Medium Format in Nikon's Future

haroldp

Nikon is not organized to market and sell very high end, low volume products at profitable margins.

The D3x space is probably the limit of their appetite for this.

Higher resolution FX size sensors, paired with higher resolution optics (look at MTF's on Leica's new APO Summicron) will continue to erode MF sales as the D3x and D800 (and Canon equivalents) have been doing.

Yes, bigger is better, but for most 'pros', IQ has to be good enough for their market, and as 35mm format cameras and lenses get better this will cross the threshold to more and more 'markets'.

In the film era, no professional would use a 35mm format camera for weddings.

I have not seen an MF digital camera at a wedding in 5 years.

.... H

ThomasHorton

shawnino .... but I can't see the Large Pocket Amateur buying a MF Hassy


They don't; they are bought buy WAGS (wives and girlfriends)

Any suggestions on where to find such a girlfriend? Just askin'

sevencrossing

ThomasHorton
Any suggestions on where to find such a girlfriend?
sorry my mate married her

kenadams

I guess you can cram a lot more pixels into a FX sensor than just now, but that doesn't give you the super shallow depth of field MF has, or the color depth beyond 14 bit I'd associate with the digital equivalents of the larger format.

I was thinking hard about getting a used Mamiya 6 or 7. But since film is all I could possibly justify spending money on as opposed to digital, I'm just not sure about follow up costs for 6x7 film. I'd really like to try it one day though...

haroldp

Going to 16 bit color (from 14 b) for FX (or DX etc.) is a business, not a technical decision and is more related to frame rates than sensor size.

It is likely a firmware decision.

DOF is a function of absolute aperture / magnification and is / will be addressed by faster high quality glass for 35mm formats.

IQ is like computer perfomance, only interesting when it is not 'good enough'.

I have shot a lot of MF film (40k plus frames per year mostly 6 X 7 cm, some 6 X 6) and my D3x or D800e with 24-70 is clearly better in all aspects except clipping highlights, and I have adjusted my habits so that I do not clip highlights.

.... H

Golf007sd

Going to 16 bit color (from 14 b) for FX....is a business, not a technical decision and is more related to frame rates than sensor size.

It is likely a firmware decision.

I have shot a lot of MF film (40k plus frames per year mostly 6 X 7 cm, some 6 X 6) and my D3x or D800e with 24-70 is clearly better in all aspects except clipping highlights, and I have adjusted my habits so that I do not clip highlights.

.... H

Harold, would you please elaborate why you believe this is a business decision and how a firmware update would be the solution in bring FX (or DX) in to the MF world.

You state that the D3x & D800e is clearly better in all aspects in relation to MF film. Would you have that same position on today digital backs? Or did I misunderstand you point :-?

WestEndBoy

We may not get much more than 36 megapixels out of FX. Sure, the sensors might get better, but what about lenses. I have been reading some articles relating to DXo Mark tests that suggest that for many lenses in the Nikon lineup (and I assume it is the same story with Canon), there is no benefit in going from 24 to 36 megapixels. I have also read suggestions that even the professional lenses would not benefit from much more than 36 megapixels.

I also seem to understand that lens quality could improve somewhat, but that there are diffraction or other related issues that restrain further improvements in lenses.

So it may be that to obtain much better image quality than today, a larger format might be the only economical strategy.

I would be interested in any comments on this.

donaldejose

I think a better idea for Nikon than investing resources in making and marketing a MF camera system is to spin off a P line for the D800. The P line would be a 50 mp AA free sensor/camera optimized for portrait work which is so often the bread and butter of medium format shooters. What adjustments could Nikon make? You don't need high fps or a large buffer. A D800P could run slow. You can optimize the sensor/software for skin tones in the 100 to 800 ISO range and sacrifice ISO 6,400 You can go to 16 bit color and sacrifice speed. You can produce some ultra sharp primes such as 35mm, 50mm, 105mm and 135mm or you can just direct customers to Sigma's and Leica's super lenses. It looks like other manufacturers will have the super sharp primes already on the market for use by the time Nikon put out a 50mp D800P body. Certainly, this route would be far less expensive for Nikon than a venture into medium format and it would likely satisfy many of the people who now shoot MF.

moreorless

I have also seen that Hassy has released many statements about going after the "Hi-End Amateur" and I can't but think it is chasing the Leica market, seeing how that company has all of a sudden turned profitable again. I just don't get Hassy with their Sony rebranded cameras with the idea that somehow there is a "High End" market like Leica users to be had. Leica's market has 60 years of nostalgia, desire, and perfection that I don't think is transferable to other brands. - See more at: http://forum.nikonrumors.com/discussion/1770/medium-format-in-nikons-future#Item_17

I think these two area's(the regular hassleblads MF systems and the rebadged Sony's) really show the problem they've got. I'm guessing the high end amatures targeted with the MF systems are in area's like say serious rich amature landscapers, targeting based on performance. You don't need to be a pro to see the benefits of 60 MP on your 40 x 30 inch holiday print of the grand canyon.

The problem is I'd say that more of the ultra high end amature market is in the smaller/bling direction, a big digital back DSLR system isn't exactly easy to lug around nore much of a fashion item. Leica's original product appeals more to this market where as Hassleblad have had to go the rebadged Sony's in in order to offer something smaller.

Personally I still think there best bet would be a digital X-pan, small enough to target the Leica market but with IQ advantages over FF camera's and a more unique aspect ratio.

paulr

As I previously mentioned I went to an Hasselblad open day, I took my out dated D3x with a 24-70 lens, I took images of their models in the studio set up with both the Nikon camera then the Hasselblad HD4 with the new 35-90 lens. I wanted to compare the difference in quality and this was an opportunity not to miss.
It was not until we compared the images on a £15,000 monitor at 400% that the fine details were better in the Hasselblad than the Nikon. My problem was that not many of my clients in Yorkshire have a £15,000 monitor and rarely crop to 400%. so could the difference practically be noticed. To invest £30,000 in a Hasselblad setup to achieve these results would mean several years of some serious overtime or I suppose I could move to London where the folding money seems to be more available.

donaldejose

paulr: great comment and information. Thanks for posting.

kenadams

There's a video out there on youtube by CameraStoreTV in Calgary, comparing a 40MP Hasselblad with a D800 in the studio. No surprise the D800 was far superior at high ISO, but the difference was noticeable at highlight recovery and colors in favor of the Hasselblad, at a Mac monitor no less. Interesting you say it was hardly noticeable, Paul.

If that is all worth anything (say, a 20k price difference), that's a whole different story though of course.. :-D

shawnino

There's a video out there on youtube by CameraStoreTV in Calgary, comparing a 40MP Hasselblad with a D800 in the studio.

I've seen the video. It's an unfair test. They are using a 50mm/1.4, which as wel all know, is a meh lens. They need to retest when the 58/1.4 comes out.

Agreed. Furthermore, I think the Hassy shooters couldn't really afford to admit that their mid-five-figure set-up wasn't a reason to hire them over a skilled D800E operator with a truly high-end lens. They are very skilled photographers, but part of their differentiation is their gear.

kenadams

Yes, no doubt, it was not really a fair pairing, which is my point. The video was not so much about sharpness though I think, but about finding a justification for a 5 figure price tag on their gear. They came off pretty snobby as well.

All I'm saying is that there's more to MF than just high MPx counts. Question is, is the difference worth 20k in price difference to anyone not in the high end fashion industry? Guess not...

Golf007sd

Here is the video we are talking about. I agree that they sure could have picked a better lens to be mounted on the D800. The Hassy had one of their best lenses on it when compared to a $450 lens. Never the less, once one of their staff found out about the price, he was so shocked he fell to the ground (joke.)

With out question, Nikon's high MP body, a.k.a the D800, gave the Hassy a run for its money and in high ISO it blew it out of the water. Their comment at the end was a bit snobbish however, in that if they go to a shoot and someone has the same Nikon or Canon body, it makes their work and equipment choice mute. [-X [-X [-X

haroldp

Harold, would you please elaborate why you believe this is a business decision and how a firmware update would be the solution in bring FX (or DX) in to the MF world.

You state that the D3x & D800e is clearly better in all aspects in relation to MF film. Would you have that same position on today digital backs? Or did I misunderstand you point :-?

I could be wrong in some cases, but I believe that the sensors currently used could (or do) support 16b color, and the 12 / 14 b limitation is in the firmware.

It is driven by trying to keep frame rates as high as possible since the market for these cameras is much more sensitive to (and understands) competitive frame rates than to color density.

As long as one avoids burning highlights, my D800 (or D3x) with 24 / 70 at F4, properly used, produces a clearly better 24 X 36 in print than I ever got from MF film. If I could put A Leica ASPH summilux or APO Summicron on a D800 it would be even better, but at F5.6 these differences disappear.

Clearly, a 36 X 48 mm digital back at 48, 60 or 80 mpx is potentially better than a 24 X 36mm sensor at 36mpx if appropriate lenses and technique are used. It is probably at least as good as 4 X 5 in film.

My main point is that 35mm digital formats do not have to be as good as MF to displace digital MF sales, they merely have to be 'good enough' for their respective markets, and in the wedding market MF digital is nearly extinct.

As they get better in IQ, and continue their advantages in handling, frame rate, auto focus etc, they will displace even more MF markets.

For landscapes, tiled panoramas (if well done) are already better than larger formats.

Regards .... Harold

WestEndBoy

Regarding the point made by Paul above, a cheap monitor should illuminate the difference in resolution that he is referring to merely by zooming in. And if it doesn't, I am sure one of the better Eizo's would be up to the job and they sell for $2,000 - $2,500. In 10 years, those monitors will be $500.

Also, I suspect that the comparison between Nikon and MF manufacturers is not completely fair. Nikon is on the sharp edge of competition with Canon, Sony etc. The MF manufacturers are in a less competitive environment. I think that he important question is if Nikon or Canon produced MF, what would it look like.

There are engineering constraints to quality and format size directly correlates with potential image quality. Weight goes up, but I recall the Mamiya 6 and 7 which was not too bad for some. The price is now astronomical, but so was DX if you go back far enough.

However, I also acknowledge that my D800 with my 85mm 1.4G or 135mm 2.0DC is DAMN good, and the 135 is a 18 year old model to boot. I might not be willing to spend much to get much better.

Everybody's comments have been quite interesting and have given me food for thought.

Golf007sd

@haroldp: Thanks for the response.

@jshickele: I agree that comparing modern MF bodies /w a digital back to the modern D-SLR, be it Nikon or Canon, is not a fair comparison at ISO's 200 or below. The dynamic range and color depth is just mind blowing on the MF when compared side-by-side in large print size (i.e. posters or billboard signs)...at least to me. Having said that, the price for such a setup is just not rational to me; moreover, the Vast majority of photographers seeking a high-level body.

I look forward in seeing what Nikon offers me in the next few years. The D800 is an amazing product for those that are seeking a D-SLR to complement their MF digital back bodies where high ISO is needed.

WestEndBoy

Yes, the prices for MF is nuts unless you are a pro. My MF speculation only becomes serious in the 5 - 10 year time frame however. MF wasn't that incredibly expensive before digital came out so the high prices now are largely due to the digital components. The price will come down.

Ade

@jshickele: I agree that comparing modern MF bodies /w a digital back to the modern D-SLR, be it Nikon or Canon, is not a fair comparison at ISO's 200 or below. The dynamic range and color depth is just mind blowing on the MF when compared side-by-side in large print size (i.e. posters or billboard signs)...at least to me. Having said that, the price for such a setup is just not rational to me; moreover, the Vast majority of photographers seeking a high-level body.

I don't know Golf. The dynamic range and color depth of both the D800/e and MF cameras exceed practically any printer in existence. I don't think any variation we can see in print will be due to either of those two factors.

Also, commercial images printed in such large sizes would typically have gone through a ton of retouching, which would likely mask any real differences between the D800/e and a digital MF.

People talk about 14-bit vs. 16-bit… but if you take a typical 16-bit Hasselblad DNG, convert it to 14-bit, and show both original and the 14-bit version to 100 professional photographers side-by-side, I'd bet near zero of them can pick the 14-bit vs. the 16-bit original better than random chance.

TaoTeJared

...I have been reading some articles relating to DXo Mark tests that suggest that for many lenses in the Nikon lineup (and I assume it is the same story with Canon), there is no benefit in going from 24 to 36 megapixels. I have also read suggestions that even the professional lenses would not benefit from much more than 36 megapixels...

Most know my opinion of DXo's lens tests (Summed up with: True "sharpness" lens tests do not change with the sensor/camera body as theirs does.) and I won't go into that but I find those claims to be warrantless and based on people who are more interested in owning "latest and greatest" rather than honing their skills. I have a real discerning eye and can pick out lenses but it comes down to NEED. Is the need for sharper lenses? Simply, NO. Long answer for those who like wishy washy: There is not an urgent need above the current path where each lens update becomes better with each iteration.

That does not mean I will not welcome better or new lenses being released (i.e. 58mm) but realistically few will ever be able to tell. If you think there is a need for higher lenses, then Zeiss has shown how much it will cost, $4,000, which is par on price (or even below) high end cinematic optics. Contrary to uninformed opinions, you don't pay for the name, you pay for what it takes to build the lenses. (The mark-up on lenses is not that great. Maybe 20-30% at most.) I'm willing to bet Nikon's 58 will register similar numbers to the 85mm as their MTFs are close, and that is probably 10-15% below the Otus. Leica's APO Summicron is $10k. At that level, If I had a 24, 35, 50, 85 kit (and just settled for the current glass for wide and tele) at $4k a pop, that whole set-up would be $16,000 vs the $6,500 of the same kit. Is 10% increase of sharpness worth $10k? Maybe for some, but not many.

Better question, how much gain/ added sharpness can you add with software.

The other kicker is, there is little difference between 16mp and 36mp once you run it through NR, above native ISO, or add filters trom Topaz, Nik, etc. All of that knocks a ton out of the image and the optimal results so many naively overly focus on.

TaoTeJared

I could be wrong in some cases, but I believe that the sensors currently used could (or do) support 16b color, and the 12 / 14 b limitation is in the firmware.

The stuff I have read is that there is a hardware change on the sensors for 16bit and it is not just firmware. If it was just firmware, you would see FX/DX sensors that could tether to computers to add that added functionality like the early Med Fmt systems. There are too many scientific users that would pay for that added functionality that if it were just the firmware, we would be seeing it deployed. The Firmware just supports what the sensor can and other hardware can do. That said, what I have read, it doesn't seem to be a huge change for the sensor but just more of a "design" decision.

To get to 16bit, all of the pipelines, processors, buffers, cards, etc. would have to be upgraded. Considering even though the main parts (buffer/processor) are the same for the D800 and D4 and we see the loss in performance going from 16 to 36mp. The added information from 14bit to 16bit would be even more data added (approx 4 times more) which would reduce performance even more.

14bit = 16,383 tones per pixel = 4.39 trillion total tones.
16bit = 65,532 tones per pixel = 281 trillion total tones.
24bit = 16,777,216 tones per pixel = my calculator doesn't go that high

Dealing with posterization effects (banding/pixelation artifacts) during editing is the real advantage.

Golf007sd

@Ade & PitchBlack: I'm will to concede to both of you on your points. I'm by no means an expert in the area of printers and their limits...as it relates to image quality or file size provided in order to have an extremely large output.

My perspective comes from the many videos I have seen, on-line, where by the editor, during the PP section, at crazy magnification (300-400%) with such amazing detail, is able to bring back details and colors that where not their before. Hence, the power of the MF sensor and its tonal range.

If any member here know of anyone one that has a MF body with a digital back, please reach out to them and get their input on the subject at hand.

Msmoto

I enjoy reading all the conversation regarding all the technical limitations of the various formats/sensor tech, color depth, etc. However, I would suspect that only a very few pros can extract the best from the equipment. And, the alterations from image capture to final product are so extensive, I would doubt I could tell the difference.

An example is in the quality of photos on the Honda Jet site....are these MF or a D800, or??? http://hondajet.honda.com/gallery-and-downloads/#1

A factor which has nothing to do with the technical aspects is what a client may expect. Dating back to the 1960's when i took the first 35mm slide to be published in one of the leading home furnishing publications, there as always a skepticism about the small formats. And, a Hasselblad does have a "cool factor" LOL

adamz

@TTJ - I agree with you that the difference is marginal considering the costs. moreover, I'll even add that if your technique is good in many cases you can shorten the distance even further. OTOH, if your technique is exceptional and you have the best tools on the market the results will be jaw dropping - the extra kick that will settle you apart from your competitors.