[biz-engineer]

What is lost by going from a 50mp to a 100mp sensor?

Viewing distance. And camera weight and size (the 100Mp MF camera is smaller and lighter weight than the 50Mp MF camera).

[hjoseph7]

At any given stage of technological development, it's tended to be true that smaller pixels produce more noise. So, assuming we keep sensor size constant, higher resolution can come at the risk of increased noise and decreased light-capturing ability (so reduced sensitivity, DR, etc.). Obviously, the situation is more complex than this because some manufacturers achieve better results than others, processing can manage noise to some extent, and - of course - sensor size is not a constant (so getting a larger sensor means the pixels don't end up as small). However, overall, it's been fair to say that making the pixels too small by packing too many of them in brings these downsides - there's generally a sweet spot between resolution and these factors.

However, the key thing above is "at any given stage of technological development". With each generation of development, it has proven possible to achieve higher resolution or lower noise (or some combination of both) than earlier generations. So this needs to be considered. The 645Z and its sensor is now about 7+ years old; that's an age in digital terms and it speaks volumes for its quality that it still stands up so well. But if you're comparing it to current 100MP sensors, those are newer. I would therefore expect that they will deliver higher resolution without any reductions in quality as a result of noise/DR; that may be so at a pixel level, but certainly will be so at the final viewing size. I say this on the basis of others' experiences, data/samples I have seen and logical conjecture, as I don't have a 100MP camera at this point. It certainly fits with what I have seen in earlier developmental stages of the technology...

Hope this helps!

Ed

Its very confusing, a few years ago technology was at a stage where bigger pixels meant more light gathering capacity and less noise. These days that is no longer true, but I wonder how long will camera manufacturers be able to defy mother nature. I remember when I jumped from a 10 MP camera, to a 17 MP (APSC) camera and I was actually somewhat annoyed by the extreme resolution of the 17 MP pixel camera. I had gotten used to the smooth transition of light and shadows and colors that the lower resolution camera offered. It seemed to me that the lower resolution camera was more true to life as far as my eyes were concerned. I could also get a pretty decent 8X10" or 11x14" print from this camera with no problems.

The higher pixel camera was sort of like "Superman" using his x-ray vision to see what other people could not see. It was a little too much for me at first. However, I soon got used to the higher meg camera when it came to cropping and printing. I could crop out 50% of an image and still come out with a nice sharp picture. I could also produce bigger prints although I never really go past 11X14" unless its a special project, because the cost of the ink and paper would be too enormous.

I have since jumped to a 24 megapixel Full Frame camera, 14 more megapixels than my original DSLR, however this is a Full Frame camera not APSC ! This camera exceeds my expectations when it comes to sharpness, smoothness, gradations of shadows, colors etcetera. And it works really nicely in low-light too. Since the size of hard-drives has increased tremendously over the past years, I have no problem downloading images to my computer from this camera either, but images of this size can still be a problem when it comes to down loading times. I once had to download 1500+ images to my computer and it took over 1/2 hour !

These days I still look at pixel sizes, but I don't think it's as important anymore. I really can't see what the advantage of owning a 100 MP camera, unless I'm printing bill-boards on a daily basis. I think there is a sweet spot to mega pixels and everything else is Marketing hype, or special needs such as as printing bill-boards. I think camera manufacturers should focus on other things, such as Dynamic range and low light photography.

[SFTphotography]

Let's just hope Pentax actually go for the 100mp 44x33 sensor and don't use on sensor phase detect.

[TonyW]

Let's just hope Pentax actually go for the 100mp 44x33 sensor and don't use on sensor phase detect.

Why not sensor phase detect?

[SFTphotography]

Why not sensor phase detect?

PDAF striping and banding. Terrible thing caused by having the AF system on the sensor. No need for it on a DSLR either - maybe on a mirrorless but not having on sensor phase detect AF is the biggest advantage of a DSLR over a mirrorless one at the moment, other than the OVF

[TonyW]

PDAF striping and banding. Terrible thing caused by having the AF system on the sensor. No need for it on a DSLR either - maybe on a mirrorless but not having on sensor phase detect AF is the biggest advantage of a DSLR over a mirrorless one at the moment, other than the OVF

Yes, I am aware of the furore in certain quarters about the banding issues. But TBH I think they have been overstated. So much so that I have gone with the Fuji GFX series. There were some issues with the first release of the GFX 100 but these have been cured by firmware updates. The work done by Jim Kasson suggest this is pretty much a none issue. There are also a few advantages to PDAF.

You could also look at this another way. Not having PDAF and CDAF on a camera is a big disadvantage as you cannot see exactly the point of focus due the disconnect of an OVF system to the sensor hence back and front focusing issues not seen until after the event. Similarly using EVF will give direct feedback from sensor level over an OVF.

IF I am wrong on this I have just made a very expensive mistake along with Fujifilm, Nikon et al. 😯

[barondla]

Yes, I am aware of the furore in certain quarters about the banding issues. But TBH I think they have been overstated. So much so that I have gone with the Fuji GFX series. There were some issues with the first release of the GFX 100 but these have been cured by firmware updates. The work done by Jim Kasson suggest this is pretty much a none issue. There are also a few advantages to PDAF.

You could also look at this another way. Not having PDAF and CDAF on a camera is a big disadvantage as you cannot see exactly the point of focus due the disconnect of an OVF system to the sensor hence back and front focusing issues not seen until after the event. Similarly using EVF will give direct feedback from sensor level over an OVF.

IF I am wrong on this I have just made a very expensive mistake along with Fujifilm, Nikon et al. 😯

My understanding of this is the af areas of the sensor are "dead" to image recording. If true, how can it be fixed via firmware? Covered up, but not fixed? Everything is a compromise. Fuji, Nikon, and everyone else may have decided people valued lightweight mirrorless with good focusing over stripes in very underexposed images. Not necessarily a bad gamble.

Thanks,
barondla

[SFTphotography]

Yes, I am aware of the furore in certain quarters about the banding issues. But TBH I think they have been overstated. So much so that I have gone with the Fuji GFX series. There were some issues with the first release of the GFX 100 but these have been cured by firmware updates. The work done by Jim Kasson suggest this is pretty much a none issue. There are also a few advantages to PDAF.

You could also look at this another way. Not having PDAF and CDAF on a camera is a big disadvantage as you cannot see exactly the point of focus due the disconnect of an OVF system to the sensor hence back and front focusing issues not seen until after the event. Similarly using EVF will give direct feedback from sensor level over an OVF.

IF I am wrong on this I have just made a very expensive mistake along with Fujifilm, Nikon et al. 😯

You can always check in live view before focusing. As a landscaper I've not had one camera with on sensor phase detect AF and not missed it. Then again I shoot stopped down.

The issue isn't overstated on some cameras, I've seen stripes across a sky because of it (off a Nikon Z something or other). It was bad. Some camera's are worse than others for it - and I'd rather not have it at all.

[TonyW]

My understanding of this is the af areas of the sensor are "dead" to image recording. If true, how can it be fixed via firmware? Covered up, but not fixed? Everything is a compromise. Fuji, Nikon, and everyone else may have decided people valued lightweight mirrorless with good focusing over stripes in very underexposed images. Not necessarily a bad gamble.

Thanks,
barondla

I guess similarly to the “dead areas” containing circuitry on a Front side illuminated sensor which sits between the lens and colour filters before It gets to the photodiodes.

Fixed and covered up can be considered to be the same. Just think about how in a Bayer sensor an image needs to be demosaiced by interpolation of surrounding areas. What is applied in firmware and how it works is not exposed to the likes of us mere mortals this us proprietary information.

What I can say with some certainty is that the GFX 100S does not by all accounts appear to have banding and the original GFX 100 banding seems to have magically disappeared during a firmware update.

Fujifilm GFX100S Shows No More PDAF Banding (and Silently Fixed also on GFX100 via Firmware Update?) - Fuji Rumors

[TonyW]

.....
The issue isn't overstated on some cameras, I've seen stripes across a sky because of it (off a Nikon Z something or other). It was bad. Some camera's are worse than others for it - and I'd rather not have it at all.

Usually the statements about banding form around issues of pushing an underexposed image by +5 or +6EV. In that case that is usually the photographers fault and trying to recover such a badly exposed image is a waste.

At other times this can be caused by using Electronic or Electronic first curtain shutter not the PDAF, the cure being use the mechanical shutter - maybe?

There is also the need to identify if one is talking about PDAF banding or PDAF striping? The striping is caused by reflections associated with flare while the banding is caused by the attempt to fix the striping by some manufacturers.

In this case manufacturers thinking prevention is better than cure, but finding that the cure turns out to be worse than the disease. Perhaps just better left to post processing solutions for the rare occurence.

Just think of the outcry when 36MP cameras announced without AA filters - 'Moire will irreversibly damage many images'. Easily avoided if spotted at the capture stage by small camera movement, and usually curable after the fact in post especially chrominance moire.

[SFTphotography]

Usually the statements about banding form around issues of pushing an underexposed image by +5 or +6EV. In that case that is usually the photographers fault and trying to recover such a badly exposed image is a waste.

At other times this can be caused by using Electronic or Electronic first curtain shutter not the PDAF, the cure being use the mechanical shutter - maybe?

There is also the need to identify if one is talking about PDAF banding or PDAF striping? The striping is caused by reflections associated with flare while the banding is caused by the attempt to fix the striping by some manufacturers.

In this case manufacturers thinking prevention is better than cure, but finding that the cure turns out to be worse than the disease. Perhaps just better left to post processing solutions for the rare occurence.

Just think of the outcry when 36MP cameras announced without AA filters - 'Moire will irreversibly damage many images'. Easily avoided if spotted at the capture stage by small camera movement, and usually curable after the fact in post especially chrominance moire.

I think the striping will be quite common, particularly for landscapers where there is sky etc. I've seen it - and it isn't pretty. I'd rather not have PDAF on sensor on a camera.

[TonyW]

I think the striping will be quite common, particularly for landscapers where there is sky etc. I've seen it - and it isn't pretty. I'd rather not have PDAF on sensor on a camera.

That is not the opinion of most including many that actually own the cameras including those with Nikon Z’s

On any examples I have looked at it is always flare induced, and reported to be worse with particular lenses. Worst case cured in two minutes in PS. But manufacturers can mitigate the effect pre demosaicing raw image - just like they do for other artefacts.

As your opinion is it will be quite common you should be able to provide some evidence of landscape showing the issues. I would love to see a link to a large image that I can download and evaluate, ideally raw please

[SFTphotography]

That is not the opinion of most including many that actually own the cameras including those with Nikon Z’s

On any examples I have looked at it is always flare induced, and reported to be worse with particular lenses. Worst case cured in two minutes in PS. But manufacturers can mitigate the effect pre demosaicing raw image - just like they do for other artefacts.

As your opinion is it will be quite common you should be able to provide some evidence of landscape showing the issues. I would love to see a link to a large image that I can download and evaluate, ideally raw please

Not got the RAW...but...

Banding in photos (IR converted camera) | Talk Photography

---------- Post added 04-01-21 at 03:34 AM ----------

PDAF striping in the Sony a7RIV

Not a lot - but it's there if you push contrast/micro contrast etc.

And I do not see an advantage to phase detect AF on sensor.

https://www.dpreview.com/forums/post/60909421

[TonyW]

Not got the RAW...but...

Banding in photos (IR converted camera) | Talk Photography

---------- Post added 04-01-21 at 03:34 AM ----------

PDAF striping in the Sony a7RIV

Not a lot - but it's there if you push contrast/micro contrast etc.

And I do not see an advantage to phase detect AF on sensor.

Why the A7III PDAF striping samples look so much worse: Sony Alpha Full Frame E-mount Talk Forum: Digital Photography Review

Well no landscape raw with sky problems, there is an oft repeated quote around these parts "you have no empirical evidence". So really what we are left with is promoting the amount of internet FUD that surrounds these issues.

It has already been shown that for the Fuji GFX100 and 100S that this is a non issue - Jim Kasson. When it was an actual issue with the GFX100 Fuji fixed it. Do you seriously think that I and hundreds of others have not excercised due dilligence prior to ordering our mirrorless systems?

Similarly Sony seemed to have done a similar job:
https://www.dpreview.com/news/6974141509/sony-striping-heres-the-fix

I know a lot of Nikon shooters all of which are landscapers, birders etc. and known to give honest opinion and answers to problems. So far no recent Fuji users are having problems after the last firmware update – at least that is the feedback I am getting.
Here are a few real world Nikon Z users answers to the question of striping and banding. So far only one has seen banding on a couple of severely underexposed images and another ‘thinks’ he has seen it twice!

Shot 5000 pics on my Z6 not seen any banding or striping

* I’ve had my Z6 for about 1.5 years now and Ive never seen it.

* Not on myZ50 either

* Checking the image count on my last photo, I've taken 170,429 images with my Z6 since late 2018. I have only seen the PDAF banding in perhaps one or two images, and only then because they were severely underexposedand I tried to increase exposure by 5+ stops. The banding was not as bad as the general noise, so I did not attempt to salvage the images anyway. I have had other images that were similarly underexposed that I was able to salvage without any apparent banding.Is this because Nikon "fixed" the issue in late firmwarereleases? Did Lightroom mask the banding with software improvements?I'm not sure, and it doesn't really matter; bottom line is that PDAFbanding is really such a rare issue that its not something to worryabout.

* It's a rare issue. I think I've seen it twice in the two years I've hadthe Z6/7 in images as Aaron describes above. Nikon isn't the onlycamera with this issue.

* No issue with the Z7 in the almost 2 years since I've had it.

* I have no such problem.

* No issue with the Z7 in the almost 2 years since I've had it.

* I haven't seen it in two years.

* Non-issue.Z7 and Z6.

* I've a Z6 and a Z7 and all I can say is is,, "What banding problem?"

There are more than a few benefits to on sensor PDAF apart from direct and precise and fast focus. For the landscaper an accurate preview of DoF can be shown prior to shooting including in some cases a display of the focus distance and DoF extents including the ability to set and display either a film based, or perhaps more importantly pixel based.

[TDvN57]

.....On your other point, I am reminded of something that an old friend of mine used to say. "Only stop when the camera is resolving more detail than the real world! Unless each pixel is smaller than a quark, it's not enough resolution! But make sure you keep massive depth of field as well!" (he was actually talking about film grain rather than pixels, but the point still stands)

And before anyone asks, yes, I did make my friend up. It was actually me who used to say that :-)

Your friend seems like a smart guy to have around. Send our collective regards.... :-)

.