The decline and fall of Reflex.

[rdenney]

I guess you ruled out the Hasselblad due to the costs rel. to the Pentax 645Z?

How do you like the way the focus falls off with distance in med. format compared to FF or smaller sensors?

The Hasselblad that used the 50mp Sony sensor (like the 645z) wasn't available when I bought the Pentax, and the Hasselblad SLRs were just vastly too expensive. But I was already in the Pentax universe, with a decent 67 kit and also a couple of 645NII film cameras that I used for occasional side gigs. So I already had a body of lenses I could put on a 645z from day one.

Format is a continuum for me. Depth of field is related to focal length, and focal length is related to format. The bigger the format, the longer the lenses, and the less the depth of field at a given aperture. (I am not saying that 165mm on a 4x5 camera is any different than on a 35mm camera--it isn't, except that the small-format lens lacks sufficient field of view to light up the larger film area). The Sonnar 180 opens to f/2.8, whereas my fastest large-format lens is f/4.5 (for a 215mm Ilex Paragon). To get the same field of view as the Sonnar on the Pentax, I need a 135mm f/2 lens on 24x36, but I don't have one that is that fast. I do have a fast 85--f/1.8--and the Pentax 6x7 105mm, f/2.4 Takumar comes close to the same magnification and depth of field on the 645z. I can make any format do what I want if there is a fast enough lens, but I don't always have the fast enough lens. Of course, print size has to be constrained by what the sensor will support, but that's sort-of like comparing a big amp to a little amp and constraining the former to keep both in their linear range.

The Sonnar has a special look, though--perhaps a byproduct of its spherical aberration that is not fully resolved.

My "full-frame" camera is a Canon 5D Mk. II (I also have a Mk. I), now a dozen years old and recently back from Canon having received a complete overhaul. My wife uses a Nikon D500. Our family is therefore peaceful--neither borrows the other's lenses. But we are also poor as a result of buying crap in parallel.

Rick "who uses the format that is most convenient and fulfills the requirements" Denney

 

[audio2design]

I guess you ruled out the Hasselblad due to the costs rel. to the Pentax 645Z?

How do you like the way the focus falls off with distance in med. format compared to FF or smaller sensors?

There is no difference in the way focus falls off, that is just a misunderstanding about the interchangeability of lens values and camera ISO and how it is far more complicated than it needs to be. With a full frame, you will be shooting with a longer focal length, and a smaller aperture than with you shot with an APS-C. However, you are actually gathering similar amounts of light, so you can dial up the ISO on the FF to get the same image as the APS-C. They tried to make it all compatible with film, but in my mind, they made somewhat of a mess and most people, even photographer don't understand it. There is no less inherent noise on a FF sensor than an APS-C sensor at readout, the high ISO capability for full frame is mainly "math", but ultimate its bigger pixels hold more charge = more signal to noise capability, ASSUMING you capture more light.

 

[Wes]

I am talking about the dF/dx, where F is the degree of out of focus (or blur circles) and x begins at the plane of focus

not about noise

 

[Chromatischism]

They did for DSLR, but their entry point for APS-C mirrorless is about $1,200 for a kit versus <$500 for their DSLR offering and about $600 for Canon mirroless, and $700 for Sony mirrorless. I have no doubt the margin especially on lenses is way better for FF, but you have to feed the funnel with new consumers.

And they aren't selling very many low-priced DSLR kits anymore because those consumers are satisfied with their smartphones. That was the primary metric that fueled Nikon's business decision to move upscale a bit into the affordable premium category, similar to what Mazda has done.

 

[Wes]

also Nikon had very poor lens offerings in DX (APS-C); nothing high quality for wide angle

 

[Chromatischism]

There is a lot of overlap on sensor design, so it really comes down to processing at this point and on that note, I would totally agree with you. Give me Google Night Sight with an APS-C and I am a very happy camper.

Yes - we can leverage both software and hardware stabilization now to do those long exposures. I would welcome this. Give me more reasons to use my camera and fewer to reach for the smartphone. That is what the industry needs.

 

[JeffS7444]

Their Pro line is really xlnt. There is a range of 3 lenses in the Pro line that are often called the Bokeh Master lenses. Oly sez they not only have ultr-smooth bokeh, but also have a smooth fall off of bokeh - or something like that....

Oh yes, they know exactly how to engineer bokeh these days if you got the $$ for it. I find that one of the joys of doing this with M43 is that I can play with insane lenses like these yet the size still isn't too bad. The Voigtlander stuff is also worth a look - manual focus, but also smaller as a result.

 

[Chromatischism]

With a full frame, you will be shooting with a longer focal length, and a smaller aperture than with you shot with an APS-C.

I think I know that you meant focal ratio and not aperture - actual lens aperture diameter is larger on full-frame lenses to achieve the same field of view and is why FF cameras produce better S/N (less noise) than smaller sensor cameras. Just for everyone following along

But you're right, there's nothing different about the sensor other than the size (field of view).

 

[Wes]

photosite size is immaterial???

 

[audio2design]

And they aren't selling very many low-priced DSLR kits anymore because those consumers are satisfied with their smartphones. That was the primary metric that fueled Nikon's business decision to move upscale a bit into the affordable premium category, similar to what Mazda has done.

DSLR sales have been on a fast decline for the last 10 years almost, with smartphone cameras being quite good for what, about 1/2 of that? It makes sense to move upmarket, but photographers tend to latch onto a brand and start at the bottom. The top selling mirrorless cameras, are all somewhat lower end, EOS-M, a-6000 mid-end, Olympus. I guess we will see where Nikon ends up in all of this.

 

[audio2design]

I think I know that you meant focal ratio and not aperture - actual lens aperture diameter is larger on full-frame lenses to achieve the same field of view and is why FF cameras produce better S/N (less noise) than smaller sensor cameras. Just for everyone following along

But you're right, there's nothing different about the sensor other than the size (field of view).

Aperture as in effective size of the opening, not the maximum possible. Let's take two examples, a FF Nikon and a 1.5x crop APS Nikon.

Let's say I am shooting a 50/F2.8 with the 1.5x crop APS. To get the equivalent image size and depth of field on a FF, I would be shooting about 76m and F/4.3. The F/4.3 will be a darker image assuming the same shutter speed hence I need to bump up the ISO on the FF to get the same overall image brightness. Assuming the same number of pixels, the final image SNR will be very close on both cameras as the total light gathering is the same on each sensor.

 

[audio2design]

photosite size is immaterial???

A bigger photosite allow you to capture more total electrons, which improves your signal to noise, and it makes for easier optics. There are two noise sources in sensors, electrical read noise (amplifier + A/D), and SHOT noise, which is a function of sqrt(total electrons). You want a good ratio between the number of electrons and the read noise, but because of the sqrt function, you need a lot more light to get more SNR.

 

[mhardy6647]

Edit: it was a Canon Canonet 19. I

DSC_0758 (2) by Mark Hardy, on Flickr

 

[Chromatischism]

Aperture as in effective size of the opening, not the maximum possible. Let's take two examples, a FF Nikon and a 1.5x crop APS Nikon.

Let's say I am shooting a 50/F2.8 with the 1.5x crop APS. To get the equivalent image size and depth of field on a FF, I would be shooting about 76m and F/4.3. The F/4.3 will be a darker image assuming the same shutter speed hence I need to bump up the ISO on the FF to get the same overall image brightness.

That's why I choose not to look at it that way, but it does illustrate that there is no free lunch and there is no replacement for displacement. ISO increase means a faster shutter speed = less light collected and lower SNR. Since ISO has nothing to do with exposure or lens optics, I would just look at aperture area, focal length, and sensor size and how the interplay affects field of view.

The more telling comparison when going from APS-C to full-frame is "what lens does it take to get me the same field of view that I had on APS-C?" The answer is a larger aperture and longer focal length, in equal parts to maintain the same focal ratio. This gives more light collecting area (less noise) and a "zoom" of the image to put the borders in the same place as they were on your cropped sensor camera. This is why you see APS-C standard zooms designed around a focal length of ~18-50mm while on a full-frame camera it's 24-70mm.

 

[Chromatischism]

photosite size is immaterial???

Of course not, but when upsizing the sensor for these comparisons, that doesn't change.

 

[Wes]

Larger format cameras have sensors with larger photosites - without exception AFAIK.

not sure what comparison you are trying to make, or why

 

[audio2design]

Larger format cameras have sensors with larger photosites - without exception AFAIK.

not sure what comparison you are trying to make, or why

Not without exception. There are exceptions. Few, but they exist.

 

[audio2design]

That's why I choose not to look at it that way, but it does illustrate that there is no free lunch and there is no replacement for displacement. ISO increase means a faster shutter speed = less light collected and lower SNR. Since ISO has nothing to do with exposure or lens optics, I would just look at aperture area, focal length, and sensor size and how the interplay affects field of view.

The more telling comparison when going from APS-C to full-frame is "what lens does it take to get me the same field of view that I had on APS-C?" The answer is a larger aperture and longer focal length, in equal parts to maintain the same focal ratio. This gives more light collecting area (less noise) and a "zoom" of the image to put the borders in the same place as they were on your cropped sensor camera. This is why you see APS-C standard zooms designed around a focal length of ~18-50mm while on a full-frame camera it's 24-70mm.

ISO increase only means a faster shutter speed if you are saturating the sensor. ISO is just an amplifier gain setting. F-Stop/Focal Length/Sensor size all play into field of view, but it comes back to my first statement about changes w.r.t. sensor size. There really is none since you are unlikely to be using the same glass, and would be adjusting the focal length, f-stop, and ISO as appropriate (and shutter speed) to give the same results no matter what sensor size you are using. The sensor size is immaterial in the composition.

 

[Chromatischism]

There really is none since you are unlikely to be using the same glass

Correct.

and would be adjusting the focal length, f-stop, and ISO as appropriate (and shutter speed) to give the same results no matter what sensor size you are using. The sensor size is immaterial in the composition.

Disagree. I explained how focal length and aperture diameter must change to achieve the same field of view when comparing cameras with sensors of different sizes. I also gave a real example of how manufacturers design lenses in that paradigm. We pay for field of view or magnification primarily and aperture diameter and focal length are what are modified to achieve that. They don't design lenses in the way you stated. Shutter speed and ISO are not involved at all - those are the same between all cameras of all sensor sizes in this example. Look at the EXIF in your smartphone pictures. With the same field of view and the same exposure, the difference is then SNR since the smartphone camera image will be dominated by shot noise due to the small aperture (note for other readers, aperture diameter is not to be confused with focal ratio or f-ratio).

 

[audio2design]

ISO which is a totally meaningless number in the digital camera world is purposely "adjusted", not exact, but close, so that it behave similarly when you take into account f-stop across different sensor sizes. That is the way cameras are designed. It is purely convention. That is why the combination of f-stop, ISO works out across various sensor sizes (approximately). f-stop, and focal length and sensor size will define field of view for a given sensor size, and f-stop/ISO, by design convention will give you equal brightness (approximately).

The small aperture has nothing to do with the shot noise, that is a factor of very small pixels on the smart phone resulting in very low full well capacity, i.e. not a lot of electrons. Yes you are light limited, but even with a long exposure on a smart phone camera and full pixels, you still with have lots of SHOT noise because there just are not that many electrons. Read noise is typically a bit higher as well.

Aperture diameter (lens diameter) is what the lens can do. f-stop setting is what it is doing while I am taking the picture. I need to adjust focal length to get the same filed of view, and f-stop (actual aperture) to get the correct depth of focus.

I am not sure what you are even going on about how lenses are designed. That was not what my comment was about or even the discussion. A comment was made about how to deal with different depth of fields for different sized sensors, and my response was that was a moot point since we are not using the same glass with each sensor and would be adjusting the focal length and f-stop to achieve the same FOV and DOF, and because of design convention, we will also be adjusting the ISO if we want to also keep the same shutter speeds.