D810, All Discussions, Questions

Ade

"Technically" Apple provides RAW support at the OS level, not through applications like Aperture or iPhoto. So you "should" be able to use D810 RAW files with the current version of Aperture, whenever D810 support comes to 10.9 Mavericks (and also when Yosemite comes out in the fall).

At some point in the future, however, Apple may update its OS RAW engine in an incompatible way with Aperture, and then RAW updates from then on will no longer work with Aperture. Hopefully, they won't break this compatibility before D810 support is provided. (Very small but non-zero probability of this happening).

EiTaro

Is this a difference in sharpness or what? (image from Dan Vojtech Blog)

nycapbarista

Hi, new to forums and new to the FX camera world. If I have posted in the wrong area, I apologize. My question/scenario is that I ordered the new D810 about 30 hours after its release (from Adorama) and now my pre-order has turned into a backorder. I've never had an item on backorder before. Is it reasonable to assume that so many people placed orders through Adorama in that 30 hour window that I could be waiting months? I've read posts from people who have been backlisted before and it took between 4-6 months to receive their gear.

proudgeek

It's quite possible, but doubtful. I think that the pent up demand for the D800 was so great that when it came out the response was overwhelming for Nikon (the tsunami probably had an impact as well on their production capabilities). With the D800 having been released relatively recently (at least the gap between the D800 and the D810 is nowhere near the gap between the D700 and D800), I don't expect to see the same level of demand. But hey, I could be wrong.

Ade

@EiTaro hard to say, if anything looks like JPEG Picture Control setting differences to me.

nycapbarista

@proudgeek The tsunami incident was horrible. The loss of life and devastating blow to the community very sad and traumatic. I was reading somewhere that Thailand does not produce the same quality as Japan, but I can't really comment on that. I'm hopeful that the D810 will be everything that it hints at being and I'm excited to be able to be a part of the FX family. Thank you for responding .

Golf007sd

@nycapbarista: I have a feeling a good majority of the pre-orders will get filled. This is not like the D7000. Many photographers that own the D800 (except those that make a living at photography) will not be making the switch over. Thus, I strongly feel the supply will be able to fulfill the demand.

Sidebar: Wish I could say the same thing about the Sigma 50 1.4 here in the states....pre-ordered on day one (April 11th)....and I'm still waiting.....arrrrr

Golf007sd

Nikon D810 Buffer Size

Ironheart

Nikon D810 Buffer Size

Now that's what I call an upgrade!

Vipmediastar_JZ

The buffer indeed is an improvement.
I just watched the videos on this blog. Looks like Miss Aniela was shooting primes (200mm)

I also listened to the shutter noise d800 vs d810. Mine sounds similar to the D810 no longer sounds like its bouncing around after the mirror goes back into position. Its all in my head I know. Im wearing a tin foil nikon hat.

SO this group focus is Like Batman and Robin sucker punching you at the same time vs Batman Punching you and Robin standing by as a backup in case his cape gets in the way with Dynamics focus? Right? Right!

WestEndBoy

The buffer size is an awesome upgrade. It won't convince me to upgrade, but I know that I am going to be thinking about it.

Wuwho

@nycapbarista : I too preordered shortly after the announcement and I'm number 4 on the wait list. The way the shop I dealt with explained it to me was basically, they'll order one for me, but its up to Nikon to deliver it to them. However, I would tend to agree with what the others have said. I too think the demand will not be high enough to create a severe drought in supplies. If this was say the D900 or the mythological D400, I think the wait will be much longer.

Golf007sd

A good overall review:

Paperman

The following is from Page 6 of ZEISS " How to read MTF curves -II " , a booklet which Ade gave link to a few days ago. It caught my eye as I was strolling down pages and couldn't decide if it made sense or not. Tried replacing the indicated screen & sensor resolutions with today's figures ( I guess it makes no difference ) but still could not conclude anything. Wouldn't have taken the article seriously had it not been by ZEISS but they are "the" lens maker of all times so here is some food for thought on the earlier sharp/not sharp discussions.

Can the below be an explanation of why we never see the sharpness we desire ( those eyelashes never seeming 100% crisp too the eye ) when we view an image 1:1 on our screens or is it totally out of context. ( Last few paragraphs ) ?

" Viewing conditions

Most likely you are viewing the images provided as examples on a computer monitor. This gives us reason to look a little more closely at how the monitor properties may influence our perception of the images.

Image size

The 12MP digital camera used here has a Nyquist frequency of approx. 1400 line pairs per image height (image height being the short side of the 24x36 format; think of a picture in landscape format). It takes at least two pixels to display a line pair made up of a bright and a dark line. The camera has exactly 2832 pixels (2x1416) on 24 mm of image height.

The monitor would have to have at least as many pixels to be able to display this image information free of losses. However we will usually have to be satisfied with a lesser monitor performance, e.g. 1600 x 1200 pixels. The monitor can therefore only display parts of the full image without losses.

If one runs Photoshop on a monitor with 1200 pixels in the vertical direction, some of these pixels are taken up by the menu bars and the net number of pixels seen is, for example, only 1036 pixels. In the 100% view, in which each pixel of the data file is represented by a monitor pixel, only approx. one third of the image with a height of 2832 pixels is seen, which corresponds to approx. 13% of the area of the image.

If the monitor diagonal is for example 21" = 54 cm, the size of the whole camera image in the 100% view is 76 x 114 cm. Even if our demonstration images are smaller in absolute units (in order not to let the file sizes grow towards infinity!) you should always be aware that you are looking at parts of a poster-sized image.


Viewing distance

If the monitor has 1200 pixels distributed over an image height of 32.4 cm, it has 3.7 pixels per millimeter. Thus the resolution of the monitor screen is approx. 2 Lp/mm.

In the (nearly) loss-free 100% view, this also corresponds to the camera sensor performance: the image with a height of 76 cm is magnified 31-fold as compared to the camera image with a height of 24 mm. The sensor's resolution limit (Nyquist) that is determined by the number of pixels is just less than 60 Lp/mm.

Magnified 31-fold, this also corresponds to approx. 2 Lp/mm.

Viewing the image on the monitor from a distance of 50 cm, the maximum resolving power of the eye at this distance is approx. 4 Lp/mm. In simple terms, this is about twice as good as the monitor image.

For this reason, images in 100% view will never appear perfectly sharp to our eye. Both the performance limits of the monitor and the giant magnification of the image for the small viewing distance give rise to a certain degree of softness of the image.

Viewing a 100% view from a distance of 50 cm is a very critical view of the image. For a more realistic assessment, the viewed distance can be doubled, for instance.

paulr

Thank you Nikon, The combination of a Nikon D4S and the Nikon 810 will just about cover all genders of photography with ther backup of Nikon lenses, and still a fraction of the cost of Medium Format cameras with the versatility of the 35mm movability.

Ade

@Paperman

I haven't really checked the math, but I think there are a few situations today:

1) With "standard" monitors / LCDs. Using the booklet example of a 21" monitor with 1200 pixel vertical resolution, an image captured at Nyquist will be displayed at 2 Lp/mm (the resolution of the monitor). But since the eye can resolve 4 Lp/mm at typical viewing distance, the image at 100% may appear slightly softer than actual, similar to when zooming-in too much.

2) With high-resolution "retina" or 4K monitors. Here we have a different situation. Suppose D810 can capture around 102 Lp/mm. If the monitor size is still 21" then we are looking at a 3.3 Lp/mm equivalent from the sensor. But the monitor will actually display the details at 4K resolution (2160 vertical pixels), which in this case equates to 4 Lp/mm (205 dpi). So the image at 100% in this case will appear sharp since it matches the 4 Lp/mm our eyes can resolve.

3) Suppose we have a very high-resolution LCD that's well above (say) 300 dpi, or close to 6 Lp/mm. A 100% image viewed on this LCD from the same viewing distance may appear sharper than actual, similar to when zooming-out. (By the way the Retina iPhone is 326 dpi).

JJ_SO

If Nikon would have decided to use a touchscreen, cursor speed would be no issue. And the split screen will become the next slow thing in this aspect. Usability is not the strongest side of Nikon.

Paperman

@Ade
"I haven't really checked the math, but I think there are a few situations today:

1) With "standard" monitors / LCDs. Using the booklet example of a 21" monitor with 1200 pixel vertical resolution, an image captured at Nyquist will be displayed at 2 Lp/mm (the resolution of the monitor). But since the eye can resolve 4 Lp/mm at typical viewing distance, the image at 100% may appear slightly softer than actual, similar to when zooming-in too much.

2) With high-resolution "retina" or 4K monitors. Here we have a different situation. Suppose D810 can capture around 102 Lp/mm. If the monitor size is still 21" then we are looking at a 3.3 Lp/mm equivalent from the sensor. But the monitor will actually display the details at 4K resolution (2160 vertical pixels), which in this case equates to 4 Lp/mm (205 dpi). So the image at 100% in this case will appear sharp since it matches the 4 Lp/mm our eyes can resolve "

So assuming ZEISS math is accurate and that most of us don't have 4k or retina screens, does this mean many discussions/complaints here for the past years regarding images not being crisp sharp at 1:1 resolutions ( supported with samples ) do not have a sound base ? I am sure most of us here ( including me ) thought viewing an image 1:1 at pixel level was seeing the image as "as is" - as good as it can be.

Ade

@Paperman

Yes and no. It will be neat to evaluate D810 sample pictures at 100% when they become available. :-)

Thinking out loud:

- At 100% we are often evaluating focus, instead of resolution. Even at 2 Lp/mm it's easy to tell if the image is out-of-focus.

- Although very fine details won't be as sharp as actual at 100%, our brains tend to evaluate the entire image in view, not just the portion with the finest detail.

- E.g., shown a picture of an eye at 100%, our brains will immediately decide whether the picture as a whole is sharp or not. I.e., its only afterwards that we notice some mid-level details like eyelashes. Then it's only after even further evaluation that we might look at the very finest details like faint patterns on the iris.

- Also at 100% we are often comparing relative sharpness vs. other pictures at 100%, not absolute Lp/mm. I.e., we don't just look at one picture in isolation, but we flip through several pictures and some will appear "sharp" to us while others will appear "soft", in comparison to each other and to other pictures we've seen in the past.

As the ZEISS booklet points out, the image we see on the monitor at 100% may be softer than actual. So one very common mistake people do is to sharpen the image for output while looking at it at 100%, before the image is sized / cropped. This can result in over-sharpened pictures if not done with utmost care.

Ironheart

Isn't the easy answer to just go to 200% or 300%?

Ade

The opposite way... stop at 50%, or step back from the monitor (to increase viewing distance).

PB_PM

Looks like a nice camera, although as suspected there is nothing compelling enough about the update to drive the need to upgrade from a D800. The price hike to $3400 is also a little much, so I'll be watching for lightly used second hand D800's in the weeks and months to come ($2k seems to be the going rate). As for those who jump on it, good on you.

JJ_SO

Has anyone seen a downloadable RAW-file (no S-type, just "ordinary" 12 or 14bit losless) anywhere? Or a reliable information, if the D810 uses a new RAW-codec?

WestEndBoy

Looks like a nice camera, although as suspected there is nothing compelling enough about the update to drive the need to upgrade from a D800. The price hike to $3400 is also a little much, so I'll be watching for lightly used second hand D800's in the weeks and months to come ($2k seems to be the going rate). As for those who jump on it, good on you.

I think that if you were a video shooter, it might be compelling. However, I agree with this statement for still shooters.

Paperman

Still haven't figured out why a videographer shooting 1080p would go for a 36Mp camera....