Today all the clouds cleared and we had a fantastic blue day:

All taken with FE 100-400mm GM and A7RV.

Took out the 16-35mm Zeiss and the A7RV for a walk today. Very nice pre-, mid- or post-apocalyptic images. 🙂

I’ve been thinking about which lens is my favourite, and I understood that I don’t have a single one. However, some of the lenses I currently own are my all-time favourites:

FE 90mm f/2.8 macro G
FE 16-35mm f/4 Zeiss
FE 100-400mm f/4.5-5.6 GM

Their common feature is that they render the image in a way that I recognise as just perfect. The colours, the sharpness, the general impression, the utility for the purpose I’m using them for – they are perfect in a sense that they just click with me.

There are such lenses that I used to own but I no longer do, because I’m out of that system, but the only one I’m actually missing is the ZD 14-54mm f/2.8-3.5, because it functions in a role that will hopefully now be filled by the FE 24-105mm f/4 G which is on the way. All my other favourites, for the most part, have a Sony equivalent that’s either currently in my kit, or might be soon enough.

The fact that I have 3 of my all-time favourite lenses on 3 important functions in my kit is amazing, and I’m very happy with that. Also, the FE 15mm f/2.8 fisheye is very close to being counted as the fourth, if I look at its image quality alone. It has strong coma wide open which doesn’t look good when I take pictures of stars, but colours and sharpness are otherwise excellent.

There’s another thing I’ve been thinking about those favourites: the way they solved my photographic problems has a finality to it, in a sense that they are just right, and I can stop thinking about it. When I put either of those on my camera, I know the image is going to be technically as good as I need it to be, and the rest is up to me, and the luck I’m having with light and the motives that day. I like that feeling of finality that comes with having the perfect tool for the job, and that goes for the new A7RV camera as well – the fact that it can create pictures that look like large format E100G slides with the dynamic range of Portra, but with no grain or crud that comes with scanning film, and it can shoot multiple such frames per second, is something of a miracle in itself. Its resolution that is basically equivalent to the 4×5″ film scanned on a Heidelberg also makes a final statement of a sort – it is as much resolution and image quality as I will ever need. If I take a picture with that camera, and one of my favourite lenses, and I’m happy with the picture, that’s something that I won’t have to revisit with better equipment at a later date.

Remember the article I wrote in praise of kit lenses, including the Sony FE 28-70mm f/3.5-5.6? Yeah, about that…

I tested my lenses on the Sony A7RV body, to see if they’re up to snuff. Let’s just say that I’m starting to get what some people were talking about. The 28-70mm is consistently less sharp than all the other lenses I tested, and while in the centre I would say it meets the minimal requirements for resolution, toward the sides and especially in the corners it is much worse.

The cause of the problem is obvious. The new body magnifies the image more, demanding more resolution from the optics. It used to be that the 35mm format was very easy on the lenses, unlike the four thirds or APS-C, which basically pull the same resolution out of a smaller circle, and a lens needs to be really great to be critically sharp on a smaller format. Basically, an APS-C body with a 24MP sensor will pull as much information from the smaller APS-C circle, that a 24MP full frame body will pull from the bigger 35mm circle.

The Sony A7RV body with 61MP pulls 26MP from the APS-C circle alone, and continues to make those demands across the 35mm frame.

I’m usually not all that obsessed with image sharpness at 100% magnification; it’s like viewing film under a microscope. It is expected that the lenses won’t draw a perfectly sharp image from centre to corners at full magnification, so the criterion is image that is good enough for printing big enough. Basically, I decided that the 28-70mm lens is the only one of my lenses that would create problems when printing big, and since my use for a lens in this range is landscapes from a tripod, which means sharp stuff that is most likely to be printed big, I decided it’s time for an upgrade. The good news is that the FE 16-35mm f/4 Zeiss and the FE 50mm f/1.8 are just fine, even though I tested the Zeiss at 35mm, which is its “softest” focal length. I didn’t even test the FE 90mm f/2.8 G macro; not only is it one of the sharpest lenses in the world, but also the closeups are generally more tolerant of magnification, because they lack the high-frequency detail such as grass, leaves or pine needles, that require sharpness on the landscape shots.

The actual reason why I finally decided to replace the 28-70mm is not this test, but the results from the last time I used it to actually take pictures, and I was shocked to find that almost nothing was sharp, and that was on the A7II. It looked as if the image stabilisation introduced some kind of an optical defect that looked like some kind of haze that blurred out the high-frequency detail, and the photos were not usable. It performed much better on the tripod, but still noticeably worse than the other two lenses I compared it with, and I decided it’s giving me too much trouble with inconsistency and I’m just done with it. I can’t rely on it to just consistently produce images to a certain standard, and instead it varies between quite nice and fucking awful, and that is simply not acceptable.

What did I replace it with? Initially, I considered the FE 24-70mm f/2.8 GM II and the FE 20-70mm f/4 G, but then decided against them both. The f/2.8 GM is excellent, but its strongest point is versatility as an all-around lens for everything, and this makes it quite heavy and super expensive. I don’t need f/2.8 aperture for a landscape lens that will be used at f/8-f/16 apertures. As for the 20-70mm f/4 G, it is much better suited for my needs, but it tries to overlap with the 16-35mm too hard and I’m not sure I need that. I ended up going with the FE 24-105mm f/4 G, simply because this is an excellent range for the intended purpose, and it is widely used and highly esteemed by landscape photographers, so I don’t expect to have issues with the lack of resolution. Sure, something will always be sharper, but I don’t need this to be the sharpest lens in the world, I just need it to keep up with my other glass and not create a blurry mess at random unpredictable moments. The FE 24-105mm f/4G is a workhorse lens relied upon to produce predictable results by many photographers and that appealed to me, because I need something that just works, and not create problems.

I keep encountering the conundrum of whether to upgrade camera or lenses first, and there’s occasionally a comparison of a high-end camera paired with a low-cost lens, against a low-end camera paired with a high-end lens, as if that’s a dilemma anyone is actually having.

I had a similar problem lately, when I decided I want to buy the FE 100-400mm GM lens, because I wanted to make a certain profile of pictures with it, but I of course understood that the autofocus on my camera isn’t capable of utilising the lens properly, and there’s our solution: camera and lens need to be seen as a unit that is combined to produce a certain result. This means that you can’t have bottlenecks that limit the effectiveness of the system as a whole, for instance you can’t have only one part of the fast autofocus system, because both camera and lens need to work together.

Also, the realistic conundrum isn’t whether to get the most expensive lens and pair it with the shittiest possible body, or vice versa. Realistically, it looks more like “should I get the 70-200 f/2.8 or 85mm f/1.8 for the portraits”, because “cheap lens” is often a fast prime, and “expensive lens” is often a zoom, where the cheap lens might actually give you better results; also, with the body it’s “do I need faster autofocus for portraits and weddings”, because that’s where the difference in price is today. If you’re shooting macro, you don’t need a body with the best autofocus, you need a great macro lens and, probably, additional lights. So, basically, the answer is to see what you actually need, where the bottlenecks of your process are, and then remove those bottlenecks. Someone else will have different problems to solve, and different money pits to fill. Sometimes the solution is counter-intuitive, for instance getting the expensive new camera body and cheap used lenses of otherwise very high quality, which looks like putting cheap lenses on an expensive camera, but in this case price is not an accurate measure of quality obtained. In any case, the lesson is to avoid formulaic thinking when solving practical problems.