Gray Fox Images Daily Musings

When it comes to selecting camera gear, one must choose from imperfect alternatives -- no brand is perfect in every feature and metric. From a strictly personal use case perspective Canon is my choice and the R5 Mk II (and R5 before it) is the preferred camera. Is this exactly what I want? No, not at all. My personal preference would be a 45MP version of the integrated grip R1. For my favorite genre, bird photography, I would pair it with a lens similar to one of Nikon’s light weight hand holdable long telephotos.

Why Canon vs Sony or Nikon? It comes down to ergonomics and lens selection. On balance I like Canon’s handling, menus and buttonology best, perhaps due in part to long time familiarity. (This is, of course, purely personal preference, and others see things differently.) For lenses Canon has an excellent selection of tilt-shifts and some remarkable light weight but optically excellent zooms ranging in aggregate from 10mm to 200mm, each well suited for travel and walkaround outings. And, the 100-500mm is an ideal mid-range zoom that can double as a moderate closeup lens.

RE long telephotos, at my age a high optical quality hand-holdable lens (4.5lb or less) capable of reaching at least 1000mm with extenders, and preferably 1200mm, would be ideal -- the Nikon 3.24lb Z600mm f6.3 in a Canon RF mount would be an instant buy. Zooms are nice but that inevitably adds weight, Therefore, I’ll deal with extenders in order to have optical quality and hand holdability. Naturally, low cost is a factor, but lenses meeting these criteria are inevitably expensive. So be it.

Anyone who finds exactly what they want is fortunate. For me, and I suspect for many, one is left to evaluate the pros and cons of alternatives and choose the most favorable compromise that serves one's photographic goals acceptably.

Bracketing for HDR blending is a standard technique for capturing scenes that exceed the dynamic range of a camera. There are, of course, several ways to set up a bracket, including step size and number of steps as well as metering approach, the latter of which we’ve discussed here. Herein we describe bracketing setup peculiarities of three Canon cameras, 5D Mk IV, R5 and R5 Mk II. Throughout, our preferred approach is based on full manual control of the exposure triangle, aperture, shutter speed and ISO.

While it is possible to vary any of the three above parameters to achieve a bracket, typically one varies shutter speed, keeping aperture and ISO constant. Why? First, merging three different apertures, with different depths of fields, is non-sensical. Likewise, varying ISO means introducing noise variations in the blended image.

One more setup parameter is relevant to this discussion, the step size for adjusting exposure, whether 1/3 or 1/2 half stop intervals. I prefer 1/3 stop intervals, as do, I suspect, most photograhers. Given the above, how do each of the three cameras above approach the task for full manual exposure. Unfortunately, there is no consistency across the three.

Sadly, only the R5 gets it right. The 5D Mk IV can be made to work but only with 1/2 step adjustment intervals. With the R5 Mk II, it simply cannot be done in full manual. One has to wonder if the developer who made this choice ever actually used the camera.

One of Canon's more problematic features is Eye Control, a focus aid wherein the camera reads where the photographer's eye is looking. This was introduced in the film era, my first opportunity to try it out being the EOS 3. The 3's AF system had only 45 points, centrally located, and as a result this worked fairly well for me -- for a while. But, when I tried to recalibrate one day it never quite worked so well again. As a result, I stopped using it -- with some regret because when effective it was a great feature.

Fast forward to the EOS R3 and the reintroduction of Eye Control. It worked for some but not for others. The feature, supposedly improved, made its way into the R5 Mk II and will also be in the announced R1 when delivered. Does this new and improved version work for me? The answer is, kinda, sorta, maybe. Why the waffling? Read on.

First, a bit about my eyes. At a young(er) age my eyes required glasses but were otherwise quite good. With a telescope I could resolve double stars to the Dawes limit -- i.e. I could make out the brightness minimum between the pair's two overlapping stars. But, time has a way of intervening, and the effects all combine to reduce the effectiveness of eye control. As time passed, I noticed that my perscription seemed to change from one day to the next -- and even during the day. Then, cataracts came along, fixed (sorta) with implants. But, that didn't fix the changing perscription problem. Worse, vision was stll blurry at times.

Finally, a truly excellent young optomologist diagnosed the problem -- a defect in the glands that produce lubrication for the eyes. Not to be confused with tear ducts, which are also a problem, leading to dry eye. The result is semi-annual treatments for the lubrication glands involving zapping the area around the eyes with high intensity light pulses -- imagine your face being stung repeatedly by bees! And for the dry eyes, drops every day. None of this, of course, does anything to cure the problem of the eyes changing shape slightly during the day. We simply adapt and go on.

Thus, it came as no surprise that the R5 Mk II's Eye Control doesn't work perfectly for me. But, does it have to? The answer is that there are situations where it can be used effectively. When there are a few subjects in the viewfinder and they are not very close together, eye control can often place focus near enough to the desired target for subject tracking to take over. I tried this out in the field for the first time in a session photographing Canada Geese on the Rappahannock River. While not perfect, it nevertheless helped prevent the AF system from insisting on finding the wrong target -- in those cases I would otherwise have had to switch to individual focus point mode and move the focus point by joystick to the right target. Eye control is much faster.

Does it always work? Sadly, not quite. But, it works frequently enough to be worth the effort.

You would have thought Canon had learned its lesson after the overheating brouhaha surrounding introduction of the original EOS R5. But no, the new and improved R5 Mk II also shows signs of thermal runaway -- despite the addition of flow-thru heat exhaust vents to the camera's base plate and non-grip side. At present, evidence is limited to one camera sample, but sadly that one happens to be mine. Maybe it's a one-off or maybe it's a harbinger of things to come -- time will tell. Meanwhile, here's the summary of our new hot off the presses -- and just plain hot -- camera.

We received our R5 Mk II on August 20, the first day of availability, and after a couple of days of setup we took it outside to try the new features, notably 30fps and pre-continuous shooting, i.e. continuous capture of .5 sec of images with a half-press of the shutter, to be written to card as soon as shutter is pushed down all the way. It was a hot summer afternoon with full sun for a couple of hours starting some time after 2 pm. Long story short, at about 4 pm the camera suddenly displayed an overheat warning and shut down completely. Dead iron.

Amid much angst, this led to the inevitable speculation as to causes. First, we observed that our add-on BG-R20 battery grip obscured the input vents built into the camera baseplate, hence no air circulation. Second, we were using pre-continuous shooting on an intermittent basis, the biggest problem being unlearning the long ingrained habit of letting up on the shutter once a bird flies -- the exact opposite of what is required to successfully capture the bird's launch with the pre-continuous feature. And finally, it was, after all, a hot August afternoon in full sun.

After reporting the problem to CPS (Canon Professional Services), we resolved to gather data a bit more systematically using the viewfinder temperature display as a metric. The R5 Mk II user manual explains that once internal temperature starts to rise a thermometer symbol appears with up to 10 bars indicating heat level. The first seven are white and last three are red. Once the red tenth bar is displayed shutdown is imminent. With this info, we set out to identify and isolate the various factors involved.

At 4 pm, full sun, ambient temperature 94 degrees, we set up on a bird bath, hoping for some action. Within five minutes of doing nothing but focusing on the bird bath and adjusting settings, and with pre-continuous not engaged, the viewfinder heat symbol came on, initially showing four bars but steadily rising. By 10 minutes from camera turn on, the display was well into the red zone, making it perilously close to shutdown. At this point, we turned the camera off to contemplate our next move.

One theory was that the battery grip's occlusion of the base plate intake vent was the culprit. So, we removed the grip and tried again. Since the camera was already hot, we immediately dipped into the red, thereafter dropping to borderline red/white but holding steady -- hot but cooler than before. That seemed to indicate that the battery grip played a role, although not necessarily a large one. Next, we engaged pre-continuous and half depressed the shutter button and held it there for five minutes (!) without fully depressing and writing data to CFexpress card. To our surprise, this did not cause an additional rise in internal heat.

Finally we replaced the grip and tried the same five minute pre-continuous experiment. Unexpectedly, this caused only a small internal rise, flirting with red but dropping back as the sun sank lower in the sky. About that time, roughly 5pm, the sun went behind a cloud and the temperature dropped two degrees. Two degrees isn't much, but it felt considerably cooler with full direct sun now removed. The camera seemed to agree, with the heat display dropping to five bars and remaining there for five minutes, once again with shutter half-pressed and pre-continuous engaged throughout.

What are the results from this little admittedly unscientific experiment? First, the BG-R20 battery grip appears to be a factor, but not the dominant one. Second, engaging pre-continuous mode without actually writing to a card seems to have at most a minor effect. Finally, this first semi-controlled experiment seems to suggest that high ambient temperature, and especially full direct sunlight is a problem. The operating range of this camera is 32 to 104 degrees F. We were within ten degrees of that value and came perilously close to shutdown without any image capture and without pre-continuous engaged -- albeit with battery grip installed and bottom vent occluded.

It is difficult to generalize at this point. Interim conclusions might be summarized as below, with more controlled testing and information from other camera users needed for confirmation

Perhaps the most disappointing aspect of this discovery is that in four years of shooting with the original R5 I never once had an overheating situation despite photographing in similar circumstances many times. It would appear that Canon's attempt to shoehorn a 45MP stacked BSI sensor into the R5 form factor may have had deleterious side effects on thermal management.

Pocketable compact cameras are disappearing in favor of smartphones, some of which now sport three cameras. But, one suspects that there's still a market for compacts at the upper end. Enter the Canon G7X Mk III Powershot. Not convinced that those tiny smartphone sensors were as good as the 1" sensors in the

better compacts, I tried the highly regarded Sony RX100 VII. After a disasterous attempt to deal with Sony's menu system and minuscule buttons, I gave up and purchased the G7X III at a considerable loss, discussed in the entry on Trashy Ergonomics.

But, this still didn't quite address the key question, namely how well does the G7X III compare to a full frame 35mm mirrorless camera. To answer that question, I took photographs of a house with an R5 and the G7X III, exposed identically and composed as similarly as possible. The results were imported into Capture One and each image was exported as printable JPEGs in two sizes, 6.67x10 and 12x18 inches. Crops were then made of a segment of a window with lacy curtains, rich in fine detail.

The conclusion is that for 6.67x10 prints (my normal size) the G7X III holds up fairly well, albeit less crisp and detailed at the pixel level. An excursion to Topaz Photo for sharpening nevertheless confirmed its viability for modest prints, bringing it nearly to R5 6.67 x 10 inch crispness level. There are disadvantages, of course. Color character for this rather flat scene is acceptable, albeit different since the G7X III sensor is a Sony product. For a scene with a greater brightness range the R5 rendering might diverge more based on a significant different in dynamic range between the two sensors. (G7X III dynamic range not available; G7X Mk II DR shown.) The G7X III also requires substantial distortion correction, which can warp shapes, e.g. circles may become ovoids. Rectangular shapes characteristic of architecture, however, seem to survive the adjustment.

Beyond modest enlargements, the G7X III rapidly loses out. The G7X III 12 x 18 inch crop reveals a visible image quality shortfall. We lack a current smartphone for comparison, so comparable output from the latest is yet to be evaluated.

The long awaited Canon R1 and R5 MkII cameras were announced on July 17. At 24MP and as a bird, nature and architecture photographer the R1 is of no interest personally. The R5 MkII, on the other hand, delivers many new capabilities, capabilities that fill a gap in the feature list of the otherwise ground breaking original R5, the industry's first unicorn camera at 45MP and 20fps. A full discussion of features (and a few shortfalls), how it compares to the original R5 and how it fits our use cases may be found here. We'll follow up with a full discussion once we have the beast in our hands.

The table below encapsulates what is currently rumored for Canon's upcoming R5 II as it compares to the R5 and as it relates to stills photography.

R5	R5 II	Advantage R5 II
45mp FSI sensor	45mp BSI Stacked Sensor	Low readout time for action
Standard EV	New EVF design *	Possible right angle finder
Standard AF	Eye Controlled AF from the EOS R3	Excellent if it works
12fps Mechanical & 20fps Electronic max burst rates	12fps Mechanical & 30fps Electronic max burst rates	50% higher frame rate
12-bit files in Electronic shutter	Unknown	14-bit would improve dynamic range
No cooling	New passive cooling design with two new vents	Good for video, possible benefit for stills
No active cooling	Active cooling grip accessory	Good for video
Battery grip accessory	Battery grip accessory	Needed for verticals
CFexpress Type B / SD card configuration	CFexpress Type B / SD card configuration	No change
Launch Price: $3899 USD	Launch Price: $3999 USD	Minimal price increase

If they'll also announce a pro quality (AKA "big white") RF500mm f4.5 or RF600mm f5.6/6.3 our birthday, anniversary and Christmas wish lists will be complete -- and our bank balance will be in the basement!

* Combining a right angle finder with eye controlled AF seems a bridge too far. However, we can hope. . .

Much angst accompanied the introduction of Canon’s 24mp R3. Why not more megapixels? Was it supposed to be the mythical flagship R1, only to be renamed at the last moment after finding out what Sony (50mp A1) and Nikon (45mp Z9) were doing with their flagships? Was Canon deliberately pounding the megapixels out of the R3 with its infamous “cripple hammer”? Or, as Canon implied but could not admit explicitly, did they not have the technology in hand at the time to produce a mirrorless camera worthy of the R1 moniker? FWIW, we’re of the latter opinion – more on that in a moment.

Now, with the rumors surrounding a presumptive July 17 introduction (finally!) of the R1 unicorn, said to be yet another 24mp camera, disappointment once again roils the Canon waters. Why bother? Isn’t this just an R3 Mk II?

The main problem is that Canon has to carry out very complex research and development for its large sensors (mainly APS-C and full format), while Sony can do this on very small sensors for smartphones. The latter is easier, cheaper and quicker to do. Sony then simply adapts this for large sensors. However, Canon not only lacks the small sensor sector for testing, but also the huge economies of scale and profits from that sector. Therefore, I think it is impossible that Canon - efficiency improvement programs / cost-cutting programs notwithstanding - will ever be able to achieve the same efficiency in sensor development as Sony. This in turn means that Canon ultimately has to spend significantly more on R&D for the same image quality. This is a strategic disadvantage.

Although Canon pioneered the use of front-side illuminated CMOS circuitry, they then sat on their backsides polishing the apple endlessly with 500nm circuitry while Sony roared past, fueled by their overwhelming presence in the smartphone camera sensor market space. First, there was backside illiminated sensors, a necessary precursor to stacked (multi-layer) sensors. Then, Sony put the ADCs on-chip, eliminating one noise source. There soon followed the Sony stacked sensor Exmor line. And now, Sony has scooped the field with the first full frame global sensor camera, the Sony A9 III. Canon meanwhile has consistently lagged behind in the evolution to smaller and smaller circuit feature size, remaining at 500nm for far too long, thus limiting what could be done.

The R3, then, was probably a belated testbed for Canon's adoption of stacked sensor architectures -- just as the R was a testbed for full-frame mirrorless technology. Going further back in time, the EOS 3 film camera preceeded the EOS 1v, Canon's ultimate professional film camera. In essence, Canon's explanation of "why the R3?" was honest -- reading between the lines they were admitting that they didn't have the stacked sensor technology needed to produce what Sony and Nikon (who uses Sony sensors) could bring to the table. Once again, because they engaged in short-sighted profit taking rather than continuing to push the envelope in sensor design and fabrication.

Observed on Internet in response to criticism of the ergonomics of Sony's A1 flagship camera:

"Trashy ergonomics? What are we talking about here? The A1 generation body and even more so the A9III generation body have ergonomics on par with an R5 but with a significantly better control layout. . . I don't think the ergonomics of an A1 are trashy in comparison. I've owned them all and can only speak from experience related to my own hand size but the control layout on Sony is a cut above the rest." [Emphasis added]

Ah, but isn't the payload contained in that last sentence. When it comes to ergonomics, aren't "trashy" and "cut above" personal to the individual? The A1 floats this responder's boat. But, for me the R5 is "hands down" the easiest to use camera I've ever owned. Admittedly I’ve never even held the Sonys or the Nikons in question. But, I know how I like to work, and those bodies run counter to that. [Emphasis added]

The Nikon Z9 is wallpapered with buttons – 29 according to long time Nikon commentator Thom Hogan. If one's inclination runs to buttontopia then the Z9 may well be the best camera ever. The Sony A1 has almost as many function/mode selection dials as the Z9 has buttons (exaggeration, of course). For those who like spinning all those dials to get into the desired configuration it's no surprise that some would think it the best camera ever.

Personally, the programmability of Canon’s M-Fn button, situated next to the shutter button, and the inclusion of three custom modes C1, C2 and C3 in addition to the normal or non-custom mode (call it C0) makes the R5 instantly responsive. A quick M-Fn push sequence puts all relevant camera settings into the configuration I want. The three top buttons across the back, set specific to each custom mode, facilitate changes during operation, e.g. from single point AF to eye AF, one shot to servo and single exposure to continuous. Together the three camera dials plus lens ring set the exposure triangle plus ambient and flash exposure compensation. The R5 isn't perfect but the shortfalls are mostly minor. But, again, that’s me. Others have different ways of working that suit them better.

Regarding menus, my only Sony experience is with an RX100 VII -- proclaimed by some to be the best compact on the market. The menus and buttonology were so arcane, baroque and frustrating that I eventually sold it to the lowest bidder just to be rid of the damnable excrescence. Turns out the Canon G7X III that replaced it (wanted a G5X II but it had been discontinued) can be set up very similarly to the R5, albeit with far fewer options. Within an hour I had it up and running with the familarity of an old shoe.

To repeat, this is all personal to the user. For me, Nikon and Sony products appear to be less user friendly than the R5. Others will see it just the opposite. Bottom line, pick what you like, realizing that no product will please everyone.

The Canon RF24-105mm f4L IS lens is advertised to provide up to 8 stops of stabilization when combined with a supported R-mount body such as the 45MP R5. But, that spec is based on a CIPA test that does not always reflect real world use. Nor does it account for individual hand holding ability, especially after a couple of cups of morning coffee!

Since I hand hold a lot, a test seemed advisable. By way of background, the old minimum shutter speed rule, 1/f, where f is the lens focal length, was developed for film. With high megapixel cameras such as the R5, a faster shutter speed is advisable. I suspect that the same is even more so with longer focal lengths, but that isn't the subject of this missive.

We did conduct such a test with the RF24-105mm f4L IS lens mounted on the R5. Two focal lengths were tested, 24mm and 105mm, one exposure for each condition. Six shutter speeds were evaluated, 1/100, 1/50, 1/25, 1/12, 1/6, and 1/3. Four conditions were evaluated, single shot vs servo modes and EFCS vs electronic shutter. The USAF 1951 resolution chart was used. The results are shown below, center target block only. The resolution numbers are taken from group 1.x and are directly comparable to the resolution test results described in the immediately preceeding article below. Bold numbers indicate relatively sharp images, regular text indicates visible shake blur rendering the image unusable. Variation from expected progression relative to shutter speed is undoubtedly due to personal variation combined with single test image capture at each point.

It seems clear that the 8 stop CIPA rating is wholly unrelated to my own (rather disappointing) personal ability. This is particularly relevant where bracketing for HDR may lead to slow shutter speeds to capture deep shadows, thereby forcing a choice between possible blur and higher than preferred ISO values. Finally, resolving tiny test chart lines is perhaps a much more stressing case than photographing many scenes -- i.e. lower shutter speeds may in fact be perfectly useable depending on subject matter.

With the arrival of Canon's new R5 Mk II, we decided to repeat the image stabilization test under conditions more informal and realistic than test charts. The prompting event was a visit to historic Scotchtown, an 18th century home that features very low indoor light levels, often illuminated by only a window and a single incandescent candle bulb. Despite relatively slow shutter speeds, many images were recorded at ISO 6400 (the limit of our increased noise and reduced dynamic range tolerance) but were still badly underexposed. Raising exposure in post is fraught with peril, including increased noise, low dynamic range and possible color shifts.

The goal therefore was to delve into the potential for "useably sharp" image quality -- as opposed to test chart/Nyquist limit sharp -- i.e. sharp enough at 100% viewing to be suitable for prints up to, e.g. 12"x18". The subject was an art portrait, indoors in dim early morning light, not so different than which might be encountered in many of the historical landmarks we visit. The lens was an RF70-200 f4L IS zoom, notable for light weight and superb image quality. The original R5 IBIS was rated at up to 7.5 stops, depending on image stabilized lens. The R5 Mk II, the camera we used, increases that to a rated 8.5 stops. Note that these specs are often difficult to match in real world shooting. With that in mind, here were the test image settings.

Images were metered for the whole scene, accounting for the ISO discrepancy between the last two test images. The bottom line is that all were printably sharp viewed at 100%, although results varied very slightly, emphasizing the need for careful hand-holding technique. Using the old shutter speed = 1/lens focal length (which should be doubled for high megapixel cameras such as the R5 Mk II), this suggests 5 stops of useful combined IS/IBIS, at least for this lens/camera combination -- a far more palatable answer than the test chart results initially reported above.

. . .or the outcome may impact the bottom line. Case in point, Canon's 24-105mm f4 zoom range -- and the rather mediocre loaner I received from Canon to test the latest version of same. In 2006 I purchased an EF24-105mm f4L IS, original version. The version II came and went without a quality improvement worthy of an upgrade. Then, mirrorless happened, which led to an RF24-105mm f4L IS alternative. Was this time to upgrade?

Difficult to say. MTFs couldn't be compared since Canon started including diffraction effects in their MTFs (a good thing, and the only one of the big three to do so), but the older lens MTFs were never redone. Online tests didn't help differentiate the alternatives sufficiently; maybe the RF was a bit better here and maybe a little worse there.

Looking for a final answer, I got a loaner RF from CPS and tested it against my beloved (if flawed) EF copy. Unfortunately, neither lens performed particularly well, especially in the corners at several focal lengths. In fact, the CPS loaner was borderline awful. So, the idea of a move to RF went on the back shelf, languishing there for a couple of years.

Then, I began replacing all the old EF zooms with RF equivalents, eventually leaving the EF24-105mm as the only survivor. Finally, I decided that it was time to gather courage (and credit card funds) and take the plunge. The new RF version arrived and was tested at 6 focal lengths and 5 apertures. Wonder of wonders, it bested both the old EF version and the CPS loaner, the failings of the latter leading to the title of this segment.

The RF version is not perfect, of course; at its price point it can't be. It is, on balance however, significantly better across the board at f4 than either the EF or the RF loaner, ample reason for the upgrade. At f8 it's more nearly a wash, with the EF having a slight edge at some focal lengths. Performance for both lenses tails off somewhat at 105mm, consistent with online lens tests. Remarkably, at 50mm and below the RF at f4 is the equal of the lens at f8. Check out the comparison here. (Tested with USAF 1951 Resolving Power Chart.)

As an added bonus color rendering is excellent: vivid, contrasty and highly saturated -- much like the RF70-200mm f4L IS and the RF100-500mm zooms. One suspects that Capture One's RF color profiles are a good deal better than those of the old EF. More recent coating technology can't have hurt either.

But for lack of a good loaner copy, this purchase was delayed a long time and might never have been made at all -- a cautionary tale for how companies represent themselves.

A previous long winded writeup described a process for metering a scene for HDR bracketing -- much too long and cumbersome as it turns out. Here's the simplified version. It makes use of mirrorless camera histograms, evaluative metering rather than spot metering and an exposure calculator from HDRsoft (web base or smartphone app).

The logic for evaluative rather than spot metering is that the histogram tells all. With wide angle lenses especially, the spot may cover too large an area for reliable results.

Anyone coming across these musings will quickly pick up on a generally negative view of Canon's current product lineup. It's a fair perception. A good summary view can be garnered in three recent offerings, Canon is doomed, lens value and camera value.

We are not the only one with such opinions. Thom Hogan, a long time Nikon commentator and observer of the camera industry in general, posted the following nuggets in a News/Views column opinion piece titled Product Envy dated 11 May 2024, wherein he took a look at what each camera maker has in their product lineup that other rmanufacturers might envy. His comments RE Canon are enlightening to say the least. (Quotes reprinted here under fair use in support of my own views.)

There is still much to like about Canon products -- e.g. the EOS R5 body, their line of tilt-shift lenses, their light weight and optically excellent f4 RF zooms, the RF100-500mm zoom and others -- but the above comments certainly ring true for me, especially as a wildlife and bird photographer. If forum complaints are to be believed, landscapers aren't too thrilled either.

As the two "value" articles referenced above indicate, Nikon is miles advanced in high end camera bodies (Z8 and Z9) and hand holdable telephotos too numerous to mention. Meanwhile, Sony continues to stay ahead of the crowd in sensor innovation (e.g. the a9III global shutter.) Where does that leave Canon, you ask? The answer is easy: with "no products that other camera companies envy."

Having completely failed to adapt to the user interface of a Sony RX100 VII, the cursed thing was sold at a 50% loss just to be rid of it. It was replaced by a Canon G7X III, the slightly less capable sibling of the now discontinued G5X II that I really wanted – sadly they were out of stock by the time I started looking. The RX100 was a mistake, pure and simple, from its featureless slab-sided body design and tiny buttons to its complex and baroque menus. Certainly the Sony brand sells well, but this camera was simply too different from the well thought out Canon bodies I’m used to.

The G7X was immediately familiar, from touch to controls to menus. Operation is simple and straightforward, albeit far less capable than a full frame body. Rather than the Sony’s smooth painted metal outer shell, the G7X is covered with an easy-to-grip dimpled rubberized coating -- in fact, it has a grip, as well as a thumb rest. There are three rings, for aperture, shutter speed and exposure compensation. Unfortunately, the rear shutter speed ring is multifunction, and it is far too easy to push down on the ring when spinning it to set shutter speed, resulting in shifting out to a different function.

On the plus side, the lens is f1.8 at 24mm as opposed to Sony’s f2.8. It makes a real difference in indoor settings. While the Sony lens is longer, 200mm vs 100mm, it turns out that I never used the longer focal lengths, so no loss there. Focus is contrast detect only, which sounds like a loss, but in practice it snaps into focus quickly enough. The G7X touch screen works for all functions whereas the Sony RX100 only allowed focus point selection by touch. It was possible to carry over many of the R5’s customized My Menu settings to the G7X, greatly simplifying moving from one camera to the other. There is one custom function (the R5 has three), which I have set up for HDR bracketing, same as my R5s.

Within a few hours, I had the G7X set up and in use, with nary a glitch. Familiarity and simplicity are very legitimate design goals, and there, for these cameras, Canon leaves Sony in the dust.

For many years, the web galleries on this site were generated using a long outdated feature of Photoshop CS5. This involved extensive modifications to templates provided with the software as well as hand modification of the gallery index code afterward. The final product is a thumbnail table index page allowing selection of larger images. Since CS5 was the last version of Photoshop we bought, eschewing the subscription trap, we have long sought a current and more modern tool for web gallery generation. After years of searching we finally tried out an oddly named product called Juicebox. Despite the quirky name, and despite the fact that development seems to have ceased, the software does a very nice job of creating carousel based web galleries that are responsive to screen size. Since the CS5 templates are already in place, we're keeping that capability and adding the the new version in parallel as an alternative.

Colonial Williamsburg's resored area visitor reference book, The Guide, has maps of the restored area showing sites of 153 restored structures with a short description of the background of each, many of which are open for tours. We've just posted a Photo Guide list providing amplifying information concering the location of each, along with a clickable thumbnail for larger images personally collected on many visits over the years.

A personal view of the state of the camera industry as it moves on rapidly from DSLRs to mirrorless (MILC) technology. All is opinion except for my personal preferences -- which are all mine and shall remain such:

Finally, yes, I get frustrated by Canon’s sometimes seemingly slow to market pace vis-à-vis the competition. What are the trends RE Sony vs Canon in the full frame high end segment (the one I care about)? What about Nikon vs Canon? I don't have access to data that would clarify that question. But, I am certainly looking with great envy at Nikon’s expanding stable of hand-holdable long lens offerings and (relatively) affordable high resolution stacked sensor bodies.

Having lamented the above, I have no desire to “switch brands”. I like the R5 for everything but action as well as the TS-Es and the many pro quality zoom choices below the supertelephoto level. I will say, though, that I am considering running two systems in order to grab one of those hand-holdable high-end Nikon telephotos. The only barrier is that I really don’t like Nikon body ergonomics. Parenthetically, having been exposed to Sony's compact RX100VII I absolutely detest Sony designs. (Update: sold to the lowest bidder -- and good riddance.)

Previously, we took a look at the relative value proposition (cost vs performance/features) for Canon, Nikon and Sony mirrorless cameras -- concluding that Canon lags the other two on features. Lenses tell the same story when it comes to distance-limited wildlife-oriented use cases. There, focal length, maximum aperture, optical quality with teleconverters and weight/handholdability are all considerations.

While all three makers offer very good budget zooms, those tend to lag behind optically when a 2X teleconverter is added. Hence, I have a preference for high quality (and expensive!) primes.

The table below lists current choices. For me personally, my older non-mirrorless EF400 f4 DO II, at 4.63 lb (plus EF-RF adapter and EF2X extender), is just about at the limit. And, a maximum focal length of 800mm is often too short. My preference would be a 500mm or 600mm lens, maxing out at 1000mm with TCs and preferably 1200mm.

Leaving aside price, and with a preference for high quality primes, it is clear from the table above that Canon has nothing to offer in the mirrorless RF lens lineup that matches my preferred parameters. While patents exist for better choices, such products do not exist at present. Sony is no better; the new 300mm f2.8 is light enough but the focal length is too short for distant subjects.

Nikon, on the other hand, knocks it out of the park for wildlife photogs -- as they did with the Z8 and Z9 high end mirrorless cameras. Even their Z600mm f4 leads the pack, including an internal 1.4X teleconverter for a very small weight penalty. Given my preferences, running a dual system would appear to be the only way forward at present.

Several versions of Canon's popular-with-birders 600mm f4 lens have come and gone, usually accompanied by significant optical and/or mechanical improvements. The first EF version, without image stabilization, appeared in 1988, weighing in at a whooping 13.2 lb. This was followed in 1999 by the first IS version (2 stops),at 11.8 lb., still quite a load to horse around. 2012 brought a big improvement with the EF version II. Tipping the scales at a much lighter 8.6 lb and yielding four stops IS, some robust photographers began to hand hold this lens. Better, optical quality proved excellent. So much so that Canon has failed to best the image quality of this gem in succeeding generations.

Image quality aside, weight reduction did not stop with version II. In 2018, version III arrived with a redesigned optical path that put a single large element at the front and everything else relocated to the back. Balance was improved and weight dropped to a remarkable 6.7 lb, hand holdable by even more photogs. IS was upped to five stops as well. When Canon went to mirrorless, the same optical path was used, but with an EF-to-RF adapter integrated into the rear of the lens.

Sadly, optical quality improvements were not included in either the EF III or the RF version. The bare lens was very nearly the equal of version II, but once teleconverters were added, image quality started to separate. Adding an EF2X III extender resulted in a perceptible difference in the MTF curves. While the EF III and RF versions showed a slightly flatter performance curve across the frame, the EF II was distinctly superior near the center, where subjects are most likely to be found, especially with birds.

The-Digital-Picture's lens comparison reflects the same performance difference. (The RF version with RF2X extender does best the EF III, making it the better choice optically.) Many photographers who have tried both affirm that the difference can be detected when viewed at high magnification. It's a tradeoff, but my guess is that many who rely on teleconverters would prefer the center sharpness of the EF II if weight does not sway them to version III or the optically identical RF optic. (Images linked under fair use)

A casual backyard photo of a dark-eyed junco shows just how sharp this lens is. Bored with a lack of suitable subjects, I clicked a single image of the junco, perched high up on a branch in a distant oak tree -- later measured by rangefinder at ~70 feet, camera to limb. The original image plus two 100% crops are shown below. Click on images for web jpeg of original and 100% crops of the isolated junco. Processed with Capture One. A pass through Topaz Photo produced an excess of sharpening artifacts. (Camera & lens tripod mounted)

Perhaps a better version will come along if the lens is ever fully redesigned for the RF mount, but for now the EF600mm f4 II remains the image quality champ, especially with the EF2X III extender. Better quality combined with reduced weight remains a mirage for the moment.

Latest Internet spec guesstimate for the anticipated R5 II, likely to be announced perhaps as early as April:

What do we make of this rather skimpy offering? Well, compared to the original R5, introduced in July 2020, it's a bit underwhelming. Not being a videographer, the only interesting tidbit in the list personally is the upgrade from 20fps (electronic shutter) to electronic shutter only at 60fps. Better autofocus would certainly be appreciated, but that entry says nothing of substance about what those improvemens are or how they might benefit photographers.

But, what is unsaid does leave room for speculation. Why only migrate from a FSI to a BSI sensor without also going to a stacked design, as Nikon has done with the Z8 and Z9? The high maximum frame rate, 60fps at 45MP, even if only at 12-bits RAW, tends to suggest a stacked design. After all, that's a lot of data to push through the pipeline. Based on specs, R5 maximum throughput is 10,800 Mbits/sec (frames/sec x megapixels x bit depth). By the same calculation, an R5 II at 60fps would yield 32,400 Mbits/sec (12-bit). That's an enormous increase for a single generation -- once again suggesting a stacked sensor. However, the list does include the caveat that 60 fps is "max burst mode." It remains to be seen what limitations that entails and how usable it really is.

In either case, what will the sensor readout speed be? The R5 is a pokey 15.5ms, which results in a bad case of rolling shutter distortion for fast action -- leaning verticals and banana shaped baseball bats, for instance -- making it totally unsuitable for certain fast action genres. If the new release is built around a stacked sensor as the previous paragraph seems to suggest, the R5 II would minimize rolling shutter, making it a full-featured all-arounder and a genuine competitor to Nikon's Z8 -- and maybe even Sony's $6500 USD a1.

So, we wait for April, as we awaited the now past February CP+ nothing burger. Will the smoke be black or white? Only Canon knows.

(* Updated sensor info in the online spec list : "We believe it'll be a stacked design" -- which would tend to validate the above stacked sensor speculation.)

Displaying flags is simple. All you need is a flag, a pole and something to attach the former to the latter. Well, not so fast -- the devil is in the details.

Starting from basics, years ago we purchased two eight foot poles for flag display at a 45 degree angle on our front porch columns. The toppers were wood, the flag suspension mechanism was an eye screw near the top of the pole, and a pair of plastic flag clips slipped through the flag grommets carried a doubled string through the eye screw, allowing the thus suspended flag to be raised and lowered.

This was our first setup. But, over time the wooden topper began to rot from exposure to the elements. Those plastic clips became brittle and shattered. The eye screws at the top of the pole tended to rust, making flag raising and lowering more difficult. The cotton string frayed and broke. Finally, because the flag was displayed at a 45 degree angle, wind always caused the flag to wrap around the pole. A six foot tomato stake served as a frequently used flag unwrapping lever.

After years of enduring the above aggrevations, especially constant flag wrapping, we decided a better solution was needed. A search turned up a rather elegant flag pole rotating ring, made from non-corroding stainless and aluminum, complete with flange and hole, allowing attachment of a caribiner to carry the flag string/rope. Flag wrapping problem solved, right? Not quite -- physics intervened to thwart a seemingly foolproof plan. With the flag string suspended at both ends, top and bottom, tightening the string to prevent flag droop placed enough tension on the rotating flange so that rotation became impossible. Oh well, at least the new hardware wouldn't rust.

But, raising and lowering flags was still difficult -- string under tension just didn't slip easily on the caribiners. So, we began to look at end caps with rotating pulley attached. Then, during the search for a good buy, we came across a customer who, faced with the same problem, instead attached a cable pulley to a caribiner held by similar hardware. A quick trip to the hardware store secured just the right size, the pesky string was replaced by paracord, and the problem of raising and lowering the flag was solved. A flag cleat secured the tightened paracord, and a solution was finally in place. Hallelujah!

With the pending and presumptive announcement of Canon's (finally at last) pro body R1 and the rumored delay of an R5 II update, it's time to take stock of how Canon's value delivered (features for price) at the high end stacks up against the competition. Nothing is known regarding specs for either so anything from this point on is speculative. But, one possible leaked spec claims that the R1 will have a 30MP sensor. If that is true it will likely render the 24MP R3 a one-and-done orphan -- perhaps analagous to the initial R vs the R5, the former apparently an early release intended to gain development time for something better. Assuming that the R1 30MP value is correct let us compare Canon's top tier to Sony and Nikon.

Each of these bodies is considered well suited for professional applications, although some are more rugged and better weather sealed than others, accounted here as "flagship" -- with the caveat that flagship is a marketing term, implying not only top features and performance but also more robust build quality and better weather sealing. Price is current but does not reflect temporary sales or discounts. Models are of different release dates, with longer running cameras often having been priced higher at introduction. Both Sony's A9II and A9III are included as the former is still in the lineup and sits at a quite different price point. All frame rates ³ are for RAW files with electronic shutter.

Sony and Nikon lead the pack with feature laden bodies. All three of Sony's top cameras come with low rolling shutter stacked sensors, the A9III also being the first full frame global sensor. Nikon's top two have the same stacked 45.7MP sensor, which gives the $3800 Z8 a uniquely high value for capability. And, Nikon's offerings are consistently priced lower. This shows up in the resolution and action value rankings, where Nikon's Z8 and Z9 and Sony's A1 score high in both categories, marking them as the best all around cameras on the market today.

At its current steeply discounted, possibly end of production price point the Canon R5 edges the three previous out for resolution value, but it opened higher than the Z8 despite having the only non-stacked sensor in the group, The R5 still holds up well to the competition for most applications, but not for high speed action photography -- think sports and birds in flight. A very visible rolling shutter effect, the byproduct of its slow 15.5ms sensor read-out time, can at times be intoleable. In the Canon line, only the seemingly orphaned R3 has a stacked sensor -- although any successor likely will also. But, the R3 trails Nikon and Sony's all arounders badly with only half the megagpixels. Its integrated grip body and pre-R1 $1000 price discount combined with 30 fps at a full 14-bits gives it the value lead over Sony's A9II.

The inference from the table seems inexcapable: Based on specs vs price, and with R1 and R5II as yet undefined, Canon offers the least value for its high end cameras. The perception is that Canon's high end bodies are sometimes introduced later and with less features than their competitors. The R5 is a notable exception -- at introduction its 45MP sensor and 20 fps (electronic shutter) set a new benchmark by conbining high resolution and high frames per second in one body.

We'll update this once the R1 announced and when more is know about an R5II update. Perhaps we'll also add a lens section.

We previously related our failed attempt to buy Blue Griffon, thwarted by failure of the merchant to provide an activation license after taking our money. We subsequently lodged a dispute with our credit card company, and after a month of investigation our purchase price was refunded. End of that story and on to the next, which was a trial evaluation of Pinegrow -- resulting in a second failure to launch, albeit in this case more a matter of mismatch between a seemingly excellent product and personal needs.

After Blue Griffon, we spent a lot of time looking for alternatives. Baseds on features, Pinegrow seemed to offer the most value. Fortunately they provide a one week trial period before purchase. As it turns out, this saved another failed purchase. Upon attempting to use Pinegrow, it became obvious that the intended market for this high end designer suite is commercial collaborative web development teams. Right off the bat, the product requires local area network changes to permit use of an "internal server." One can't proceed without this feature.

Next, activity must take place under the umbrella of a "project", which imposes its own organization on development. This also is a no go as I have a very robust existing file structure, upon which the entire site is based. Parenthetically, the same logic applied when I began to catalog my image collection. Capture One provides two approaches, managed and referenced. In the former, the catalog manager decides where to put images, leading, we feared, to data corruption or loss vulnerabilities. The referenced approach leaves the existing file structure completely intact and untouched -- far safer to my way of thinking.

There is more, but the intrusive nature of these constraints was enought to convince me that for a small solo operation with an existing file structure and an approach that requires no network revisions this wasn't a good direction. In the end, there seemed to be no easy path for migration into the Pinegrow envelope. Therefore, we passed on Pinegrow, although we suspect that it is a very good match for commercial collaborative developer teams.

So for now, we continue with the long discontinued MS Expression Web editor augmented by an ancient version of Photoshop for web galleries. The search for a modern editor will continue, but two of the most promising candidates have fallen by the wayside.

Postscript: Just took a look at an editor called Sublime Text. Looked promising until it emerged that it is incapable of displaying HTML code and formatted output side by side. Another no go.

". . . Since they're already hovering at just below 50% market share, it means they [Canon] probably didn't break that threshold but also means their dominance of the market is unchanged."

Said before, but Canon has publicly stated that their corporate goal is to hold a 50% share of the camera market -- successfully accomplished according to the above stat. To that end, it appears that their strategy is to emphasize two segments: the low end market, where volume can be had, and the high end professional market, where performance drives sales that in turn deliver high profile visibility at prestigious events.

Parenthetically, that low end success has another benefit besides pumping up market share. It provides an affordable gateway for entry into the ILC world for new photographers wanting more than a smartphone. Maybe a few will progress to something more. That's certainly helpful if the ILC market is to sustain itself over the long haul.

On the lens front, the segment that's largely missing, that provides perhaps best value for performance and that many hobbyists would be most interested in, is the middle range -- reasonably priced optically excellent lenses a notch below the ultra expensive, sometimes exotic pro offerings. Where manufacturers don't offer them third party makers have in the past provided alternatives. But, Canon hasn't allowed companies like Sigma to offer full-function RF mount lenses.

One example will suffice, the lowly 50mm prime. Canon offers two choices, a $200 RF50mm f1.8 lens that is, charitiably, worth what it costs, and an optically excellent but heavy RF50mm f1.2L beast, one that is sharp all the way into the corners wide open -- and is spectacularly expensive, way north of $2K. Missing is an affordable 50mm f1.4 that provides both wide aperture and excellent performance. In the EF mount, Canon had such a lens, albeit an optically compromised one. Sigma offers a sharp and affordable f1.4 optic for less than half the price of Canon's RF mount f1.2 offering -- albeit sadly not in the EF or RF mount. The lack of such choices, either from Canon or other makers, has been the subject of much dissatisfaction among Canon R-mount owners.

Canon's 50% market share doesn't mean that there aren't segments where other makers can compete and excel -- witness Nikon's plethora of Z-mount telephoto choices and excellent for value Z8 and Z9 bodies, all of which, on specs, should appeal to wildlife photographers. (As a bird photographer, I would already own a Z800mm f6.3 PF or Z600mm f6.3 PF if either would mount on my R5. And, may yet run a dual system to get one or the other.) Further, Sony's rise in recent years, largely at the expense of Nikon, demonstrates that their lineup has sufficient performance and value to attract sales.

Bottom line for me: Buy what meets one's needs, or preferences, and tamp down expectations from posturing on Internet forums. If one must complain, or boast, try the one place that might, might make a microscopic difference -- i.e. the makers themselves. Goodness knows, the dear departed Chuck Westfall must have dreaded opening my emails in days long past. And I'm not bashful about letting CPS know what I think -- not that it has made one iota of visible difference in subsequent products. But, at least I know who butters the bread.

The recently published R1 spec list may or may not be real. But, assuming it is, and given that the 30MP resolution is too low to be of interest, one can start to speculate on how the next generation of silicon technology might effect a hoped-for R5 upgrade. In effect, a speculation on top of a seculation. The starting point is a rather interesting value for sensor readout time -- 0.8 milliseconds. If true, this is a significant improvement over previous cameras, including mechanical shutters. The benefit of short readout times is greatly reduced rolling shutter effects -- banana shaped baseball bats, for instance.

It is difficult to obtain accurate information from all manufacturers, but a readout speed comparison chart available on Internet, if valid, helps give scale to various cameras. The Canon R5 value is listed at 15.5 ms and the R3 at 5.5. The Sony a1 and Nikon Z9 best the latter at 4.2 and 4.0 respectively. Mechanical shutters are listed at 3.0 ms.

The R1 will undoubtedly have a stacked sensor but, if form follows true, Canon will eschew this technology in an R5 II, likely opting instead for an updated reprise of the current R5 front side illuminated single layer sensor. If this becomes fact then an R5 II will also display visible rolling shutter effects, the extent of which will depend on readout time. However this may turn out, it is still possible to estimate the total throughput of each sensor in megabits/second. Throughput is the product of sensor megapixels, bit depth and frames per second. Maximum frame rate is achieved with electronic shutter.

Bit depth is difficult to find for the R6 II so 12-bits is assumed, same as the R5. Of course, the R1 numbers are just a possible leaked (or fabricated) list at this point, but it is likely that it will be of a different generation of silicon than the first three, all of which are in the same ballpark for throughput. Since the R1 is an unknown both cases are shown, albeit it will likely be 14-bit if it follows the R3 example.

The above formula can be run in reverse to estimate possible frame rates for the rumored R6 II, likely to be announced long after the R1. This, of course, assumes that the downstream processing pipeline throughput of the (presumptive) R1 can be shoehorned into an R5 II. The values are based on 12-bit image files, same as the original R5, and on two possible resolutions, 45MP and 60MP. These values are sometimes mentioned as possibilities on rumor sites.

Posting absolutes on an Internet forum is a sure-fire guarantee for a woodshed hiding. Such is the case when someone says everyone in a certain category needs a particular type of camera. Post that, and you'd better duck -- the antibodys will be swarming in three. . .two. . .one. . . The best one can say is that such pronouncements represent preferences, not needs.

I suspect we all see "need" (i.e. preference) through the filter of personal use. In one such case, a particularly viriulent antibody asked the rhetorical, and sarcastic question, how did we ever capture worthwhile images with the 18MP 1DX? Well, for that individual (and for me in some uses) the 18MP 1DX did the job.

But for others, again speaking personally, when one is already at 1200mm and can't get any closer more megapixels becomes a "need" (for the cropping ability) in the only terms that really matter to me: will I spend my own personal cash for something less in that one metric, no matter how spectacular those other specs may be -- particularly when the gaudy features may cost 50% more than a camera that may be lesser in many ways but that meets my "need" in the one metric that's most critical for my use?

Hint: when the pre-release specs for the R5 solidified I sold both my 20MP 1DX Mk IIs in anticipation of a pair of gripped 45MP R5s -- a decision I have not regretted for even one instant. While I would much prefer the 1-series form factor (and while I look with envy on Nikon's big glass lineup), the MILC experience has relegated DSLRs to the past and 45MP has sent significantly lesser models to the back of the line.

Just to be clear, one of my all-time favorite photographs is of a snowy egret taken with a Minolta 8000i and a 200mm APO lens on Kodachrome at almost reach out and touch range. There's plenty of detail for modest prints. But, one can't always count on being so fortunate or so close. I also have an image of a pileated woodpecker high up in a tree (line of sight distance 100 feet measured with a rangefinder), captured with the R5 at 1200mm. The image, a vertical cropped from a horizontal frame, shows feather detail in the subject. We've kept the EF600mm f4L II precisely because it can resolve that kind of detail with a 2X extender.

PS: more megapixels also resolves more fine detail in mountain overlook images. Individual leaves at a distance become individual leaves in the photograph rather than blurred blobs. The R5 reveals both, i.e. detail in birds and detail in landscapes -- at 20 fps electronic if you need it.

That doesn't mean I would ever denigrate anyone who makes a different choice, and I would hope that sentiment would be reciprocated -- sadly perhaps an unrealistic expectation given human nature.

We've been waiting a long time for something more than idle speculation regarding the near-mythical Canon R1. Up to now, guesses about specs have been all over the place. But, Digital Camera World just published a list (since picked up by Canon Rumors) that appears a lot more concrete than anything before. The list could be a legitimate leak or it could be a complete fabrication (especially considering it is from a single source with a poor track record), but it has certainly sparked discussion. The anticipation is that all will finally be revealed with an announcement at Japan's CP+ toward the end of February. Until then, prudence suggests reserving judgment.

With that caveat, these specs, if true, say a couple of important things about Canon. First, Canon is remaining true to it's tradition of not following other manufacturers -- at a reported 30MP there's no attempt to match Sony's 50MP a1 or Nikon's 45MP Z9 and Z8. Rather, a camera built to these specs would advance the state of the practice in a number of areas other than resolution, including frame rate, dynamic range and sensor readout speed (with accompanying flash sync speed) -- the readout speed being the most difficult to believe.

Second, the list (once again if true) suggests that the long time market leader is staking a lot on succeeding in its chosen market -- professional genres where sensor resolution is perceived to be less important than class leading performance and ruggedness. A camera positioned according to the list has in the past competed well in the sports and photojournalism market -- and should do so against Sony's just announced global-shuttered A9 III. And, if higher DR proves to be a fact wedding and event photographers may find it appealing as well. That would be quite the scoop in pro photographer market positioning.

And yet, the spec list does not appear to be sitting well with everyone. The main sticking point is the 30MP sensor -- underwhelming to many given Sony and Nikon offering -- a sentiment I share given a strong interest in bird photography. There, distance considerations as well as the need to capture fine feather detail puts a premium on high resolution and the ability to crop to a suitable composition. If the 30MP spec is true I'll wait until later in the year to see what an expected R5 update brings to the table -- hopefully at least the 45MP of the original R5, and perhaps more. (While we're wishing, a stacked sensor like the Nikon Z8 would be nice. . .)

There is also an unenviable implication for the R3, which at 24MP could well become a one-and-done product. Or, perhaps, morph into something completely different, moving to another niche in the Canon ecosphere. Worse for Canon, heavily discounted sales of the thus marooned R3 could well depress early R1 sales numbers.

Is the above really what the future holds? Nothing for it but to wait and see. The presumed February R1 announcement can't come too soon for us long-suffering Canonicals.

Good question. Followed by another, which answers the first: what exactly am I looking for in a long lens? The answer to that is a high optical quality hand-holdable telephoto that gets to at least 1000mm and preferably beyond with extenders and comes in at 5 lb or less.

With that in mind, how does the new zoom stack up? First off, at $1900 the RF200-800 is certainly affordable. And, at four and a half pounds it is definitely hand holdable. For a non-L lens image quality is tolerable. But, leaving aside issues such as the extreme extending design and reported long throw and stiffness of the zoom ring, there are more fundamental concerns. Given that our supertelephoto goal is to get to 1000mm or more, does the lens plus 1.4X extender measure up optically?

Canon has yet to publish an MTF chart showing the lens with an RF1.4X extender, so without actually trying one out it's impossible to know image quality for sure. But the indications aren't good. From The-Digital-Picture's lens tests the bare lens at 800mm doesn't best my EF400mm f4 DO II with EF2X extender. Add an RF1.4X extender and one has an f13 slow and dim lens that must inevitably lose fine detail. The 400 DO with 2X is right at the bottom edge of what works for small or distant birds. Given the TDP test, we simply can't see buying this one.

Another consideration is the slow maximum aperture combined with the added weight that a zoom design entails. The lens produces a maxiumum focal length of 800mm at aperture f9. That results from a front element diameter of 89mm, rather small as supertelephotos go -- hence the dim f9 aperture at the long end. Even so, the weight with hood is still 4.7 lb, same ballpark as the EF400mm f4 DO II. The latter's front element based on focal length and aperture is 100mm. Further, the EF400mm is an older design and could surely profit from recent weight reduction techniques.

The implication of all of this is that my particular interests would best be served by a prime lens rather than a zoom -- used with extenders when necessary. It all comes down to optical quality, front element diameter and overall length, the latter two implying weight. The Nikon Z800mm f6.3 PF, with a front element of 126mm, nevertheless weighs only 5.2 lb. Still, that's a big front element and a long barrel at 15+ inches. While the weight might be tolerable, the poundage combined with overall length and large front element implies a very tiring arm-extended hand holding posture.

Nikon also makes a Z600mm f6.3 lens with conventional optics, 11 inches long and weighing in at 3 and a quarter pounds. While better, a 2X teleconverter still puts it at f13, a consequence of its 95mm front element. My guess is that something with a front element in the range of 110mm would be the best compromise between weight, focal length and aperture. An RF 600mm f5.6, possibly in a DO design, might fit the bill. The front element would be 107mm, and a 2X exentder would produce 1200mm of focal length at a not so dim f11 while maintaining a shorter and more hand-holdable package compared to the Nikon Z800mm PF.

Canon, can you hear me? [listening, listening for a response. . . Sadly, only silence. . .]