The importance of "good" glass?

[davehicks]

Asking my original question was not so much about exactly what I should do with my near-term purchases. It was more about educating myself. I have often wondered, hypothetically, if I always shoot at 28mm and f/8 - f/11, is there a reason to buy a "better" lens, or is the "cheap" lens that will do those settings going to work just the same? And now I understand that the answer is "it depends" because the wrong port/dome could make the better lens perform worse overall. It's the combo - the whole package - that must be evaluated.

A cheap lens may not have the autofocus ability of a better lens. No, you don't need F2.8 very often underwater but a really junky lens is likely to have distortion effects (Cromatic Aberations) or poor focus seeking and lock. Look for feedback from other photographers and the Housing / Port vendor you are using. They generally call out the good combos that are popular and effective.

 

[Hoag]

The 28-70mm Sony FF "Kit Lens" is surprisingly good. That is what I still use on my A7iii (again, I use it above water). Although fairly slow f3.5-5.6, it is not unreasonable for most uses especially if you plan on shooting mostly in the f8-f11 range. Autofocus is actually fairly fast and accurate. (I would suspect that it will be even better on the A7Riv than on my A7iii.) Colour is good and the lens is relatively sharp. Here is an example of an image that I shot of sunrise at Mt Whitney and the Eastern Sierra Nevadas with that lens mounted on my A7iii:

The only concern that I would have using it on an A7Riv is that with 61Mpix, the sensor might exaggerate any softness in the lens. With this camera I would definitely consider the Sony/Zeiss 24-70mm f4 at some point, but if you have ordered your A7Riv with the 28-70mm then I think that you will be very happy with it.

WRT your question about major manufacturers making bad lenses, the lens that I bought that was as sharp as a spoonful of peanut butter was for my crop sensor A6000. It was a 55-210mm. It was so bad, that I owned it for less than 24 hours before I returned it. For full frame camera, most lenses will be pretty good, but some will be better than others. As an example, Sony makes (I think) 3 lenses that are about 50mm. I have the entry level 50mm f1.8 lens. It is relatively good for its price ($250USD) but autofocus is slow and of the 3 options, it is the least sharp. I don't use a "nifty 50" that often, so I was willing to accept some limitations with it. They also make a 50mm f2.8 Macro lens that is sharper than the f1.8 but is very slow to autofocus, but is fairly sharp. Then they make 1 that stands in a league of its own: The Zeiss 55mm f1.8. If I shot at 50mm f1.8(ish) more often (medium portraits), I would have gone for the 55mm Zeiss, because it is so much better than the lens I have, but I shoot it so rarely that I could not justify the difference in the cost. I am willing to accept a slower autofocus and a lens that is relatively sharp but certainly not razor sharp.

 

[stuartv]

@Hoag, thank you for all that.

I have not ordered the Sony yet. But, my current understanding is that the Sony 28-70 is "the" lens to use with the WACP and the 24-70 will not work with the WACP. So, my choice would be the 28-70.

You also touched on something that I have been pondering on. I have felt for a while like there must be a difference in lenses that would have a tangible result of producing an image that looks the same on a 24MP FF sensor and on a 42 (or 61MP) sensor versus a better lens that would actually produce visibly more detail on the higher MP sensors.

I recently learned that there actually IS a concept of lens resolution. Excellent! So now my question is, how do you know what the resolution of any given lens is? I don't recall ever seeing that as a spec from a manufacturer, but maybe it just didn't register because I had not grokked what the term meant. Is there an independent tester that publishes resolution specs for lenses? Does DPReview have resolution specs for lenses?

Now I'm curious if the Sony "kit" lens (for an a7r IV) and a WACP has 61MP (or more) resolution.

Also, I never knew that the lens itself had any real effect on focusing speed. Barring the lens not feeding the sensor enough light (i.e. not being fast enough), I would have thought the camera was solely responsible for focusing speed.

So, another question: How do you know how the focusing speed of different lenses compare to each other? Is there a spec or a database of reviews that contain focus speed evaluations? Or do you just have to get one of each lens you want to know about and compare them yourself?

 

[davehicks]

All of these lenses have motors that spin to move and align to to achieve focus. Like what you do in manual focus mode. Different lenses allow various levels of light in for the camera to use to achieve focus in dark (underwater) conditions. Depth of field and focus points factor in. Sure the camera picks the focus but it literally has to depend on and manipulate the lens to do it. So the whole system combined factors in.

 

[Hoag]

@Hoag, thank you for all that.

I have not ordered the Sony yet. But, my current understanding is that the Sony 28-70 is "the" lens to use with the WACP and the 24-70 will not work with the WACP. So, my choice would be the 28-70.

You also touched on something that I have been pondering on. I have felt for a while like there must be a difference in lenses that would have a tangible result of producing an image that looks the same on a 24MP FF sensor and on a 42 (or 61MP) sensor versus a better lens that would actually produce visibly more detail on the higher MP sensors.

I recently learned that there actually IS a concept of lens resolution. Excellent! So now my question is, how do you know what the resolution of any given lens is? I don't recall ever seeing that as a spec from a manufacturer, but maybe it just didn't register because I had not grokked what the term meant. Is there an independent tester that publishes resolution specs for lenses? Does DPReview have resolution specs for lenses?

Now I'm curious if the Sony "kit" lens (for an a7r IV) and a WACP has 61MP (or more) resolution.

Also, I never knew that the lens itself had any real effect on focusing speed. Barring the lens not feeding the sensor enough light (i.e. not being fast enough), I would have thought the camera was solely responsible for focusing speed.

So, another question: How do you know how the focusing speed of different lenses compare to each other? Is there a spec or a database of reviews that contain focus speed evaluations? Or do you just have to get one of each lens you want to know about and compare them yourself?

Those are some very good questions, and to be honest, I am not really qualified to answer most of them. Having said that, I can give you my opinions, but please do not read more into them than that.

I'm going to draw a parallel that I am not 100% sure is an accurate one, but it is the best one I can think of. People will often comment after they buy some really high end speakers that the purchase had the opposite effect than they were expecting. Their stereo actually sounds worse. The reason for this is that the high end speakers reproduce all of the sound, and small flaws that may have previously been hidden, are now suddenly being reproduced. Lenses and camera systems will tend to work in a similar fashion. If you have a 20-24MP sensor (like on my A7iii), then small shortcomings in the lens may not be apparent, but with 61MP, you will see it all, the good and the bad because the resolution of the sensor will pick it up.

I honestly don't know how the 28-70mm kit lens will work on the A7iv, but I suspect it will be fine especially at the apertures that you plan on using. DxO has pretty in depth camera and lens reviews, but for the average enthusiast (as opposed to professional) I think that the 28-70mm Kit Lens will be fine. It does have a reputation (and I have seen it in practice) for being a little soft in the corners. As for quantifying the lens' resolution, I honestly do not know if most reviews do that (although many do publish "MTF Charts"). I think of it more subjectively rather than numerically.

Here is a review that you might find interesting.

WRT the lens having an effect on the speed of the Autofocus, the camera will do the bulk of the work and all of the "number crunching", but the lens comes into play because there is a motor in the lens that once the camera has done the number crunching has to physically move an element of the glass inside the lens to achieve focus. The type and power of that motor determines how fast it can move the element and therefore how fast it can achieve focus.

I hope that helps, and like I said, don't treat any of this as more than simply my opinion.

 

[Chris Ross]

You can find resolution tests of various lenses on sites like DXO, Photozone, The Digital picture among others. However these are tests in air. As others have said some lenses seem to work well in domes others do not. I suspect the issue may have something to do with how the lenses handle virtual images - the focal plane for lenses is generally not completely flat and could conceivably curve up or down. For an image taken in air the edges will be within tolerance however when focusing on the curved virtual image of a dome the curved focal plane may either help of make things worse. This might be why some very good lenses in air don't perform so well underwater.

The WACP is another issue again which has it's own set of limitations as to what will work including a minimum of 28mm full frame equivalent focal length and I also recall a limitation on the size of the front element for the lens.

I think you can really go on actual underwater tests of lenses. The charts produced by the housing manufacturers are certainly a good starting point.