gdanmitchell Offline Upload & Sell: Off
|
PhilH wrote:
Strong disagree particularly to higher ISOs when elevating the noise floor. In practice the biggest gains are gradients in the low end as well as saturation. It is indeed noticeable in print and digital display. Especially for flatter looks. There's also subtle chromatic gains, but I would say hard to detect, but something that could be visibly seen with tricky light sources for sure.
Also, when positioning potentially a camera like the R1 where it will likely be marketed as a viable option against middle and medium format 16-bit sensors, I can see that value add and reason for the feature. Of note, when toggling the modes on the GFX and X2D, I can see noticeable differences at the cost of filesize, but if you're doing work for a client and focused on quality imaging, certainly worth exploring the feature. But I agree it will be a new thing for Canon to pursue.
I also agree that a medium format Canon would be a very, very long shot. But sub-3.2 micron pixel pitches at the moment will only get you so far. 2-2.2 is viable, but there's a lot of things not ideal with current tech surrounding light gathering with photosites in that realm. Which is why I think refining a current pixel pitch/design with incremental improved hardware is potentially a really good way to go.
I don't think we'll get a viable 6x9 sensor anytime soon, but new "close to" 645 should be incoming shortly, which hopefully will be a nice refresh in the PhaseOne world.. There's two new sensors already available that haven't shown up in commercially available cameras of that size. You can keep the camera package relatively small still with a 645 imaging plane with modern optical design, especially in the mirrorless era.
If they do go medium, we have a couple years for sure as FF35 and APS-C are still growing for RF Mount. And it's somewhat uncanny that just a few days ago they dropped a new lithography line stepper that can handle 300mm wafers and in a single go also produce an sensor with a ~70mm diagonal:
https://global.canon/en/product/indtech/semicon/fpa5550ix.html
I do want to chat pricing, which is a moving target with current market and financial times. I will be curious about where market tier exists for the R1. 1Ds existed in the $8K body realm. R3 is currently $6K. R5 is a lot of camera for $4500 in my mind and literally can be had now for $3700. Fujinon has the GFX100s $4700. Hasselblad will always be on the high side as well as Leica, different thing going on there. Sony A1 is their premium at $6500.
Former owner of all the 1Ds bodies, the OG D30 and D60 (then when the flipped the naming around again), and all the full frame bodies until fairly recently with the entry level ones and R3. R5 compared to much of that is an incredible value and much improved IQ. Only a few directions to go from here while remaining in the FF35/135 still format size.
...Show more →
There's a lot to chew on here.
I think that 16-bit files (as jMan13 suggested), while producing an objectively quantifiable difference in file content, would not lead to any appreciable difference in final image quality in all but a few unlikely edge cases. I suppose that the arguably advantage is a bit more "malleability" for more extensive manipulation of images in post before the file falls apart, but not sure how big of a deal that is.
My feeling is that 16-bit is one of those "well, sure, if you can do it without performance or financial cost, why not" things than something that calls out for the feature.
I confess that I haven't closely followed the R news since the R5 was introduced and I decided that for my photography it wasn't a compelling upgrade to my 5DsR. I decided to hold off for the time being and to continue the watching and waiting process. So perhaps I'm off-base on this, but my assumption has been that a R1 would not be the sort of camera to be positioned as a "viable option against middle (?) and medium format 16-bit sensors." It seems to me more likely — virtually certain, actually — that it would be the successor to the 1-series cameras, which have less-than-maximum resolution but which focus on operational speed (burst, AF, etc) for what I think of as "practical" purposes. That is roughly the opposite of the focus on MF and miniMF digital systems these days.
We agree that a Canon miniMF camera would be an extreme long-shot. The IQ and performance capabilities of FF, plus the availability of a huge ecosystem of lenses and other accessories, outweighs any potential image quality improvements for almost all photographers when we compare miniMF systems and FF.
On the other hand, a company that strives to provide close to the level of lenses/accessories/performance found in the FF world to cameras with the marginally larger miniMF sensors could justifiability claim an advance beyond FF performance that would appeal to some buyers. Because miniMF isn't that much larger than FF (unlike traditional MF formats, even including the small 645 format), it would be possible to make high performance lenses and other gear for such a system. I wonder if anyone will x9 exploit that? Fujifilm could, but they haven't — so far their excellent miniMF systems seem constrained by design/performance notions based on old school film MF ideas.
As to "close to 645" MF digital sensors, they already exist, though the costs are extremely high... for IQ advantages that are small to invisible. And the significantly larger lenses, etc. are not going to be attractive to very many folks. Any 6x9 digital sensors are likely to remain extraordinarily rare for a whole bunch of reasons.
From the perspective of film photography, the idea of a 6x9 digital sensor might look pretty appealing. But most photographers I know who use or have used larger formats, including LF film, look at this from almost the opposite perspective. Rather than wanting to use digital to recreate the familiar larger formats, they are excited about being able to use smaller digitals systems to get the quality that formerly required the very large systems. (A friend of mine who was a long time LF photographer documented this process over a decade ago in older issues of Luminous Landscape. He got access to various Phase One backs early on and did lengthy and exhaustive comparative testing of that technology against his LF film system... and eventually arrived at a point where he was convinced that the Phase One system produced equal or slightly better quality than he could get from scanned 4x5 film. These days he uses a Fujifilm GFX system.)
I shoot FF and APS-C. I have been using the Fujifilm x-trans system for over a decade for half of my photography — the half where the smaller system has photographic benefits. But I continued (and continue) to use the larger format for tripod-based photography that may end up being reproduced at large sizes. But I recently picked up one of the new 40MP Fujifilm cameras... and I'm convinced that the boundary for using the smaller system has moved significantly, e.g. I can get some very good IQ from that system in situations where I wouldn't have previously relied on the smaller format.
And, of course, if we can do that at 40MP on APS-C, FF can be pushed further in that direction, too — I'd argue that it should easily be able to do what a 50+MP miniMF system could do a few years ago. And that was regarded by some to be pretty darned good. Thus the real-world benefits of pushing to larger formats becomes smaller and smaller over time.
YMMV.
|