Are we going to need new lenses?

[Akaru]

Been thinking about the immediate future of MF digital photography in regards to lens design. Let’s say Phase One comes out with a digital back that is 300 megapixels in a few years (which, I would think, would assuredly and finally match—perhaps surpass—8x10 *dodges rocks*).

Will we need all new lenses to take advantage of it? Will any manufacturer be willing to put in the work? And would we be able to afford them if they did? And if we did, would diffraction have us shooting at f/4?

I wonder if, knowing this, Phase One will keep the resolution at this limit for a while, at least for another generation, focusing on other value adds. But 35mm and crop MF may need to keep upping the resolution to stay competitive. Perhaps Fuji and others have the money to spend on all new lens designs.

Is this as good as it gets for MFDBs?

[JoeKitchen]

If Sony and P1 took the current micro-architecture of the IQ4-150 sensor and put it to use in a full frame MFD back with 60 MP, I would buy that in a heart beat. 

I doubt that will be the case, so more then likely I will purchase the IQ4-150 when refurbs come on the market and that will be the last back I ever buy.  To be perfectly honest, lens cast is the only reason I want to upgrade, but beyond that, there is nothing I see as being ground breaking in the digital back market. 

After this, any other camera I buy will be a mirrorless system. 

[George_Cleansman]

With 300 MP we would have a pixel stitch of approx. 2,7µm that leads to the max. aperture 4 for avoiding diffraction issues. It is evident that that's no deal for landscape photography. So we do not need other lenses. What we need are less MP's with improvements in quality, better high iso noise, etc.

The actual price for a back with 150 MP is not interesting for me in the normal medium format market. I like my Hasselblad H6D-100c especially for the very fine colors and hope that Hasselblad like to develop a better firmware in 2020.

[araucaria]

I hope they come up with a 6x8 sensor.

[Akaru]

I hope they come up with a 6x8 sensor.

I’ll take a digital 4x5 any day of the week 

[ErikKaffehr]

Hi,

Sensor development goes in small steps. AFAIK it takes something like 3 years to develop a new sensor. So, I don't think we will see 300 MP for a good while. Just as an example, the Sony A7rII sensor at 42 MP was introduced something like June 2015 and the 61 MP arrived in mid 2019.

Lenses are definitively good enough for 300 MP.

Increasing resolution of the sensor helps with all lenses except the worst one. Of course, an image from a 300 MP sensor viewed at 1:1 will not be as sharp as an image from a 150 MP sensor viewed at 1:1, but if the lens is anyway sharp it will render better.

Best regards
Erik

Been thinking about the immediate future of MF digital photography in regards to lens design. Let’s say Phase One comes out with a digital back that is 300 megapixels in a few years (which, I would think, would assuredly and finally match—perhaps surpass—8x10 *dodges rocks*).

Will we need all new lenses to take advantage of it? Will any manufacturer be willing to put in the work? And would we be able to afford them if they did? And if we did, would diffraction have us shooting at f/4?

I wonder if, knowing this, Phase One will keep the resolution at this limit for a while, at least for another generation, focusing on other value adds. But 35mm and crop MF may need to keep upping the resolution to stay competitive. Perhaps Fuji and others have the money to spend on all new lens designs.

Is this as good as it gets for MFDBs?

[ErikKaffehr]

If Sony and P1 took the current micro-architecture of the IQ4-150 sensor and put it to use in a full frame MFD back with 60 MP, I would buy that in a heart beat. 

I doubt that will be the case, so more then likely I will purchase the IQ4-150 when refurbs come on the market and that will be the last back I ever buy.  To be perfectly honest, lens cast is the only reason I want to upgrade, but beyond that, there is nothing I see as being ground breaking in the digital back market. 

After this, any other camera I buy will be a mirrorless system.

Hi,

If you put the micro-architecture of the IMX 411 in a 54x41 mm back you get 151 MP back. The closest sensor for 60 MP would be the Sony IMX 410 sensor used in the A7III.

I asked about the costs of reusing an existing pixel design, like IMX 410 for a larger sensor, and it would probably incur like a dozen million dollars in NRE costs, in addition to manufacturing costs, which would probably be similar to the 151 MP IMX 411 in cost.

Best regards
Erik

[BAB]

Hopefully newer up and coming sensors will go in a direction other than mp. The future has so many more options and so does my imagination. Capturing light in a much faster, steadier way and incorporating pixels to extend not only to a higher dynamic range but also deeper focus ability. However photographic camera manufacturers seem to be just a small fraction of the core sensor business.

[JaapD]

I see serious advantages in more MP because in nearly all cases we’re currently undersampling (sampling at a lower frequency w.r.t. the incoming analogue signal, resulting in artefacts). Let’s for once start doing oversampling, just as what’s customary with audio for a long, long time.

Of course I’m aware of the negative side benefits of shrinking pixel size but the lost s/n ratio can be regained after downsampling near the end of the image processing chain.

For completeness I need to say that there is a certain optimum in sensor design. At a certain moment there is too much silicon lost at wiring in between the pixels; pixel size vs. interconnection losses.

Regards,
Jaap.

[Jim Kasson]

Been thinking about the immediate future of MF digital photography in regards to lens design. Let’s say Phase One comes out with a digital back that is 300 megapixels in a few years (which, I would think, would assuredly and finally match—perhaps surpass—8x10 *dodges rocks*).

Will we need all new lenses to take advantage of it? Will any manufacturer be willing to put in the work? And would we be able to afford them if they did? And if we did, would diffraction have us shooting at f/4?

I wonder if, knowing this, Phase One will keep the resolution at this limit for a while, at least for another generation, focusing on other value adds. But 35mm and crop MF may need to keep upping the resolution to stay competitive. Perhaps Fuji and others have the money to spend on all new lens designs.

Is this as good as it gets for MFDBs?

My simulations for Bayer-CFA sensors indicate that at least 800 MP for a FF camera is required to resolve what existing lenses can lay down on the sensor, at least on-axis. More for larger sensors.

Jim

[ErikKaffehr]

Been thinking about the immediate future of MF digital photography in regards to lens design. Let’s say Phase One comes out with a digital back that is 300 megapixels in a few years (which, I would think, would assuredly and finally match—perhaps surpass—8x10 *dodges rocks*).

Will we need all new lenses to take advantage of it? Will any manufacturer be willing to put in the work? And would we be able to afford them if they did? And if we did, would diffraction have us shooting at f/4?

I wonder if, knowing this, Phase One will keep the resolution at this limit for a while, at least for another generation, focusing on other value adds. But 35mm and crop MF may need to keep upping the resolution to stay competitive. Perhaps Fuji and others have the money to spend on all new lens designs.

Is this as good as it gets for MFDBs?

Hi,

For one thing, Phase One, Hasselblad and Fujifilm don't decide on sensors. They use whatever sensors that sensor makers sell.

In many cases, I would suggest that it is rather the photographer than the lens that is the limiting factor:

• Good lenses perform optimum at say f/4-f/5.6. Stopping down farther reduces sharpness. It does little sense to make better lenses if we throw away sharpness stopping down.
• Precise focusing is needed, preferably at shooting aperture.
• Obviously, factors like stable tripod, release technique, use of electronic first curtain and other things matter a lot.

My point may be that getting the best of present lenses needs a lot of perfection from the photographer.

On the other hand, it may be that we don't need image quality at all, that is a pretty good argument. Once an image is good enough, adding image quality will yield diminishing returns. Can I make good 30"x40" prints, I am quite happy. Would I make large prints intended to be viewed close, that would be another thing.

I would argue that there are two reasons to increase resolution:

• Reducing aliasing in the images. Have correctly sampled image instead of low resolution artifacts on detail.
• Printing large.

Adding a small point. A boring image is a boring image, regardless of technical quality. But a great image may deserve excellent quality.

Best regards
Erik

[George_Cleansman]

Erik Kaffehr gets the point 

"Reducing aliasing in the images. Have correctly sampled image instead of low resolution artifacts on detail"

I think the better demosaicing process with increasing pixel numbers is the most important reason - until we get diffraction issues in case of smaler pixel stitches (<3,76µ) and f-stops (<5,6) on the actually largest sensor 54x40mm.