"I doubt we'll see this sensor in any future model..." —multiple comments on Canon's announcement of a 410mp image sensor
By way of explanation, 410mp is 24K video, or 24,592 x 16,704 pixels.
That specific image sensor? No, we're not likely to see it in an upcoming Canon camera. The announcement itself is coordinated with SPIE Photonics West (a conference for optics and photonics work), and is really speaking to two things: Canon is not only fully capable of smartphone level photosite sizing, but more importantly, pulling immense amount of information off the image sensor quickly. In this case, that 410mp sensor can achieve 24K video at 8 fps, or 12K video at 24 fps.
However, don't rule out really high pixel counts. More sampling over the same area opens up many options. As an engineer friend pointed out in response to something I wrote earlier, at about 4x the pixel density you not only can improve the phase detect focus performance, but you also start to be able to develop a useful depth map. You're still essentially at the same final pixel count, dynamic range, and noise-at-standard-output size, but you have more information to do more things. I've written it for a couple of decades now: I'm always taking more sampling, all else equal.
What's really happening at Canon is this: they have prototype chips that allow them to investigate what they might be able to achieve with more sampling. When they figure out what that is and it imparts a tangible user benefit, you can bloody well bet that Canon will have a camera with a megapixel monster inside.
"Your AI edits may interact with each other and require updating" —post by Greg Benz
You bet. Moreover, this is a variation on the "in what order should I do things in processing raw files" question that has been around since the beginning. I've been reticent to write much about that because it keeps changing as new tools and workflow appears. I have enough things I need to update that don't get updated fast enough than to put a "raw workflow order" article on the site ;~).
However, Greg has done some of my work for me. I invite you to take a look at his AI order article and video, as it explains at least a subset of that fairly concisely. Short version: if you're going to use AI Denoise, do it first. Then learn the order of the other AI tools and do them in order.
"OM [Digital Solutions] told us that the OM-3 is effectively the replacement for the much-loved Olympus Pen-F...explaining it was impossible for them to make the Pen-F design fully weatherproof." — Photographyblog
I find that statement to be incredible. This would effectively be saying that creating an off-set viewfinder camera makes it impossible to weatherproof, while a centered viewfinder is. Sounds like a made-up-on-the-spottism explanation, to me.
I mean, I get it: OMDS decided that their more retro looking camera would be SLR design, not rangefinder design. Not that there's all that much difference between the two in actuality, other than where the viewfinder sits and whether you have a viewfinder hump or not. Other than that, I see no functional differences between the Pen-F design/controls and the OM-3 design/controls. So in that respect, I suppose, OMDS could indeed claim "we did the OM-3 rather than a Pen-F II because we liked that idea better." Both ideas play off older Olympus designs—particularly in panda cladding—and both feature the flat soap-bar-in-leather front instead of a real hand grip.
I'm going to go out on a limb here and suggest why OMDS chose OM-3 instead of Pen-F II: the SLR-type design is about a half inch wider than the rangefinder design. It's highly likely that when someone disassembles an OM-3 you'll find that OMDS didn't really redesign the OM-1 Mark II digital board, and it needed a wider area.
Frankly, I'd say that we haven't yet seen OMDS design a new camera since taking over from Olympus. We've seen tweaks and refits of existing designs, but nothing particularly new from them. I'll also double down on my claim that Olympus/OMDS has missed a significant opportunity, a m4/3 compact camera to take on Ricoh's GR and Fujifilm's X100. And a Tough m4/3, too. Those have been clear opportunities to expand their volume and up their average selling price while still reusing the most expensive parts (m4/3 image sensor and TruePic processor). Total miss on Olympus/OMDS's part. Instead we get almost an identical camera in a different body at (currently) the same price.
