[Matsu]

Most people who compare digital to film output have to admit that, at least for enlargements and detail, digital punches a few sizes above its weight class compared to film. With the right lenses, at least two stops, or conventional sizes of film, and sometimes more. m4/3 equals or betters 35mm film. APSC/DX nibbles at 645 film, and 35mm sensors seem to do what needed 6x7 in the past. But that leaves a few people that may still want a larger than 35mm sensor. There's always someone who wants to print larger or see deeper into the dark, and what's coming off Sony's 50MP 44x33 CMOS is amazing in that regard.

A new from the ground up digital field camera platform from a mainstream manufacturer could be interesting in medium format digital, as one part of a mirrorless medium format system/mount. An attachment bellows with front and rear movements, electronic contacts and a set of compatible lenses... and if they can pull it off, on sensor contrast detect AF, if only for convenience? But would also need a really nice focus assist/peaking system on the LCD. Such a machine could replace large format cameras, at least 4x5, for landscapers, and find many happy users in the areas of food and product photography.

[Eugene]

My issue with Nikon continuing to ignore the professional video industry is precisely that Sony, Canon and Panasonic are actively pursuing it. Modern digital still camera hardware is inherently capable of video. Merely keeping up with the rest of the pack would at least help retain some of that middle ground.

[PB PM]

Nikon is a little strange when it comes to video. The D810 and D750 have come with decent video features (zebras, 1080p 60FPS etc), but stuck with 1080p as the market has started to transition to 4k. Why? The Nikon rep at Photokina this year simply said something like, "nobody really needs or uses 4k yet." Right, because having cameras that will be on the market for 2-3 years using last generation (1080p) video really makes sense, when in 2-3 years 4k will be the norm. Strange considering that like GPS, Nikon put out the first DSLR with video (D90, albeit only 720p 24fps with no controls), to now when it has taken them 4 years (D800-D810) just to get the basic 1080p stuff right.

Nikon needs a highly capable 4k camera out in 2015, not 2016. I'm taking D5 here. It needs to have everything, zebras, peaking, 4k with decent frame rates (24&30FPS), along with all the needed still photography stuff. That said, I don't think Nikon needs to make a dedicated video camera, not yet anyway. Should they start developing one? Hell yeah!

[Eugene]

Actually I don't consider better video a priority for the D5. Those shooters have very specific needs, and video really isn't one of them. They'll also likely be toting around extra bodies if they do occasionally want to shoot some video.

When Nikon implements 4K or higher video capture, it will almost surely be in a mirrorless product smaller than a typical DSLR.

[Matsu]

A Nikon 1 product with a proper fast wide ratio zoom could be an excellent video platform.

[PB PM]

Originally Posted by Eugene Actually I don't consider better video a priority for the D5. Those shooters have very specific needs, and video really isn't one of them. They'll also likely be toting around extra bodies if they do occasionally want to shoot some video. When Nikon implements 4K or higher video capture, it will almost surely be in a mirrorless product smaller than a typical DSLR.

So you don't think Nikon should put high end, professional video features into their high end professional cameras, after saying Nikon needs to target the professional video market. If anything the type of people that will be shooting pro video, with a DSLR, are likely to be the people who have a D4 type camera, photo journalists (who are becoming more and more so video-journalists). They are the type of people who will have two camera bodies, one of which could be used for stills, while the other could be used for video. Makes perfect sense, in my mind, to put the best stuff into the best camera. Should it be the only camera that Nikon puts 4k video features into? Of course not. Should 4k first appear in the D3400? Nah.

Should Nikon make a dedicated video camera? I think I already answered that question.

Originally Posted by Matsu A Nikon 1 product with a proper fast wide ratio zoom could be an excellent video platform.

If Nikon ever decides what they are going to do with Nikon 1, that could work. I would not be surprised if the Nikon 1 bodies (AW series aside) disappears in favour of fixed zoom compact cameras (aka high end Coolpix).

[Eugene]

Originally Posted by PB PM So you don't think Nikon should put high end, professional video features into their high end professional cameras, after saying Nikon needs to target the professional video market. If anything the type of people that will be shooting pro video, with a DSLR, are likely to be the people who have a D4 type camera, photo journalists (who are becoming more and more so video-journalists). They are the type of people who will have two camera bodies, one of which could be used for stills, while the other could be used for video. Makes perfect sense, in my mind, to put the best stuff into the best camera. Should it be the only camera that Nikon puts 4k video features into? Of course not. Should 4k first appear in the D3400? Nah. Should Nikon make a dedicated video camera? I think I already answered that question.

I'm saying the market is huge and their future hinges on targeting the upper echelon of professional video as well as its middle-ground. A Nikon equivalent of the EOS-1D C is ultimately a niche product occupying the middle-ground. Maybe the D5 does get 4K+ first, maybe it doesn't. At the heart, it remains a PJ's camera and video is still mostly an afterthought. If a PJ wants to shoot video, there's nothing stopping him from using an A7s, GH4 or one of Panasonic, Sony, RED's dedicated rigs. If you have an associate, it means you get both photo and video at the same time...

PJs are going to buy D5s whether it has 4K video or not. Even most D810 class buyers aren't going to base their purchasing decision on whether the body supports 4K capture. People seriously looking into video are perfectly willing to invest in a second, more video specific camera like a GH4 or A7s. Nikon can therefore kill two birds with one stone by introducing mirrorless and 4K in the sub-$2K price range at the same time.

Meanwhile in the film/cinema world, a D5 won't make a dent. That's what I meant by saying the specific combination of 4K and a D5 is a low priority.

[Dorian Gray]

Are video people on the ground asking for 4K? The data requirements would be breathtaking. Even high-end TV series use Apple ProRes (PDF) instead of raw video to save space, so I have to wonder whether putting 4K in amateur cameras makes practical sense (of course, whether it makes sense may not matter to Nikon if buyers are asking for it).

I suppose you could just compress the heck out of the data in-camera with H.264 or H.265 (the latter is used in the iPhone 6 for FaceTime over mobile networks), but that greatly limits your editing possibilities – it’s much worse than editing JPEG stills.

[Matsu]

There's a series of very interesting DPR post Photokina panel discussions on YouTube. The small camera big sensor discussion features a Sony, Fuji, Panasonic and Lytro guy, along with a couple of spokespeople-photographers, many concerned with moving the right amount of data around for hybrid video/still functions.

There's two parts to the 4K question. One is as a delivery medium - it finally brings us into good enough territory for large still image displays. We'll get better of course, 5,6, and 8k in the future, but there's something to seeing a large 55-80" wide still image fully illuminated on a screen. It has just enough detail to look impact-full at a large size. A screen is more forgiving, suddenly, 8MP is enough to display large images to an audience that will never look at prints anywhere near that size. And for pulling of still frames for 8x10, or newsprint, or web, it's got enough resolution to do a credible job - depending on conditions.

[Dorian Gray]

I watched those DPReview videos as they were broadcast live. Several of them were indeed interesting.

Data rates for 4K video are pretty scary. Consider the extreme case of 4K video at 12-bit colour depth, full 4:4:4 colour, and 60p. That’s 4096 × 2160 × 12 × 3 × 60 = a staggering 19.1 Gb/s. So an hour of such uncompressed video would take up in the region of 8 TB of space (or about all the storage I have at home!).

Compression schemes are obviously of paramount importance here (even for the likes of the BBC). Even so, I doubt many amateurs are realistically set up to deal with the backup and archival of 4K video, even if they can just about cope with editing it.

I don’t really do video at all, so perhaps I don’t count, but I’m in no rush to jump aboard the 4K train. The advantage in picture quality is pretty useful (though not as useful as the jump from SD to HD), but it’s not yet practical (I have no 4K display, for example). And if anything, the average display size has gone down recently, with smartphones and tablets replacing desktop computers.

Of course 4K is coming whether I’m on-board or not.

[Matsu]

In some ways the discussion around lens technology was even more interesting. I had taken a lot for granted regarding how lenses have to be designed for both stills and video or one versus the other, as well as PDAF vs CDAF systems. They almost all commented that changes in lens design were a major element for making mirrorless systems work.

[PB PM]

Originally Posted by Dorian Gray Data rates for 4K video are pretty scary. Consider the extreme case of 4K video at 12-bit colour depth, full 4:4:4 colour, and 60p. That’s 4096 × 2160 × 12 × 3 × 60 = a staggering 19.1 Gb/s. So an hour of such uncompressed video would take up in the region of 8 TB of space (or about all the storage I have at home!).

Not scary if you are working with a large number of video files already. Even 1080p in low compression formats eat drive space quickly. Cameras like the GH-4 are selling well enough, on the 4k feature alone to amature film makers. While most are currently downscaling to 1080p (because it is sharper than native 1080p capture), 4k itself has some other advantages already covered in this thread.

While I wouldn't buy a camera for 4k video alone, since I mostly shoot stills, I wouldn't mind having it available. I do find myself shooting video more and more simply because capturing the "desisive moment" with stills can be difficult in some situations. Video lets you work with slower shutter speeds, thus lower ISO settings, which can be beneficial. I still don't like pulling stills from video (too much blur IMO), the ability to do so is nice.

[Dorian Gray]

If 19.1 Gb/s doesn’t scare you, you’re a better man than I am. (I admit most of us won’t be working with bit rates quite that high even if we get 4K cameras soon.)

Unrelated to anything above, I just got my Nikon D60 back after a multi-year loan to a friend. She’s an artist, and she used this thing until the textured plastic was polished smooth by her hands. But it’s still ticking along nicely.

And you know what? It is surprising how little progress has been made in six years of Nikon SLRs. Sure, image sensors got significantly better, video (and live view) became possible, and LCD displays moved on considerably. But the basic operation of today’s Nikon SLR is nearly identical to the D60, with the same usability flaws. Nikon’s menu system is still the same old yellow, black, and white mess, lacking even the basic dignity of a proper font. The same crude seven-segment displays are used in the viewfinder.

Already with the D60, Nikon had solved the problem of unresponsive operation when taking a digital photo and reviewing it. So going from a D800 to a D60 doesn’t feel much slower (there’s a tiny bit more shutter lag and mirror blackout).

By contrast, since 2008:

• mobile phones have changed enormously (apps, etc.)
• mirrorless cameras have gained enormous capability (focusing speed, entirely new control schemes, scores of new lenses, etc.)
• whole systems have risen and fallen, e.g. the innovative Ricoh GXR, whose basic concept may yet prove prescient
• compact cameras have been transformed with faster lenses, bigger sensors, faster operation, and electronic viewfinders.

I can’t say what Nikon should have done differently, if anything, but it’s certainly understandable why many SLR users have adopted other ways of taking photos in those intervening six years.

Still, the D60 was already a capable picture-taking machine in 2008. I may just use it as my small camera while I wait for things to shake out in the mirrorless world.

[PB PM]

The newer cameras don't use the seven segment display, pickup a in store demo D810 or D750 and you'll notice the difference right away. The colour of the displayed info isn't green anymore either.

Yes the menu system is "old school", so what? It's usable, far more so than the pre-2006 cameras, and for what it does it works. Honestly the biggest UI changes I'd like to see is backlit buttons on all models across the board, not just the D4/D4s as if only people with such cameras dare to use them in low light conditions. Second, touch screens across the board, which would make accessing items in the menu much faster. I can just imagine not having to waste time scrolling.

[Dorian Gray]

I had forgot about the OLED data displays in newer Nikon viewfinders (I think that started with the D7100). So Nikon has made some progress there, after all.

The problem with the old-school menu system is that it’s slow – your touchscreen idea would speed it up a bit – and confusing, i.e. not very functional. I don’t think I’ve ever met a photographer in the wild who used Nikon’s memory banks. In fact, how can they be used? They’re basically broken. To change the camera’s setup you have to change memory banks in at least two places (Custom Settings Bank and Shooting Menu Bank – both idiotically labelled with the same letters A, B, etc. to sow more confusion), which has to be done in the menu system (there are no hardware buttons for this). Then, once you change a camera setting, that change is immediately saved to the bank you’re currently using. So you can never trust the camera to revert to a known state by switching banks. What’s more, you can give the banks names – e.g. ‘portrait’, ‘tripod’, etc. –  even though merely using the camera in one of those banks will override the settings in it and make ‘portrait’ or ‘tripod’ useless for portraits or tripods the next time you select it.

I bring this up not to show Nikon makes imperfect cameras – so does everyone else – but to show that Nikon has been lazy with development. Many of the things that are wrong with Nikon SLRs are unrelated to technology. We’re not waiting for radical new technologies to fix these memory bank problems. We’re just waiting for someone at Nikon to give an hour’s thought to how a camera should work.

[PB PM]

You know someone who uses the memory banks now, I use extensively for different situations, have since the first camera I got that had the option. To make them more usable I either access them via the info button or my menu. Would I like to see the banks improved? Yes. Do I want a chunky mode dial with C1-3? Not really, since I rather like the mode button.

Would I like the menu and interface improved overall, yes. What would I change? No idea. Once I have the camera setup I hardly use anything other than my menu, which really is the saving grace of the system.

[Matsu]

For all the talk of innovation, what's really happened in the last ten years? Both for phones and cameras?

For Phones: Smartphones have become better cameras both for still and videos. Phones themselves actually got "smart" and a cloud emerged that readily accepts (and makes available) content from said phones. They compete with a range of consumer products and services as a result, not just cameras, but MP3 players, retail media stores, books, magazines, even booking services, credit services, and legal tenders. There are a lot reasons for them to be with you all the time; unless you're passionate about making photographs, the same can't be said of cameras.

For Cameras: Film died as a mainstream platform and digital took over. Video got into everything, and after the failure of all but two traditional SLR systems, a generation of mirrorless systems has emerged.

But, apart from being smaller and new, how much innovation is there really in mirrorless? You could argue that most of it is aimed at catching up to the capability of SLR focus and viewfinder systems, which they will undoubtedly accomplish and surpass at some point, probably soon.

I find it interesting to read comments from camera makers that reveal what they believe is the best way to tackle a problem, for them.

The mirrorless makers are all selling "new-generation". They have to, they simply don't register as part of the existing generation. But that new generation is simply a smaller version of the old. Enhancements to post capture data management and sharing are as absent there as anywhere. Right now they're all working on a value proposition that says "good enough performance, smaller package". They call this a new-generation, but in many ways, it isn't.

Canon and Nikon are selling established systems. They should, theirs are the most robust, though the others are getting closer.

They need to make those systems smaller, though, and decide whether they do that with their current mounts or whether they can start all over.

For large sensors, AKA full frame, I see no need for new mounts. Take a look at what's out there. Sony's FE is actually too small for full frame. It shows up in some designs. Fast, wide apertures, are mechanically shaded by the narrow mount and short flange, they have to make wide lenses slower or make the lens physically longer (and more retrofocal) to keep the right performance. Nikon's mount is in fact narrower, but it's not as close either which let's it accept some faster designs more easily. Neither is perfect.

If Nikon simply drops the mirror, they have a few degrees of freedom to move the rear elements close to the focus plane, probably somewhere between 38-30mm is close enough to make Leica M sized primes without ditching the mount. They might need electronic diaphragms on some new lens designs, but we see these creeping in already. Keep the couplings on the camera mount for continuity - keep selling the robust system while moving it forward to a "new - generation" Does mount register need to be smaller? I doubt it, but the cameras certainly do.

Nothing has been done to significantly shrink full frame DSLRs. They only need to do it once to set the foundation for future work. Integrate all the electronic bits onto one board and shrink wrap the camera body. It would be virtually the size of an A7. And, while shrinking cameras isn't exactly a next generation change, it tells the future buyer that these aren't dinosaurs. Buy with confidence both the future and the past are covered.

[PB PM]

Looks like Sony is starting develop a type of sensor I was thinking about a while ago (a year or so), one that can measure light at the pixel level. We could be looking at, in effect, in camera HDR in RAW images, not multi-shot jpgs.

http://petapixel.com/2014/11/05/sing...ixel-exposure/

[Dorian Gray]

Originally Posted by PB PM Looks like Sony is starting develop a type of sensor I was thinking about a while ago (a year or so), one that can measure light at the pixel level.

They all measure light. This one is just claimed to measure it a bit better! I’ve never understood how that technology could work – it’s not enough to merely not clip; for it to be useful, you’d also need to know how bright the pixel should be, i.e. by how much your new technology retarded clipping – but I guess that’s the trick that Sony believes to be possible.

I’m still mulling over Micro Four Thirds. Gspotter, if you see this: do you have any problems with shutter shock with your E-M1? This issue of shutter shock is off-putting. Maybe it’s not as annoying as it seems.

[Matsu]

Is this an issue with mirrorless cameras not having enough mass to adequately damp mechanical shutters? The A7 seems prone to issues according to some reviews. If I recall correctly, in his review of the Olympus OM-D, Ming Thein thought that between the extra depth of field and sensor based stabilization, that the camera had a very generous shooting envelope. Surely on camera stabilization should be able to dial out shutter shock, unless I'm missing something.

[Matsu]

Well that shows how long since I paid attention to the market for a new camera in any serious way. Looks like there's an issue between 1/160-1/200 of a sec visible on some cameras depending on lens. Rumor has it that firmware updates have reduced the problem by introducing an electronic first curtain. curious.

[Dorian Gray]

Yeah, the problem seems to arise from the first shutter curtain slamming home at the end of its travel, which of course coincides with the exposure (since the second curtain hasn’t yet begun its journey or is still zooming across the focal plane).

The Panasonic GM1 and GM5 have purely electronic first curtains so they seem to be immune to this problem. But the servo-driven mechanical second curtain in these cameras limits the shutter speed to 1/500 s. If you need faster, you must use a completely electronic curtain mode which incurs other problems (rolling shutter artefacts and only 10-bit colour depth in raw files).

The risky shutter-speed range on cameras like the E-M10 coincides with my most-used shutter speeds, which is why this is so off-putting to me. If it was something that happened at 1/4 s or 1/4000 s I wouldn’t really care.

I’m not personally interested in the E-M1 (I’m looking at the E-M10), but I suppose Olympus’s solution to the problem is similar on all its cameras. I’m finding it hard to pin down the disadvantages of Olympus’s electronic first-curtain mode.

[Matsu]

There might not be any? I don't know. Does it work in continuous mode? Have any effect on recorded bit depth? Olympus might have a better implementation here, and also a better shutter, notwithstanding the shock at some speeds. I am curious if they simply couldn't program the IBIS to cancel it out - if it's present in a completely predictable way. Maybe it's hard because the shock happens exactly at the moment of exposure? Even a flash pulse would have some trouble freezing it out; giving a slight double edge image instead of a clear blurring?

[Dorian Gray]

Originally Posted by Matsu Does it work in continuous mode? Have any effect on recorded bit depth?

It doesn’t seem to work in burst mode, so that’s a limitation (though not one that particularly bothers me). Dunno about bit-depth.

It may be difficult to model the shutter shock and correct it with a programmed sensor-shift movement, because the shock appears to be lens-dependent, i.e. blur is caused by lens elements moving around (there being some ‘slop’ in autofocus lenses) rather than or in addition to the whole camera moving. To model that with useful accuracy you’d need to know the starting position of the elements, which may be impossible to know. Even then it wouldn’t be trivial (but correcting handshake via sensor-shift isn’t trivial either, and by all accounts Olympus managed to do that astonishingly well. This is the kind of problem Japanese companies seem well-equipped to solve).

Hopefully the next batch of Micro Four Thirds cameras will have electronic first-curtain shutters and no need for the little shutter-dance Olympus cameras do when using their ‘0 sec’ anti-shock setting.

[Matsu]

Hmmm. Lens dependent or resonance dependent? Maybe it's just visible at different speeds because some resonant frequency in the mechanism is best exposed at a particular shutter speed and degree of angular movement - it would naturally correspond to some focal lengths, because we'd typically look for certain shutter speeds to freeze motion, usually 1/x focal length. It could be about slop in the lenses too, I just don't know enough about that to guess.

I'm curious to see some more low light work from m43 camera shooters to see if the combination of depth of field, contrast detection, and stabilization helps in getting the shot when the light is really poor. Wide open, high ISO work is kinda poor on full frame DSLRs too, until you add a flash to freeze motion. So, maybe the smaller sensor isn't as much of a penalty if you just end up adding some light anyway.

[Matsu]

Just the other day I got back from Future Shop - the Canadian big box electronics dealer owned by Best Buy. I visited the Home Outfitters next door looking for a new coffee maker and just walked in on a whim to see if the Retina iMac might be on display - it wasn't - but I saw the camera display counters. On strictly aesthetic terms, it was the first time I was in an electronics store and generally impressed by the ILC offerings overall.

They look both smarter and smaller than entry level DX DSLRs, which look sort of bulbous and dumpy by comparison. The DSLRs may be great to actually hold and use, but they look old next to even retro inspired ILC cameras like the OM-D and Fuji X. The counter was well appointed with cameras to handle, batteries charged, low-profile theft deterrents attached in a sensible way that you actually handle the cameras, and a variety of kit lenses attached to the Sony, Fuji and Olympus cameras. The Sony FE cameras were under glass, without kits lenses mounted, so you couldn't really compare those.

Overall, I just got the impression that this Christmas might be something of a breakout season for "mirrorless" cameras amongst consumers - even if that only means they hold steady while other markets contract... The m4/3 and smaller APSC cameras are usefully smaller with the right lenses. Some are loaded with helpful features and strike a nice balance between the way people used point and shoots and the way we use larger cameras. You could actually use an E-M10 and 14-42 in full out arms out mode - you shouldn't, but you could.

[PB PM]

Having handled and fiddled with most of the mirrorless cameras out there (everything from several m4/3s, Sony FE, Fuji), I'm still not impressed enough with the overall packages offered. They all just seem to be 4-5 steps back from a comparably less expensive DSLR. Small is nice, but is it really the be all and end all of existence? As for looking smarter? Okay, if a piece of electronics can look intelligent...??? I don't think there is anything wonderful about the designs that makes them look better than the average DSLR. In any case all of them have major issues that would keep me from considering them:

1) Sony FE - poor interface, battery life, control layout, and overpriced slow lenses (when F4 lenses cost as much as Nikon/Canon FF F2.8 lenses something is wrong). I guess Sony needed to make the A7s to make up for those slow lenses.
2) Fuji - poor movie recording. They have interesting sensors, but it takes a lot of work and a bunch of different PP software to get the most out of them.
3) M4/3s aka Panasonic/Olympus - Sensor is a little on the small side and flash support could use some work (without extras like pocket wizards etc)
4) Samsung - They have no clue what do with imaging products beyond making a pretty looking spec sheets. Not much of a lens lineup, even though the cameras have been in the market for a while now (long than Fuji).

Of course all of them having good features as well

1) Sony - best sensors (full frame), bar none.
2) Fuji - best camera and lens designs, and post sale support (aka firmware updates). If you love primes, go with Fuji.
3) M4/3s Panasonic/Olympus - Best lens and camera (in terms of variety of types) selection, bar none
4) Samsung - Spec sheet (APS-C)

I pick up these mirrorless cameras and really want to like them, but I always think, "maybe the next generation will fix what's missing," but none of the manufactures have pulled it off yet. In any case I don't think the mirrorless cameras will be protected from the decline in camera sales either. While DSLR sales have fallen sharply, mirrorless sales have not increased. To me that says that yes mirrorless cameras are attracting buyers, but without some new sensors from m4/3s or Fuji I cannot see them continuing to gain market share, although I'm sure current users will continue to upgrade just as some DSLR users do.

Mirrorless cameras have some nice features and designs, but for a primary camera I'd rather have a more balanced DSLR (D750 size) with enhanced features (hybrid viewfinder tech for example) than use any of the current mirrorless models. Could I put it in my pocket? Nope, but then I couldn't put most of the mirrorless cameras (without a pancake prime) in my pocket either.

[Matsu]

They look smart, like someone might say a suit looks smart, it's just an expression, but a relevant one, since cameras, like phones, are devices that can be equal part fashion accessory, and it's something that drives buying decisions. Also, I don't think the deficits are too significant compared to entry level DSLRs, and some things, like live exposure previews, IBIS, and a smaller size overall, might make the mirrorless offering a better choice for the casual user.

It's the first time that I've been in a retail store and seen the camera counter dominated by mirrorless offerings, and that the whole thing didn't look like a neglected mess. If anything, the DSLRs seemed the merchandising afterthought.

EDIT: They had the full Fuji Line-up, X-A1, X-M1, X-E2, X-T1. I didn't see the X-Pro, but I didn't look too carefully for it. They had all the Olympus OM-D models - E-M10, E-M5, and E-M1 - but only one PEN, the E-PL5. No Panasonic models. Sony models on hand were A5000, A5100, A600, and A7. Except for the A7, which was in the display cabinet with the more expensive DSLRs, the rest were all free to hold and try with a variety of kit lenses.

No Sony a-mount product to speak of, only DSLRs from Nikon and Canon. Rebel T5i and SL1 and a few vintages of Nikon D3xxx and D5xxx along with a with D7100/70D to handle. Some full frame kit under glass and a few lenses. Lots of boxed/bundle product under the counter, locked.

Now if you go online to the same store you'll see tons of DSLR bundles: camera and lens; camera and super zoom; two lens kits; camera plus bag and tripod and memory cards and that kind of stuff... So, if a buyer walks in looking for a camera based on online deal hunting, there's a good chance they're still pushing a lot of entry level DSLR out the door compared to mirrorless. Someone reading online might decide that one or the other is more their speed depending on what they know about cameras and what they want to do or imagine they might want to do. And of course it might also depend on what the store has decided to push.

But I also think that, for the first time, someone who just walks in and says to themselves, "I want a camera for me, for kids, for family, for trips, etc..." That person is going to handle what's on the counter and be impressed by it. They feel like real compact cameras, they have nice finishes and tactility to both the bodies and the lenses. The entry level DSLRs aren't just bigger. Some by a lot, some by just a bit. On primary inspection, they feel and look older, and a little less "finished" - not looking as "serious" as the full frame kit locked under glass display cabinets.

I still think that the customer will eventually talk to a sales person who may say words like, "if you're planning to expand" or "add lenses" or "get serious about it" or "grow" or "become a professional" that such ideas still resonate. But they aren't really up-selling you on a camera anymore so much as an idea about the kind of cameras you should buy if you want people to believe you understand photography. That still moves entry level DSLRs, but I think that another group will begin to give equal weight to the advantages of mirrorless cameras.

[Matsu]

Hey Dorian, as long as you're looking for an m4/3 option, have you considered the GX7? Basically as good as the Olympus OM-D models. It's even got IBIS and some fast electronic shutter modes. Not weather sealed, but otherwise spec'd like an E-M1, and recently available for about $700 if you shop around.

You could add a nice prime or two and a kit lens to get started on the system, and not be out very much money. I myself am very tempted by this option, as it's not a lot of overhead to run it alongside an FX kit. And, no doubt you've ready considered this, but you also have access to focus peaking that might make decent use of some of your nicer manual focus F mount glass.

[Dorian Gray]

The Panasonic GX7 did catch my eye, yes. Magnum photographer Thomas Dworzak, whose work I like, uses this camera. (This kind of advertising is effective on me. )

If the GX7 didn’t have the shutter-shock problem it would probably convince me. (Am I worrying too much?) The electronic shutter cures this but at the expense of rolling across the frame very slowly (1/15 s), which means straight buildings can look wobbly with handheld exposures (see the clock faces down that page).

The GM1 and GM5 have slightly faster electronic shutters (about 1/25 s, I think, but at only 10-bit colour depth compared to the GX7’s 12-bit), but more importantly they have a mechanical shutter too (second curtain only; the first curtain remains electronic). The miniaturised mechanical shutter is slow by mechanical standards but still a lot faster than a rolling electronic shutter – somewhere in the region of 1/100 s, I believe. Enough that buildings look straight. Still too slow for a decent flash-synch speed, and limited to a top speed of 1/500 s without flash, but I live in a grey place and might get away with that.

The Olympus E-M1 and E-M10 (but not E-M5) have electronic first-curtain options that trigger the mechanical first curtain, wait for about 1/40 s, and then trigger the electronic first curtain to start the exposure. That small delay is apparently long enough to prevent shutter shock from affecting the image. Why fire the mechanical first curtain at all in this mode? It seems to be a limitation of the shutter design, which is spring-driven and requires both curtains to be fired and recharged in tandem. From my recent research it sounds like this Olympus solution to shutter shock is the best one. But the E-M10 is more expensive than the GX7.

I quite like the idea of the incredibly small GM5 at the moment, but it feels overpriced. I’m not ruling anything out. Perhaps I’ll get over my worry about shutter shock and buy a GX7 and be happy.

I like Micro Four Thirds because it can be a lot smaller and lighter than full-frame. The APS-C mirrorless systems, less so. I’m not sure how my F-mount lenses would perform on a small Four Thirds sensor. Perhaps well enough to be useful for some work. If I get a Micro Four Thirds cameras I’ll certainly get an adaptor and find out!