The 36 megapixel announcement has been made, but it is interesting to note that Calumet say the sensor is 'newly developed Nikon FX' - it may still be Sony at heart and not a Nikon-own development.

The 800E version has no AA filter, this is an interesting development combined with 16-bit image processing as it makes the D800 equal to many medium-format digital backs in theoretical performance.

Full audio control with line input option (as I have repeatedly pressed for in reviews and on web pages) is implemented with 30 manual volume levels, Sony please take note, and headphone output and visual sound level monitoring. The HD video crop to DX from FX is good too, identical video performance whether you use 1X or 1.5X crop!

Canon's new 24-70mm f/2.8 (the existing one was really poor but try getting any Canon user to see it - I knew pros who switched to Sigma from Canon for this lens) has no IS. This is just remarkable when the Tamron new model has VC for Canon. Yet they have revised the ancient 24mm and 28mm f/2.8 designs (nice move) and included IS, claiming the world's first stabilised prime wide angles.

Of course, as Sony/KM owners know, we've had stabilised wide angles since 2004... and on the A77, we've got em for video too.

David

I suppose if you attempted to argue the point with Canon about the worlds first stabilized primes not actually being 100% correct, I bet they would just say ‘we said the lens was stabilized, not the camera’….and I guess they’d be right.
But then we know what is really going on, (you have to get up pretty early in the morning to pull the wool over our alert little peepers) if you were too limit the argument parameters to address only ‘image stabilization’ then that’s a whole different kettle of fish.
Greg

So, here we are, into the real of supremely high resolution, testing our lenses to their limits.

I like that. I always want my lenses to be the limiting factor, because then I know I can catch as much detail as I can, and can always buy better lenses to capture more, IF i feel the need.

Actually, I'm content with 14 MPs on APS-C, but will probably opt for a 580 when funds permit. I'd also like a 900, but can't find $2700 right now in my budget!

36 MPs means we then have the ability to shoot wide and crop, one of my favorite things to do on Medium Format or 4x5. (Not really much choice, long lenses are few, expensive, and relatively short in reach.)

If Sony WERE to come out with a FF NEX, at 36 MP, it, too could be used with short lenses and cropping to keep a compact package at hand to do more shots than the present combo really allows.

Dusty

A few comments...

36Mp with typical FF DSLR lenses is simply not enough, as those lenses normally have much higher resolution, at least in the central part of image circle. So moiree is guaranteed with certain targets.

Then some measly 22Mp without AA filter on MF may be OK due to generally lower manufacturing precision of MF lenses, and 36Mp on MF is almost always OK. So the MF lenses provide the necessary blurring in absence of AA filters.

But the blurring effect of AA filter could be far better achieved by very small high-frequency sensor vibrations. A cheap ultrasonic piezoceramic or quartz actuator shaking the basically static sensor assembly during exposure would easily do the trick. And there would be no need to ever use any stupid AA filters. But I guess those camera manu's are simply too stubborn...

Now having in-lens stabilisation in the primes simply defies the very purpose of primes. The image degradation from taking exposure while the active element is decentred immediately makes the use of primes unnecessary, as the more versatile zooms already produce the results every bit as bad and no worse. And then the only advantage of the primes is the possibility of larger max apertures like f/1.4-f/2.0; but an f/2.8 optically stabilised prime is just there next to the oxymoron.
The only way to properly stabilise a prime is to move the sensor, or better yet to mechanically stabilise the system as a whole.

..................

Well, that's enough venting. Anyway, those developments will bring some good to some users. But I really hate to see how brain-dead are the approaches taken

Thom Hogan says "The sensor itself is a Sony EXMOR-derived one, which is mostly good news. Since the sensor has the ADC circuitry on-board and this results in low read noise and good linearity, we should get the strong shadow capabilities we've seen in cameras like the D7000 and NEX-7."

http://www.bythom.com/d800intro.htm

He is usually pretty accurate about Nikon .....

Andy

allowing for the typical ~30% discount from SRP that Canon lenses typically settle at after the initial demand is satisfied you will still be looking at ~£500 for the 2 f2.8 stabilised primes & £1900 for the non-stabilised zoom. Unlesss it's a phenomenal lens it makes a Nikon (£1200) or ZA (£1400) look good value.

It would be nice if Sony updated the 28 & 20mm f2.8 primes & re-introduced an f2.8 24mm, one can hope ...

Stabilised primes sound nice but..
I read this
Available June 2012
EF 24mm 1:2,8 IS USM 829 Euro
EF 28mm 1:2,8 IS USM 799 Euro

Erm a tad pricey I think and I'm not entirely sure how essential stabilisation is for wider angle lenses.
Canon really need to update their 35mm and 50mm primes too

I expect to see more stabilised primes but at those prices I think they can keep them!

Ha ha agorabasta, a double shake thanks
What amplitude and frequency would you be thinking of, half pixel pitch or what exactly? And does the sensor need hf vibrating only horizontally or vertically or both at once? What about interference, would that be a problem? I'm thinking of a frequency that might interfere with a particular co-incidental wave of light.
And the traditional anti moire filter, wouldn’t that likely be vibrating and emitting already (at nano scale) from the heat coming off the sensor? If so I’m not sure if that would be additive or subtractive for some wavelengths, maybe it emits in IR only, I'm not sure.
Greg

Greg,

The period of vibrations should be a few times shorter that the fastest shutter speed. So if max shutter speed is 1/8000s, the freq about 30kHz should be enough. Such frequency seems to match the sensor size pretty well. So two waves travelling in orthogonal directions at freq's about 30kHz would produce rotational shift of the sensel. The momentary energy value of such vibration would be about 10mJ - totally negligible if it's not immediately dissipated for friction. So having a moderate Q-factor for such an oscillator at about 10^4 means additional power consumption about 30mW at sensel shift amplitude about 2um.

The regular passive AA filter is then totally unnecessary, as you have the blurring vibrations on for AA action or off for no AA.

To my taste, I prefer the D800 than the D4. and the 800E sounds very interesting. I'm starting to see a next Leica M10 ... with this sensor or equivalent ... ISO performance should be equivalent or even little bit better than the NEX 5n.

I'm thinking about if I sell all my Sony stuff ... probably I can get a D800 with some good prime ... But, my question to David, which good prime or Zoom Nikon lens you think can support this 36MP for the moment with no stress?

Actually I know most of my Minolta lenses (after acceptable Aperture reduction) are not stress by the NEX sensor where those one are equivalent to 36MP sensor. So, I'm curious about Nikon (probably better ?).

Frank

If I was to buy a D800 it would be D800E, and I would buy a 14-24mm f/2.8 because it's exceptional for what it is; a 45mm f/2.8 PC-E tilt shift macro lens (because I could not afford all three); a 50mm f/1.4 new AF-S; and a 70-200mm f/2.8 VRII plus the 1.7X TC.

Nikon's new AF module is now as sensitive as the A77 (down to -2EV, Sony achieved a world best with the low light sensitivity of their A77 module and the photo press generally failed to notice) and functions with lenses to f/8 which also means its performance with wide angles will be improved, and its accuracy with all lenses will also be better. That's because a higher resolution sensor is needed within the AF module to do this.

Knowing Nikon and Sony and the way this works, I would guess that the f/8 modules sensitive to EV -2 are closely related to the A77's new AF innards because it can do both of these things too (unless the lens locks out AF by telling the camera to use manual).

David

I might be wildly off the mark here but I thought the 500mm f8 mirror lens was killed off possibly because the SLT models could not AF with it? (A55 etc not A77 it was stopped prior to this)

Any truth in that or was it just not a "worthwhile lens" for the system?

The Sony 500 mm mirror lens can autofocus since the optical system is f:6.3 or something like this. The obstruction caused by the secondary mirror makes this a T:8 system, not f:8. This should be old Minolta info, I think.

That is what I have read way back when.

tom

Correct, the 500mm will focus with models after the date of its introduction - AF is disabled, by its reported aperture, on the original 9000, 7000 and 5000 only. From the 7000i on, all models except the 3000i operate with the 500mm because they are programmed to ignore the reported f/8 aperture.

Minolta AF sensors from 1986 to 2004 have a virtual aperture window of f/7.1, which is the set-up aperture used on the special AF calibration 50mm lenses. It's often said that f/6.3 is the smallest usable aperture but actually the Sigma 500mm f/7.2 will AF, and as long as a converter/lens combo reports shy of f/8, it should AF.

I believe the sensors in the A77 are now a true f/8 but still disable f/8 lenses; the Nikon sensors are a similar type, f/8 and EV-2, but perhaps they don't disable f/8 lenses - we'll have wait and see.

David