Nikon D800 - seems harder to justify Medium Format

[ErikKaffehr]

Hi,

I don't see 36 MP as a giant leap forward, it is a significant improvement above 24 MP, about 22% on a linear scale. The giant leap is that Nikon has an affordable camera above 20 MP for the first time. A larger format does have advantages. It collects more photons and the lens is less challenged regarding MTF on fine details.

On the other hand, modern (read Sony) CMOS technology greatly reduced readout noise, greatly improving image quality in the dark details.

Diglloyd has presented some real world test images comparing Leica S2 with Nikon D800, where the Nikon definitively came out on top. In other tests we may see different results.

The way I see is, DSLR technology is moving upscale. Todays DSLRs are better than 135 SLRs using film of yore, at least in my view. My experience is also that my 24 MP DSLR is on top compared with my Pentax 67 using Velvia. So DSLRs have come a long way. No question, MF digital has some areas where they are better. MF digital is moving upscale, too. Many users have replaced 4x5" film with MF digital, and the present 80MP digital backs may have made inroads in what used to be 8x10" film.

Best regards
Erik

It is often unwise to phrase issues as all-or-nothing questions like "is MF justified now that we have the D800?"
Instead, I am curious as to how developments in 35mm format (and in improved performance or price from DMF, but we have not seen so much of that lately) changes the use cases for medium format. That is, when is there an advantage to DMF over the D800 or D800E, and which hMF gear gives that advantage. For example, it is hard for me to see any sensor performance advantage of 22MP DMF over the D800/D800E options (yes, even in DR and tonal gradations), but other aspects like the lenses and leaf shutter lenses might still give a different and preferred look.

[Dick Roadnight]

Meanwhile, actual testing by Bart van der Wolf in another thread here indicates that the AA filter reduces resolution by about 1%, so this seems (how should I phrase it?) a rather pessimistic estimate of the effects of the AA filter.

Sorry,

This seems to be gravely wrong. What you are saying that you need 40MP to replace 2MP HD-camera.

Here is a screen dump from Zacuto Bigchip Video Camera shootout 2011 (part II)

Left is Arri Alex right is a raw image from a DSLR (I don't know which, but a Canon I believe) which is sharpest? Both images are HD.

Best regards
Erik

On the (un-scienticic) assumption that an AA spreads the light destined for one pixel over about ten pixels, I calculated that you would need to down-sample a DSLR image by up to 20 times to get optimum per-pixel quality.

Video looks better than a still from a video, as the human eye averages and blends several frames.

¿Was it a still image from a DSLR or an image from DSLR video?

Thank you both for the info... it would seem that AA filters do not spread the light destined for one pixel over ten pixels... but, when I down-sampled pictures for the web, I was amazed how much better the Hasselblad H4D60 (studio flash) shots were than the GH2 shots (hand-held at 800 ISO).

[Ben Rubinstein]

The real problem here Ben and not me but folks THINK they can compare a 36mpx 35mm cam to a 60 mpx warhorse. I know better why I have both. LOL

All I have done is show how the Nikon D800 really upped its game and that YES you can get closer to MF. Believe me if it was better I would be putting 40k back in my pocket in a heart beat but also the Nikon is the best game in town in 35mm or one of the best. Hate that word best, its like better than what.

To answer your second part of the question (D800 make it harder to justify medium format of the same resolution) the answer i still think is no it can't if you care about color, tonal range, DR and everything else MF is known for and that look that mystifies us all. Its still 35mm format photography and we keep coming back to exactly the same damn thing we always say bigger is better and it has not changed when we put a sensor in than a piece of film. Justifying it is a money issue and we all have different levels of that. I would love to put that money back in my pocket but it would not solve the quality issue, certainly make my wife one happy girl that I can tell ya. LOL

I think the real down the throat choke up question is what quality level are you willing to pay for. It ALL comes back to this question and no one has the answer except what you want.  Believe me I ask myself this question what do I really need or want here. I'm in the same boat as the guy willing to spend 40k just on different sides of the ship.

I always was naive Guy, didn't realise anyone would want you to make that comparison!

We're working with an Aptus II-8 now in the studio that I've set up and manage, photographing ancient manuscripts, I did suggest to the boss (guy funding it) that it does bother me that we don't have backup for when I'm flying abroad to shoot private collections, have to rely on rental if there is a problem and because we can't afford a 2nd set of MFDB gear, that the D800 although probably not as good, could well be good enough to be both backup or for use sending outside the studio with one of the photographers while the main beast is tied to the shooting stands. As good? probably not. Pretty close, I wouldn't be surprised in the slightest. I wouldn't have mentioned the 5D2 for that in the past, I've seen the 5D2 shooting our stuff, it ain't pretty.

We are ISO 80 tethered studio shooters with the camera bolted to a shooting table pretty much permenantly. Thank goodness, I'd hate to use this gear out in the field, on a D800 the 'close enough' at low iso and 'heck of a lot better' at higher iso would probably make me wonder whether the extra colour depth/tonality which let's be frank many pro's would have to look twice to see, is worth that slow focusing '90s camera technology with a back whose screen would disgrace a decade old p&s on the MFDB gear I have at present.

I suppose the problem is that I don't have a digital point of comparison. I know what 40 megapixels of MFDB looks like photographing hundreds of year old ink on parchment and it looks a heck of a lot better than a 5D2. I've never used the MFDB shooting stuff like portraiture or landscape where I have a frame of reference from a decade of shooting DSLR's. I'd take the Aptus out for a spin but as it's not insured off site and as I'm not the one funding this project....

Have to be honest though, the camera I really want personally is a Nex 7 at the moment!

[ErikKaffehr]

Hi,

The two problems with your assumptions is that that they have little founding and don't match experience.

1) It seems that AA-filters reduce image contrast just a few percent.

2) I have seen some theoretical discussion indicating that the optimal AA-filter would spread the light about 0.2 pixels. Anyway the very physical evidence I have seen is that Nikon D800 does surpass Leica S2 in the corners on a flat surface architecture surface with both camera using top of the line macro lenses at optimum aperture, Leica S2 using macro Elmar 120/2.5 and Nikon D800 using Macro Planar 100/2.0. (http://www.diglloyd.com/prem/prot/DAP/NikonD800/compare-LeicaS2-mosaic.html )

On the other hand, the Zacuto tests indicate that DSLRs have lousy resolution compared to real movie cameras, I'd presume that this depends in part on the DSLRs not having optimized low pass filters for video but use software filtering instead.


Best regards
Erik

On the (un-scienticic) assumption that an AA spreads the light destined for one pixel over about ten pixels, I calculated that you would need to down-sample a DSLR image by up to 20 times to get optimum per-pixel quality.

Video looks better than a still from a video, as the human eye averages and blends several frames.

¿Was it a still image from a DSLR or an image from DSLR video?

Thank you both for the info... it would seem that AA filters do not spread the light destined for one pixel over ten pixels... but, when I down-sampled pictures for the web, I was amazed how much better the Hasselblad H4D60 (studio flash) shots were than the GH2 shots (hand-held at 800 ISO).

[gerald.d]

Let me try to ease this and explain the context for the "heavy" comment, which may have been misunderstood: we all agree that the D800 is lighter than the MFDB setups, significantly so. In fact, carrying MFDB setups all day is no joy, although ones with WLF (which I prefer) have an advantage here as they can be held at waist height, especially useful for those of us less young (!).

The point was rather about use of the overall package: by the time you figure the need to hold the camera up at the eye, stability and lens requirements .... the use pattern isn't that different than what is required for a MFDB. While Nikon starts with each separate aspect suggested as lighter/smaller/cheaper, etc., when considered as a whole, these things all add up. To get the most out of the D800 will likely requires a MFDB sensibility; high quality results may not be achievable with the more casual "point and shoot" technique.

Geoffrey -

Is this view based primarily on the resolution of the D800 sensor?

Would you say the same thing regarding, for example, the Canon 7D?

Regards,

Gerald.

[theguywitha645d]

Both the D800 and a medium format camera are big. Neither is particularly small:

http://camerasize.com/compare/#291,211

From the camera size site:

Nikon D800E is 6% (10 mm) narrower and 5% (6 mm) taller than Pentax 645D.
Nikon D800E is 32% (37.5 mm) thinner than Pentax 645D.
Nikon D800E [1000 g] weights 32% (480 grams) less than Pentax 645D [1480 g] (*inc. batteries and memory card).

I have carried my 645D around all day. I shoot with it hand held. It is just a camera. The D800 is just another camera. Neither require superhuman skill to use. Certainly you don't need to weld it to a steel post and only use 1/250s at f/11. The difficulty in shooting with MFD and the D800 are really over exaggerated.

[BernardLanguillier]

From the camera size site:

Nikon D800E is 32% (37.5 mm) thinner than Pentax 645D.

That part hurts though. Together with the larger lenses, it makes a significant difference in really world packability.

In a 50/60 liters pack, you basically go from being able to store a jacket in parallel with the camera to protect it, to a situation where the camera pretty much occupies a whole layer in the pack.

Cheers,
Bernard

[theguywitha645d]

In a 50/60 liters pack, you basically go from being able to store a jacket in parallel with the camera to protect it, to a situation where the camera pretty much occupies a whole layer in the pack.

Just because of that 37mm? I have a Mountainsmith lumbar pack as a camera carrier. The Nikon would not be saving me lots of space. And the difference would not prevent me from getting anywhere. Besides, the D800 needs those enormous CF cards.

[BernardLanguillier]

Just because of that 37mm?

No, those 37mm and the extra lenght of most lenses.

Cheers,
Bernard

[Geoffreyg]

Geoffrey -

Is this view based primarily on the resolution of the D800 sensor?

Would you say the same thing regarding, for example, the Canon 7D?

Regards,

Gerald.

The thread was moving so very nicely, I was hoping this could fade to the back....  

There is no simple answer to your question. Its a very slippery slope. It just seemed that the resolution of the D800 was putting the camera in direct competition with a different group of photo equipment than the 7D. The 7D seems like a very good DSLR and fits that niche nicely. We've had one around the house for a few years, and it has a lot of use with a variety of techniques, not as disciplined as the good habits one has to learn for MFDB.

It just seemed that the D800, with its big sensor, Nikon's cautionary notes on how to get the most from the camera, along with a users likely desire to "pull the most" out of it, led to this sense that this was a camera working in a different category than other DSLR's . So no, its not just resolution, but also how we manage our expectations.

[Graham Welland]

I think it's plainly obvious to everyone that there will be a segment of the gear buying public who might have made the stretch to affording a medium format digital system that will now be perfectly happy with the D800 and the excellent system that goes with it. If you think that this isn't going to happen then I would respectfully have to say you've got your head buried in the sand.

Is it the end of medium format digital? Obviously not, for all of the well articulated arguments listed here and all over the web. If you want the very best image quality then you are simply going to have to shoot with a current generation back and digital glass. If you want the flexibility to use the same back on multiple systems then ditto. (Obviously if the D800 is 'good enough' for your needs/desires then none of this applies).

Will there be folks who are only using medium format because it WAS the only game in town to provide the image quality, resolution or colour rendering needed for their work - sure, and some of them will be reassessing whether it is the best use of resources vs a cheaper and arguably more flexible Nikon system. It happened with other formats such as LF & MF film and will doubtless happen now too.

All I know is that I can't put my Schneider or Rodie technical camera glass on the front of a D800 nor stick a D800 on the back of my Alpa (at least to shoot wides) - something VERY important to someone who shoots for personal enjoyment and not to put bread on the table. I also know that it's important to me personally to use the very best equipment that I can afford to enjoy my photography - that's just the way I am, and I know for a fact many others feel that way about personal photography and other pursuits. I could afford to put a D800 body on every one of my lenses and still have change left over vs the cost of my IQ back but it still wouldn't make using a D800 a better system - but hey, that's just for me.

Now if you're just shooting grip 'n grins, high school seniors or mass production weddings then I'm sure that it's a no-brainer that a 35mm DSLR like the D800 (and all the lenses, flash, automation etc etc) is a better choice today, at least as of April/May 2012.

(Oh, and as a long time Nikon shooter yes I ordered a D800 too )

[gerald.d]

The thread was moving so very nicely, I was hoping this could fade to the back....  

There is no simple answer to your question. Its a very slippery slope. It just seemed that the resolution of the D800 was putting the camera in direct competition with a different group of photo equipment than the 7D. The 7D seems like a very good DSLR and fits that niche nicely. We've had one around the house for a few years, and it has a lot of use with a variety of techniques, not as disciplined as the good habits one has to learn for MFDB.

It just seemed that the D800, with its big sensor, Nikon's cautionary notes on how to get the most from the camera, along with a users likely desire to "pull the most" out of it, led to this sense that this was a camera working in a different category than other DSLR's . So no, its not just resolution, but also how we manage our expectations.

Well the reason why I asked was because the 7D has a higher pixel density than the D800. I'm not looking to 'bash' any particular camera here - they are all capable of creating great images in the right hands.

To go back to the point you were making -

The point was rather about use of the overall package: by the time you figure the need to hold the camera up at the eye, stability and lens requirements .... the use pattern isn't that different than what is required for a MFDB. While Nikon starts with each separate aspect suggested as lighter/smaller/cheaper, etc., when considered as a whole, these things all add up. To get the most out of the D800 will likely requires a MFDB sensibility; high quality results may not be achievable with the more casual "point and shoot" technique.

For the same focal length, it's the pixel density that necessitates stability, not overall resolution. If you're comparing equivalent fields of view, then yes - larger sensors require more stability due to the longer focal lengths required compared to smaller sensors.

Example FoV at the pixel level - same focal length lens:
50mm lens on 7D - 17.8 arc-seconds/pixel.
50mm lens on D800 - 20.0 arc-seconds/pixel.
50mm lens on IQ180 - 22.3 arc-seconds/pixel.

Example FoV at the pixel level - same FoV of the image projected onto the sensor:
31mm lens on 7D - 28.4 arc-seconds/pixel
50mm lens on D800 - 20.0 arc-seconds/pixel
80mm lens on IQ180 - 13.9 arc-seconds/pixel

The lower the number of arc-seconds/pixel, the more critical stability of the system is.

Obviously an MFDB sensibility and approach will produce higher results regardless of the equipment being used

[HarperPhotos]

Hello,

I was just talking to the Nikon rep in New Zealand and he tells me that they have 200 orders already for the Nikon D800 and there first shipment of 12 arrived today.

I think that is a very good indication of how popular this camera is.

If Nikon don’t announce a Nikon D4x soon I will buy a Nikon D800 and D800E camera with MB-D12 battery packs to replace my Nikon D3X.

Then I will put my Mamiya 645AF kit up for sale and use my Leaf Aptus 75 on my Mamiya RZ system which I would never part with.


Cheers

Simon

[ErikKaffehr]

Hi,

If your technique is perfect than you'll be OK with any system. If your technique is less than perfect why care about having the best lenses and the most megapixels?

Best regards
Erik

Well the reason why I asked was because the 7D has a higher pixel density than the D800. I'm not looking to 'bash' any particular camera here - they are all capable of creating great images in the right hands.

To go back to the point you were making -

For the same focal length, it's the pixel density that necessitates stability, not overall resolution. If you're comparing equivalent fields of view, then yes - larger sensors require more stability due to the longer focal lengths required compared to smaller sensors.

Example FoV at the pixel level - same focal length lens:
50mm lens on 7D - 17.8 arc-seconds/pixel.
50mm lens on D800 - 20.0 arc-seconds/pixel.
50mm lens on IQ180 - 22.3 arc-seconds/pixel.

Example FoV at the pixel level - same FoV of the image projected onto the sensor:
31mm lens on 7D - 28.4 arc-seconds/pixel
50mm lens on D800 - 20.0 arc-seconds/pixel
80mm lens on IQ180 - 13.9 arc-seconds/pixel

The lower the number of arc-seconds/pixel, the more critical stability of the system is.

Obviously an MFDB sensibility and approach will produce higher results regardless of the equipment being used

[tsjanik]

That part hurts though. Together with the larger lenses, it makes a significant difference in really world packability.

In a 50/60 liters pack, you basically go from being able to store a jacket in parallel with the camera to protect it, to a situation where the camera pretty much occupies a whole layer in the pack.

Cheers,
Bernard

Hi Bernard:


The Pentax 645 lenses are not much larger, and are in some cases smaller, albeit slower.  When I considered buying a D3x, I looked at the Nikon lens offerings and a weakness is an absence of slower, lighter and smaller lenses in the focal lengths I wanted.  For example the 70-200 zoom is only available as the huge f/2.8 version (1540g, 209x85mm).  There is no exact equivalent for the 645D, but the 80-160 FA (1010g, 85x136mm) and 150-300 FA (920g, 132x198mm) are both lighter and shorter.  Nikon offers only an f/2.8 at 400mm (4600g) vs the 645 f/5.6 400mm (1260g)!  Yes the lenses are slower, but that is a small price to pay for the lower weight.  Canon has much better offerings for my purposes, but they didn’t have the D3x and now the D800.

Tom

[BJL]

For the same focal length, it's the pixel density that necessitates stability, not overall resolution.

I agree with the fact, but see no practical relevance: the only comparisons that make sense to me are formequal compositions, with equal angular FOV, amd then it is angular resolution that matters, as in your secodn set of numbers. That is because a given amount of camera or subject movement will produce a given amount of angular shift of the scene framing, so the effect will be similar with a 50 mm lens on a Canon EF-S camera with its "1.6x format factor" and 5 micron pixels as with an 80mm lens on a 35mm format camera with 8 micron pixels (and so the same pixel count on the same subject framing.)


And the angular resolution you get of a particular composition is measured, roughly! by pixel count. For example, the traditional minimum shutter speed guideline of "1/f" need to be scaled inversely with increase in linear pixel count, or better, by resolution in lines per picture height.

[theguywitha645d]

No, those 37mm and the extra lenght of most lenses.

Cheers,
Bernard

As Tom pointed out, only if you cherry pick your facts. The point is, the large 35mm DSLRs today (the D800 is no Nikon FM) are large cameras. The difference between that and a MFD camera is not so great to be really significant.

This is not a case of which is "better," but exaggerating the differences is not really useful.

[Guy Mancuso]

As Tom pointed out, only if you cherry pick your facts. The point is, the large 35mm DSLRs today (the D800 is no Nikon FM) are large cameras. The difference between that and a MFD camera is not so great to be really significant.

This is not a case of which is "better," but exaggerating the differences is not really useful.

One issue and I know I have done this for years is you wind up having more 35mm lenses in your bag over having a DF kit which got to 4 lenses tops. The Nikon kit I have right now is 5 and looking for 6 so total bag weight and space is really a toss up. I certainly would not buy a Nikon thinking its smaller and lighter , just go pick up a 24-70 lens from either Canon or Nikon and its bigger and heavier than any DF lens I had except the 300 mm of course. Now MF cameras are more boxy style than 35mm but a D4 for instance is no small camera either just a different shape.

[BernardLanguillier]

As Tom pointed out, only if you cherry pick your facts. The point is, the large 35mm DSLRs today (the D800 is no Nikon FM) are large cameras. The difference between that and a MFD camera is not so great to be really significant.

This is not a case of which is "better," but exaggerating the differences is not really useful.

It goes both ways... there are 35mm lenses that are compact also. 

Cheers,
Bernard

[Guy Mancuso]

It goes both ways... there are 35mm lenses that are compact also. 

Cheers,
Bernard

Very true but my bottom line is what I have to put on this 55 year old back so it really comes down to what's the bag weigh in total more than anything at least for me. LOL

Now I will totally admit a DF with back and 110 for instance shooting handheld there simply is not a lot of joy in that after awhile either. Some of this is just not fun.