The Rhetoric Upgrade

The most advanced camera in the world

The end of DSLRs

The DSLR killer

A game changer

The first Canon/Nikon killer

These are all headlines that were used last week when the Sony A9 was introduced. One of those headlines was a direct lift from the press release. 

My only problem is that we’ve had the same sort of hype-level headlining with the A7 models, with A7 Mark II models, with the A99II, with the GH5, with the E-M1 Mark II, and now again with the A9. I guess eventually it might be true of some camera finally toppling the DSLRs, but that generally isn’t how giants go down. 

I’ll let you in on a little secret: one reason why click-baiting is up is that affiliate revenues for Web sites have been strained lately, what with a sensor shortage, declining overall sales, and much more having changed the market dynamics. That’s not the whole story, of course. Just as politics has become polarized and pushily partisan, it seems that everyone now has to make choices of horses to bet on, and do so publicly and vociferously. Pepsi versus Coke. Ford versus Chevy. Macintosh versus Windows. AT&T versus Verizon. Plus the long-standing Canon versus Nikon. Now it's Mirrorless versus DSLR.

It seems that in limited-player markets, we all devolve into devotees of one player, and the player we back can do no wrong. Indeed, the SOP (standard operating procedure) is that we must amplify and redistribute the message from the company we follow. 

Let me say this right up front: the Sony A9 is probably a very good camera. What little chance I had to handle one so far tells me it’s a lot like the already very good A7 Mark II models, only some of the things we users have been complaining about have been addressed. 

But DSLR killer? No. Game changer? Not really. Most advanced camera in the world? A very arguable point given that the camera uses USB 2.0 in a USB 3.1 world (and yes, we have a USB 3.1 mirrorless camera already). Some numbers are impressive on the A9, sure. But others aren’t. 

Personally, I’m still agnostic about the A9. I don’t know if it does anything useful for me or not. While the name of the game in sports cameras has tended to be “up the frame rate,” I know a lot of guys still shooting with 12mp 8 fps cameras for a very good reason: when you shoot thousands of images in a game and your first deadline is at the first intermission (and your second within an hour or two after the game) you really don’t want to generate more data than you need. Even 24mp is too much really. That produces a 20” print on a high-end photo printer. We’re not really shooting for that use. So 20 fps at 24mp is one heck of a lot of data to grab and deal with on deadline.

Moreover, Sony chose SD for the A9 rather than XQD. I’m not sure why. I can tell you that I have no UHS-II SD cards or card readers that can come close to keeping up when downloading thousands of images compared to my XQD cards and reader. Since Sony emphasized sports and speed so much, I have to wonder if they designed the A9 correctly. 

Still, I’ll withhold a real discussion of that until I’ve been able to use an A9 at an event I’m covering. 

But let’s go back to those DSLR-killing headlines for a moment. I’m going to tackle this two ways. 

First, Sony long ago telegraphed their punches. If Canon and Nikon don’t eventually have strong responses, then shame on them. I can’t remember when I heard a Sony executive first say it, but at some point it made it into their presentations at conferences and product releases. Their goal is highly technology infused: fastest, most, highest, lowest. As in fastest processing pipeline and focus, most pixels and dynamic range, highest frame rates, and best results when used in low light. 

There once was a question whether you could build an image sensor that could do all those things simultaneously, but given my discussions with others in the industry and the camera companies, it’s been a long time since anyone questioned that. Sony, however, is in a unique position to make that happen, as they both make sensors and cameras. But I’d be remiss not to mention that Canon can claim that Sony isn't in a unique position, that they, too, make both sensors and cameras. Canon just isn't pursuing trying to be first and most at everything because, as market leader, they don’t have to. Doing so is expensive, and may not pay off as well as you might think. 

Canon and Nikon executives had to have heard the same things out of Sony’s mouth as I and others in the press corps did. They had to know that Sony had the ability to pull that off, and Sony has been sneaking up on their goal for a while now, so there’s also a progression you can look at. That’s the thing about tech: it’s fairly predictable. An unknown breakthrough doesn’t happen out of no where. You start to see the hints and early iterations of it long before it pops onto the market. 

Thus, global shutter was something we talked about long before it became truly usable. On-sensor ADC was talked about and experimented with before it became a reality. On-sensor phase detect actually came from Nikon and Aptina back in 2011. The list goes on.

You generally have a clear view about two years forward, which is good, because that’s a pretty full product development cycle. We have a pretty good view about five years forward, which is good because it is predictive of follow-on models. Where things get fuzzy is out there in the 10 year time frame.

So to my first point: a camera announcement alone doesn’t kill the DSLRs, nor does the sensor inside that camera. If indeed DSLRs are now today toast—which I don’t believe they are—it would be something that would have been predictable.

Which brings me to my second point: DSLRs had a “use before” date stamped on them a long time ago. Every DSLR camera maker had compact cameras, so they knew the benefits of “live view” use of the sensor. They also knew the implications and complications. The bean counters at those DSLR companies could also tell everyone the cost of maintaining that optical path versus the cost of an EVF, especially since that EVF declines predictably in price while the mechanical/alignment issues of the DSLR don't follow that same path. I can promise you that there are spreadsheets at every DSLR camera maker that say “we’ll be making mirrorless instead of DSLR—at least for most of the product line—after X date.” Why? Because of product cost if nothing else. Fewer parts. Fewer alignment steps in manufacturing. Simpler designs. 

That Fujifilm, Olympus, Panasonic, and Sony placed huge bets on mirrorless early was their hope that Canon and Nikon didn’t have that spreadsheet, or that the Canon/Nikon spreadsheets had the wrong use before data on them. Moreover, while development costs for those four mirrorless companies was very high—probably higher than they’ve recovered to date—their production costs picked up benefits from being mirrorless designs. Basically, the four of the seven dwarfs that were the most viable all dreamt of catching the duopoly giants sleeping: they couldn’t take market share in DSLRs, but maybe they could get to the next iteration of ILC cameras first and grab market share there.

To some degree it worked (they got there first). But it hasn’t worked as well as they thought (Canon actually increased their market share, so they only nibbled a bit at Nikon while increasing the size of the ILC market somewhat from what it likely would have been). 

Now I’ll give Sony’s marketing team full credit for launching every missile they’ve got at the DSLR duopoly. I’m not convinced that the A9 is anywhere close to a replacement for a 1DxII or D5 for reasons from size to control placements to lens availability, but Sony’s PR team has managed to get enough of the photography press saying they are that there certainly is splash damage.  

What we know at this point is this: Sony executives weren’t kidding when they said they’d pursue having the best specs in a number of key categories, most of them driven from the image sensor. That produces new cameras with lots of PR-worthy items to promote. 

What happened last week with all the Sony A9 hyperbole isn’t new, and it wasn't unpredictable. 

Stay calm. Relax. The world didn’t end. A new world didn’t begin. Instead, a very nice camera was introduced.  Happens several times every year, and some of them have been and will continue to be DSLRs. 

A bit of an aside, but pertinent to this discussion: it seems that a lot of people don’t understand the whole notion of what exposure is and what sensors can and can’t do. 

Exposure = Light filtered by Aperture filtered by Shutter Speed

Fail to understand that and you can’t make good conclusions about new cameras. I keep getting people telling me that I can use the new Sony 100-400mm f/4.5-5.6 lens instead of my 400mm f/2.8 for sports. Nope. Either I need four times as much light or I need to set a shutter speed two stops slower.  Since I’m in fixed lighting conditions and need fast shutter speeds to stop motion, I can't do anything about the light and don’t want to be changing the shutter speed “filter.” My biggest changeable item is to find lenses that are faster.

Which brings me to the second part. The next argument from those folks is that “better sensors” mean I can just boost the ISO two stops instead of using a faster lens. Well, not really. In almost all of my sports shooting I’m not particularly limited by the sensor ability, I’m much more limited by the randomness of photons. Most of the noise generated in my night and indoor sports shooting is limited by the exposure, not the sensor. 

This is why the Sigma f/1.8 zooms were such an interesting development for the D500. There you have a highly capable camera that’s far lower cost than the A9 and very usable for sports. The f/1.8 lenses get you back what you lose from the smaller capture area in terms of exposure. So yes, you can pretty much think of the D500 as the mini-D5, at least if you’re careful about lens selection (e.g. Sigma 50-100mm f/1.8 instead of 70-200mm f/2.8, Nikon 200mm f/2 instead of 300mm f/2.8, 300mm f/2.8 instead of 500mm f/4). Of course, Nikon’s PR department wasn’t selling the D500 as the world’s lowest cost sports camera, probably because they also sell the D5 ;~). 


text and images © 2017 Thom Hogan
portions Copyright 1999-2016 Thom Hogan-- All Rights Reserved
Follow us on Twitter: @bythom, hashtags #bythom, #dslrbodies