Wednesday, November 08, 2023
Painting complete. Client (me) happy as a clam. House and office nicely refreshed.
Global Shutter arrives. World radically changed within hours. Early adopters already ensconced in the Photography Hall of Fame. All previous cameras rendered useless.
Here we go. The GSO. The global shutter onslaught. Breathless bloggers and video influencers have also changed history by declaring that Sony has invented a whole new way of capturing images that no one else in the world had the foresight or technology to create.
Well.... maybe not so fast. I think a small company that designs and produces imaging sensors solved the technical stuff of G.S. well over a decade ago. For that matter Arriflex, Red, and Panasonic have been using global shutters in high end video cameras for.....a long time. Quite successfully. Sony basically fine-tuned the sensor and imaging pipeline technology in order to help you fill up memory cards much more quickly, in a consumer camera, and with many nearly identical frames. Just like video.
Don't get me wrong. I think global shutters are, in theory, a great idea. They solve some problems. But whether they are the problems we needed to solve is yet to be determined. Is a super high frame rate a good thing? Maybe, but maybe it creates a worse problem by allowing one to generate so many nearly identical files that memory cards quickly fill up, hard drives get overwhelmed and editing time extends egregiously which cuts down on time for human enjoyment.
Here's a prediction: As soon as global shutters trickle down to affordable, consumer digital cameras, and the "feature" of an "endless frame rate" arrives with them software companies will almost immediately come out with "AI" software that learns your taste in photographs and automatically winnows down your take in a folder. Effectively homogenizing your vision based on what you already did in the past. Work that falls outside your software's learned "taste" parameters gets dumped. Even if it's something you thought to try in a new way. The software will require faster processors and bigger, faster SSDs to operate. In one camera purchase you've created a new need to upgrade vital parts of your computer and your image processing.
A fact check: Loss of quantum efficiency. Global Shutter sensors can't currently take advantage of BSI technology and depend on older semiconductor tech. Current sensors used for global shutter photography are between one and two stops less efficient which equals one or two stops more noise.
And then there is the engineering workaround to delivering speed; and speed is what lies at the base of a global shutter implementation. Each file that comes off a sensor; each frame, must be processed by the camera. Hundred thousand dollar movie cameras get around the need to process their files in camera by dumping all the raw information off the sensor and directly into large scale, high speed memory. Super fast SSDs. The raw results (flat, dark, low saturation, etc.) are only made "visible" and workable in post production. Or, in movie cameras, review outputs that have their own adjustable Log profiles.
Current users of high end still cameras have been spoiled. With current frame rates from mechanical shutters cameras have time to take raw data and profile with white balance settings, contrast, saturation and other settings in order to make a pleasing and (hopefully) representative smaller file to delivery to your camera's rear screen or EVF for your immediate viewing or preview and review. The files have been profiled, your camera settings incorporated, etc. It happens to every frame that comes through your camera. Otherwise you couldn't really make much of an evaluation when reviewing the images on your camera screen.
But here's what is almost never mentioned about cameras and speed, it's the compromises between the throughput of the camera and the amount of time and processing power being lavished on each frame. In order to give you the faster frame rates that many desire something has to be streamlined. To keep costs down parts in the camera (sub processors, main processors, GPUs, etc.) have to be evaluated and compromises have to be made to meet budgets. You could have better color and lower noise out of current sensors if you were willing to pay in terms of battery use, overall cost and increased camera body size (to handle thermal issues...) but too many consumers only see the top level specs. Those are resolution, frame rate and PRICE.
Price being the ultimate "deal killer". But also the compromises required to hit a price point have a direct effect on image quality. As far as speed goes, all else being equal, the faster you pull frames through a camera's pipeline the less time the camera's working guts have to apply processing to each image. If cost and frame rate are fixed conditions for a camera (to hit the all important price point) then the maker cuts cost in the invisible to consumers areas. They can take a slower imaging processor (the unit that applies corrections and writes output data after files come off the sensor) and reduce the number and/or complexity of the operations the imaging processor undertakes. Do you wonder why very high end Phase One cameras work in the realm of 16 bits while many popular Sony full frame cameras work in the pedestrian realm of 12 bits? It's because churning out highly detailed and data dense 16 bit files requires more (and more expensive) parts and processing. It's also a reason why cameras with very high degrees of color discrimination process files at a slower speed than the one's which paint files with a broad and not very discriminating "brush."
Color discrimination is the process of breaking file's colors into finer and finer differentiations of tone and hue. The finer the color discrimination the more accurate and nuanced the color coming out of the camera. Less color discrimination and you get a less accurate representations the colors in your images. Would you personally rather have more speed or more color discrimination, the ability to work with files that have higher bit depths, and files in which noise is treated with a scalpel instead of a sledge hammer?
Granted, faster and faster processors will narrow the gap. But only if the makers of cameras earmark the speed increases for file quality over speed and throughput.
So, yes, the new camera with the global shutter from Sony is an engineering marvel but like any camera it's host to a number of compromises. Higher noise, vis-a-vis current top line 24 megapixel BSI sensors. Lower dynamic range. And almost certainly a processing pipeline that includes 12 bit processing as the default for all raw files. But you do have the nice features of being about to shoot with flash at any shutter speed and the ability to use super high shutter speeds. But with much more limited dynamic range.
Oh, and I almost forgot, the ability to fill a 256 Gigabyte memory card in the blink of an eye.
Processing. As in most computer based endeavors how processing power is used is a continual trade off between consumer candy features (speed) and real, state of the art image quality. Just sayin'
If you are heading to the Olympics next year to shoot still images for Sport Illustrated or Obscure Sports Quarterly and you are horribly unsure of your skill, timing, etc. A new camera with a global sensor might be just what the editor ordered. Hold the shutter button down long enough and you'll have something for A.I. image selectors to choose from. Every new technology that trickles down into photography has some positive use case, but that's separate from everyday value to most photographers.
I'm sure the new Sony A9III is a miracle of sorts. But every photographic miracle comes with some sort of balancing compromise. Old saying.....there's no free lunch.
Tuesday, November 07, 2023
OT: Crazy Times at the Pool. What was I thinking???
I'm happy to be swimming with a competitive team now that I'm 68 years of age. Every workout seems sweeter. Every little victory delicious. Just finishing a hard set seems more fun. More rewarding. And, interestingly, I'm not slowing down. Much.
Today we were coached by Jenn. She's our toughest coach.
The warm up was standard fare. 400 yard swim, 400 pull, some kicking sets, A thousand yards with which to loosen up the muscles, find one's rhythm and get the heart rate up. But today's main set was intense.
We did two sets of 20 x 50 yards. That's not so unusual but the intervals were. We were asked to choose a "challenging" interval for the first set of 20 ( times two lengths...) and then choose an interval that was five seconds faster per 50 for the second set. My lane mates chose 45 seconds per 50 for the first set and then 40 seconds per 50 for the second set. Mercifully, there was a break between the two sets. But a "break" at swim practice still means doing yardage. For us it meant a 300 yard dolphin kick. But that was restful compared to the main sets....
Some lanes chose to do their sets on more relaxed intervals but no lane selected faster intervals. We got into a pace we could sustain and which, for the first set, gave us about 5 or 6 seconds rest between 50s. The second set was almost touch and go. A few deep breaths between 50s. A second or two of rest.
In the end we knocked out about 3300 yards in our hour workout. But man, those were some fast yards with micro amounts of recovery. I'd feel more smug about finishing a set like that except that our lane leader today is just a bit older than I am... There is a lot to be gained by not succumbing to our culture's perspective on age and performance. We can do a lot more than most people imagine.
Now, back to the house painting adventures... (going well).
Monday, November 06, 2023
After a weekend of pulling out stacks of paper, moving filing cabinets, rolling up carpets and other prep work for house upgrades there was a lot of pent up desire to be outside with a fun camera.
One Danger in Dealing with Retirement. Or even the Expectation of retirement.
I'm a sucker for clouds. Puffy ones stuck in contrasty blue skies, especially.
A reminder that everything here at the blog might slow down from now until Wednesday. The painters are here. The computers are moved. Everything is chaos. Pretty much normal...
I have been checking in on the availability of the Fuji X100V for a year now. It is still unobtainable. Does it still exist as a viable product choice?
The Fuji X100V is a very nice camera. Well worth its retail price of $1399. Its fixed 35mm equivalent lens is a very good performer. It's lightweight and easy to carry around. The battery life is acceptable. But there is one major, overriding problem with this particular camera. It has been unavailable for over a year now.
Sure, you can go to EBay or Amazon and find a vendor who will sell you an "open box" or lightly used one for anywhere from $2,200 to $3,000. For those prices you can do better. In fact, it puts you closer to the prices on used Leica Q2s, which are substantially better picture taking devices.
I have no idea why there is so much demand and so little inventory for the X100V. It would seem that Fuji could sell every single camera they could churn out for its list price and, if the past is any guide, Fuji is capable of mass producing cameras. So what's the deal? Did they just get tired of having a wildly successful product or are they hellbent on driving medium format sales to the exclusion of everything else?
If I think the camera is a good value then why did I sell two of them? Easy. They are not built to the same standards as more expensive fixed lens compact cameras from Leica. The finders are not as good. The overall feel is plastic-y. The menus are not as well thought out. The optical finder combined with the digital finder is gimmicky and not instinctive in use. But for the vast majority of people, especially those who have never tried out a better finished product, it's a wonderful choice.
But it's not the "end all" and "be all" of compact cameras. Not by a long shot.
So, I wonder if Fuji is in the process of retooling and getting ready to launch an upgraded product that is a bit more expensive but also feels more substantial and incorporates features such as image stabilization.
Your guess is as good as mine.
As far as I'm concerned, if budget isn't an impediment, a good, used Leica Q2 is a lot more fun to shoot with. But you may have other metrics for cameras than I do.
If you ever see one offered with a true warranty, new, and at the retail price of $1399, you might consider snagging it. On the other hand that could be a speculative investment bundled with all the risk of every other bet.
Just thinking about that today as so many commenters rushed to laud the X100V as a substitute for a Leica. Maybe it is. But only if you can get your hands on one.... And you are willing to accept some compromises...
Sunday, November 05, 2023
Resurging interest in Leica rangefinder cameras? Fad or Trend? All aboard or waiting for the next big thing?
It may just be the way the algorithms present stuff to me personally but I've been noticing more and more "influencers" on YouTube, and even a few well known and long-tenured photographic blog writers, (myself included) are creating and presenting more content about Leica Q and Leica M series cameras than ever before. Some of it can be tied to Leica's recent product announcements of the Q3 and the M11 and M11-P cameras but those are hardly mainstream, mostly unaffordable by vast swaths of photo enthusiasts, and made in such limited quantities that taking possession of one is as hard as being able to afford one.
Friday, November 03, 2023
Shaking up the geography of personal existence. Thinking about the sweet spot of cameras. Belaboring the obvious. Planning the next trip.
I posted this image two days ago. For some reason readers could see the image "inline" in the post but couldn't see it in the windowed gallery. I dug down to see why. Because of the detailed grain the actual size of the Jpeg file was 25 megabytes. That's a lot for a 4000 x 6000 pixel file! But the devil is in the details.
Have you ever noticed that very grainy files can be much more detailed (in appearance) than smoother files? The requirement to resolve and represent accurately a very fine grain structure allows for much less Jpeg compression. Smooth files are a breeze. Noisy files require machines to exercise a higher level of finesse. I dropped the resolution of this file down to about 12 megapixels and, when I look at the files at 100% I can see an obvious difference in how convincing the larger files is, as a "real" photograph, than the smaller version. Something to think about as you consider the level of compression at which you save Jpeg files. No wonder some raw files can look so much better...
Painters are coming on Monday. I need to denude the studio of delicate and breakable things and also remove heavy stuff that impedes moving filing cabinets and six foot high shelving around. The painters need access to all four walls and while they will move heavy furniture they are allergic to handling things like computers, cameras, lighting gear, rare books, etc.
I started my part of the prep process by shutting down my main computer, disconnecting all the hard drives I had connected to it. Disconnecting the big, wide carriage, Canon inkjet printer and clearing out all of the power strips that had been multiplying under the desk I've been using for the past 24 years, without a break. It was an oddly emotional process since I've been sitting in that corner writing and doing image post processing for more than two decades. I got acclimated to the certainty of that spot and never expected to have an emotional response to the physical disruption.
When I woke up this morning I felt unmoored. Adrift. A bit lost. I now have my computer (with no drives attached) sitting on a mid-century desk in an unused bedroom in the house. It feels weird. There is a window in front of me and that's weird. The desk is smaller, and that's weird. I can hear B. typing away in a room down the long hall from me, paying bills online and answering email. I miss the sense of absolute privacy of action. After all, my studio/office is mine alone and I can go out at anytime and feel nicely isolated from....everything else. Working in the house I suddenly feel guilt for going to the Boston Leica Store website to take another look at a very nice condition, chrome M 240... That's a new feeling.
The painting in the office, and the rest of the house, will be completed by Wednesday evening but that seems like such a long time away. I guess I'm just remembering the cadence of work life in years past when the idea of shutting down business resources for any length of time seemed perilous.
I worry about "workspace creep." The idea that it might be nice, after restoring my "work" computer to its rightful place back in the office, if I might be tempted to enjoy still having computer in the far off bedroom for all those days when it's too cold or hot to venture into the studio. Or I become too lazy to walk the twelve extra steps from the house to the office. Maybe a nice, 16 inch MacBook Pro for the smaller desk. A complement to the 13 inch laptop I already keep on the kitchen table to catch up on stocks and news over breakfast and coffee. But I remind myself that the studio shelves already look like a laptop graveyard with older, now obsolete units stacked chronologically from Powerbook to Blueberry iBook (which was Ben's first computer ---- nostalgia alert!!!) to more recent models in a tower of eclipsed tech, a few feet high.
Interesting how, at the beginning of the career there were no computers, no cellphones, and no internet and we spent our time optimizing the dark room and the studio lighting instead. When the phone rang it rang on the office desk and if I was in the darkroom and slow to repond an answering machine recorded incoming messages for me on magnetic tape.
I'm sure I'll get everything back into order but I'm noticing, as a I clean up the studio space that it's almost like an archeological dig; layers upon layers of stuff once deemed absolutely necessary now gathering dust under newer stuff now considered...obsolete. I haven't hit "hoarder" status just yet and B. doesn't tolerate the clutter of redundant stuff in the house but still....it's a visual indicator of the waste one can generate with a changing industry.
Camera Sweet Spots. There are any number of experts on the web who can make decent arguments leading one to believe that every advancement in camera technology is a welcome, vital and necessary one. I suspected all was not temporally linear when, after having much imaging joy with "old" used Leica SL cameras (24 megapixels, non-BSI, non-stacked sensors and 2015 tech all the way) I picked up a much newer SL2 complete with, at the time, a state of the art 47+ megapixel sensor and, ostensibly, a more advanced imaging pipeline. Why then, over time and with quantities of experience, have I always liked the images I get out of the older models better? The simple answer is: The Sweet Spot. Which is also "the Ultimate Compromise Theory."
The bigger geometry of the individual pixel wells on the older camera was a sweet spot for making a certain kind of file which just happens to align with my taste in how photographic images should look. Crispy and detailed as opposed to endless resolution and a certain, cloying smoothness. I felt the same when I compared the older Nikon D700 (12 megapixels with enormous pixel wells) versus the Nikon D800 with its much higher resolution. Sure, each of the files from the more modern and higher resolution cameras could be enlarged to a much greater extent without visible stair stepping and disintegration but did they look better when viewed in normal circumstances? Sadly, no.
I recently had the same experience when I compared my recently purchased, decade old, 24 megapixel Leica M 240 with a much more recent Leica M10R camera that has 40 megapixels. The difference is not as obvious as the SL versus SL2 comparison or the Nikon comparison but it still holds true. The less pixel engorged sensor of the M 240 makes files I like the look of better than the M10R's 40 megapixels. Both are very good. And I'm sure the M10R has more dynamic range and --- maybe --- less noise but it lacks a certain crispy detail look that I can get easily from the older camera. Which pushes me to look for at least one more clean iteration of the older technology before it disappears altogether from the marketplace.
What am I talking about? Think of this example when thinking about super high resolution. You know how good an iPhone photo can look when you see it on an iPhone screen? How the electronics multisample and interpolate and "enhance" the look of the photographic images? Now, take that phone image and blow it up big. Or take an image in low light and then look at that on a big monitor. The images start to fall apart, the colors look thin. And using a shadow slider not only generates noise but almost instantly generates banding. The size of the pixel wells still matters. It's physics even if software provides at convincing image when viewed in undemanding media. Or in less rigorous sizing.
A few of my friends have purchased Leica M11 cameras and those cameras are interesting. I'd conjecture that about 30-40% of the final image quality is mostly dependent on in-camera software manipulation. Don't get me wrong. I think everyone else's high res cameras are largely beholden to software for their final appearance as well. Not just Leica. But it's a crapshoot to be too dependent on software engineering for ultimate quality. At this point anyway. And just think, the real power in software enhancement is doing it at speed. If you are doing any file processing in camera you are always weighing the quality of results against the speed of the process. Otherwise you lose the P.R. wars around operational responsiveness and frame rate. The bigger the file the faster processing you need in order to get good results. In order to just match the quality of smaller files.
Imagine instead if we took the same ultra fast processors and, instead of applying their power and speed to enormous 61+ megapixel files we applied those resources to a 12, 18 or 24 megapixel file instead. If we kept the frame rates the same we could apply four to eight times as much processing to overall image quality instead of compromising between high frame rates AND large files. How much better the images from the less crowded sensors would be done that way. Which inevitably reminds one of the engineering matrix Sony landed on when they came out with the A7S camera line, downshifting from a "standard" 24 megapixel sensor to a 12 megapixel sensor, and delivering a low noise champion in the process. But it's not just noise that can be affected; you could make different compromises that might entail things pickier traditional photographers value such as color discrimination and high bit depths.
So ----- hey Leica! Give me a 24 megapixel, new M camera with all the cool stuff like USB charging and .....USB charging. And use all that hard won software expertise not to make the sensors more densely populated but capable of delivering discernibly higher image quality for real world use.
Belaboring the obvious. I've been taking advantage of the new A.I. stuff in Lightroom and PhotoShop. I'm not using it to create wholly new photographs but rather I use it in the fine tuning of my usual photographs that I deliver to clients, or to the blog. Things like "depth blur" filters which create a 3D map of an image and can selectively blur the images from back to front. It's much more realistic for projects on which you have a separate background image (like an architectural interior) and you need to realistically composite an image (portrait of a CEO?) on top of it. Another wonderful filter which is mostly non-intrusive is the use of A.I. in noise reduction via the DeNoise filter in Lightroom. The tools are here and now and if you use them to create something you were able to visualize before but didn't have the tools to perfect then these features are more of a step forward that a complete re-imagining of photography.
The same cohort of "you kids get off my lawn" photographers who thought digital imaging would never displace analog/film photography are now grousing and posturing their discontent for anything A.I. But they are missing the distinction between "generative" A.I. and more basic enhancement A.I. which, in many forms, can be nothing more than convenience calculations aimed at allowing one to do something they already do in their workflow but with more speed and better overall results. Not a "deal-killer" by any stretch of the imagination.
So, put down your Speed Graphic cameras, change out of your Lands End shearling slippers and into some walking shoes, fire up that Buick sedan, put down your paper copy of USA Today, and takes some time to make some test new and improved test photographs, and then use them to try out some of the new tools. Life in the photography world is not completely about printing long scale landscapes on double weight printing paper in your chemical darkroom. Broadening one's horizons is a good thing. At least that's what I'm trying to convince myself in my new, temporary office.
Planning the next trip. I like to travel. The weeklong adventure a month ago in Montreal whetted my appetite for it. So now I'm trying to balance the weather, the season, the flight availabilities and whatever obligations I have locally in order to plan the next one. My friend, Andy, just got back from a couple weeks in Japan. He wrote about it and it sounded wonderful. Ben is currently hopscotching across Japan for two weeks and I can hardly wait to hear his stories and recommendations. But a silly part of my brain is bent on seeing Montreal again but this time in the dead of Winter. I'm thinking of a quick re-trip in January to see what a "brutal and savage" Winter really looks like. And how the city operates when everyone heads indoors; into the subterranean city.
And why not give it a try? The off season rates for hotels and even flights are so low for that time frame that it seems like they are paying you to come up for a chilly visit.
It's not like I need to choose only one place to go. But it's so rare I've been north in the Winter months that it's almost like science fiction to me.... Well, except for that blizzard in Toronto in 2017....
I guess I should wrap this up and head back out to the studio to keep the pre-paint process rolling. Monday morning will be here before I know it.... Oh, and there is a nice, noon swim practice available today. Might be good to store up some extra, natural vitamin D.






























