Origami101

I can’t tell whether that’s an embarrassed admission they’ll miss their publicly announced 2021 release date or just an awkward statement the release isn’t now. I’m reading it as the former, but it’s hardly clear.

Actually taking a little more time in the oven to bake their next gen camera might be a good thing for Fuji. The 26MP sensor is pretty much state of the art, and they’ve rolled it out to the X-T3, X-T4, X-Pro3, X-E4, X-S10, and X100V. What they could really use now, imho, is: 1. Improving the consistency, predictability and reliability of the autofocus, particularly for face recognition. 2. Rethinking the menus, which are serviceable but not ideal. I think they’re also a legacy of the original X100. The cameras have evolved significantly in features since then and the menus should reflect that. 3. Redoing the Q menu, which is oriented towards JPEG settings. You can switch the defaults, but the entire concept should be rethought. 4. Making touch more pervasive. This goes hand in hand with the two points above. I don’t think you currently can adjust any menu settings with the touch screen. Most or all of that can be done in firmware. For the next gen body, I’d like Fuji to think who they really want to be when they grow up. The ergonomics on, say, the X-T4 seem a mishmash of Leica M homage - from the original X100 - early 2000s Nikon DSLR for the dials, and hybrid videography camera. With the recent GFX models they’re pushing towards a more modern, streamlined interface. To me that’s a good thing; there’s a reason Canon and Nikon evolved in that direction: it works. I’d like Fuji to reflect on and really hone their interface for whatever X series camera comes next.

Exactly. The beatings will continue until moral improves — or put another way, we’ll neglect m4/3s till you switch to L-Mount — is hardly a sound strategy. Since we’re all speculating here, here’s mine: much of Panasonic’s apparent disdain for m4/3 and attraction to ‘full frame’ is driven by one of the classic business motivators. No, not profit, don’t be silly. Ego. I swear the execs were at some industry conference and heard the Canon and Nikon guys snickering about ‘half-frame’ and felt inadequate. Now we’ve got the gargantuan S series bodies and lenses. Nothing unmanly about those! As a corollary, that’s why stupid terms lead to stupid thinking. There’s nothing special about 135 format, either in size or aspect ratio, that lead to it being placed on a pedestal. At most it’s the largest format that tends to be on the latest sensor tech. Plus Canon and Nikon had a lot of legacy lenses in that format they’d prefer you to keep using, at least till you upgrade to their latest version.

Let me point out a few things that could have been updated without needing a new sensor (and which, other than I think Canon for HEIF, none of the other manufacturers support, so this is hardly exclusive to Panasonic): HEIF as an option or in lieu of JPEG. That’s easy to add since HEIF is just a subset of H.265. DCI P3 or ProPhoto RGB image space for HEIF or JPEG. Apple’s ProRAW, a demosaiced DNG with tags for tone mapping, as a raw format. One thing specific to Panasonic I’d like to have seen — other than a new sensor, phase detect autofocus, and the viewfinder and monitor from the S1 — is Bluetooth 5. They’re still using Bluetooth 4.2 from 2014.

If they’ve got a GH6 coming, they’ve got to be crystal clear about that. It’s coming next year, will have a brand new sensor, and will be priced substantially above the GH5 mk2. Otherwise people will draw the obvious conclusion: Panasonic’s pretty much done with m4/3 and this warmed over offering exists solely to milk a few more dollars out of their existing customers. The thing I don’t get is the strategy here. Two short years ago m4/3 users were 100% of Panasonic’s customer base; now, they’ve effectively abandoned them. Most people looking for a 135 system will choose a less fickle partner to invest their money in.

That'd be a terrific camera, the only problem, based on the current rumours, Panasonic has no interest in making it. Where I think they goofed on their 135 adventure is having no cross-compatibility with m4/3. m4/3 has a short flange distance, just 19.25mm, so it would have been tricky, but maybe with something like Z mount, only 16mm, it would have been possible.

Panasonic’s logic is baffling. They seem to think by neglecting m4/3 users they’ll incentivize them to switch to L-Mount. Instead, they’ll drive them to Sony, Canon, Fuji, or Olympus. At this point I wonder if their camera business is viable at all. L-Mount certainly hasn’t been a roaring success, which was completely predictable considering they foolishly chose to take on Sony, Canon, and Nikon where they’re strongest, 135 format. M4/3, at least, was a defensible niche.

Ouch, that sounds unfortunate. I’d focus on documenting exactly what’s going wrong so you can confidently say to Canon here’s the problem and I need you to fix it. From your description it’s not clear whether the problem’s with the R5 or the 800mm lens. I’d start by narrowing it down. Take a known good lens, EF mount if you still have one, and shoot a series at different apertures and focus distances; repeat with the same lens and a known good body, eg your 5D mk3. Chimp as you go and repeat anything that looks soft to make sure it’s a genuine problem, not user error. Next, shoot a similar series with the 800mm. You won’t have a control body, so maybe try different AF settings to see if that makes a difference. You should now be able to clearly point to which body, lens, and settings trigger the problem. Send that information with the supporting and control pictures (to show what it should look like) to Canon and see what they say. You might as well copy B&H though I think it’s past the point they’d be required or even reasonably expected to do something.

I'll go further and say there should be a future for M4/3 but might not since Panasonic's backed themselves into a strategic corner. To start with, why should M4/3 hang around? Well, aside from the obvious benefits that it generally yields smaller and lighter camera kits —especially for telephoto—has sufficient image quality for both stills and video and is broadly supported, there’s two other distinct advantages. The first is, for any given sensor generation, a M4/3 sensor will read data off faster. This is where I think the traditional camera makers have gotten lost. Did the stunning jumps in iPhone image quality come from bigger sensors? They’ve gotten marginally larger, but no. The jumps have come from throwing massive amounts of computational power at the data, which means reading it as fast as you can. Blown highlights? Merge multiple exposures. Terrible low-light performance? Same! But with different algorithms. Excessive depth of field? Depth map! A smaller, but still sufficient, sensor like M4/3 fits the computational imaging paradigm better. And as a bonus, they’ll also hold an edge in IBIS, heat generation, and battery drain due to the smaller size and mass. The second is smaller sensors—or more accurately, smaller image circles—have distinct theoretical advantages for lens design. I don’t pretend to understand the theory, though I do know the complexity of making lens elements increases with the cube of its diameter, so all things equal you should get better, smaller, or cheaper lenses for M4/3 for any given field of view versus larger formats. Which leads to the tragedy of Panasonic going the L mount route (in the dumb business decision sense, not Romeo and Juliette). I get the rationale of wanting to consolidate their cinema and Lumix lines on one mount, and the logic m4/3 was too small for the former. The problem was they picked a sensor size, 135 format, putting them in direct competition with Sony and Canon, who offered, or were inevitably going to offer, soup to nuts mirrorless solutions, picked a mount designed for APS-C, and stranded their entire current user base. Heck of a job, guys. Worse, since L mount is now their halo product, they’ll be loathe to undercut it by offering better features on m4/3, even if they’re technically feasible. So a GH6 that substantially outperforms the S1 or S1H? Probably not going to happen, even if S series sales continually underperform. But I would love to be proved wrong.

Unless we see some signs of life from Panasonic on the m4/3 front soon, that's probably the right conclusion. It's been three years since their last substantially new m4/3 camera, the GH5S. Compare that against the four S series bodies and new lens line launched and the attention they've lavished on that mount. Here, for example, is a description of their promised future firmware update. What irritates me is there's no business case for the L mount, at least not for Panasonic. Assaulting Canon, Nikon, and Sony where they're strongest is simply foolish. To succeed, the S series cameras would have to not just be better but dramatically better, and they're not. Good specs aren't sufficient, particularly when there's no existing users or lenses (discounting the dozen or so Leica users), the autofocus is subpar, and they've done a bang up job alienating their m4/3 user base signalling they're bored with that and would like them all to migrate. I'm sure they will, just more likely to Fuji or Sony or Canon than L mount.

The idea for internal storage is to free up space in the camera -- for a larger battery, bigger heatsink, even a more powerful processor -- while pushing the need for any additional storage to the Thunderbolt ports, which can link to fast SSDs. How often would you record more than 512GBs in a session and not be using a rig and external monitor where you could easily slot whatever storage you need in?

This is just a Wishlist, not even speculation what a potential GH6 might have. But if I were Panasonic -- which I'm not! -- I'd go bold: micro 4/3s is always going to have a perceived, though not necessarily actual, image quality deficit versus 135 and APS-C formats. So play up your strengths and use the cost savings from the cheaper sensor to bundle in better tech. For the body, pick up the refinements from the S series like that gorgeous 5.7m dot viewfinder, highres touchscreen, and 8 way joystick. Also sweat the details, for example, calibrating both the EVF and touchscreen from the factory. Emphasize connectivity. I'd like to see the latest Wifi 6 and Bluetooth 5.1 or 5.2. Then add 2 thunderbolt 4 connectors. Aggressively ditch the legacy stuff. Cut the card slots and instead offer 128 or 512GBs of internal storage. Make transferring data off via thunderbolt or wifi simple, fast and dependable. Push the boundaries on tech. How about H.266 as an option? For stills, 10 bit HEIF and Display P3. And please, toss the proprietary raw and go with DNG. Adopt the latest 1.6 spec and use that semantic and tone mapping Apple's done on the iPhone 12 Pro (ProRAW is just a DNG file). For autofocus, I doubt Panasonic will go with phase detect. But maybe this'll be the update that makes DFD great. Drive that at 240 hertz with improved algorithms and processing power, especially for subject recognition, and it could be pretty good. While you're at it, profile the Olympus lenses (I suspect they may have done this but don't advertise the fact. But if that's the case, make it official. You want m4/3s to be as attractive a platform as possible).

One thing to note, we're not seeing the steady trickle of rumours from Panasonic you'd expect if the GH6 were imminent. All the 4/3 rumors post that kicked this thread back to life shared was that, yes, Sony's listed a new m4/3 sensor in their catalog. It's pure guesswork that's heading to a GH6. Where're the hints and teases you'd expect from the Panasonic side ('Breaking! the GH6 features [shiny new spec]!'?

Would the IMX 492 even be suitable for the GH6? It’s got a max readout speed of 31fps, which wouldn’t be fast enough to adequately drive DFD autofocus. Perhaps it can readout faster at lower resolutions; for video, 6k, 4K, etc.; for stills, selective sub-sampling that’s higher than the viewfinder resolution, which is really all you need before capturing the frame.

What in particular do you prefer about the Fujis over the Z6?

It’s hardly inspiring that Panasonic seems to have lost interest in m4/3s and would prefer to focus on the L-mount. Those resources could all have been put towards the GH and G series and new or revamped lenses (even something as simple as refreshing all the primes to give them similar form factors and features). When a company seems to be putting its interests over its users, its fair for those users to wonder if they should stick around.

Panasonic’s L mount push still seems quixotic to me. Canon, Nikon, and Sony users won’t move as the benefits aren’t sufficient to justify repurchasing their lens stable, and the AF’s second rate. And there’s no cross-comparability with m4/3, so what’s the incentive there? Strategically, I’d say going all in on m4/3 makes more sense. The smaller sensor has significant advantages, so long as they can capitalize on them: better IBIS, faster readout, deeper DOF, more compact body & lens packages, and, hey, less heat. Leave 135 format to the other three.

Cliff, are you basing this on any firsthand knowledge? Or is it just speculation that, of all the companies in the industry, Sony Semiconductor refuses to sell PDAF sensors to Panasonic?

I find the lack of credible rumours a bit disconcerting. I mean we’ve now seen the R5 and A7S III, know the Z6S’s coming, and there’s not even a whisper about a GH6? Not a great sign.

However nice the cameras may be, I have no idea how Panasonic plans to make a success of this system. The most likely target market is high-end Nikon and Canon shooters, but they’re used to their respective brands, have a stable of lenses, and want autofocus that works. Micro 4/3 shooters moving up? Maybe a few, but there’s no interoperability between systems. They could just as easily move to R or Z mount, or even Fuji. That leaves video shooters, but are there enough to make this viable?

So, anything on the GH6? NAB is April 18-22, are we going to see it then? Or Photokina in May? Though I think NAB‘a a much better audience for this camera, especially if Panasonic gives it all the bells and whistles. Also, with the E-M1.3 failing to deliver a new sensor, what are the odds Panasonic will come through with the rumoured ~40mp one?

The iPhones actually take the first image before you press the shutter (i.e. they’re constantly buffering). Android phones might do the same, I don’t know.

I’ve got a G9 and noticed the distortion immediately but forgot about it within minutes. I suspect it’s something the vast majority won’t be troubled by while a small group will find it unpleasant and a deal breaker. If you’re playing the odds, assume it won’t bug you. If you have to know, rent or borrow one for a day. I suspect you’ll find which group you belong to within a few minutes.

So, from perusing Wikipedia, it seems this is a backlit (Exmor R), but not stacked (Exmor RS) sensor, unlike the IMX594CQR, which is backlit plus stacked. Curious why Olympus would go with this one when there seems a better and otherwise very similar one, which Panasonic’s choosing, available. Also interesting that both these m4/3 sensors are listed for surveillance. Are Olympus and Panasonic going to be propped up - that is, there being enough volume to justify future sensors - by a completely separate market?

It's not particularly surprising Panasonic's sales are disappointing; it's hard to see who their intended market was. Breaking it down, potential buyers are current Sony, Canon, Nikon, or m/43 users, and for m4/3, read mostly GH5 and GH4 users. Sony users won't be tempted. The only compelling upgrade the Panasonic offers is ergonomics, and if they cared about that, they wouldn't be Sony users. On the minus side, the Pansonic lens selection is limited and pricey, and the autofocus worse or reputed to be worse, which is the same thing to potential purchasers. Nikon and Canon users won't be tempted for one reason: lenses. One of the main draws of the Z and R mounts is compatibility with the existing EF and F mount lenses. People looking to get into a $2,000 plus 35mm format mirrorless system are almost universally going to be experienced shooters with a drawer full of lenses. Sticking with their current brand allows them to preserve that investment and transition in stages if they choose using the mount adapters; going Panasonic deprecates that investment and doesn't seem to offer enough compelling in return. You could, theoretically, use mount adapters for the L-mount, but whereas the Canon and Nikon mount adapters are almost seamless, the L-mount ones are less than ideal. A kludge rather than a well engineered solution. And GH5 or GH5s shooters would upgrade to this why? Ok, a marginal improvement in image quality, but a huge increase in cost, size and weight. Plus no using existing lenses. All that's before we get to price. I just specced out roughly comparable systems on B&H. A Z6 plus 24-70, 50mm 1.8, FTZ and 70-200 f/4 (F mount) clocks in at $4,300. You can shave $600 off that if you want the slower, but longer and still excellent, 70-300 AF-P. The S1 plus 24-105, 70-200 f/4, and 40mm Sigma 1.4 is $6,300! Add $1,000 if you want the Panasonic 50mm. Oh, and the Sigma's not available yet.