Why are modern 50mm lenses so damned complicated?
dpreview.com
dpreview.com
When I started Lensrentals I had a lot of conversations with service centers that went like this. Me: "That lens you repaired still sucks". Person at service center: "No, it’s within specs". Me: "What are the specs?" Service center: "We can’t tell you". One day, after I raised hell with a factory service manager, he patted me on the head and said, “testing lenses is complicated; you don’t have the background to understand.”
Any of you who has ever seen a physician after someone says something like ‘you wouldn't understand; it’s complicated’ knows what happened next. I had no option but to spend a couple of years buying testing equipment, offering internships to really smart optical engineering students, and developing a lens testing center and methodology that was as good as anything in the industry.
He is also capable of writing about hard topics in a way that most people can understand them.
Anyone interested in camera lenses should read also his previous posts https://www.dpreview.com/members/1329131098
I also like cameralabs' lens tests [1]. Not as scientific as lensrentals, however, it has real samples and images of charts for comparison, so you can get a sense for what the different resolution means in image resolution. It also tests things beyond resolution. Such as out of focus rendering, coma, sun stars, etc.
[0] https://www.lensrentals.com/blog [1] https://www.cameralabs.com/camera-lens-reviews/
Recently I looked at some very old ones I haven't used in a long time. I thought they are faulty (soft).
After some diagnosing I found that it is not the fault of lenses but rather in the meantime I have started looking at images at 1:1 in Lightroom and gotten used to ridiculously sharp images from new cameras.
Have you ever watched an old TV show and despaired how unwatchable it is because of low quality of image? Yes, that's the same thing -- our tastes change.
I've had the opposite reaction - I'd watch something I've seen before in 4k remastered or something and it would be a worse experience because the extra details reveal the low quality tricks you can get away with on lower reproduction quality. I've even noticed poor acting a lot more on higher def. It took away from immersion IMO.
Every time I go back visiting parents, which have TV with this on, anything on TV feels surreal and not in a good way. Which isn't bad, watching TV instead of interacting with my parents is stupid waste of time anyway and TV itself actively helps with that
I only use it at about 10-20% though, not at 100% which I think is the default setting and I agree is vomitive.
I think the people complaining about the first are whiners and as a digital video expert I will allow you to leave it on.
Reading the manual doesn't help when your TV doesn't have a fast mode. And many don't.
If the source happens to have more frames, then show them of course. But otherwise I'd prefer to see the same frame again for the extra 1/60th of a second.
Also, when there was a cut to a new scene, for a few seconds my brain thought the movie was playing faster. Like those old black and white movies of the early 20th century.
But clearly we learned something or there was something special about Hobbit, because that hasn't necessarily been my experience with all 4k / high-frame shows or movies...
I saw them in both normal frame rate and HFR as they were coming out and it looked kinda low-budget, weirdly lit/color corrected, and generally just clunkier than the LotR films did in the moment.
I would describe the effect of HFR as pushing immersion past suspension of disbelief into reality. You can no longer see them as characters, you see them as you would as if it wasn't through a screen - as you said, like watching actors on a stage.
Fortunately(?), humans may get used to HFR as we have with previous advancements in video technology. Remember when people were freaking out about 4K, 3D, HD, Color, "The Arrival of a Train at La Ciotat Station"?
I have worked for Canal+ a long, long time ago and I have been on the set couple of times. When the studio was lit you felt hot even though there was air conditioning.
It is basically the same as meeting sunshine on a winter day with no air movement. The air is cold but the sunshine on your skin makes you feel much hotter than it really is.
It's a shame there is no 1080p HDR content. 35mm movies would benefit a lot more of higher color bit depth that higher resolution. Anyone knows if this is a codec limitation?
Typically, you as a consumer, will be receiving an H.264 or H.265 stream that your cable box or streaming device decodes. These have all been updated with levels/profiles to allow for things like 3D, 4K, HDR, and 10bit encodes. The shoehorn used to shove new features into these formats has been put to heavy use.
I don't have any examples on hand. But with large amount of time I have found myself with over the past year to watch a lot of content, I too have noticed even some of the last episodics to be shot on film (2005-2010 range) are very noticeable. I have also seen some features from the 90s that looked really clean in comparison. Lots of things go into that, from how much the camera department "cared", what film stock was used, what film processor was used, etc. Towards the end of mass film production, there were fewer and fewer labs left. During they heyday, the soups used in the processing where in constant use. As demand lowered, the soups kind of stagnated especially in the shops financially strapped.
Film is fickle to be sure.
Here's a good list to check whether a film is real 4K or not.
https://www.digiraw.com/DVD-4K-Bluray-ripping-service/4K-UHD...
I've now remembered that the X-Men films from 20 years ago did look pretty good in 4K HDR. And apparently they are "real 4K". Also the 4K remaster of The Fifth Element.
Best to leave modern productions (like Nolan) out of the conversation, IMO. It can give an unrealistic picture of what resolution is achievable. Take Lawrence of Arabia for example: it's possibly the paradigmatic example of a 70mm film, and yet in most scenes the resolution is no better than an upscaled 1080p. This is due to a combination of problems: film stock degrades relatively quickly and was generally of lower quality 50+ years ago than today, and focus pulling was not precise enough to take full advantage of the available resolving power of 70mm film.
Unfortunately this is also true for many scenes in 2001: A Space Odyssey. Take a look at these comparisons here: https://caps-a-holic.com/c.php?go=1&a=0&d1=12517&d2=12509&s1... You're looking at an upscaled 1080p film versus native 4k (tonemapped to have roughly comparable colors on SDR screens). There's little if any more detail in the 4k image.
On the other hand, 35mm film certainly can have details visible only in 4k, especially more recent stuff. Check out this comparison for Casino (1995), the difference is unbelievable: https://caps-a-holic.com/c.php?d1=13389&d2=13388&s1=134076&s...
I agree with you about HDR though. It's certainly the most important improvement that's come up in the last decade, much more than the resolution increase.
> It's a shame there is no 1080p HDR content. 35mm movies would benefit a lot more of higher color bit depth that higher resolution. Anyone knows if this is a codec limitation?
Yes and no. The standard for Blu-ray (which uses H.264) is not able to accommodate HDR or 4k, so it was not possible until the release of the UHD Blu-ray standard. Players with the UHD standard support 4k, HDR, and the HEVC codec instead of H.264. HEVC (aka H.265) can support HDR at any resolution, but since you're releasing a "UHD" disc anyway that only people with supporting players can use, you may as well put it on there in 4k anyway. Better for marketing, probably. So it's a bit complicated. If you check bittorrent trackers, you'll probably be able to find 1080p films with HDR (encoded with HEVC), but no one in the industry releases films that way.
(Also, the higher bit depth is not quite as important as the extended dynamic range.)
Long time ago, when dinosaurs roamed the Earth and AMD K8 was the pinnacle of technology, I invested in a basic home theater setup. A decent AV receiver, a subwoofer, two studio-quality monitor speakers. A friend warned me that the speakers I had picked were "crisp, but unforgiving".
All of a sudden DVDs sounded wonderful. The audio tracks in TV programmes became clear. The warnings made no sense.
Until I watched a classic TV show episode from 1980's. Bad mixing. Flipped stereo channels. Stuffy voiceovers. Everything sounded ... wrong.
Suddenly, it all made sense. I understood.
There is no generally accepted theory of stereo sound, instead there are at least three theories:
(1) a method for mastering sound recordings intended to be played by two speakers that are right next to each other (two sides of a TV, see https://en.wikipedia.org/wiki/Sound_Retrieval_System for some theory)
(2) mastering sound records so that 'left means left' and 'right means right' which means your speakers are on opposite sides of the room and you're near the middle. That can cast sounds effectively to the left and right but falls down casting them to the middle
(3) Ambisonics, which doesn't really work because the coherence in natural sound sources badly damages the timbre of natural sounds and the concept of 'represent natural soundfield' doesn't really mix with 'record instruments individually in dead environments', 'add reverb', and 'repeat' which is how 'quality' sound in the judgement of the industry is attained.
Often DVDs were mixed under the assumption that they couldn't trust the receiver (e.g. no center channel); blu-rays sound a lot better not just because they allocate more bits to sound, but because the mixing is more aggressive... they don't feel they have to copy dialog on the L and the R because they know many people don't have a working C.
I am delighted to be one of today's lucky 10000: https://xkcd.com/1053/
I watched Gemini Man in 4k HDR and some of the scenes seemed a little like a video game. I think there was sort of an "uncanny valley" sort of thing where you could see detail into the deepest shadow and it looked fake.
On the other hand, 2001 in 4k was awesome.
People here blame it on age, but very high resolution always screams it was shot on video. I don't need to see every pore on an actor's face. I'm usually wanting to get away from reality when watching a movie, or a tv show too.
I am very content with the visual quality film gave us.
I love black, and white film for the same reasons, and others, even more.
https://www.flickr.com/photos/acomjean/35581239466/in/photos...
https://www.digitaltrends.com/home-theater/what-is-the-soap-...
Basically, we (over the age of about 25 anyway) are used to associating 24 FPS with “high quality” content like movies and higher FPS with cheap content like soap operas.
I'd especially emphasize the 'tricks you can get away with' in low resolution television. A few months ago I was watching a rerun of an early 1960s B&W sitcom on a TV sub-channel running at normal 1080 HD resolution. The program was My Favorite Martian which is a spoofed up science fiction comedy series. The program features a Martian who can disappear and levitate objects so it's not unusual to see things either moving or floating in midair.
Anyway, the tricks employed to levitate things were embarrassingly obvious in standard HD/1080 resolution as it was dead easy to see that they had been suspended on nylon fishing line. When the show was first aired the most common TV standards were 405, 525 and 625 lines so the series producers were betting that the limiting resolution of these TV systems would hide the lines—which clearly they did. BTW, as a young kid I used to watch the show [on 625 B&W], and if I'd have seen that fishing line then I've made a big deal out of the fact.
Incidentally, the early series were in black & white and filmed on 35mm film which meant the original negatives would have been dazzlingly sharp in comparison to the then TV systems. It's interesting to note the present video quality of this series—unlike so many others from the time—is excellent (and that includes the later color episodes). Clearly, the recent transfer/digitization was done from negatives have been archived well. It's especially obvious in the color series which are of exceptionally good color quality given the passage of time (there's not a trace of fade, cross-coupling of color dyes or color bleed which plagues so many old films).
I don't use autofocus, pretty much ever, so that isn't a consideration. But I do want quality precision mechanics, and I feel that shouldn't be much to ask for.
My old lenses feel like quality scientific instruments, in comparison, and have long focus throws that allow me to focus very accurately.
I now own a Canon body and zero Canon lenses except for an FD 300/2.8 that I modified to fit EF.
For a great example of a sharp modern L, check out the 16-35 f/4 IS. I'm not an ultra wide photographer by nature, but that lens impresses me each time it is called-upon.
Choosing lenses, in the long run, is such a personal choice. My favorite first came to market in the early 1990s (and spent this morning with it) -- glad you have found good fits, too :).
35mm f/1.4 - super sharp, great quality lens.
100-400 f/3.5-5.6 II - amazing lens. Very sharp and focuses very fast. I often use this with a canon multiplier as well for wildlife shots. For the price it’s amazing.
16-35 f/4 IS - this is such a great landscape lens.
Wide open in heavy snowfall at night, there's nothing I've experienced like it. Not quite life-changing, but it's real good.
Couldn't justify a purchase, but I may have purchased a used 1.4 a month later..... The focus-falloff, rendering, and out-of-focus elements aren't the same, but it's still wonderful in the dark and inexpensive-enough to be replaced if I kill it. Someday, the stars will align and I'll work with a 1.2 again.
My desert-island lens is a (used) 400 f/5.6. Someday I'll try the 100-400, but worry about the telescoping assembly ingesting moisture and dust in the field.
But it is a consideration. Modern lenses are all designed for autofocus bodies. Yes, you can flip them into manual, but it is still a manual adjustment on a gear designed for computer control. A decent manual focus lens will typically have 270 degress of motion in the filter ring allowing much better control for humans. A autofocus lens typically only has 90 degress of motion (and then infinite turning drek). Much hard to get critical focus manually on an autofocus lens than a manual focus lens.
Also, your older lenses probably weigh much more than the modern lenses. Lots of people prefer the lighter barrels.
I do a fair bit of astrophotography, and the modern Canon glass is great for that. Crip, clean, and no need for focus once truly focused on infinity (varies on each lens, but typically "close" to where it is marked). The older lenses just don't have the sharpness.
So there's a place for each of the older and newer lenses in my kit.
Interesting viewpoint! I do a lot of astrophotography and I found AF-capable lenses to just not work for me at all because a couple of shakes and some play in the gears would slop the lens out of focus, and the short throw and lack of infinity precision make it really hard.
For astrophotography I just love being able to focus on a star and either calibrate the lens stop to be at infinity, or slap some gaffers tape and it's good for at least 2 months until the night temperature outside changes significantly.
Of course, yes, sharpness is an issue for many older lenses, but the better ones of the 70s and 80s have very decent IQ. Honestly I just wish modern lens manufacturers would make more manual focus lenses. Laowa, Samyang, and Zeiss are all awesome for continuing that tradition.
I would buy an entire kit if they would release modern glass with 100% manual including aperture. There are places that will rehouse your lens so that it is full manual. The cost pretty much doubles the cost of the lens, but I have been tempted to do that for the 70-200.
FWIW, I've noticed that Hasselblad lenses have the longest focus throws of any lens I've ever used.
I know some will vociferously disagree with this statement, but I’d like to see that backed up by proper side by side comparisons.
We know for sure that people do rationalize the purchase of expensive gear with reference to spurious properties (such as the mythical “medium format look”).
The preceding considerations show that the medium format look, if it exists, must be a property that goes away as soon as you crop the photo. But none of the candidates for this alleged look (e.g. 'rendering' of out of focus regions) has that property.
In principle, medium format should offer shallower depth of field. However, in practice, medium format lenses rarely go beyond f2.8 whereas crazy ~f1 lenses are available for full frame. So if you really want ludicrously shallow depth of field you're probably better off with full frame.
All this being said, it's obviously the case that any given medium format lens might look different from any given full frame or crop sensor lens. Perhaps there are even generalisations about the 'look' of commonly used MF lenses vs. commonly used FF lenses (though I have never seen any evidence for this).
In-lens leaf shutters can also make a real difference, and are more common on medium format cameras.
On an anecdotal level, I'd add that I shoot a bit of 4x5 film as a hobby and the photos 'look' the same as the photos I take on my phone with its tiny sensor (apart from the expected differences such as resolution, grain, depth of field).
This site has some side-by-side comparisons of FF and MF: https://srussenschuck.com/review-fujifilm-gfx-50s/
The world is not just 35mm in size.
An exception would be old kit zooms from the film era, which really are awful, since they were designed for people who rarely even went up to 8x10 prints. Modern kit zooms are usually very sharp lenses, though.
Any smartphone these days can give perfectly good sharp photos. I lug the beast around to get shots you can't get with a smartphone.
Agreed on the shots where you're at 1.4 and pushing the limits of your shutter speed.
DC stands for “defocus control", as distorting the bokeh forward of backward.
How do you feel abbout the DC function? I've found R settings don't really focus correctly with AF, so it's best to focus manually. Is that the case with the 135mm version too?
These lenses stopped down are better more for artistic photography as they have interesting bokeh patterns and lots of character.
First film is much less sensitive to the angle at which light strikes it. There are some great, sharp old lens designs from leica, zeiss and others that show heavy light fall off in the corners on digital sensors. This is a big part of why even f1.8 or 2 lenses need to be more complicated.
Next film grain itself has sharp edges which effectively sharpens images a bit and also limits the resolution and generally has a pleasing effect when viewed at high magnification. With digital you don't get that and lens flaws get a lot more obvious.
What gets me is that lens distortion is a fully reversible convolution. It's not lossy, at all. Or rather, it's only lossy in the sense that the existing quantization in the sensor is already lossy: pixel boundaries, spectrum of the color sensors, etc... all have straightforward analogs in both space and frequency domain.
Basically: why hasn't someone come out with a camera with a junky single-piece lens and a fancy image processing backend and won the market?
Canon's M-series was similar, they didn't go all in on "pro" features, instead focussed on a consumer market that was shrinking rapidly.
The problem wasn't the technology, the problem was the fear of cannibalizing their own DSLR sales.
I assume part of this is that the camera companies are Japanese and refuse to pay any engineers more than $20k a year.
For real pretension you have to get a Leica.
It looks like they still don't really plan to add it: https://www.dpreview.com/interviews/9195902169/interview-son...
but at least if they had better bracketing options we could do it ourselves.
Not better tech (better as in takes better photos), rather more convenient tech which is good enough.
Once you are in the 2k$ usd lens space the buyer is looking for ultra premium stuff either as an enthusiast buyer or as some one who makes their living that way.
Secondly, we already do correct distortion and chromatic aberration in software, but this doesn't really make up for loss of resolution that you can't fix. Information is being lost.
Lens design is about minimizing this.
Even if reconstruction is possible, it might amplify the noise by a substantial amount.
The error you get out is of the order of the sampling error you put in. That's true of all lenses.
Information is lost when a ray strikes somewhere it shouldn't. Things get blurred and cannot be recovered. You can't fix this with software.
Can't you? If you're dealing with known hardware, then you can apply a deconvolution. Blurring can literally be undone, people do it all the time. It's harder when you have to estimate the kernel, but much easier when you already know the lens's exact properties. Also when you apply it directly to RAW data.
So not exactly sure which information you're referring to you when say information is being lost?
Sure artifacts can be introduced from noise, etc., but that's all just tradeoffs. If a simpler lens is letting in more light, or you put money towards the sensor rather than the lens, the end result may well be better, no?
And there's a variety of sharpening tools that claim to do this! DxO is a company that sells a raw processor for precisely this purpose. People still buy better lenses.
I wouldnt say the camera companies did not see this coming. They did and chose to not do anything about it. I shoot videos with some pretty low end DSLR cameras and they are often lacking the feautures of any modern flagship phone. It's a joke. We once used a whole Iphone shooting at 120fps and made some plain scenes look amazing in post. A revelation since most pro videographers pooh pooh phone cameras without realizing that they are growing with alarming speed.
I believe it’s called the iPhone.
Such corrections are all but essential for affordable zooms, but I'm still split on the potential for over-use in primes. I'll point to the Sony 16/2.8 E as an example; the use of correction in JPEGs is pretty evident compared to shooting raw, and you can see the correction softening the corners.
For example Panasonic has a 25mm f/1.7 lens that is 125 grams and costs $150 brand new.
Of course you can also buy similar focal length lenses that are huge and weigh over a pound, while spending $1500+. But I don't know how much advantage those really give except in specific rare situations.
For 150g on full-frame you can get a 50mm 1.8, at a cost of 150$ brand new. So on full frame at a similar weight and at the same price you get superior optics.
If you really care about the weight, the 40mm 2.8 has the same weight and the same price but superior optics again.
All modern FF cameras do distortion correction too. M4/3 lenses are not especially decided to trade-off distortion for sharpness anymore than full-frame.
Noise is decided by the quality of the sensor, and the solid angle of light that hits each point on the sensor. The solid angle as it happens also decides depth of field.
So it is always true that you should apply crop factor to aperture. ISO is just not the correct factor to describe gain because an inverse crop factor is applied to it in effect.
So at the same exposure and shutter speed, at f/3.4 on FF you will get the same image with the same amount of noise as you would at f/1.7 on M4/3.
I'm interested to learn more about the effect of light angle on noise. Do you have a source that talks about this? I didn't find anything useful in 2 minutes of searching the web.
As for light angle, it's a consequence of the second law of thermodynamics, that translates in optics as the conservation of étendue.
Basically what this means it that the amount of light captured from a light source is equivalent to the solid angle of light from that light source onto the entrance pupil - whose size is the focal length divided by the f number.
So basically, what determines in the same scene how much light is received by the sensor depends on the size of the aperture in milimiters.
So on a sensor of the same efficiency, it's the physical size of the aperture that determines how much light is captured.
> what determines in the same scene how much light is received by the sensor depends on the size of the aperture in milimiters
That's true, but the amount of light is not the intensity of light.
Yes, you will need four times the ISO. But you will get exactly the same amount of noise.
This is what I thought (it's not really convolution, but yes it is correctable), until I realized darktable/lensfun doesn't have lens profiles for any of my lenses... :(
https://pixls.us/articles/create-lens-calibration-data-for-l...
that's what phone cameras are...
And a lot of lens aberrations not correctable robustly. LCA is basically impossible to correct, field curvature is impossible to correct, bokeh quality is impossible to correct (within reason), wide depth of field is impossible to correct, dynamic range is impossible to correct (without bracketing), coma is impossible to correct, flaring is impossible to correct and the list goes on.
Unfortunately, you don't. You have a sensor that captures only amplitudes, on a spatially quantized grid, with some noise. It loses a lot of information. If your lens has very low aberration and not too much distortion, most of that information is information you don't care about. If your lens has a lot of aberration and distortion, a lot of the information you care about is encoded in exactly the stuff that the sensor is losing.
It's the photography equivalent of "if you were a real audiophile you'd only use pure analogue signals, digital destroys the sound".
A great example of this is DiffuserCam [1], which uses a diffuser as a “lens”, combined with a carefully measured point spread function (PSF) of a single point for each axial plane to estimate the “true” PSF that actually varies throughout the entire volume. So although in this work they estimate the reconstruction by making a translational invariance assumption, this is done for practical reasons and quality is lost.
To fully recover a lossless image as you suggest, one would have to measure a PSF for each point in the volume.
DPreview largely seems to write for the crowd that doesn't even care about the art aspect and just wants validation that they have better photo gear than others.
These new 50mm lenses often defeat the point cause they're enormous, heavy, very expensive, and have little benefit over the slower versions of old lenses since most of the time we don't walk around shooting everything at f/1.0-f/1.4, and when we do we're actually looking for some artistic weirdness.
I've mostly had Canon.. I've owned the old f/1.8, the f/1.4, currently have the newer f/1.8, and have rented the f/1.2.
Even the Canon f/1.2, it's older, and not as big/heavy as some of these more recent ones, but it's already so big it feels silly most of the time compared to the smaller/cheaper ones.
It was useful as a rental for specific scenarios, and the new huge/expensive ones are likewise. They're great though if you're going to compare your gear collection instead of your photographs though.
DPReview, FStoppers, PetaPixel, and back in before the web the Popular Photography and Modern Photography magazines. There's a big market for prosumer gear that costs a lot, has ideal technical specs, and hangs around the owner's neck like a prize. Of course these outlets, and the magazines before them, get almost all their money from advertisers, and run lots of review articles (totally objective of course /s) and are of almost zero interest to anyone working in the industry.
Great photographs can be taken with any camera.
It's big enough and has a diverse enough community that it can avoid the worst excesses of pay-per-review.
Their writing is very much in the vein of "your gear is inadequate if a new version has come out."
They are really the bottom of the barrel. They continually write about how some new piece of gear is so revolutionary because it allows you to take a particular photo and it was impossible before, as long as you didn't know about the 50-100 year old technique to get around the issue that actually works better even if you have the brand new camera.
Haha... I was getting sucked into this, still am in a way. I spent about a year (2019) with a now cheap Sony Nex-5n with an 18-55 kit lens... but wanted to upgrade, more MP, higher video quality... got an A7II (24MP, 1080P) and then an A7RIII (42MP, 4K) ... I bought a 12-24mm F4 G lens at $1.3K __USED__ which to me is insane and that's not even that much in this space. It's just my car is $2K... I still can't comprehend it... I've been poor most of my life till recently, I still am poor but yeah... just interesting climbing the levels... Since the last month or two purchased 6 other lenses recently all under $1K but yeah... it's like "I need that 55mm" or "I need "12mm" just funny... then you think, this money is just sitting on your shelf not invested... idk.
I am self aware though, FIRE and make more money in general, try to avoid the above.
$2000 could easily get a 15 year old Honda Civic, Toyota Corolla or similar compact car with a reputation for longevity, without major issues. Shopping in that price range can, however be risky if you're not mechanically inclined enough to inspect the condition of the car and don't have the aid of someone who is.
Not saying you can't get a good car for that price but yeah.
All my stuff is used, the latest Sony Alpha 1 is $6.5K ha a bit out of my budget.
Anyway the talk of money is bad/pointless, I saw one cinema lens that was $100K like wth... levels in life.
Most people putting out high-production-value videos regularly aren't doing it for fun or to make the world a better place; they're closely watching how much money they make from it.
In good light, great photographs can be taken with any camera.
The trick is knowing what can be made with available light.
Kai W (& Lok) was solid when he was there, since he would trash things that deserved it and his juvenile objectivity was humorous. Now they seem more generic in the "this marginally better camera is 100% worth buying" sense.
I've used a bunch of cameras ranging from smartphones to high end point-and-shoots all the way up to full frame "prosumer" mirrorless. While I have no doubt someone whose work gets featured in say, NatGeo, can take an iPhone4 and make wonderful shots, I can't help but feel the rest of us Average Joes and Janes get a tremendous boost from having better gear.
And even that NatGeo featured pro would produce much better pictures with a FF DSLR than being limited to a iPhone 4, no?
Define "better"
I do like reading Roger's work (here and at his LensRentals blog) just for curiosity about how lenses actually work, though.
However, the entire thing feels artificial. It's hard to describe. I don't want to go back, but getting shots is so easy now that it's less of a rewarding feeling nailing a shot. Messed up ISO? Doesn't matter, sensor is ISO-invariant, just adjust exposure in post at no cost to quality. You cannot mess up focus anymore. At 10fps, you cannot miss a shot anymore (mostly).
I recommend analog to people who want fulfillment from the process itself.
I like good pictures, so I find it more rewarding when a shot succeeds and I can move on to other shots or try a harder shot for a better picture. I can play video games when I need extra fake challenge because life is too easy.
The movie from the 70s had so many out of focus shots, tracking shots were a bit hit or miss and colors were rather poor (mostly just flat but also color balance issues between scenes).
The modern indie movie looked almost like any triple-A movie, with crisp focus, smooth tracking and with lovely natural colors, even in low light.
Not like it came as a surprise as such, but just the back to back experience highlighted the contrast. Really made me sit and appreciate the technological advances.
Imagine if content can be generated, real-time, by an AI that monitors your brain for signs of engagement. Holy shit, what if that’s what this life actually is!
To be fair, the author of this article is Roger Cicala of LensRentals fame [1]; optical testing is his bread and butter. He's also the first to point out that optimal lens performance is neither necessary nor sufficient for good photography, and is famously snarky in his writing towards "spec sheet warriors" and sharpness-obsessed gearheads.
He rents lenses, and a big point of what I'm saying is these fancy lenses are very often specialist tools that are much better to rent from someone like him occasionally when you need something very special, but that doesn't make them better items to own vs the less expensive & smaller/lighter 50mm lenses.
I am pretty glad dpreview does these kind of technical reviews, it nicely completes other photographer reviews who'll dig more on the experience and real world impact side of things.
The excessive weight of these modern 50mm lens can be a real detriment to getting better photographs. It makes you less mobile, and you can only carry so much weight. If you're carring 10x more weight for your 50mm lens than an older design that is 95% as high quality you inevitably have to leave something else out of the pack.
Your point on lens weight is valid, but I don't see it as something that should be in every single lens optic review, all the more so the use case of 50mm lenses are not just street snaps for instance.
I think the closest analog might be someone that reads the reviews, buys the latest-and-greatest computing hardware... and then uses it to surf the web and play Tetris.
Besides, some people like the technical optical engineering aspect of it and are happy to follow the latest developments in the same way that some people feel the urge to read articles on the latest HPC Xeon. Gatekeeping and condescension is completely unwarranted.
The sneering is pretty typical for any hobby that lumps practitioners and collectors under the same heading. The former group tends to see the gear as mere tools, to be used as needed to achieve some other end. For the latter group the gear is the whole thing - or perhaps talking about the gear.
Neither is bad, but it would probably work better to call the hobbies by different names. Perhaps "photographers" and "camera collectors" or something. But then, being a collector doesn't have quite the same cachet as being an artist.
I agree somewhat, which is why I mostly use a cheap lens, and an ancient EOS 60D. However, I think the $3,000 lens is a red herring. The vast majority of lenses sold are kit lenses or cheap plastic. Because whilst there is an awful lot of average photographers, not all of them are millionaires. Now, to actual millionaires $3,000 are probably worth less than $300 for me, so I don't see how I would have a moral high ground.
Honestly, this is a made up problem. Some people prefer working on composition, some are better at capturing subtle lights, and some have a technical mind. These people still buy camera and take pictures, as a hobby. So why wouldn't they be called "hobbyist photographers" like the rest of us? And why does it have any importance whatsoever?
>So why wouldn't they be called "hobbyist photographers" like the rest of us?
I feel like I explained my thoughts on that in my previous comment. Basically if you're in it for the love of the gear rather than to make photographs that's fine, but your hobby is collecting rather than photography. Or, to put it another way, photography gear rather than photo-making.
>why does it have any importance whatsoever?
Sometimes the photo-makers get annoyed at being lumped in with the gear collectors. And a fair number of people get taken in reading the gearhead commentary and end up buying expensive lenses that won't actually help that person take better photos.
Yes, you did. Sorry, it shouldn't have sounded so personal. My real problem is with the community in general.
> Basically if you're in it for the love of the gear rather than to make photographs that's fine, but your hobby is collecting rather than photography.
The thing is, it is a spectrum between pinhole photographers and gear heads. There are different optima for different people along the way, but fundamentally they all do the same thing: taking pictures. So where do you draw the line, and who gets to choose who is branded a legitimate photographer instead of some weirdo with too much money?
I have never seen any evidence that "collectors" (in the sense of people who buy modern gear without taking any picture with it; there are plenty of legitimate collectors of old cameras) really are that common, or any more common than soccer moms who buy a camera to collect dust. Every time they are mentioned it is as bogeymen on a photography forum. It is puzzling that a community would show so extreme sentiments towards one of its subsets. And I am a railway modeller, so I am not stranger to strongly opinionated hobbyists.
> And a fair number of people get taken in reading the gearhead commentary and end up buying expensive lenses that won't actually help that person take better photos.
Sure, that is a problem for some people who are eager to give themselves the right tools and get to overspend. And to be honest most of us are willing to fall into that trap.
Yeah, for sure, I didn't intend to make it sound like as much of a clear cut dichotomy as I ended up doing.
>who is branded a legitimate photographer instead of some weirdo with too much money?
You ask the Internet Photography Police to render a verdict, of course! But yeah, it's not a clearcut distinction in most cases.
- Photographers - Modern Gear collectors - Vintage Gear collectors
You can have a little bit of each of these 3 you're interested in, and you won't necessarily always have the same interest in each area.
But people who are 100% interested in one of the gear collecting aspects are pretty common and are very far from the artistic side.
I do agree consumers don’t deserve condescension for geeking about technical details, but I still see it as at least ironic when somebody buys an overengineered product.
It's perhaps my second most frequently used lens, after the 28/2 "Hollywood" Distagon (which isn't really all that sharp off-center, and sports very strong vignetting, but produces a beautiful, dramatic quality).
A few from eBay, some good deals pop up every now and then if you're patient for a few months and set up notifications based on keywords rather than thinking "I need this lens right now".
Sometimes if you see an auction on eBay with a high starting price and no bids, it's possible they have been re-listing the item for a long time, and you can message the seller and say you'll buy it right now at a lower price if they turn on that option; many would be happy to be "done with it" and sold. Sometimes if the seller is local you can convince them to cancel the listing and do an off-eBay, in-person sale for cash at a discount (eBay charges a lot of seller fees).
If you bid on an actual eBay auction, place your bid with the highest price you are actually willing to pay and ONLY in the last 10 seconds, or preferably last 3 seconds. By doing that you will either lose OR win but get it at just slightly higher than the next highest bidder's max price and significantly less than your stated price, while leaving that second bidder with no time to re-think and re-bid. By doing the single last-minute bid, you effectively turn the auction into a sealed-bid, second-price (Vickrey) auction for yourself, which is most efficient for you in a game theoretic sense.
KEH has very reasonable prices on most old lenses. Not cheap, but very reasonable.
Also, if you're willing to take apart a lens with a slight amount of haze/fungus, sanitize and clean the hell out of it (not for faint of heart), and put it back together, you can get some very, very steep discounts. Or if the lens has a couple of scratches, you can get steep steep discounts due to loss of antique value, you don't need to take anything apart, and it will have zero noticeable impact on wide aperture images.
As a hobbyist photographer and a reader of DPReview, I can tell that you can find what you're looking for. There also guys who are in for the art, in for the studio photography or just for gear.
These kind of technical articles makes me happy and piques my interest deeply, because I love to know how things work, and why they're designed this way.
I'm personally aware that better gear makes my photography better, however I love to know theoretical capabilities and limits of my gear. Understanding my gear made my photography better and DPReview's both articles and forums helped me a lot TBH.
For the most light and scenarios, f2.8 is very suitable from my experience, esp. with latest generation of sensors (A7III, Z6/7, etc.), however in extreme dim light (I shoot tango dancing, BTW), f1.8 is really helping. Since you're away from subjects too, DoF becomes very acceptable too.
I agree that unless you've a specific need, f1.4 and f1.2 lenses are halo products. As a Sony shooter, that 2.5/40mm lens [0] is very compelling for street photography from perspective.
At the end of the day, new lenses are giant leaps when compared to older ones in terms of resolution and sharpness. Older lenses indeed have character but, newer ones are not always devoid of it.
[0]: https://www.sony.com/electronics/camera-lenses/sel40f25g
That said, it works magic at night.
This is a pretty broad generalization. There is plenty of photography that is nothing but technical. Also, the technical side of photography is what drives the innovation to enable the creative side. Do hobbyist landscape and portrait photographers need to be pouring over test data like this or even benefit in appreciable ways from these advances? Not likely. But the R&D that is done in optics begins with some of the most technical fields and trickles down.
(OTOH if I open a 200mm as wide as it goes I definitely do prefer it sharp - different scenario.)
What has happened to the iPhone camera in the last 10 years has also happened to handheld cameras too, but they were starting from a much better place, most people just haven't realised as the iPhone is good enough for almost everything.
If you are unsure what to get next, I suggest looking at this list of every E-mount full-frame lens:
https://phillipreeve.net/blog/manual-lenses-sony-a7/
Plus, a comprehensive list from the same folks for all the E mount lenses:
https://phillipreeve.net/blog/fe-lenses-sony-comprehensive-i...
Generally speaking all of the M42 mount Soviet Zeiss clones will be interesting and relatively affordable.
I'm looking now to buy Tamron 28-200 that is basically best super zoom available on any system (including first party zooms). Excellent for vacations.
I shoot concerts with it on an A7ii. The lens is great. The camera body however is showing its age which is unfortunate, but I'll upgrade in a generation or two.
Sigma have been top of DXOMARK's scores for the last five years.
Then there are new requirements like AF and optical image stabilisation. With classical prime lenses, focussing often would move the whole lens forward and backward. Or at least some large lens group. That makes the optical construction easier but prevents fast AF. Modern lens designs would use a rather small lens in the middle for focussing. With the reduced mass, the focus speed goes up. Image stabilization also asks for a small lens or lens group to be moved.
But one big difference shouldn't be overlooked either: the abundance of compute power. Today, you can easily simulate a lens system of any complexity. Leica did actually use raytracing to optimize their lenses in the 50ies - but in the absence of computers, they had a room full of people who did the calculations manually. So with a months worth of work, they managed to calculate the path of like 5 rays through their given lens design. Which had like 6 elements. Modern prime lenses can have up to 20 lens elements, zoom lenses up to 30.
For a good comparison about different lens design approaches, one should look at Leica lenses. Because Leica both still makes classical manual focus lenses for the M system, which typically have like 5-7 lens elements, and absolutely modern style lenses for their SL system, which have twice or more lens elements. In each design philosophy, they are making some of the best lenses, cost usually not the top priority :). Which makes a good base for comparisons. For example, they quite recently both introduced a new 35f2 lens for the M and the SL mount.
Personally, I am quite fond of the prime lenses with few lens elements. While modern glass coatings can pretty much eliminate reflections on glass surfaces, every single glass-air boundary does create reflections. So while the modern designs are technically close to perfect, I think those designs with few lens elements have a special image rendering. Very contrasty, lots of character.
I shoot with a nice, but old, MF camera and prime lens on a tripod and then do wet prints (usually) or drum scans (rarely). Based on close examination of drum scans, I estimate my setup gives images of around 60-80 megapixels. Knock that down somewhat for color. This is a bit lower than the quoted resolution for the film I use, if you read the spec sheet.
Rough comparison of digital vs film leads me to conclude that you can’t just compare the camera resolution against the film scan resolution. Maybe it’s the low-pass filter / Bayer filter, but a film scan will have the same image quality as a raw digital photo of somewhat larger size (say, 50%).
So if you take a high-end full-frame digital camera with a 24x36mm sensor, you get 50 megapixels, and my film setup will get you around the same number of pixels when shooting color, but the film scans will be sharper. The film is 4x the area—56x70mm. So it’s close, and I wouldn’t be surprised if a top-of-the-line full-frame digital camera could beat a 645. But 645 is the smallest of the medium format sizes, and digital wouldn’t be beating it by a huge margin. It’s a bit subjective, because the transfer functions for digital and film are so different and there’s no obvious correct way to compare them mathematically.
That said, digital will easily win at higher ISOs.
The point when digital became competitive with medium-format was an important milestone--it's just that if you did events, sports, wildlife, or photojournalism you were probably already on digital.
Digital is in an awkward spot where semiconductor yields make larger sensors impractical, so they have to focus on hyper-optimizing lenses. This ends up being very expensive. And, there are diminishing returns; the wave nature of light introduces interesting artifacts the smaller your pixels get, so there is unfortunately a hard limit. Compare this to a big piece of plastic dipped in some chemicals, and you wonder how much we really benefited from the switch to digital. It's great for the snapshot market, it's great for people that need to produce 1000s of wall-sized photos a day, but for the large middle ground, it feels like technology for technology's sake. (I of course own a DSLR, being very into technology for technology's sake!)
The particle nature as well. Because light arrival is a poisson process, your pixels need to be big enough that their “backing capacitor”, so to speak, can fit at least a few tens of thousands of electrons for acceptable noise levels.
If your pixel capacitor maxes out at 100 electrons, your SNR will never exceed 10, even under optimal conditions (or, at least, you have to downscale, and then you would be better off with bigger pixels in the first place).
But the problem with 50mm prime lenses is they all seem to lack image stabilization. (The one exception I could find, a 45mm from Tamron[1], has bad chromatic aberration which really bothered me when I tried it out.) Maybe OS isn't essential in all situations, but not having it just seems like a needless downgrade for no benefit.
At the low noise levels that you get with high ISO on a D750 you really don't need stabilization w/ an f/1.8 50mm. Turn on auto-ISO and cap it it ISO 6400 or 3200. Even in low light you'll get reasonable shutter speeds at f/4.
While these can be stabilized using IBIS on mirrorless cameras, I have just that combination (Nikon Z6 + 50/1.8G) and the VR is less useful than you'd think, because being around f/2.0 already gets you to shutter-speeds where it's not really necessary, and a 50 mm can go pretty slow without stabilization anyway.
That's 1/50 for a 50mm 1/200 for a 200mm, etc.
For everything else, there's a tripod.
You can get a 50mm f/1.8 for about £90 used, you can find the 50mm f/1.4 D for a similar price too.
Minimal size and weight that still gets the job done is mostly my take on this. small specific cameras for the task at hand. I have a very similar to this, and it's very compact and gives me multiple cameras and angles. Sound quality is often more important than video quality anyway. And you can get good video quality by just using a tripod and an external mic like the rode micro.
1. g7x (mark iii) f1.8 lens and canon colors for talking head videos or pictures. Sony equivalent zv1 would be good too.
2. sony fdr3000 for walking around. Wide angle, and natural looking stabilization. or gopro.
3. dji mini drone. The only drone that is legal to fly in many places.
4. phone for everything else
5. Rode video micro (microphone)
6. Small Tripod (Ulanzi MT-16)
________
~1kg and around 1.5k to put together
I actually have more cameras but for Travel I think this is ideal. The only thing I would add is possibly a bridge camera like lumix fz82 or fz1000 for zooming.
There's a lot of perspective lost on some of this ultra high end handheld digital camera gear these days.
A $500-1000 drone can easily be worth $50k worth of ground camera for landscape photography for example because it can erase massive amounts of effort and barriers to getting a shot. The expensive handheld gear can't get the shot period and can be wasted money.
I have a DJI Mini and it's a fantastic photo tool, in it's case it weighs less than a DSLR/Mirrorless body. A Smartphone or very light DSLR Body/Lens kit + the drone can outperform a much heavier & more expensive kit for outdoor photography.
Super-experienced people backed up by a small army of people doing ray-tracing by hand.
Wikipedia might give you starting points to find modern answers: https://en.wikipedia.org/wiki/Optical_lens_design#Lens_optim... https://en.wikipedia.org/wiki/History_of_photographic_lens_d...
And if you want to get a general feel for how the modern designs work, Bill Claffs optical bench is a nice starting point: https://www.photonstophotos.net/GeneralTopics/Lenses/Optical...
(tbh it sounds a lot like the EDA space, and CAE in general)
Computer ray tracing has been used for decades. I'm not sure what the state of the art tools are however.
I also had a modern plastic 18-55 mm kit lens, high f (starting at 3.5).
But the colors on the modern zoom lens were so much better. Hard to explain, but they just looked better, kind of like a DSLR photo has "better" colors than a phone photo.
It was very confusing, I was expecting differences in sharpness maybe, but not in colors.
It really is incredible how much a lens affects the final image.
I had a Canon 40D a while ago, and at first I was using a 17-40/4. I absolutely loved the colors I would get. Gradients would have a certain "creaminess" to them.
Then I figured I'd change it for a 17-55/2.8, to get some extra light at night. That was pretty much when I stopped going out taking pictures altogether. I just wasn't getting the pictures I wanted anymore. Everything looked just... weird and unsatisfactory. I then sold everything a when I realized I hadn't touched the camera in a year.
A few years later I got to do a similar comparison again, between the Olympus 17/1.8 and the Panasonic 20/1.7 on a Pen-F. The panny was OK, sharp, etc. But there was just something missing that the oly had. A kind of delicate rendering of gradients that just appealed to me.
This was when I realized what had happened with the Canon change earlier. I had believed that I had just lost interest in photography or that for some reason I just "didn't have it" anymore. I had never thought that a lens that was rated better in practically every aspect would actually be so uninspiring to me.
Even just fixing the vignette some lenses add is enough to make pictures no longer depressing. If the lens adds chromatic aberration that's quite difficult to fix though.
I actually tend to add vignetting in post-processing, so that's likely not the cause. The issues I had were with color rendition, and in particular with subtle variations in tone.
I don't remember there being any strong chromatic aberration either. I remember that both the 17-55/2.8 and the 17-40/4 where quite highly rated at the time, with the former having a better measured quality than the latter, even on a crop sensor.
Don't get me wrong, I don't remember there being any "technical" issues with the 17-55. It was plenty sharp and aberrations were kept to a minimum, especially since I never pushed my luck with strong backlight, etc.
But maybe that's the thing, images were maybe too "clinical". I wouldn't say the pictures were "depressing", they were just... "meh". Like random snapshots. Sharp, well-exposed snapshots, but snapshots still.
---
Edit: Regarding editing, I should add that I experimented with DxO optics, which had profiles for both lenses. The corrections were quite amazing, especially at the wide end. But given my usual subject matter, I never considered the change meaningful, so it didn't bother with it in my regular workflow.
[Edit: Found https://www.lensrentals.com/blog/2011/08/lens-geneology-part..., which looks like a promising start]
Examples of flaws include swirly bokeh in the background, a glow around high-contrast edges, and flares from lights.
[1]https://lightartacademy.com/tutorials/defects-in-lenses-all-...
Phones effectively killed the small camera market, but anything on the prosumer level or higher will run circles around them.
For a lot of shots I want the specific look of ƒ/2.0 on a decent prime lens. I don't actually want an aperture wider than that.
Its interesting how much effect the founder can have on a business even after its grown large. You can tell if the company exists primarily to fulfill a mission or just as a money making exersize.