Sadly Leaving Nikon!

[Jakov Minić]

Somewhat is not a word to be used in Physics

[Bjørn Rørslett]

Agree to that

One might as well argue that with larger formats, (exceeding the FX), lenses increasingly tend to be slower and thus the finder image gets darker not brighter. Any old hand of the view camera era will be familiar with that and usually having had the experience of shooting "wide open" at f/8 or f/11* ... Having a finder groundglass of 4x5" doesn't mean it is blazingly bright !!* my long Nikkors at that time were the excellent T-series Nikkor ED lenses; 270 mm f/6.3, 360 mm f/8, 500 mm f/11, and 720 mm f/16. Those were the days.

[bjornthun]

I didn't have problems using the D800 for manual focus using the live view, as most of my manual focus work is with tripod and stopped down for large depth of field, but the D810's live view is much better (higher resolution, less noise in lower light, two subsections of the image can be used to adjust tilt and focus in some cases). The D810's OVF is also with better coatings giving a clearer image which more pleasing to work with. The D5 is a further improvement in viewfinder quality in my experience.  I don't routinely use it for manual focusing hand held though, since I use autofocus lenses now for that kind of shooting. For the tripod based shooting typically I shoot with the D810 and aim for extended depth of field (stopping down to f/9-11, using tilt to maximize the near-to-far sharpness, I find the OVF useful for initial adjustment of the tilt and focus, to be refined by LCD). If I wanted to use manual focus with fast primes, I too might feel strained eyes, but when trying the A7R II, I noticed that I wasn't able to focus using the unmagnified EVF even remotely well enough to get usable results with the 135/2 Apo Sonnar. By using the zoomed view focusing was made possible but then I can't monitor the subject's and the image's overall expression at the same time if I have to view a zoomed-in subsection of the image. This kind of approach of hand holding the camera and zooming in for a partial view is not an acceptable way of working for me and I wouldn't be able to concentrate of the image content and timing doing that. I by far prefer to use an OVF to monitor the subject's emotion and clues to changes in it, and let the camera's continuous autofocus do the focusing. Subtle details of facial expression is something I can't see through an EVF, and without the use of zooming in, manual focus would not be possible either, so for me the EVF concept is unfortunately a non-starter. It doesn't give me anything that I could use to work with.  However, I completely understand that my subjects are not everyone's, and people have widely varying preference regarding viewfinders and what kind of camera design works best for them. This is what is great about the situation today: There is something for everyone.

The D800's AF gave me plenty of headaches with fast primes; the D810 was a significant improvement which solved most of the inconsistencies that I had. The D5 AF is phenomenally fast and sensitive in low light and in particular, it allows me to shoot moving subjects at f/1.4 on a routine basis with very few out of focus images. Especially with longer distances the earlier Multi-CAM 3500 series cameras had problems which I do not see with Multi-CAM 20k. However, time will tell how the new module will perform on a high resolution sensor. For me 20-24MP is plenty enough for most of my practical applications, and I've been very happy with the D5's resulting image quality, though not as good as D810 at low ISO.

In 2012 there is a breakthrough in one area (sensor and image quality) but because of production problems due to the 2011 natural disasters I understand that they had difficulty making correctly functioning cameras. This is regrettable but Nikon's products are far too valuable to me to not forgive them for their errors. I can understand that when there are large-scale casualties, evacuations, no working infrastructure, it can be difficult to concentrate on manufacturing cameras with precision requirements. Things have improved greatly since and I'm very happy with Nikon's quality control from the last few years. Also their progress with the D5 autofocus is really amazing.

Using liveview would have meant I could only have used the D800 on a tripod, and then with something like a Hoodman loupe. In total I found that setup cumbersome. Handheld use with LV would haveant holding a DSLR away from me as if it were a smartphone. Practical use steered me toward Sony mirrorless.

I found the liveview and EVF on my Sony to be of higher quality than that of the D800 with line skipping. When the D810 came with improved LV, I had already moved to mirrorless.

[simsurace]

Even such a claimed authority cannot get basics correct;

"brightness is somewhat proportional to the total amount of light captured, too: bigger formats simply capture more light"

forgetting that this is a circular argument as a larger area means "more" light but as area increases, intensity remains constant as there is a larger area to cover.

Oh well. Simple advice is to keep the salt shaker easily accessible.

Well spotted. Interestingly, he writes immediately after that:

which if condensed into an output projected image of a fixed size will be much brighter than a smaller format ‘expanded’ to cover

Is anyone aware of an SLR camera where that is the case?
If there were such a camera, his argument would have some merit.
But my understanding is that the viewfinder size is pretty much determined by the size of the mirror, which in turn must be pretty close to the size of the sensor.

There really should be some kind of peer review process for bloggers
Independent of popularity of the blogger, there are a lot of false statements out there, and critical comments tend to be suppressed.

Somewhat is not a word to be used in Physics

Right, at least not in conjuction with 'proportional'. Either it is proportional (to some statistical significance level) or not.

[Roland Vink]

What he's getting at is that smaller format cameras (DX, original 4/3) with TTL optical viewfinders usually have small pokey "tunnel vision" viewfinders compared to FX and larger format cameras. These cameras are generally fitted with slow zooms projected over a proportionally smaller format, so the viewfinder image is also proportionally smaller. The viewfinder could be larger but the light would be spread over a larger area making the image rather dim. You could solve that by using a faster lens, but a fast lens on a DX camera with a big viewfinder would be considerably bigger and more expensive, which really defeats the advantages of the smaller format cameras.

His explanation may not be quite right, but the general point is correct.

[Bjørn Rørslett]

A poor finder system can mar any format, but that is not a causality for the strange explanations given.

An f/1.4  (or f/8 or any other aperture) lens will illuminate the camera sensor with equal intensity at all formats (within its projected image circle, and ignoring small differences in transmittance by different optical designs; if necessary use T-values instead). Finder brightness is in turn set by incoming  light intensity. If not, it is the design of the finder that is flawed not any correlation to the format. Thus one is grasping at straws here.

For the nth time, at the same set aperture, all formats receive the same intensity of light.  That is the basic of photography that hasn't changed over the last couple of centuries to my knowledge.

[Ilkka Nissilä]

The D500 VF has 1.0x magnification; the D5 VF 0.72x.

[armando_m]

I think the D4, D800 were very significant milestones for the nikon dslrs.

The 36mp and dynamic range of the D800 still amaze me today.

But it has it's defects, all my my af lenses, including the primes needed af tuning on the d800, once that is done , af is accurate, speed is fast with all lenses except the 150 macro and that is to be expected.
Live view is ugly, in video mode is dlightly better.

my 800 is now 5 years old, it has problems wit the left af points, and a broken chasis, glued so it can be used on a tripod, but I still use it often and get the  results I want
 I use a vf magnifier , dk something, that makes the vf acceptable

I had originally  planned to use it for 3 years, if I had to replace it today I'll get a D810

[simsurace]

The D500 VF has 1.0x magnification; the D5 VF 0.72x.

If I am not mistaken, this means that the the viewfinder virtual images would be roughly the same size, but the D5's about one stop brighter. Can someone verify this with a light meter?
One stop is probably too small to notice because the eye will immediately adapt.

[simsurace]

For the nth time, at the same set aperture, all formats receive the same intensity of light.  That is the basic of photography that hasn't changed over the last couple of centuries to my knowledge.

It' worth pointing it out. But this is not where you disagree with Ming. It is the additional assumptions that are not disclosed.
As Ilkka suggests, the additional assumption seems to be that the size of the viewfinder image is similar across formats. While this seems approximately true of the D5&D500, my guess is that there are many counterexamples as well.
The possible assumptions about which lenses is usually fitted to various cameras would also be possible, but this would be too far-fetched in my opinion since it has nothing to do with optics.

[Bjørn Rørslett]

All is part of the confused mess masquerading as 'bigger is brighter', 'equivalence' or what have you.

I  mentioned the overlooked fact that the incoming intensity is independent of recording format. The finder adds a scaling either up or down from unity so as to make the image on the groundglass/panel comfortable to view at a virtual distance of approx. 1 m while at the same time (one must hope) letting eye discern sufficient fine detail.

In practice a very small format has a magnification and a larger format might have a reduction of the projected image format. Thus the final outcome is an interplay between format and finder that could make the finder appear brighter or darker according to the criteria used for its design and whether or not measures such as Fresnel microlenses are employed.

For a (D)SLR, the choice of the focusing screen is absolutely critical with regard to its perceived finder quality. On the FX/24x36 format, easy to evaluate due to its long history and diverse development stages, one can have finders so dark they are almost impossible to use for focusing to those so bright that your eyes are blasted yet still not being able to focus the image. The latter type, claimed to be optimised for 'AF' are commonly implemented today and makes it well-nigh impossible to focus  fast lenses visually. The dark often coarse screens of the earlier SLRS needed eye accommodation under anything than bright daylight, but did allow pin-point focusing accuracy even with slower lenses. The intermediate and more optimal approach perhaps was to have devices such as split prism or microprisms integral to the ground glass to make breaking up of the aerial image more complete and hence allow better visual focusing. With earlier F-series models, up to F5(6), a selection of replacement finder screens and  was available to optimise focusing under a wide range of shooting situations. Finder heads also could be replaced (not on F6? never used that model). I do wish that approach had survived to the current days.

Medium and larger format cameras were increasingly burdened by dark finders and for the view cameras, focusing the slow lenses in use frequently necessitated a finder loupe.

[simsurace]

Thanks for the detailed explanation.
This was my understanding: there are so many additional assumptions (same type of screen, finder rectangle the same size etc.) that have to be satisfied in order to have the correlation that Ming is trying to argue for, that his explanation does not hold much water.

Also, some of the cheaper DX cameras do not even have a pentaprism, but only mirrors. This may also account for some light loss and as a consequence, dimmer finders.

On a similar note, it is clear that it would be easy to make a very bright viewfinder image by making the viewfinder rectangle very small 

[Bjørn Rørslett]

On a side note, not every member of these discussions has used a sufficiently wide range of formats over the years to appreciate the true differences in existence. Switching between small and really big formats was a daily encounter with reality in my earlier days as a nature photographer.

The above might explain not excuse the strange argumentations washing around the 'net these days.

[simsurace]

This concern would be addressed by simply restricting the generality of the statement.
I understand that Ming Thein is not making universal statements, but only such statements that concern his own universe.
Of course it would be much nicer for him to make a disclaimer. For people who do take him as an authority (I don't, partly because I have seen him perpetuate much more severe misconceptions about optics, and not correct himself after the errors had been pointed out by others), I think such a disclaimer is important.
But even within the range of cameras he is discussing, his statement is questionable IMHO.

[Bjørn Rørslett]

Thus the inevitable conclusion is that Ming Thein is "somewhat questionable"

Time to move on to more entertaining topics.