New 247MP sensor by Sony

Perhaps we will see this in a new phase one camera.

2 Likes

Clearly I need this now. What is the delay in getting it to market??

6 Likes

they give you some time to save money for the gpu youā€˜ll need to process those images ā€¦

6 Likes

This is essentially the Fuji 40 MP sensor (manufactured by Sony), scaled up. Iā€™d also expect to see a 90 MP full-frame sensor before too long.

2 Likes

247 megapixels ā€“ I thought that was a typo, initially. 24 megapixels would be a significant upgrade for me. :smile:

3 Likes

For sure this is on the way. I guess for some people, you can never have too many megapixels because it just allows for insane cropping, but processing those files must get very tedious.
Iā€™m happy with my Fuji 40MP and probably wonā€™t go much higher than that. But I never say neverā€¦

2 Likes

I need it now. Now!

Nah ā€¦ with my 24" 2K monitor and no printer, Iā€™ll stick with my 3.4 MP Sigma camera. :slight_smile:

1 Like

Iā€™m happy with my Fuji 40MP and probably wonā€™t go much higher than that.

Yeah, seriously. Agreed. I just upgraded to a Fuji X-T5 as well, and itā€™s nice to have 40 MP for the extra cropping. And, at least for the primes and the zooms I have, everything seems to fully resolve that extra 40 MP (even lenses that arenā€™t officially rated to).

However, itā€™s really on the upper end. The files are larger, so every photo I snap means Iā€™m just a little closer to having to upgrade my storage yet again.

I donā€™t think an APS-C can go much above 40 MP before thereā€™s diminishing returns on what lenses can resolve. And I donā€™t think most photographers even need 40 MP generally, although itā€™s nice. (And scale up the megapixel count with the same density for full frame and medium format, as @bastibe mentions.) Weā€™re probably close to hitting the wall of what we can do with light, physics, and lens manufacturing at this point?

2 Likes

Whoa brother, the 45mpix files from my Z7ii cropped to a 4:5 aspect rato is just enough for a 16x20in @ 300ppi. The resulting prints are wonderful and putting them 1mm from my nose and seeing all the detail brings me great joy.

It sounds like I am being sarcastic but I assure you I am not.

4 Likes

MFT has used a 20 MP sensor for years, which has the same pixel size as the 40 MP APS-C sensor. The newest MFT sensors are physically 80 MP, but pixel-bin to 20. So that seems to confirm this conjecture. (there are good technical reasons for pixel-binning: it reduces blooming, compared to a lower-resolution sensor. I wonder if the same thing will come to Fuji eventually)

However, both MFT and Fuji support high-res multi-shot modes that do eke out a good bit more information from the same lenses. So perhaps the limit is not so much one of physics or optics, but more one of processing and practicality. Even the 40 MP Fuji is a good bit slower than the 26 MP Fujis, after all. Practical use cases for more than 20 MP are rather rare, too. I kind of wish my Fuji had a low-res raw mode, like some Leica cameras do.

2 Likes

Wow! Is there a link to one, please?

Itā€™s a quad pixel design: OM Digital OM-1 Review: The Best Micro Four Thirds Camera Ever Made | PetaPixel

1 Like

Thanks ā€¦ also found this:

2 Likes

I think that for most lenses, lens resolution is the limiting factor. I am not sure that you will get near the resolution of a decent tele lens just by cropping.

An interesting point. Does anyone know of such a comparison and, if so, how was ā€œresolutionā€ measured?

I have tested my tele converters against cropping. For the 1.4x TC, the difference in resolution was almost imperceptibly in favor of the TC. For the 2x, it was a solid win for the TC.

This is relevant, as tele converters optically crop into the lens FOV, and magnify that view to cover more pixels. If the TC shows more resolution than cropping, then the lens has some unused potential that a higher resolution sensor could exploit.

My tests were with a 26 MP Fuji on the 70-300 lens and the TC benefit was very minor. I would therefore presume that there is no TC benefit at 40 MP.

Purely anecdotally, I do find that I can crop into the 40 MP images quite a bit more than the 26 MP images without them falling apart. Which is contrary to my first impressions that I shared at some other point on this forum.

I now find that this enhances the usefulness of prime lenses noticeably. My 23 f/1.4 is surprisingly usable up to a cropped 50mm (f/3). At that point, Iā€™m pretending my APS-C camera has a 1", 10 MP sensor. Incidentally, so does my Nikon J1. Now thatā€™s a fun comparison :thinking:.

1 Like

Thanks @bastibe, thatā€™s very interesting to me, as I own the 70-300mm Fuji and have had the 1.4 TC in my shopping cart for over a year. My hesitation in buying it has been whether I need it with the 40MP X-T5.
I guess you can also crop in with the TC and therefore get even closer, so itā€™s not like it becomes redundant.

Iā€™ve heard that the x1.4 is a better buy than the x2 TC for optical reasons and because of light gathering (too much compromise with the x2). Would you agree?

The resolution gains of the TC were minor on 26 MP. Iā€™d wager theyā€™d be all but non-existant on 40 MP. However, the cost in AF speed is severe, and the sheer hassle of taking your lens off, then screwing in the TC, then reattaching the lens, was just not worth it to me in any way.

I found an additional thing, though: at 300mm, the TC often seemed completely worthless. No advantage at all to cropping. I interpret this as being resolution-limited by the lens. However, at 250mm, the TC did give a slight advantage, and at 200mm, even the 2x looked absolutely brilliant. To my understanding, this means weā€™re completely at the mercy of the lens. Where lens resolution is good, the TC can magnify the image a lot. Even the 2x is fantastic in that case. But as my 70-300 goes towards 300mm, its resolution drops, and the TC gets less and less effective. The optimum in terms of reach was around 230ish mm 2x.

So from my experience, Iā€™d say both the 1.4x and the 2x perform perfectly. I saw no evidence that the 1.4x is ā€œgoodā€ and 2x is ā€œbadā€. But theyā€™re very much limited by the lens, and not every lens has enough resolution to benefit from the TC.

With the 2x, my X-T3 hunted focus quite badly, and had to run the ISO uncomfortably high for moving subjects. I found the 2x unusable for that reason. And the 1.4x just didnā€™t give enough benefit over cropping to see much use. Also, 1.4x is a surprisingly small crop actually. Itā€™s really not that much more.

So in the end, I sold them both. A fun little experiment.

1 Like

How was the ā€œresolutionā€ measured may I ask?