The Future at Nikon

[MSNWatch]

Lifelong photography enthusiast who went from Minolta to Nikon and now exclusively use sony FF mirrorless. Smartphones have progressed due to computational photography which is something the conventional companies like Nikon and Canon lag Google and Apple. By a lot. That and other software features that bythom and others have been talking about for so long now like connectivity. Unless they strongly beef up their software they will become niche companies.

 

[mansr]

Software can never get around the diffraction limit of tiny optics.

 

[Objectivist01]

Software can never get around the diffraction limit of tiny optics.

AI may be able to predict and draw it accurately in some time.

 

[GXAlan]

Software can never get around the diffraction limit of tiny optics.

Yes and no. Software can never get around the limits of physics. Big sensors will always be better than smaller sensors. I say this as a medium format digital shooter. But software makes tiny optics better. Software makes large optics better.

Not the best examples, but the current generation of diffraction compensation is pretty good

Canon's
https://snapshot.canon-asia.com/art...s-aberration-correctiona-close-up-look-part-1

Nikon's
https://nps.nikonimaging.com/technical_solutions/z7_z6_tips/small_apertures/

Fuji's
https://www.fujifilm.com/products/digital_cameras/x/fujifilm_x_t1/features/page_02.html
(and still used in the GFX100)

Capture One
https://learn.captureone.com/blog-posts/compensate-lost-sharpness-small-apertures/

AI may be able to predict and draw it accurately in some time.

I actually think the AI in iPhone (as an iPhone user) is both great for what it is, but also adds a high level of artificiality to many images. Over time it may get better, but this may be the equivalent of auto-tune or "loudness" equalization in hi-fi.

Getting back on topic, Nikon has some exceptional lenses, and some exceptional patents. The ergonomics of the Nikon professional line are field proven. Their diffractive optics have surpassed that of Canon's. Their camera brand/technology is a very good target for acquisition esp. if Nikon is under duress.

 

[GXAlan]

Fewer high performance pixels are better than a few more pretty ordinary ones, too. It is not just numbers.

But a lot of high performance pixels is best.

 

[JeffS7444]

Lifelong photography enthusiast who went from Minolta to Nikon and now exclusively use sony FF mirrorless. Smartphones have progressed due to computational photography which is something the conventional companies like Nikon and Canon lag Google and Apple. By a lot. That and other software features that bythom and others have been talking about for so long now like connectivity. Unless they strongly beef up their software they will become niche companies.

How do you compete against mobile devices which are so beloved that many people are never more than an arm's reach away from theirs, and which are even now being strapped to people's wrists?

 

[GXAlan]

How do you compete against mobile devices which are so beloved that many people are never more than an arm's reach away from theirs, and which are even now being strapped to people's wrists?

It’s the same challenge as hifi.

The MP3 revolution 20 years ago has led to somewhat of a new renaissance of desktop headphones and DACs. As millennials move into the next phase of adulthood, they will likely invest in better home systems.

For photography, it is still about the content. Smartphone enthusiasts have often upgraded to dedicated APS-C/FF as their photography matures.

Just as some music may not really need an audiophile system, some photos may not need a high quality image...

 

[Objectivist01]

Yes and no. Software can never get around the limits of physics. Big sensors will always be better than smaller sensors. I say this as a medium format digital shooter. But software makes tiny optics better. Software makes large optics better.

Not the best examples, but the current generation of diffraction compensation is pretty good

Canon's
https://snapshot.canon-asia.com/art...s-aberration-correctiona-close-up-look-part-1

Nikon's
https://nps.nikonimaging.com/technical_solutions/z7_z6_tips/small_apertures/

Fuji's
https://www.fujifilm.com/products/digital_cameras/x/fujifilm_x_t1/features/page_02.html
(and still used in the GFX100)

Capture One
https://learn.captureone.com/blog-posts/compensate-lost-sharpness-small-apertures/


I actually think the AI in iPhone (as an iPhone user) is both great for what it is, but also adds a high level of artificiality to many images. Over time it may get better, but this may be the equivalent of auto-tune or "loudness" equalization in hi-fi.

Getting back on topic, Nikon has some exceptional lenses, and some exceptional patents. The ergonomics of the Nikon professional line are field proven. Their diffractive optics have surpassed that of Canon's. Their camera brand/technology is a very good target for acquisition esp. if Nikon is under duress.

I said in future not right now. It already got lot better over the years

 

[mansr]

I said in future not right now. It already got lot better over the years

No matter how advanced your software, bigger optics will always give sharper images.

 

[Soniclife]

No matter how advanced your software, bigger optics will always give sharper images.

This is similar to statements about audio bit depth, after a certain point most people correctly stop caring as good enough has been passed. In good light good phones are very good now, as soon as you move out of their comfort zone they get worse quick. But the comfort zone slowly expands each year, but they are a long way from replacing a proper camera if you want to take more than snapshots.

 

[Koeitje]

Fewer high performance pixels are better than a few more pretty ordinary ones, too. It is not just numbers.

If you are talking about less pixels = better low light, that is a myth.

 

[mansr]

If you are talking about less pixels = better low light, that is a myth.

Is it a myth that larger pixels capture more light? Do tell.

 

[mansr]

This is similar to statements about audio bit depth, after a certain point most people correctly stop caring as good enough has been passed. In good light good phones are very good now, as soon as you move out of their comfort zone they get worse quick. But the comfort zone slowly expands each year, but they are a long way from replacing a proper camera if you want to take more than snapshots.

Software can create a convincing fake sharpness. That's not the same as the real thing.

While I agree that recent phones have become very impressive indeed, final picture quality isn't all that matters either. Speed and ease of use have a very long way to go before they can compete with a DSLR.

 

[Koeitje]

Is it a myth that larger pixels capture more light? Do tell.

You don't look at individual pixels. 4 pixels on a 20mp sensor capture the same light as 1 pixel on a 5mp pixel. Scale them to the same size and performance will be pretty much the same, but the 20mp can capture more fine detail. For low light performance the actual sensor generation and technology like BSI is what matters, not MP count. And most importantly, sensor size. Sensor size is the number one indicator of low light performance (For example: 5MP and 20MP 35mm sized sensors will both capture exactly the same amount of light).

More pixel do decrease dynamic range I believe, but since dynamic range is getting better and better and already surpasses film its not really an issue.

 

[LTig]

You don't look at individual pixels. 4 pixels on a 20mp sensor capture the same light as 1 pixel on a 5mp pixel. Scale them to the same size and performance will be pretty much the same, but the 20mp can capture more fine detail. For low light performance the actual sensor generation and technology like BSI is what matters, not MP count. And most importantly, sensor size. Sensor size is the number one indicator of low light performance.

Same sensor size. Yep, see the DXOMARK results of the D800. When it came out it had the highest resolution ever, and also nearly beat the D3 in SNR compared at print size view.

That was the reason I bought it - you have the best of both worlds. With enough light you get 36 MP high res pics, in very low light just scale down by 2 and get 9 MP pics with similar quality as a D3.

 

[mansr]

You don't look at individual pixels. 4 pixels on a 20mp sensor capture the same light as 1 pixel on a 5mp pixel. Scale them to the same size and performance will be pretty much the same, but the 20mp can capture more fine detail. For low light performance the actual sensor generation and technology like BSI is what matters, not MP count. And most importantly, sensor size. Sensor size is the number one indicator of low light performance (For example: 5MP and 20MP 35mm sized sensors will both capture exactly the same amount of light).

More pixel do decrease dynamic range I believe, but since dynamic range is getting better and better and already surpasses film its not really an issue.

I think I might agree, but you're expressing the situation in a very convoluted way.

For a set number of pixels, a bigger sensor gives better performance because of the bigger pixels. At a fixed sensor size, fewer but larger pixels again give lower noise at the pixels level. Of course, here you can do averaging in software and (more or less) simulate the same thing. That's assuming the smaller pixels have the same coverage as the large ones, which isn't always the case.

 

[Koeitje]

I think I might agree, but you're expressing the situation in a very convoluted way.

For a set number of pixels, a bigger sensor gives better performance because of the bigger pixels. At a fixed sensor size, fewer but larger pixels again give lower noise at the pixels level. Of course, here you can do averaging in software and (more or less) simulate the same thing. That's assuming the smaller pixels have the same coverage as the large ones, which isn't always the case.

We don't want to talk about a set number of pixels, we want to talk about total surface area to capture light. Which is dependent on sensor size and nothing else (well apart from the tiny tiny tiny space between pixels). You want to capture more light for better low light performance? Get a bigger sensor, not less pixels. For equal sized sensors the actual sensor design matters far far far far far more than any difference in coverage due to space between pixels.

I'd argue that averaging might actually be better!

The main reason why denser sensors aren't noisier is that in modern sensors the gaps between each sensor site ('sensel') have been minimised to a level where their effect on output is negligible. As a result, the sensels effectively cover the entire sensor area, regardless of how many or few of them there are. Consequently the light-gathering power of the sensor is entirely dependent on its surface area, and very nearly independent of how finely that surface area is divided up.

 

[mansr]

We don't want to talk about a set number of pixels, we want to talk about total surface area to capture light. Which is dependent on sensor size and nothing else (well apart from the tiny tiny tiny space between pixels). You want to capture more light for better low light performance? Get a bigger sensor, not less pixels. For equal sized sensors the actual sensor design matters far far far far far more than any difference in coverage due to space between pixels.

The coverage issue was solved with micro-lenses in front of the sensor elements.

Anyhow, I'm not seeing any phones with 35 mm sensors, let alone medium format. Physics still favours "real" cameras.

 

[Soniclife]

Software can create a convincing fake sharpness. That's not the same as the real thing.

This looks enough to take great pictures with, it's from raw, so less camera trickery.
https://www.dpreview.com/files/p/articles/5666550435/Samples/GooglePixel2-DNG-ISO50.acr.jpeg

Lots of great pictures have been take with worse than the above, I still take much better pictures with a proper camera, but resolution at base iso has nothing to do with it.

 

[Wombat]

I have always been bemused by the 'pixel wars'. It seems to be an amateur preoccupation. Most amateurs are, generally, financially constrained to A4 sized prints with a few able to have the means to produce A3 size prints and fewer still going larger. The arguments over pixels tend to be moot.

Pros. just get what they need to do the job and don't bang on about it unless they have a commercial barrow to push, a review site or a web/blog site - most don't.

Reminds me of audio.