Nikon D800E - Amazing Resolution ...but "Houston, We Have A Problem!"

[torger]

Here comes my story why I'm so curious about the pixel peep performance of the D800.

In the past it was quite well-defined what a camera could do. Each genre had its camera. 35mm for photojournalism, 6x7 for portraits, 4x5" for still life and landscape. This is no longer true, the D800 may be the first camera that truly extends of the the whole range from 35mm to 4x5". It is fast, it can do short DOF it can do high resolution.

A few months ago I started to think if I should change system. I had a Canon system. Landscape photography my main interest and also an interest in large and/or high-resolution prints. To support my prints I'd ideally like to have something like 60 sharp megapixels. I did stitching at the time but I prefer one-shot images. My mentality is such that I always want to shoot with the best gear I have, so when I happen to shoot an image with artistic quality enough for putting on the wall the technical quality should allow large prints.

D800 was on the way, but not released yet. I did a simple experiment with my TS-E 24mm II and an APS-C 7D, shifted to the full-frame corner and tested how sharp it was, that would correspond to 45 megapixels. It was then clear to me that one of the sharpest 24mm lenses on the 35mm system would not deliver even 40 sharp megapixels. Wide angles are known to be hard to make sharp corners with though.

Another aspect I had discovered is that tilt and shift had become almost an necessesity for my shooting style. With increased resolution the tilt function becomes even more important. Also my shooting style is very slow so I did not really make much use of the quickness of the DSLR format. And where's the 35mm wide tilt-shift lens, my most important field of view? The best performing solution today seems to be a TS-E 24mm 2 with 1.4 III tele converter, or use APS-C. A DSLR more and more seemed like a compromise, although a very affordable one.

I started to look at 4x5" film and scanning solutions. The long turn-around times with film was a problem though for me as an amateur with limited shooting time, I really gain from immediate feedback of digital in my artistic development process. I looked at scanning backs for 4x5" but those were a bit too limited even for me. So for fun I started to look at MF digital and tech cameras. Since MF guys like to say "it is so much better than 35mm digital" all the time it was a bit tricky to find out its limitations. Also MF digital tech cameras have problems with wide angles, all formats have. But to a lesser extent. This format seemed to support my desire for those ~60 megapixels in the long term, and of course you get tilt/shift for all available focal lengths, the shifts are often a little smaller than possible on DSLR shift lenses though (which is not too big a problem since tasteful pictures cannot swallow large shifts anyway).

As we all know, MF digital pricing is crazy though. Oh well, Schneider "large format digital" lenses are actually quite affordable, and the tech camera bodies no worse than a pro DSLR body, but the digital backs... ouch! For an amateur there is the option to buy second hand though, then an older 33 megapixel back can be had for 1.5x the price of a D800, and the whole mint second hand system ends up 2x the price of corresponding new D800 system with tilt-shift lenses (3x if you compare to second hand D800). If new digital backs costed say $5000 instead of $20,000 I think many more amateurs would choose the tech camera path ahead of a DSLR.

I ended up chosing a second hand MF tech camera system rather than buying a new D800 system. Still kept large parts of my Canon system though which I use for more action-like photography.

But now when the D800 is out I am still indeed curious about how all the available lenses performs with it. If the DSLR systems get too good I may very well switch back. I'm already a bit in love with the 100% mechanical tech cameras though, it will be hard to part from it.

[ErikKaffehr]

Hi,

Perfectionism has a price...

From where I stand I see 36 MP more like an incremental upgrade from 24MP, and I would suggest that a lens is good enough for 24 MP will also be good enough for 36 MP. It seems that some of the Canon TS lenses actually work very well with MF, as demonstrated several times by Stefan Scheib.

I foresee an evolution in mirrorless cameras, like a pro style NEX camera with FF sensor, such a camera would take essentially any lens, Leica R, Leica M, MF and LF. Lens cast would still be an issue. I would expect something like that show up at Photokina. It's just common sense...

Best regards
Erik

Here comes my story why I'm so curious about the pixel peep performance of the D800.

In the past it was quite well-defined what a camera could do. Each genre had its camera. 35mm for photojournalism, 6x7 for portraits, 4x5" for still life and landscape. This is no longer true, the D800 may be the first camera that truly extends of the the whole range from 35mm to 4x5". It is fast, it can do short DOF it can do high resolution.

A few months ago I started to think if I should change system. I had a Canon system. Landscape photography my main interest and also an interest in large and/or high-resolution prints. To support my prints I'd ideally like to have something like 60 sharp megapixels. I did stitching at the time but I prefer one-shot images. My mentality is such that I always want to shoot with the best gear I have, so when I happen to shoot an image with artistic quality enough for putting on the wall the technical quality should allow large prints.

D800 was on the way, but not released yet. I did a simple experiment with my TS-E 24mm II and an APS-C 7D, shifted to the full-frame corner and tested how sharp it was, that would correspond to 45 megapixels. It was then clear to me that one of the sharpest 24mm lenses on the 35mm system would not deliver even 40 sharp megapixels. Wide angles are known to be hard to make sharp corners with though.

Another aspect I had discovered is that tilt and shift had become almost an necessesity for my shooting style. With increased resolution the tilt function becomes even more important. Also my shooting style is very slow so I did not really make much use of the quickness of the DSLR format. And where's the 35mm wide tilt-shift lens, my most important field of view? The best performing solution today seems to be a TS-E 24mm 2 with 1.4 III tele converter, or use APS-C. A DSLR more and more seemed like a compromise, although a very affordable one.

I started to look at 4x5" film and scanning solutions. The long turn-around times with film was a problem though for me as an amateur with limited shooting time, I really gain from immediate feedback of digital in my artistic development process. I looked at scanning backs for 4x5" but those were a bit too limited even for me. So for fun I started to look at MF digital and tech cameras. Since MF guys like to say "it is so much better than 35mm digital" all the time it was a bit tricky to find out its limitations. Also MF digital tech cameras have problems with wide angles, all formats have. But to a lesser extent. This format seemed to support my desire for those ~60 megapixels in the long term, and of course you get tilt/shift for all available focal lengths, the shifts are often a little smaller than possible on DSLR shift lenses though (which is not too big a problem since tasteful pictures cannot swallow large shifts anyway).

As we all know, MF digital pricing is crazy though. Oh well, Schneider "large format digital" lenses are actually quite affordable, and the tech camera bodies no worse than a pro DSLR body, but the digital backs... ouch! For an amateur there is the option to buy second hand though, then an older 33 megapixel back can be had for 1.5x the price of a D800, and the whole mint second hand system ends up 2x the price of corresponding new D800 system with tilt-shift lenses (3x if you compare to second hand D800). If new digital backs costed say $5000 instead of $20,000 I think many more amateurs would choose the tech camera path ahead of a DSLR.

I ended up chosing a second hand MF tech camera system rather than buying a new D800 system. Still kept large parts of my Canon system though which I use for more action-like photography.

But now when the D800 is out I am still indeed curious about how all the available lenses performs with it. If the DSLR systems get too good I may very well switch back. I'm already a bit in love with the 100% mechanical tech cameras though, it will be hard to part from it.

[torger]

In other words, not interested in trashtalk but very interested in discussing and finding out the system's limits.

The D800's reach into MF territory is similar to comparing teleconverters plus crop bodies as an alternative to ultraexpensive superteles. If your requirements are fulfilled there is a lot of money to save.

[BernardLanguillier]

I foresee an evolution in mirrorless cameras, like a pro style NEX camera with FF sensor, such a camera would take essentially any lens, Leica R, Leica M, MF and LF. Lens cast would still be an issue. I would expect something like that show up at Photokina. It's just common sense...

That'd be cool! Let's hope that Sony keeps pushing the enveloppe.

It will also be interesting to see how Canon decides to enter the mirrorless market.

Cheers,
Bernard

[ErikKaffehr]

Hi,

Yes, that would be cool. In my view it is just common sense. I have no insight in Sony plans. Making an FF mirrorless camera just makes a lot of sense to me.

Bst regards
Erik

That'd be cool! Let's hope that Sony keeps pushing the enveloppe.

It will also be interesting to see how Canon decides to enter the mirrorless market.

Cheers,
Bernard

[Fine_Art]

I agree.

Reply to thread title: The cup is not forever half empty.

IMO the economic downturn is forcing all these companies to provide what they have rather than doling it out in dribs and drabs. Its the only way they can keep earnings going. This is highly democratizing to the field. It used to be that there was a clear distinction in quality between the top gear and the stuff a consumer could buy at a regular store. This was a barrier to entry for new pros. The solution was throw money at it. I think those days are over. Now it is clearly at the point that the togs make the equipment look good. Why? Becasue all the equipment is very good.

There is a guy in eastern europe that has been using the A900 since it came out. infrastellar.net He is willing to sleep in a tent at the top of a mountain to get the morning glory shot. He will always have stuff others can't touch. The same for Bernard making nikon look good. Or Micheal with whatever he happens to be using. etc for hundreds of people on this site.

The power has shifted from the manufacturer to the photographer.

[CalvinHobbes]

"The power has shifted from the manufacturer to the photographer."

Do not kid yourself, the people with the power are those at the top of the wealth ladder. At the moment there is some trickle down from the top cameras to the lower tier cameras. The high end DSLR/MF(or other high end camera) market will change. Canon and Nikon will realize as the wealth differential continues to change in the industrialized world, the wealthy will pay exorbitant amounts of money to have the top of the line gear. This gear must be marketed as significantly better than what the general population can buy.

This is already manifesting itself in the high end home audio industry. Just last year 20k+ for speakers was considered expensive. Now some manufacturers have 3 different models of speakers  above $100,000.00.  At a show here in socal one of the demo rooms is proudly displaying the price of a pair of speakers, preamp, cables cd player and amplifier for a grand total of $275,000.00. Dealers are locating there stores near where the mass of the super wealthy live.

Why try and sell many $5,000.00 speakers, where one has to work to convince people it is worth the money to upgrade from their HTIB. Much easier to sell $50,000.00 speakers and associated crap to the uninformed who need it to validate their own existence and ego.

[David Watson]

"The power has shifted from the manufacturer to the photographer."

Do not kid yourself, the people with the power are those at the top of the wealth ladder. At the moment there is some trickle down from the top cameras to the lower tier cameras. The high end DSLR/MF(or other high end camera) market will change. Canon and Nikon will realize as the wealth differential continues to change in the industrialized world, the wealthy will pay exorbitant amounts of money to have the top of the line gear. This gear must be marketed as significantly better than what the general population can buy.

This is already manifesting itself in the high end home audio industry. Just last year 20k+ for speakers was considered expensive. Now some manufacturers have 3 different models of speakers  above $100,000.00.  At a show here in socal one of the demo rooms is proudly displaying the price of a pair of speakers, preamp, cables cd player and amplifier for a grand total of $275,000.00. Dealers are locating there stores near where the mass of the super wealthy live.

Why try and sell many $5,000.00 speakers, where one has to work to convince people it is worth the money to upgrade from their HTIB. Much easier to sell $50,000.00 speakers and associated crap to the uninformed who need it to validate their own existence and ego.

Sure there are "vanity" products around which some people will pay silly money for but is that a bad thing?  It is all economic activity and whilst a little trivial it is still better than working 6 days a week, 12 hours a day down a coal mine.  Each camera that is sold at whatever price pays wages and taxes and creates employment and the higher the price the more added value goes into creating employment and tax.  I would much rather that rich individuals recycled their wealth in this way than put it in an off shore bank account and sat on it.  $100,000 MF camera - bring it on.  Some rich fool will buy it. LOL

[BernardLanguillier]

Sure there are "vanity" products around which some people will pay silly money for but is that a bad thing?  It is all economic activity and whilst a little trivial it is still better than working 6 days a week, 12 hours a day down a coal mine.  Each camera that is sold at whatever price pays wages and taxes and creates employment and the higher the price the more added value goes into creating employment and tax.  I would much rather that rich individuals recycled their wealth in this way than put it in an off shore bank account and sat on it.  $100,000 MF camera - bring it on.  Some rich fool will buy it. LOL

Although the trend is global, there are huge geographic differences.

A good indicator is the ratio btwn average CEO vs base employee salary. The US is at around 400 while Japan is at 15.

The question then becomes that of markt prioritization. It appears that the needs of the Japanese market still have a very heavy weight in most Japanese companies marketing/product planning decision process.

Asca rule of thumb, Elite good pricing has to be 5 to 10 times more than general market high end to appeal to the super rich, and performance along the relevant metric probably 50% higher. The relevant metric differs per domain. For high end audio, it is clearly physical appearance rather than sound.

What is it for photographic equipment?

Cheers,
Bernard

[theguywitha645d]

Some rich fool will buy it. LOL

You mean the idiot who has managed to make more money than you and can actually afford to drop a ton of money on something that would give him or her pleasure?

[David Watson]

You mean the idiot who has managed to make more money than you and can actually afford to drop a ton of money on something that would give him or her pleasure?

Yes quite right too.  I guess we all validate our financial success by spending at an "appropriately" high level - why else did we we use our lives grabbing all that loot in the first place?  Just kidding?

[kers]

Coming back to Housten or Houston…
It seems that the 24mm PCE is a mysterious lens… and not very good and a lot of sample variation…

Now i only have only one example and mine can be bad ..and good.
if i focus in the middle - the corners are soft , but if i focus on the side the corners are a lot better and the middle is still sharp..
I think it all has to do with field curvature, with the sides focussing closer. The great depth of field at f10 supplies the center sharpness. ( indeed you need f9-f11 when you shift)

I took this one shot -d800e- so it may be better but in any case not worse.
Very difficult situation with no light and 3 minutes exposure… (no wind!)
I think that the pce lens could be better, but it is not as bad as the sample in the thread. I like the mechanics of the Canons a lot.

I agree that the Liveview of the d800 could be better in low light and that red light that illuminates the focus marks spreads too much light in the viewfinder.
Still i could focus.

btw I find the Zeiss wide angle lenses usually very good but not so much in the extreme corners- You need d11 usually to cover those.

[Ray]

To get the topic back on track, below are some comparisons between my D800E and D700. The shots were hand-held, using the 14-24/2.8 zoom on each camera at 14mm, the focal length at which the lens is sharpest.

I used F11 because this seems an appropriate aperture if one wants everything reasonably sharp from foreground to infinity, and especially if one wants maximum corner sharpness. It also reduces concerns about misfocussing.

Both images have been processed similarly, and the same amount of sharpening applied in ACR, which was: Amount 50, Radius 1 pixel, and Detail 100. No noise reduction or masking was applied.

As regards corner resolution, bottom left corner, the hand-held procedure has resulted in the D700 having a slight advantage because the FoV was just a bit more towards the left and a little bit lower, so the corner detail in the D700 shot, in the comparison, is just a little bit further away from the extreme corner. I always like to give a slight advantage to the underdog.

The day was dull and cloudy, but almost totally calm, so any resolution differences cannot be attributed to subject movement.

However, I should mention a disclaimer. Two of the three pieces of equipment used in this test, the 14-24 lens and the D700 body have been completely submerged in water for at least two seconds, when I recently stumbled whilst crossing a river, waist deep. There's no doubt that water entered the camera, because the LCD screens later misted up with condensation from the inside. I had to dry out the camera body in the sun, all day long, with lens and body cap removed. The camera is still not completely functional but is useable. The lens seems unharmed.

Of course, the message here is for those who are concerned that the 36.3mp of the D800 are of little use unless one has superb lenses. Lenses are certainly not sharpest at F11, however good they may be, and are certainly worst in the extreme corners, however good they may be.

The D800E delivers better resolution even in the extreme corner which is representative of a very poor lens.

[BernardLanguillier]

Of course, the message here is for those who are concerned that the 36.3mp of the D800 are of little use unless one has superb lenses. Lenses are certainly not sharpest at F11, however good they may be, and are certainly worst in the extreme corners, however good they may be.

The D800E delivers better resolution even in the extreme corner which is representative of a very poor lens.

Indeed, the kind of gap shown here reminds me of what we typically saw when comparing 4x5 to medium format scans.

Cheers,
Bernard

[ErikKaffehr]

Ray,

I miss out on a couple of issues:

1) Images both scaled to D800 resolution?

2) In my view the 14-24/2.8 is a lens of fine reputation, it gets glowing prise from Photozone.

So I'd say yours is a valid comparioson of the D700 and the D800 using a very good lens. F/11 is probably not optimal but I'd regard it OK.

Best regards
Erik

To get the topic back on track, below are some comparisons between my D800E and D700. The shots were hand-held, using the 14-24/2.8 zoom on each camera at 14mm, the focal length at which the lens is sharpest.

I used F11 because this seems an appropriate aperture if one wants everything reasonably sharp from foreground to infinity, and especially if one wants maximum corner sharpness. It also reduces concerns about misfocussing.

Both images have been processed similarly, and the same amount of sharpening applied in ACR, which was: Amount 50, Radius 1 pixel, and Detail 100. No noise reduction or masking was applied.

As regards corner resolution, bottom left corner, the hand-held procedure has resulted in the D700 having a slight advantage because the FoV was just a bit more towards the left and a little bit lower, so the corner detail in the D700 shot, in the comparison, is just a little bit further away from the extreme corner. I always like to give a slight advantage to the underdog.

The day was dull and cloudy, but almost totally calm, so any resolution differences cannot be attributed to subject movement.

However, I should mention a disclaimer. Two of the three pieces of equipment used in this test, the 14-24 lens and the D700 body have been completely submerged in water for at least two seconds, when I recently stumbled whilst crossing a river, waist deep. There's no doubt that water entered the camera, because the LCD screens later misted up with condensation from the inside. I had to dry out the camera body in the sun, all day long, with lens and body cap removed. The camera is still not completely functional but is useable. The lens seems unharmed.

Of course, the message here is for those who are concerned that the 36.3mp of the D800 are of little use unless one has superb lenses. Lenses are certainly not sharpest at F11, however good they may be, and are certainly worst in the extreme corners, however good they may be.

The D800E delivers better resolution even in the extreme corner which is representative of a very poor lens.

[Ray]

Ray,

I miss out on a couple of issues:

1) Images both scaled to D800 resolution?

No, only the D700 was scaled to D800 resolution. 

[Fine_Art]

No, only the D700 was scaled to D800 resolution. 

Erik means both samples were scaled to D800 resolution on the D700 shots, which is what the title bars say. I'm sure you know it will degrade the image.

It does show the D800 makes good use of the lens.

[Ray]

Erik means both samples were scaled to D800 resolution on the D700 shots, which is what the title bars say. I'm sure you know it will degrade the image.

Yes, I know. Just having a bit of fun. Hope Erik forgives me. 

The following image is a comparison of the bottom left corners showing the D800E downsampled to the D700 size. I've also shown the full corners, edge to edge, which demonstrates the slight advantage I've given to the D700.

Downsampling a higher resolution image throws away image information. Upsampling a lower resolution image retains all the initial data. I think upsampling is the more truthful comparison.

[Ray]

What surprises me about the D800E is the amount of sharpening the images can take before appearing oversharpened. After all the discussion about the pros and cons of having no AA filter, I'd got the impression that one could achieve very similar resolution from the camera with the AA filter by applying more sharpening, and that such increased sharpening would be inappropriate for the other image taken without AA filter.

A sharpening amount of around 40 or 50 in ACR is generally what I use with images from my D700 and D7000 which has the same pixel density as the D800E, so I'm therefore surprised that a sharpening amount of 70, without masking, seems quite okay for the D800E, at ISO 100.

The following crops of the centre of the scene are from reconverted images using a sharpening amount of 70, 1 pixel radius, and detail 100. Instead of interpolating the D700 shot, I've equalized the size by applying greater magnification, comparing the D700 at 172% with the D800E at 100% on screen.

Both crops are sharper and better than my previous comparison, but from the same RAW files.

[hjulenissen]

Wondering if lensezone.de will use a D800E to test new lenses. Aliasing can be used as a measure of lense sharpness if the input signal is known beforehand.

-h