The Latest Ask Thom Q&A

From time to time I try to answer questions I'm getting multiples of in my In Box with a quick post here on dslrbodies.com. While I continue to try to answer individual questions I get, sometimes broadcasting an answer is the better way to go.


I really want to get to 1000mm. How well does the TC-20E work on lenses like the 500mm f/5.6 PF?

I'm not a fan of the pursuit of this. You're simply not close enough to the action, whatever it might be. Not being close enough is going to cause all kinds of other issues. Ambient air temp may be varying between you and the subject, and if so it won't matter if you can focus well or fast enough when that happens, you'll get odd or poor results. You'll have a terrible time keeping 1000mm aligned and on target with proper support. You're stealing light from the focus system. You're adding glass—and in particular air/glass transitions—to an already complex formula, and you never get "equal" let alone "better" optical performance when you do so. Veiling flare will increase. You'll be tempted to turn on VR to solve the support issue, and that, too, may start stealing acuity from your result. 

A couple of sub-questions come out your question, so I should probably answer them. I don't own a TC-20E any more—which should tell you something—but I do have the TC-14E. My experience with three different 500mm PFs and three different TC-14E's is that this combo works better than expected. With a caveat. It works better than expected on the Z6 and Z7, which can autofocus at f/8 just fine in reasonable light. On a D5, D500, or D850 the problem is that the lack of light getting to the focus sensors is a real issue. You'll lose some focus capability even in good light, as IIRC you only have five cross sensors and a maximum of 15 available sensors being used (out of 153).

But optically, the 500mm (and 300mm) PF with a TC-14E seems to work respectably well. And focus performance stays good on the Z6/Z7 until you're in low light.

One reason why amateurs always want to find the magic "more reach" combination has to do with their approach. To get close with wildlife takes an investment of time, energy, and good guiding. Yes, there are times when I can't get to where the animals are doing something interesting. That's why I spend a month at a time in Africa and keep going back (or Alaska, or any other place I shoot wildlife). It's not "go once for a short period on a budget tour and get the shot of a lifetime." Oh, that happens by chance sometimes, but it really is by chance, not by the equipment someone is carrying. 

Realistically, you have to go to somewhere where the likelihood that you can get close is higher (e.g. not the Serengeti, but maybe the South African private preserves instead), you need to spend more time in the right positions and not chase after the shot, plus you need to understand and anticipate animal behavior. You may need to create a bird-friendly area within your yard and/or build/use a blind. You need to find where the lions are hunting, understand how they'll hunt, position yourself properly, and sit. 

1000mm to me indicates an impatient photographer. Personally, I'm very patient and perfectly happy with 400mm ;~), as in the following shot: 

bythom INT Africa Bots ChobeSavuti 7-2017 D500 28579


Or 70-200mm ;~)

bythom INT BOTS Savuti 2019 Z6 74383a


You get images like these not by buying more expensive equipment and then adding TCs to it, but by spending more time where such images can occur, being patient, and learning what the animals are likely to do and how to properly approach them doing it.

Nikon needs a 60mp camera. When will they introduce one?

I'm not sure you need a 60mp camera ;~). Having now spent some time with one (and a 100mp medium format one), for most people we're well into the declining improvement range with 60mp.

This question is sort of related to the first question: some people want 60mp to have way more cropping flexibility. They could get the same pixel density from a 24mp APS-C camera if they're looking at 60mp for "more reach", though. 

One of the things I wanted to look at with 60mp was just how much benefit that really produced in landscape photography. (Caveat: landscape photographers will always claim they can never have "enough". Of anything. Pixels, dynamic range, you name it. That's why we used to have folk wandering around with 8x10" film cameras and larger. ;~) 

My current favorite landscape lens is the Nikkor 19mm PC-E, not just because it's sharp, but also because it is flexible. Pano stitching is simple with that lens, and the plane of focus and perspective point can be controlled, too.

So what does the 60mp Sony A7Rm4 and 16-35mm f/2.8 give me over my D850 with the 19mm? Basically, a more impromptu, less workflow intensive capability with some very modest enlargement capability. Even that latter is a bit tricky, as I find Sony ARW files a little more difficult to fully optimize processing for than Nikon NEF files. 

But let's start with the premise that you're looking at making large prints. I've updated the table in my How Big Can You Print? article to now include 60mp and 100mp. Using the criteria I established over ten years ago after a lot of deep testing and examination, 60mp gets you to "good+" compared to "good" for a 45mp camera when you print at 36" maximum width. It takes a 100mp camera to get you to "excellent" at 36" print size.

Coincidentally as I was writing this response, a new question popped up in my In Box from someone asking if they should sell their 36mp D810 now before it lost any more value. Uh, why?  The D810 is a remarkably capable camera even today. Was this person planning on printing beyond what the desktop inkjet printers are capable of? No. Then they don't need to sell their D810 and wait for a 60mp Nikon.

I think I've been clear about this for a long time, maybe even the entirety of this century: I personally will always take more pixels because it gives me more sampling of the subject. However, the pragmatic benefits of more sampling is going down with each generation of sensor. And we're not tending to output large sized images as much as we used to. That combination means that pixels alone is no longer a huge motivating factor for someone to upgrade. 

I haven't released my Sony A7Rm4 review yet because I'm still working on it. But personally I had to make a decision: upgrade from my Mark III model? My answer was yes, but it was not even remotely influenced by pixel count. The real decision came because of handling changes to the body design and some EVF, focus, and write performance benefits. Sony fixed or improved a number of things that frustrated me with the previous model. The extra pixels was a side benefit of upgrading.

I suspect that a lot of Nikon shooters are going to fall into that same category when Nikon upgrades the D850 and/or Z7 models to 60mp. It won't be the pixel count that is the compelling upgrade feature, it will be other things.

Why do we need yet another card type (CFexpress)? SD cards work fine.
Why do we need yet another connector (USB-C)? What we have works fine.

I've grouped (and rephrased) these questions together because they derive from the same notion (status quo is okay). I could probably add a whole bunch of other similar questions (e.g. why do we need 4K or 8K?), but I'm going to concentrate on these two.

One thing I've noticed more and more about others in the Baby Boomer generation—which is the group most of the serious photo enthusiasts belong to—is this notion of "enough is enough, stop all the improvements" is now in full bloom. You'll note that all of the questions today revolve about trying to do more, which is the opposite.

So let's start with that last word in the quote, improvements. Yes, CFexpress is an improvement. Indeed, CFexpress Type B is an improvement over XQD, and the smaller CFexpress Type A (not yet available) is an improvement over SD (Secure Digital). 

What most don't realize is that cards have been mimicking ongoing internal personal computer engineering improvements for quite some time. CFexpress brings cards up to the PCIe 3.0 level, which means they can be quite fast, as fast as what's happening inside the bus of your PC. 

299MBps (Sony SD Tough UHS-II card) seems like it should be fast. It is, at least compared to where we started with SD. However, I've noted two things about SD versus XQD (and eventually CFexpress) that don't get talked about a lot. First, that 299MBps generally doesn't hold up under sustained writes. Fill the buffer, and you're in sustained write mode. This is one reason why SD isn't necessarily the right choice as we move to faster cameras with more pixels. 

The other has to do with getting those images off the card. The fastest current SD cards tend to be 300MBps read speed, and you can buy readers that can get that from the card in sustained reads to your PC (but not with USB 1.x, and often not with USB 2.x ;~). Current XQD cards tend to be 440MBps read speed, and CFexpress can double that (two-lane PCI instead of one). That means that you'll get those big 60mp images from your camera transferred faster to your computer. Those of us that shoot large quantities of images appreciate higher transfer speeds.

Which brings us to the USB side of things. Let's be clear, USB is a confusing mess. Some standards committees work decently to make clear the progression of tech and what will work with what. Others don't. USB is one of the better examples of total consumer confusion. 

We have USB 1.x, 2.x, 3.x, and yes, 4.x (not yet available). We have multiple "gens" (generations) with USB 3, which have different capabilities. We have naming conventions that were rendered meaningless ("Full Speed" is not at all what most people would call "full" these days, nor is "High Speed" actually all that "high"). We have Thunderbolt overlaid on top of USB. And we have a whopping 10 different USB connectors (plus some additional proprietary ones!). There's even two different power delivery options that can get in your way. None of this is clearly labeled, nor is it well understood by customers. Plus all this messiness happened in a bit over 20 years.

One of the goals of USB was to "simplify the interface between computers and peripherals." Yeah, didn't happen. 

Nevertheless, the need for what's happening in the underlying tech behind USB was indeed necessary. You may know about Moore's Law—the idea that the number of transistors doubles every two years—but there's been a similar trend in storage, often referred to as Kryder's Law. Our computers have been getting faster and beefier on predictable paths. Moreover, there's been a huge cost reduction curve, too. My original 5MB hard drive was the very first made by Seagate, literally, as in serial number 00001, and cost US$1500—$4600 adjusted for inflation—while today I can find 5TB drives that cost US$100.

While I was in the 0.01% with my desktop computer in 1980 (2MB RAM, 5MB drive), my needs grew exponentially as I did more with writing, photography, graphics, video, and publishing. Considering that a typewriter was considered state-of-the-art and had limited household penetration when I was in high school, the march of tech has been relentless in moving that bar upwards. 

The files that run my Web sites are nearing 20GB. And that's using small, compressed JPEGs. Put another way, I couldn't have fit these sites in 1980 onto my personal computer even if the Internet had existed then. 

Personally, I'm glad tech keeps moving the bar. It's allowed me to do more things, do it at higher quality, and in less time. I don't see that changing any time in the future. So I'm going to do what I've always done, and stay near the forefront of what's possible. If you want more beyond what you've got today with your cameras, you need to embrace this constantly moving progression of cards and cables, too.

If status quo is okay for you, then you're probably on Last Camera Syndrome. I'd advise you to purchase extra accessories and disposables now (such as batteries and storage cards), before you need them in the future and new ones are more difficult to find. Maximize what you get out of your current gear and be happy with it. No need to complain to others about standards moving on. You got off at a station and have decided to only explore its surroundings. 

Realize, however, that things change. In tech, they change pretty dramatically, fairly quickly. Once cameras started embracing electronic technology with auto metering and focusing in the 60's and 70's, it was inevitable that they'd have to be on the fast track to keep up with the rest of the world. As it turns out, the cameras companies are now running a slow train on a fast track, so are actually falling behind other competing and complimentary technologies. Some of us pros are struggling to keep up with our clients because of that. 


Aside: Note I wrote "[tech allowed] me to more things, do it at higher quality, and in less time." Let's apply that to recent cameras. More things? Uh, it seems that the imagination of the camera engineers is dwindling. Sure, we got focus stacking and pixel shifts recently, but it feels like these are not-quite-polished capabilities and we should be getting more of them, and more fully fleshed out. Higher quality? Yes, that's probably been the thing that the camera makers have done the best job on, though for the mass market there's the question of how high a quality is enough? We're above that bar. Which is one reason why MP3's ate CD's and now smartphones are eating dedicated cameras. Finally: in less time. The camera makers are failing big time on this. Menus sprawl. Communication standards lag our computers and mobile devices. Image processing workflow is still stuck in the one-hour processing lab. Even camera customization sucks big time. Why can't I store multiple "camera settings" on my smartphone and blast one over to the camera to instantly change the way the camera works?


text and images © 2019 Thom Hogan
portions Copyright 1999-2018 Thom Hogan-- All Rights Reserved
Follow us on Twitter: @bythom, hashtags #bythom, #dslrbodies