Sony 24-240: yikes. Avoid at all costs.

After Sony announced the A7III and A7RIII, prices on the A7RII dropped pretty dramatically – dramatically enough that I decided to pick one up as a full frame mirrorless option.

There were several reasons for this impulse buy. I had several international trips on my calendar (Thailand, Cambodia, New Zealand, Portugal), and I didn’t want to carry my full D850 kit with me, particularly as some of my destinations involved a bit of hiking. But I’d been spoiled by the ridiculous image quality of the D850, and wanted something a little more than my lightweight M43 setup could give me. The A7RII seemed like the obvious answer: a relatively small and light body with image quality that rivaled the D850. But as I’d pointed out to many people, the problem is always the lenses. Small and light full frame lenses aren’t really a thing.

I dropped by my local electronics store, which happened to have an A7 mounted with Sony’s 24-240 super zoom. Having used super zooms on other platforms in the past, I can assure you that my expectations for this lens were well under control. I was not expecting Zeiss levels of IQ. But my oh my were those expectations not met.

To be clear: I almost certainly got a bad copy of this lens. I’ve been pretty lucky with most of my lenses in the past, but this one was a disaster. It wasn’t bad at all focal lengths, and I was able to end up with some decent shots, but note, for example, this shot from Porto:

full shot
1:1 at center of the frame

Or perhaps this shot from Lisbon:

Full Shot
1:1 at the edge of the frame, showing clear transition in quality

The bigger issue for me, I think, is that it calls into question Sony’s entire QC procedure. Sure, this is one of the cheaper lenses in the FE lineup, but at a retail price of $1000, it’s not exactly inexpensive. Coupled with Roger Cicala’s statement that Sony couldn’t find a good copy of a lens to send to reviewers if their future depended on it… I’m not sure I’ll be buying their lenses moving forward.

My current plan, at least until Nikon’s new Z series is widely available and the price comes down a bit, involves getting older manual focus lenses (primarily Minolta Rokkors) to use on the Sony. It’s been a fun, interesting experiment, and I may post some of my thoughts / samples at some point in the future. 

The practical takeaway? Don’t buy this lens. Definitely don’t buy it used – there’s a good chance it’s on the second hand market for a reason.

Skylum, ON1, and half-baked products

For me, it started with ON1 Photo RAW. For those of you not familiar with ON1, it’s a company that sold a Photoshop competitor oriented toward photographers. It had a niche following, was reasonably priced and somewhat well regarded, providing a significant amount of Photoshop’s functionality in an accessible workflow for a fraction of the cost. In the wake of Adobe’s subscription-only pricing model for Lightroom and watching photographers increasingly use RAW files, ON1 saw a market opportunity. For months, they talked up their new product – ON1 Photo RAW – that would combine the features of their previous product – Perfect Photo Suite – with a powerful, modern, built from scratch RAW engine that promised lightning fast speed and editing without Lightroom’s bloat. They enlisted Matt Kloskowski – a YouTube personality who made his name selling presets in Lightroom – to talk about how ON1 was so amazing that he would never use Lightroom again (spoiler alert: he’s still using Lightroom). So what could go wrong? Everything, as it turns out.

ON1 accepted orders for Photo RAW for several months, and after customers began to suspect it might be vaporware, the company set a  release date of November 2016. When it became clear they weren’t going to meet that deadline for the final product, they issued a “prerelease” version on November 23, promising a “full” release in late December.  Calling the November version of Photo RAW “prerelease” was, in my opinion, extremely generous. It had the feel of an early alpha build, an unfinished work product that no developer could have felt good about shipping. Not only did the prerelease version not deliver on its promise of a lightning-fast workflow, it crashed frequently and lacked core functionality (e.g. the crop tool was listed as “coming soon”). An early December build improved a lot of things, and ON1 promised to fix “all” the bugs in the two weeks before the final release (there were over 100 active bugs / issues when the early build came out). When release day rolled around, it was still clear that Photo RAW wasn’t really there, and ON1 laid out a series of updates they would roll out over the following months to fix everything and add all of the features they’d promised. By the time Photo RAW 2017.6 was released in August, ON1 had, mostly, delivered what they’d promised, but by that time a lot of photographers had moved on. And by that time, the hype train for Photo RAW 2018 was well underway, and it was clear the same thing would happen again – an early order cycle, an aggressive list of new features, and a series of “updates” to add core functionality well into summer.

ON1 isn’t the only company guilty of this, and actually I’m not sure they’re even the worst offenders. Skylum – at the time Macphun – employed a similar strategy with their followup to Aurora HDR 2017. There were dozens of promises made: a Windows version, a new HDR engine, the ability to save 32-bit raw files, and more. Skylum acknowledged explicitly that the Windows version wouldn’t have feature parity with the Mac product at launch, but promised that in “early 2018” the all marketed features would be available in both versions and there would be complete feature parity. I’m writing this article in September of 2018, and it’s safe to say those promises have not been fulfilled. The latest update – version 1.2 as of this writing, is still not feature complete, and the versions do not have feature parity. Meanwhile, every month or two, Skylum sends out a new email about how they’re spinning up a new project – an AI partnership, a DAM for Luminar, Loupedeck integration! – essentially fundraising for and diverting resources to their next product without completing their last / current one. And today, I received my first announcement for Aurora HDR 2019 which may, hopefully, but may not, actually have some of the features I was supposed to get in Aurora HDR 2018. 

Here is my problem with this business model: it’s basically Kickstarter for photo processing, and not in a good way. You know those companies that pop up every now and then that are always founded by “[insert prestigious school here] engineers” offering a breakthrough! product that is going to disrupt the whole industry? They’re able to generate a ton of support and startup funding, and inevitably find out that it’s a lot harder to produce a product than it is a prototype. Their project runs into issues, delays, and funding problems, and they end up six months to a year behind schedule, if they deliver at all. This, it seems like to me, is the basic business model Skylum and ON1 are using, though they’ve managed to hang around a couple of cycles longer than I would have expected. They focus on gimmicks and flashy-sounding features while neglecting the core functionality of the product. Having an “AI powered filter” is more important than, say, building a decent demosaicing engine. 

Companies like Skylum and ON1 may say their products are “buy once, enjoy forever,” but the reality is that their business model relies on selling new versions of their software. I’m obviously not privy to the sales numbers at either of these companies, but my guess is that the majority of sales are people buying in on the pre-order (kickstart), and that the long tail is pretty low. Which means, really, that they have every incentive to talk about how awesome next year’s version is going to be, and not a lot of incentive to deliver on any missing features from this year’s version; after all, they’ve already got your money.

Here’s the deal: with both of these companies, and any others promising features “coming soon in a future update,” you should always evaluate the software as if that update will never come. Because there’s at least a chance that it won’t. You’re buying the product as-is, and the company is under no obligation to make good on their promises. Say what you will about Adobe – and there’s lots to say – but they aren’t in the habit of making wild promises about future versions of their software, or failing to deliver on those promises. 

To be clear, not all small software outfits producing photo software are like this. For example, I’d give a big shout-out to Serif, the developers of Affinity Photo. No software package or developer are perfect, but I’ve been consistently impressed with the quality and project management I’ve seen out of Serif. 

My approach to ON1 and Skylum: caveat emptor. I’ll never pre-order a package from either company again, and I wouldn’t recommend anyone else pre-order either. There’s no guarantee the “feature list” will be present at launch, or ever. Only hand over your credit card if there’s a real, shipping version of the software that does everything you want it to do. Anything else is funding a prototype that may never materialize, no matter how badly you want it to.

Chattanooga Waterfront

A couple of months ago I was in Chattanooga for a meeting, and managed to bring my camera along for the first time. I was lucky to get a couple of interesting sunsets, though I was probably in the wrong place both nights to really get the pictures I wanted.

To blog, or not to blog

Long ago, before Facebook became the default method for posting everything, I had a blog. A couple of months ago, I stepped away from Facebook, but found myself without a venue to post pictures. I’ve decided, at least for now, to start a small blog, with a few pictures, some random thoughts on photography, and possibly some more in-depth pieces on my photographic process.