannibalize their cinema line up but I think that is going to happen in any case when the hybrid market essentially has still photographers subsidizing video features.

Actually Fujirumors specifically states that the XT-3 will 'not' have IBIS. Sort of messy spreading their best video features over two different models,

M43 offers both limitations and advantages for video. However, their C-AF for video is totally unacceptable when compared to the competition and if they dont sort this out properly, I think the will continue to lose market share. I realize a lot of people here dont use or care about af for video but the average hybrid buyer will take it if he can get it. Panasonic has absolutely no excuses - they were the first company into mirrorless. I dont see a future for Olympus in the camera business. They are forecasting to lose US$60m this year on sales of US$600m (their 8th loss in the last 10 years.) To put this in perspective - Olympus will lose US$150 on each of the 400,000 ILCs it expects to sell this year. This is not a business - it is a charity. Sony has really stirred things up by releasing the A7iii - with such a strong feature set - at US$2000. They pretty obviously did this to position themselves pretty aggressively against the new competition they expect from Canon and Nikon. However all the other mirrorless manufacturers have been moving relentlessly upmarket for the last few years. According to CIPA the average shipped price of a mirrorless ILC has doubled in the past 6 years (while the average price of a DSLR has increased 20%). But a side effect of Sony's US$2k FF is that it somewhat pulls the rug out under the US$1500-US$2000+ market for ILCs with smaller sensors (including Sony's soon to be released a6700.) Still photographers tend to be sensor size snobs. Meanwhile I think Canon will continue to do well at the bottom end of the market, because it is the only manufacturer who focuses there. Mirrorless made up 36% of ILC shipments in the first half of 2018 - good chance it will be over 50% in 2019.

Have you tried HLG on the Sony A7iii? Just wondering. You mostly use Cine4 right? I used to use Cine4 but ?I find it can be problematic with highlights (in terms of easy to blow and hard to reclaim in post (although it does amount to user error.)

I think we have to be fairly careful when we go down this sort of path. I am 100% behind the concept that we should give visionaries room to breath if we are going to progress but that doesnt mean they should be immune from criticism. My personal opinion of Elon Musk is that while he is a clearly a visionary (and more importantly is incredibly adept at getting financing to turn his vision towards reality) he also skirts the borders of 'vision' and 'hubris'. Take the fact that he launched a 'Tesla' into space. Fabulous but should we celebrate this as a 'hubris' or the 'pinnacle of human achievement'. And before anyone says achievement of anything at is worth celebrating, I would say that it is almost exactly 50 years since the US achieved landing the first person on the moon which largely proved that many things arent really worth achieving in the first place. But we have a more significant problem. You see I seriously dont have any problem with Elon Musk doing exactly what he pleases with his money (on acid or not) rather like I do not have any problem with Trump being rather eccentric (with an unhealthy obsession with his daughter) - nbut I dont think it is wise to make him President of the US. So once, Elon Musk starts using other peoples money - he is CEO of a US$50bn public company - he has to be held to certain public standards or at least within the Law. Once someone is managing your capital 'who cares?' or 'he was probably on acid' just doesnt really cut it. Personally, I think there is a serious disconnect here. A 'visionary' who is essentially looking say 50 years in advance, is probably uniquely unqualified for running the day to day operations of a public company or being CEO of a company reporting quarterly earnings

Looks pretty good to me. The control and button layout looks remarkably A7x to me but that isnt a bad thing. I much prefer Nikon's placement of the PASM on the left shoulder compared to Sony's placement. The LCD looks very nice and doesnt seem to suiffer from the massive Sony bezels. I find the absence of a control wheel on the back rather strange.

My experience too. You get reasonable DR, reasonable latitude to push files and colors and you can shoot down to iso125 which makes switching between stills and video a breeze.

I am looking forward to next Thursday's Mavic 2 release. We do know a couple of details courtesy of Argos 1) Flight time increased from 27 to 31 minutes over Mavic 1. Range increased 1km to 8km. 2) Mavic 2 will have active tracking 2.0 which will preumably be much better than active tracking 1.0 (which I havent found to be particularly useful. 3) Omnidirection obstacle avoidance sensors 4) Mavic 2 will have APAS. APAS was introduced on the Mavic Air. This looks great - demonstration here... I am sure we will get some more flight modes. A 'hyperlapse' mode would be awsome and save a lot of time. Another possiblity is that DJI will also announce a small handheld gimbal to attach the detachable camera unit. We have seen drawings and what looks like a photo.

Sensible guy gets overexcited and dissolves into cliches.... home run game changer 24 bit hit it out of the park revolutionary WOW factor

It's interesting what the DP is quoted as saying about the lack of 'rolling shutter' and 'quasi global shutter' being so important. I am pretty sure that this will be the key feature of the A7siii with its stacked cmos tech and what is likely to make the camera pretty expensive (US$4,000?)

Well yes and no. Assuming that the 'pro' is a 1 inch 20mp sensor that shoots 4k (8mp), it is quite possible that you can zoom 1.4x digitally without any loss is resolution. If you shoot in 4k and zoom in post you are reducing the resolution.

The Argos catalog that released the 'leaks' on the Mavic 2 (Pro and zoom) implies that the 'dolly zoom' is restricted to the 'zoom' version which is expected to be a 1/1.7" 12mp sensor with a 2x optical zoom. Seems a shame as the 'Pro' is expected to be a 1" 20mp sensor which has plenty of on sensor 'zoom' capability (at 1080) to also do a dolly zoom.

I am totally onboard as to why pro photogs are very conservative in their camera choices - I just dont get why camera companies are so conservative with their tech. I am sure that pretty much everyone has taken a photo of someone with their camera and then shown it to them on their lcd. They squint at the LCD (which hardly matches a first generation iphone) and has even bigger bezels - it aint much of a viewing experience. Is it so much to ask on a US$3000 camera for an LCD as good as a US$200 cellphone. Oled? pinch to zoom? scroll photos with a swipe? it isnt 2005. And sending a raw file to my phone shouldnt be asking too much?

I think it is a heroic leap of logic to conflate 'Samsung may well be producing a FF BSI CMOS sensor' to the likelihood they well re-enter the digital ILC market. Producing high end CMOS sensor for dedicated ILC cameras makes a lot of sense, becoming a digital ILC camera manufacturer (again!) doesnt. According to Thom Hogan there are precisely three manufacturers of large CMOS image sensors for dedicated digital cameras - Sony, Canon and Tower Jazz. http://www.dslrbodies.com/newsviews/nikon-2017-news/july-2017-nikon-news/confirmation-bias-and-image.html Once you rule out Canon (in-house, pretty ancient tech) and (at the moment) Towerjazz (busy upgrading Panasonic plants) you are left with Sony. Forget Renasys (acquired by Sony) or Toshiba (acquired by Sony) or Aptina (acquired by ON Semi) or Panasonic (acquired by Tower Jazz). Making large CMOS image sensors isnt easy - from what I remember you get something like 20 FF sensors, 100 M43 sensors or 1000 smartphone sensors on a 300mm wafer - which makes yields tricky the larger you go. I very much doubt that any camera manufacturer is enamored about buying their sensors from 'Sony' who is pretty much the most aggressive competitor in 'their' space. (Nikon appears to be looking towards Towerjazz but that is a year or two away.) Panasonic/Olympus seems to be working on Sony crumbs (is it really their choice to not have ospdaf?). So for a major very high tech player like Samsung to come in and offer large sized CMOS image sensors makes a lot of sense for Samsung and would likely be welcomed with open arms by pretty much every dedicated camera manufacturer (apart from Sony).