Also the desktop I have now for ML work is powerful enough to do a full 3d simulation of a lens using the full Maxwell equations from first principles. I remember doing the back of the envelope calculations that I'd need the Bluegene/L back in my undergrad days to do it for real, well: https://bnnbreaking.com/arts/video-gaming/nvidia-geforce-rtx...
What a time to be alive.
Thanks for encouraging me to search about it, and quickly realize Bolte Bridge is a natural target due to relative level of architectural sophistication as well as being easily accessible.
https://en.wikipedia.org/wiki/Bolte_Bridge#/media/File%3ABol...
Takes high ISO settings to get something out of that, but it works and is dead simple.
When I borrow my dads camera for the odd shot at times, I totally relly on AF, if I went by the view finder I am almost blind... Same thing the other way round!
"Fingerprint coated" haha !
"...this lens fails to deliver persuasive arguments in nearly all situations..."
LOL
after you had taken a pic, you had to rush into the darkroom, develop the paper and then reverse print it (cannot remember how, or even if i did).
all a bit weird, but it kept me amused back then. i haven't been involved in photography for nearly 40 years.
https://www.timhunkin.com/a198_goodbye-cibachrome.htm
Sadly, this marvellous paper - direct positive colour and easy to develop - is no more.
Direct link to the video since (for me, right now) the embed in the above doesn't work. https://youtu.be/5AOlPuTQt-M?si=q7RibENicPH0Be9m
It’s a great way both to explain the whole premise behind photography (your fancy camera and lens is just an elaborate way to project an image onto a bit of light-sensitive stuff that you otherwise keep in the dark) and to give students some early hands-on experience developing prints without all the intervening steps involved in making a print from film.
It's a pretty common trick photographers and videographers use to get a sort of dream-sequence effect.
I did learn that they must be used judiciously. After shooting most of a personal music video with a Sweet 80 wide open, my wife described it as "like watching an ocular migraine".
I don't know anything about photography. Is using specialized lenses like the ones you mentioned better than postprocessing software?
Edit: CARIBOU - Hello Hammerheads https://www.youtube.com/watch?v=2N_VvfPX1Bc
...
With mere hours of experience in the art and science of optical design, the team at SUPERCHROMAT remain novices in the field. As such, this prime lens, with 6 elements over 4 groups, provides inferior optical performance at a price affordable to few.
Regardless of whether it’s a matter of selective focus in the close-up range, high-contrast available light applications or landscape shots with immense depth of field, this lens fails to deliver persuasive arguments in nearly all situations unless stopped down to f/22.
2X 200mm biconvex lenses BK7. Uncoated.
-700 / 075 X 3 used eyeglass lens 1.67 / Vd 42. Fingerprint coated
+30mm aspheric achromatic triplet LAK14/SF57/Aspheric polymer. VIS 0º coated
Because that's basically what happened in his last 2 abominable movies. He found some really unsuitable thrift lenses and thought it was a master stroke of genius to use them in a professional movie. That's why virtually every single frame is horribly blurred except at the exact center.
Luckily these lenses are pretty cheap (The USSR made many of them), I purchased two in late 2022 that shipped from Ukraine
A video about it: https://www.youtube.com/watch?v=lYvWpavSXeE
YT, voice copying demo: https://www.youtube.com/watch?v=TNwuAjeDOVE
arXiv, voice cloning detection: https://arxiv.org/pdf/2308.12734.pdf
kaggle, Deep-Voice (support files): https://www.kaggle.com/datasets/birdy654/deep-voice-deepfake...
Has voice cloning using Retrieval-based Voice Conversion and then detection for: Ryan Gosling, Joe Biden, Elon Musk, Barack Obama, Margot Robbie, Linus Sebastian, Taylor Swift, and Donald Trump.
Apparently overly-optimistic detection results reported based on discussion, although they're still not bad at 86% after removing possible training data issues.
https://camerapedia.fandom.com/wiki/Radioactive_lenses
I own a Super-Multi-Coated Takumar 50mm f/1.4 and can confirm that the glass browns with age. UV exposure (sunlight) does clear it some.
The so bad it's actually good :) Lovely song, too
Improvements are mainly in terms of coating (reflections, ghosts and such) as well as zoom ranges and auto focus systems. That vintage lenses are not sharp is simply not true, having a lot less of glass in a lense is actually an advantage.
If you talk about lab test numbers, especially around the corners of the frame, modern lenses sure beat vintage ones. Not that you would realize any of that in real life (art replication, detailed macro work and other specialized stuff nonwithstanding).
My universe broke when Sigma introduced an 18-35mm f/1.8 zoom. An f/1.8 zoom. Wow. And it was optically brilliant.
Seventies lenses are super-sharp, but that's because they're mostly slow primes. Any modern prime stepped down to f/5.6 -- even cheap consumerific ones -- will be super-sharp.
There's nothing in seventies technology which allows lenses to have the aperture, zoom range, and aberrations of modern ones.
I have a very acclaimed old design 50/1.5 lens (but was bought new), but it just sucks compared with a modern much cheaper 50-70/3.5. The colors in particular, they are just bad. I'm not sure what test would pick that, a color accuracy kind of test. Modern coatings truly do wonders.
But lenses can do things that postprocessing can't do, strictly speaking (though it can often emulate well enough to fool most people).
The input to a lens is a field of photons that are:
1. Striking the sensor at different places. In other words, which pixel they hit.
2. Coming into the sensor at different directions.
3. Coming in at different wavelengths.
The 2D image captured by a sensor completely discards 2 and it drops a significant amount of information from 3 on the floor by bucketing all wavelengths into red, green, and blue amplitudes designed to mostly mimic human visual perception.
Lenses have access to the whole shebang.
https://news.ycombinator.com/item?id=24196650
So, maybe it wasn't "kept" but did it at least once, so now it's ingrained as a thing to poke fun.
https://wandel.ca/homepage/yashicamat.html
https://wandel.ca/homepage/wetcamera.html
TLDR: The first one is about using a 20-year-past-expiry film in a format I didn't otherwise use; the second is about developing a drenched E6 slide film in b&w chemistry and trying to get prints.
The answer was the worlds fastest super computer at the time. I was a bit shocked that he claimed that was the state of the art for lens manufacture in industry too - no idea how true that was. But figured that if you needed that much computation it made sense.
Well I have 2x that under my desk now and I use it to do local development before I push to cloud machines with 100x the power where the actual work happens.
Not entirely surprising. You want to be able to run stuff a bunch of times to optimize it.
https://www.synopsys.com/content/dam/synopsys/optical-soluti...
What those old lenses have so, is build quality. They are machnical master pieces, as oppossed to modern day plastics. I like that. Also, close to no electronics that can fail.
Those wide zoom ranges, and large appertures, do have other downsides so. Everything is a trade off, and some things are sacrificed to achieve a 18-35/1.8 lense. Still impressive. Or the latest Canon (?) patent on a tilt-shift-macro-zoom...
It's possible due to things like:
- Much better lens coatings (allowing for more elements without flare)
- Computer modelling
- Being able to better machine aspherical elements
On integrated cameras, like an RX100, we can do even better due to digital. We don't need low-distortion, but just well-characterized distortion which a computer can invert.
Those old lenses -- at least the ones still working -- do have build quality. Adjust their prices at the time for inflation, and have a look at modern Zeiss, Leica, and other premium lenses. You'll see similar build quality.
For fun, next time you're looking at old lenses, pick up a Quantaray too, for fun. Or most older (before around 2000) Sigma lenses. Or other off-brands. You'll see unparalleled build quality, but in the other direction -- eighties-era plastic. The optical quality is bad. Not "fun" bad, "vintage" bad, or "whimsical" bad, but just bad.
When we look at older stuff, we tend to look at stuff which stuck around, but even that Quantaray is high-end compared to an older consumer camera. Most people who owned a camera had a non-SLR one. You looked through a little window on top, and the pictures were shot through the main lens. Those were even worse. Or a disposable (a single plastic element lens).
f/2.7, sharp at fully open, parfocal, zero focus breathing, fully manual zoom lenses...
and what used to be...
f/5.6, sharp only when stopped down even further, massive focus breathing, prime lenses...
well, vintage lenses start to look like toys.
Of course, if it's a F/1.8
Roughly the same optical design, limited to f/2.8, will be better in all respects.
A good way to think about this -- oversimplified obviously -- is you have a bunch of functions you're trying to cancel with e.g. a linear combination. If you look really close to the center, it's already a flat line. If you step out a little, two are adequate (have the slopes cancel). As you go further and further out, things become increasingly wonky. That's why you have super-complex designs for an f/1.8 zoom -- to get that cancellation right -- but even a single element works fine at f/22.
There actually were f/1.8 zooms all along, but for applications which didn't demand that sharpness (TV and CCTV). You can pick those up cheap and see what happens. They're sometimes fun on a real camera too (many will span a μ4/3 sensor with just a tad of vignetting).
Color rendition is also impacted by the sensor, assuming digital cameras.
Since I don't know which lenses you talk about, hard to tell. I do have some really old ones, 80-200 f 4.5 from the late 70s and an equally old 300 f4.5. Both render color just fine, no difference between those and new Nikkor lenses. Sharpness wise, those old ones are easily as sharp as any new one, lab test confirm that. And the limited amount of glass gives them, a totally subjective, clarity new lenses don't have. Bot that I would d be able to tell just from looking at a printed or processed picture.
Particularly since the old lens was designed for film, perhaps even black and white film. The choice of film has a much larger impact on color rendition than the lens would have. Also, unlike sharpness, color rendition is highly subjective and easily corrected. If you're shooting to JPG and you don't like how the camera is interpreting the colors from a lens, most cameras allow you to customize the white balance.
If you take a 100% crop from a corner so, well, that is different. I'd argue so, that in this case, you should have composed your shot differently in the field. And again, you have to print huge to actually see the difference.
I'm primarily doing videography, but most of the vintage lenses I've tested are too soft to crop a 1080p vertical "short"-style video out of a 4K landscape video (even though the landscape video may look fine).
It's not just my old konica minolta lenses, though, the more affordable Sony lenses from 2015 are just as bad.
I'm now using contemporary fujinon cine lenses and they're tack sharp all the way to the edge, even fully open at T/2.8
That said: If you've got hidden gems, or like the less clinical look of vintage lenses, go ahead, have fun!
And yes, not all vintage lenses are good. One has to go look for the good ones, usually the top line pro lenses of the day, those are still very good.
The one, cardinal thing about gear is so, once it hits a certain threshold of performance, e.g. "good" 4k video, whatever makes you feel good is great gear. Doesn't have to be the newest and shiniest, nor a specific brand or something.
My vintage glass so is neither soft nor sub-par, for my purposes, optically. Well, they do have a certain clarity to them, especially for blavk and white, that I like. I also know it is in my head, as I actually cannot tell those lenses apart from new glass by just looking at the pictures. I do like the incredible mechanical build quality so, they are master pieces of precision mechanics in all metal, and I like that feeling using them.
While not always true, there is a connection between price and quality: cheap / affordable products are generally less good than expensive ones, at the same time there can be incredibly good cheap products and crappy expensive ones, also the expensive ones can be bad value by virtue of being overpriced.
I heard good things about Fujinon, never tried them so.