Nikon D850: Announcement of an Announcement of Development

[BernardLanguillier]

How do they compare to jpegs out of the D810?

I don't think the D850 samples are out of camera jpgs, I believe they are most likely raw files processed through the Nikon raw converter.

To my eyes ISO64 seems even cleaner than D810, and cleaner than the files I am getting from my H6D-100c. But I do of course agree that we'll have to wait for the raw files analysis to know for sure.

Cheers,
Bernard

[Bart_van_der_Wolf]

Some interesting aspects selected from the Imaging-resources website's review:

11) The Nikon D850 uses a backside-illuminated (BSI) sensor, but not for the reason you might think.

    The D850 uses the first backside-illuminated sensor in Nikon's DSLR lineup, but when we asked about it, it turned out the reason wasn't to provide better low-light performance (its pixels are big enough that there's not much gain in ISO speed by moving the wiring to the back of the chip), but rather to give more flexibility in the chip's wiring, to achieve the high speed they were after. Interesting...

Might be Marketing speak?

14) Nikon says dynamic range will be as good or better than that of the D810, despite the higher pixel count.

    They've stated that there is no trade-offs to be made in balancing dynamic range at base ISO vs. higher ISOs, and that this sensor resolution represents the optimum balance for performance and image quality.

QED

Cheers,
Bart

[Jim Kasson]

Some interesting aspects selected from the Imaging-resources website's review:

Might be Marketing speak?

QED

Can't wait to get my hands on it and see if it's got DR-Pix, or something similar.

Jim

[BJL]

I was wondering. We'll soon be able to derive the actual information from Raw files ourselves, but aren't Back-Side Illuminated (BSI) sensors better in Quantum Efficiency (QE) but somewhat worse in Dynamic Range (DR), compared to front side illuminated CMOS devices?

I do not see any fundamental reason for BSI sensors to have lower DR; indeed it seems to have the potential for greater well depth leading to increased DR. But it might be the case with implementations so far. The tentative good news on DR is that the base speed is still the same low 64 ISO, with "pulls" down to 32.

[Jim Kasson]

I do not see any fundamental reason for BSI sensors to have lower DR; indeed it seems to have the potential for greater well depth leading to increased DR. But it might be the case with implementations so far. The tentative good news on DR is that the base speed is still the same low 64 ISO, with "pulls" down to 32.

BSI certainly didn't seem to hurt the a7RII DR.

[Bart_van_der_Wolf]

I do not see any fundamental reason for BSI sensors to have lower DR; [...]

I'm not sure, but wouldn't having the transfer gates behind the photo-sensitive areas, reduce the real-estate for deep "wells"?

Cheers,
Bart

[Ray]

One interesting comment that caught my attention in the Imaging Resource review was the claim by Nikon that the image quality of the D850 will be a full stop better than the D810 at higher ISOs.

"While they didn't attribute it to the BSI sensor, Nikon told us that the D850 should produce the same image quality (both JPEG and RAW) at twice the ISOs as the D810, a full-stop improvement. That is, the D850 at its top "native" ISO of 25,600 should deliver the same image quality as the D810 did at ISO 12,800."

Now, I very rarely use ISOs as high as 12,800. However, if that increased image quality also  at applies to ISOs such as 400, 800, 1600 and 3200, I would consider a full stop of improvement at such ISOs to be a very appealing upgrade.

However, I imagine the improvement over the D810 will vary according to the ISO chosen. Perhaps it will be a maximum of one full stop better at very high ISO's, and perhaps only a 1/2 stop better at ISO 400 or 800 or 1600; but that's still a worthwhile improvement.

This camera's going to generate an enormous amount of discussion because of the multitude of significantly upgraded features, compared with the D800E and D810.

[Christopher]

Missing some the best ISO... http://www.photonstophotos.net/Charts/PDR.htm#Nikon%20D810,Nikon%20D850

[Ray]

Missing some the best ISO... http://www.photonstophotos.net/Charts/PDR.htm#Nikon%20D810,Nikon%20D850

I'll wait for DXOMark's results which I've always found to be reasonably accurate according to my own rigorous testing.

[shadowblade]

I'll wait for DXOMark's results which I've always found to be reasonably accurate according to my own rigorous testing.

PDR and DxO data are closely correlated, with small differences due to sample variation. In fact, PDR values can be directly calculated from DxO data (not the DR value alone, but the whole DxO dataset).

But the PDR data on the D850 so far is incomplete - I'd wait for the complete dataset before passing judgment.

[davidgp]

I'm not sure, but wouldn't having the transfer gates behind the photo-sensitive areas, reduce the real-estate for deep "wells"?

Cheers,
Bart

Not really... before the electronics were in the way of the light path... BSI increases noise - ratio performance and DR... especially in smaller sensors... in bigger sensors like full frame, the benefits are smaller, since the wiring was already very small with respect the area of the pixel... there is probably a benefit, but an small one...



http://dgpfotografia.com

[davidgp]

question,

if i buy a perpetual version of Lightroom will it cover the d850?
or will it only be covered in the CS version.

Yes, unless Lightroom releases version 7, version 6 is updated like the CC verdion, new features will be disabled, but camera, lens support and bug fixes will be available for both...


http://dgpfotografia.com

[shadowblade]

Not really... before the electronics were in the way of the light path... BSI increases noise - ratio performance and DR... especially in smaller sensors... in bigger sensors like full frame, the benefits are smaller, since the wiring was already very small with respect the area of the pixel... there is probably a benefit, but an small one...



http://dgpfotografia.com

That doesn't actually affect the maximum DR, but the rate at which light can be collected. That affects high ISO performance (more light hitting the sensor in the same space of time means a higher SNR above base ISO and less amplification needed to achieve the same ISO), but moving the circuitry from the front to the back doesn't, in itself, increase the SNR or DR at base ISO. Only increasing the FWC or reducing the noise will do that.

In a way, it's a bit like changing the aperture on a lens - opening it up will let photons in faster, but won't actually change the maximum number which can be recorded by the sensor.

You can see it in comparing the A7r and A7r2 sensors (although there's more than just BSI to that). The A7r2 has similar DR at base ISO, but significantly more at high ISO. Since more photons are being collected in the same space of time (with fewer being blocked by circuitry/hitting non-photosensitive areas) the SNR achieved at any given ISO level above base is higher than for the non-BSI sensor.

[kers]

Yes, unless Lightroom releases version 7, version 6 is updated like the CC verdion, new features will be disabled, but camera, lens support and bug fixes will be available for both..
http://dgpfotografia.com

Ok thanks- that will be my upgrade path from CS6.
A lightroom 7 with a new Raw engine (2012!) would be even better...
Main problem now is for me is moiré and false colour in general.

[davidgp]

That doesn't actually affect the maximum DR, but the rate at which light can be collected. That affects high ISO performance (more light hitting the sensor in the same space of time means a higher SNR above base ISO and less amplification needed to achieve the same ISO), but moving the circuitry from the front to the back doesn't, in itself, increase the SNR or DR at base ISO. Only increasing the FWC or reducing the noise will do that.

In a way, it's a bit like changing the aperture on a lens - opening it up will let photons in faster, but won't actually change the maximum number which can be recorded by the sensor.

You can see it in comparing the A7r and A7r2 sensors (although there's more than just BSI to that). The A7r2 has similar DR at base ISO, but significantly more at high ISO. Since more photons are being collected in the same space of time (with fewer being blocked by circuitry/hitting non-photosensitive areas) the SNR achieved at any given ISO level above base is higher than for the non-BSI sensor.

Thanks for the info!


http://dgpfotografia.com

[davidgp]

Ok thanks- that will be my upgrade path from CS6.
A lightroom 7 with a new Raw engine (2012!) would be even better...
Main problem now is for me is moiré and false colour in general.

Even if you don't want to buy Lightroom 6, you can download adobe DNG converter to convert the Nikon Raws to DNG (when it supports the D850) and use then with your CS6... the main problem will be if you don't like DNG format or want to change to Capture ONE in the future, they don't have good support of DNGs as far as I know...

About moire, I know Lightroom CC has a tool to handle it, not sure when it was introduced, with 6 or previously...

Good news is that, with more MPx less probability of moire in your pictures


http://dgpfotografia.com

[BernardLanguillier]

I would give C1 pro a try, it works wonders on Nikon files.

Cheers,
Bernard

[kers]

I would give C1 pro a try, it works wonders on Nikon files.

Cheers,
Bernard

I did - a few times- but somehow did not like it-  maybe i will try it again with the next upgrade...
i like the uniform sharpness of ACR in the images ; for instance if you have a picture of grass ( football field)  capture-one mashes up some parts ( discontinuity) while ACR gives a more uniform sharpness from close to far.
For colour i would choose Nikons NX- but is miss the sharpness i get with ACR....

(Speaking of grass- i really am sensitive to that colour green on photographs often find it a very ugly colour... while in reality it is not an issue)

For sensitive photos i have a work around to deal with the colour issues in ACR. On the whole i like the workflow of ACR and also use it for 6400 asa images.
I test everything printing it on my 44 inch printer 150 dpi to check.

[Michael Erlewine]

Here is a question I have. The Nikon D850 uses a new battery, the EN-EL15a. But somewhere I thought I read that you can use the EN-EL15 batteries from the D810, but have a little less power. Does anyone know the answer to this, please?

Added later to this post: It looks like both batteries will work. Correct?

[Rob C]

I did - a few times- but somehow did not like it-  maybe i will try it again with the next upgrade...
i like the uniform sharpness of ACR in the images ; for instance if you have a picture of grass ( football field)  capture-one mashes up some parts ( discontinuity) while ACR gives a more uniform sharpness from close to far.
For colour i would choose Nikons NX- but is miss the sharpness i get with ACR....

(Speaking of grass- i really am sensitive to that colour green on photographs often find it a very ugly colour... while in reality it is not an issue)

For sensitive photos i have a work around to deal with the colour issues in ACR. On the whole i like the workflow of ACR and also use it for 6400 asa images.
I test everything printing it on my 44 inch printer 150 dpi to check.

How odd; I find exactly the same problem. I also feel a bit negative to yellow in the same way.

In fact, I am getting to the stage where I don't really like colour photography much at all. There are some images where monochromatic colour is used to wonderful effect, but on the whole, bright, strong colours turn me right off. It's as if the picture can't make its mind up.

Rob C