Despite the rather extreme length of the prior post on lenses, some parts were heavily abbreviated. In particular, in the section considering whether it’s worth thinking about the upgrade path to full frame when buying lenses today, I omitted one huge reason for not doing this – you can’t get wide enough on a crop camera (DX) with full frame lenses alone.  I did mention that there are really no wide primes for crop cameras (on Nikon), but this is mostly true of full frame zooms also.

Setting aside the very special purpose (and very expensive) 14-24/2.8 for a moment, it’s temping to say “not true, what about the recent 16-35/4.0 VR?” (or its Canon L equivalent)?  After all, the widest all-in-one zoom is the 16-85mm – so this gets at least as wide, right? On the surface of things, this is a true statement, and indeed 16mm (or the more standard 18mm starting point for most kit zooms) may be as wide as you need to go. Thus, the titular question – how wide is wide enough?

The problem is that a lens like the 16-35/4.0 is an ultrawide lens on full frame, but merely covers the wide end of the range on a crop camera (24-52mm equivalent). And here’s where you’re stuck in a “heads you lose, tails you also lose” situation:

• If you don’t need to go wider than 16mm (on a crop camera), then it seems like you’re set. But then as soon as you do upgrade to full frame, you’ll find you spent $1000 on an ultra-wide full-frame lens that covers a range that you decided you didn’t need.  Oops!
• If you do need to go wider than 16mm, then no full frame lens will do that for you.  Sure, the 14-24/2.8 will get you a hair wider for $1800, but that’s really not a viable solution.  Eventually, you’ll break down and buy a DX lens that won’t work on full frame – and also might not complement your other choices of full frame lenses.

The core of the above issue is that while lenses like the 24-70 are useful (IMO) on a crop camera, and the 70-200 even more definitively so, an ultra-wide full frame zoom covers an awkward and narrow range on a crop camera.  The “pro” recommendation to skip midrange zooms and go for a combination such as 16-35/4.0 + 50/1.4 + 70-200/2.8 is thus great for full frame, but significantly lacking when used on a crop camera.

Still, full frame aside, the question remains – how wide is wide enough? 18mm? 16mm? Less? This is a very personal question. I believe that most non-photographers don’t really have a need to get below 18mm – and certainly not below 16mm. It’s at least advisable to read something like Ken Rockwell’s article “How To Use Ultrawide Lenses” before deciding, since as he notes it’s tough to get good results.  And frankly, when you want people in the picture (which is often a main goal for the non-photographer) it’s even more difficult to use an ultra-wide well.

I’ve still got a ton to learn in this department, though I will note that for the non-photographer, “getting it all in” (criticized in the above article) can still be quite useful even if it creates a somewhat ugly picture.  As an example, consider all those indoor kids birthday parties:

I actually picked this image, shot with the Nikon 10-24, because (a) it was an enclosed space where it would have been impossible to capture the whole group @ 18mm, and (b) it shows how brutally distorted people at the edges of the frame become (not due to lens errors, just due to perspective).

Given the effect on people, I find I use the ultra-wide more often to capture places I’ve been rather than people (unusual, since about 80-90% of my photos are of the people I’m with).  Still, over time, you get to learn how to incorporate people into the frame – though that’s something I’m still working on.

Here’s an experiment creating an embedded SmugMug gallery of wide angle shots; everything here is wider than 18mm and most shots are closer to the 10mm side of the equation.  You won’t see this in your RSS reader (you’ll have to open the post on the actual site to see this); clicking the image outside the buttons will take you the gallery itself on SmugMug:

Almost everything above was shot with the Nikon 10-24 which I own, but #16 and #17 were with Wen’s Tokina 11-16 – a very nice lens that’s a bit more restricted in range, but faster (f/2.8 vs. f/3.5-4.5) and I think perhaps sharper than my copy of the Nikon. The Sigma 10-20mm was another strong contender. Overall this category of lenses is pretty pricey, but quite a bit of fun – you’ll be able to capture your experiences in a somewhat unique way!

The last post introduced the unsolvable question of picking the right lenses, suggested starting with what you have / a kit lens, and using the amazing capabilities computers give us to understand how we are using what we have.  You can’t tell what you would have used if you had it, of course, but it’s still a good starting point.  At least, it was for me.

Part 2: Understand The Basic Approaches

Having read a lot of online discussion before getting my initial D60 + 18-200 – and far more after jumping in – I came to believe that everyone comes at things from a particular perspective, and usually makes recommendations from that perspective as to what is “right”.  Of course, they’re all right – everyone is recommending what genuinely works for them, and your task is to figure out which recommender feels closest to you.

However, this discussion is often a little stacked; those serious enough to head online and give advice to others (myself included if you count this as such) are almost by definition more serious about photography, and this often weights the opinions that you’ll see in a particular direction.  Thus, if you’re a beginning non-photographer (that is, just moving beyond a compact/cameraphone but certain that your intent is better memories, not selling images or launching a new career), it may be helpful to understand how these approaches apply to you.

Note that these approaches are not mutually exclusive, but it’s typical to at least start with one or another.

The examples here are all Nikon specific, but I’m pretty sure anyone with knowledge could translate this directly into Canon or other terms pretty easily.

Approach 1: One Lens to Rule Them All

Just because you can change lenses with a DSLR, does not mean that you have to.  The benefits of near-instant response time when you press the shutter button, fast auto-focus, better image quality especially in low light, and others apply as much regardless of whether you have one lens or twenty.  Indeed, anyone with a DSLR wishes that there was the one universal lens that did everything they need, but alas, no such thing (yet).

Occasionally, you’ll see the philosophy that a single fixed (i.e. can’t zoom) wide angle lens is all you need.  In fact, anyone coming from a cameraphone as their main equipment will likely be quite familiar with this idea, and indeed by adjusting your perspective it’s possible to achieve quite a range of results despite a fixed focal length.  An example of this is the attractive, retro-styled Fujifulm X100, pictured below (the image is Fuji’s, and links to their X100 site). I can’t bring myself to pay even close to the astronomical $1,200 asking price – even though I like the concept, and like the controls (which I just realized are almost exactly what I asked for in the M/A/S/P post):

More often, proponents of the one-lens approach recommend a zoom that covers a decent range from the wide end through to at least a mild telephoto.  It’s much more typical to find yourself wanting something wider than it is to lack reach when dealing with one-lens situations (especially on anything less than a full-frame camera, which an all-in-one shooter is almost certainly going to be using). S0 18mm (on crop sensors) is usually the minimum you see here.  Nikon makes a whole boatload of lenses for this purpose; the more popular current ones are:

• 18-55mm f/3.5-5.6 DX VR – $100 kit lens that adds very little to the cost of the camera.  It’s wide enough, but it’s reach on the long end is pretty limiting, which is why it’s designed to be paired with something like the 55-200mm or 55-300mm telephoto zooms.
• 18-105mm f/3.5-5.6 DX VR – $200 kit lens that just about doubles the reach of the entry level kit lens, giving you a very usable range.  I recommended this as a good starting point to figure out what you really use.
• 18-200mm f/3.5-5.6 DX VR – $800 “superzoom” that covers an absolutely enormous range, but comes with a hefty price tag to match.  This is the only lens in this category that I’ve personally owned.
• 16-85mm f/3.5-5.6 DX VR – $700 lens that gets 2mm wider at the expense of some reach – but that is respected as having the best image quality in this class.
• 24-120mm f/4 VR – $1200 full-frame lens that gives similar coverage to the 16-85mm on an FX camera; there’s also a smaller, cheaper, variable aperture version.  Neither version seems to be too popular with users.
• 28-300mm f/3.5-5.6 VR – $1050 full-frame lens that’s equivalent to the 18-200mm but for full frame.  Again, only viable on a full-frame camera (28mm is not nearly wide enough on DX), where it’s unlikely that this will be your only lens.

I only mention Nikon offerings above, but many 3rd-parties like Sigma, Tamron, Tokina, and others have decent offerings as well, usually for much less money than Nikon (kit lenses aside).

The major appeal of this category is that it’s convenient (no changing lenses around and missing things while doing so), relatively light (vs. carrying multiple lenses), and inexpensive.  What you give up for this is wider apertures – almost all of the above are zooms that hit f/5.6 on the long end – and in most cases, some image quality. Bokeh (background blurring) in particular tends not to be great on almost any lens in this category.  And while the range of focal lengths is good, you don’t get that wide, or that long.

I started in this category, before growing out of it.  Per my last post, I think it’s a great place to start as you learn your way around and figure out what you really need.  In my case, I tried and loved the increased sharpness of primes, which was my first step out of this camp.  Also, in the D60 era, going anywhere above ISO 800 noticeably reduced image quality, so I tended to try and stick to ISO 400 or below, and this meant the lack of wider apertures hurt quite a bit.

I think my personal error was not starting with this category – it was definitely the right category for me at the time – it was that I should have gone with a $200 kit lens instead of spending four times as much on the 18-200. The 18-200 is supremely convenient and I still take it when I need a one-lens solution, but with the difference in price I could have been experimenting with primes and different lighting options instead.  Or having some really nice meals.

Still, there’s many pictures of the family – including nearly the entire first year of Olivia’s life – that the 18-200 captured for me, and despite having acquired quite a bit more gear it’s still a definite candidate when I’m getting on a plane and don’t want to bring multiple lenses along.  Here’s a rather poor picture of my actual lens – highlighting how what a pro photographer knows is still vastly different from what I know:

Telltale non-photographer signs above (besides the lack of overall quality):

• Using a blue polka-dot baby blanket to reflect some light back from my one and only external flash, which you can clearly see in the front element reflection.  Sorry, it’s all I had!
• An F75 film SLR standing in as an “arm model”, like the owner of an anonymous arm used in a Rolex commercial who is there just because they need something – anything – to attach the watch to.  I’ve only got one DSLR body!
• Black background with the plastic tripod visible.  I have nothing resembling a studio (and don’t need one, as that’s not where my kids play); the above was taken in my kitchen!

Approach 2: A Set Of Primes

Don’t worry, this one isn’t a math question about prime numbers.  Prime lenses are just another way of referring to lenses that operate at just a single focal length (like the Fuji X100 above) and don’t zoom.  If you want the subject smaller, you step back; if you want the subject larger, you step forward.  Of course, stepping backwards isn’t possible in an enclosed room, and stepping forward when you’re shooting the Grand Canyon can certainly bring you closer to its rock formations in a hurry, but probably not in a manner that you’d like.  Plus if you’re shooting landscapes, you’ll need a fairly impractical number of steps to adjust your perspective, but hey, if you’ve never climbed Everest, go for it!

How many primes you need depends on what you shoot and how many of those aforementioned steps you’re willing to take. Nikon seems to think that the answer is “as many as you can carry”, and thus you see 14mm, 16mm, 20mm, 24mm, 28mm, 35mm, 50mm, 60mm, 85mm, 105mm, 135mm, 180mm, and 200mm+ primes available, with the 24, 35, 50, 85, 105, and 200 being quite current.  There’s many others in between if you look, like the Nikkor 58mm f/1.2 Noct, which seems to fetch almost $4,000 used.  200mm+ primes  are usually rather expensive (thousands of dollars), and bulky enough that only pros with specific needs venture there.

Primes typically give you the best image quality – and historically, a much lower per-lens weight – at the focal length they cover.  They are also typically quite fast; few primes are “slower” than f/2.8, and the fastest are f/1.4.  A shot that’s ISO 200 @ f/1.4 would be ISO 3200 at f/5.6 – the difference is deceptive, because light absorbed is proportional to the square of the aperture.  However, the fixed focal length means you better be prepared to switch lenses!

With that as context, a few comments on going this route as a non-photographer, based on my own personal experience:

• I started with a very standard 50mm prime; I got the 50mm f/1.4D ($350), but the 50mm f/1.8D ($130) works just about as well.  On a crop camera, this is actually pretty long – it’s useful for portraits but you’re not going to get a group shot with a 50mm lens.
• I received the 85mm f/1.8D around the same time, and found that this was essentially as long as I wanted/needed to go on DX.  I used to joke that if I’m shooting my kids @ 200mm, then I’m too far away from them!  The 85/1.8 is a great lens and also only about $350-400.
• A prime lens that no Nikon shooter with a DX camera should be without is the 35mm f/1.8 DX.  It doesn’t work on full frame, but it is a very useful focal length on a crop sensor, and produces great results for its low price (now $280 new) and size.  I pretty much stopped using the 50mm f/1.4D after this came out; the 35 + 85 combination was just a better match for most things.
• Sadly, the wide end is a real problem if you’re using any DX (i.e. consumer) camera.  The widest prime you can economically find is probably the 24mm f/2.8D, and that’s not wide enough on DX.  Thus while I liked primes a lot, I got the 10-24mm DX ultrawide zoom to cover the wide end. Thom Hogan frequently calls for Nikon to wake up on this one, and I hope that they listen to him and do.
• If you think primes sound interesting, buy a D90 or above.  In fact, with the recently released D5100 having the same sensor and image quality as the D7000, and pretty much every modern zoom having AF-S and working just great on the D5100, one of few major reason to get something better is if you want to use older primes that require a focus motor.  The 50/1.8, 50/1.4D, and 85/1.8 all still require that the camera has a built-in focus motor; otherwise, you’ll be manually focusing them.

Nikon has modernized some of their primes recently, but for some reason they seem to think that only people made of money are interested in them (though an upcoming 50/1.8 AF-S addresses this at least a little).  Their 3 most recently updated primes – the 24/1.4, 35/1.4, and 85/1.4 – all cost $1700 or more, each.  That makes this set containing all three a “bargain” at just 4,900 Euros (photo origin unknown, I found it on Nikon Rumors but Nikon is likely the original source):

So, get a camera with a focus motor and get some older, cheaper primes. Unless you ARE actually made of money.

Approach 3: The Holy Trinity

On a crop sensor, covering the entire range of meaningful focal lengths requires 10-12mm on the wide end, and 200-300mm on the long end.  This offers everything you’d need from an ultra-wide perspective through to shooting faraway things (like a bank account that allows you to retire, after buying all this stuff).  No all-in-one solution does that, and on crop sensors, no set of primes does that either – and long primes cost a veritable fortune (the 300/2.8 is $6,600).

When I started looking beyond the 18-200, there was a full frame set of three high-end zooms – dubbed “the holy trinity” – that covered a wide range while providing image quality comparable – and in some cases superior – to equivalent primes.  Pulling this 0ff was something of an engineering marvel, as prior to this zooms had generally been considered a convenience that was going to cost you in terms of image quality.  Not anymore!  The only compromise was a maximum aperture of f/2.8, but that’s pretty reasonable for most subjects.  There were three members of this “holy trinity”:

• 14-24mm f/2.8G AF-S
• 24-70mm f/2.8G AF-S
• 70-200mm f/2.8G AF-S VR

Even used, each of the above goes for about $1,500, and they’re considered pro zooms.  But they’re a benchmark for what you can get without compromise.  And they’re expensive particularly because they are full frame lenses. Still, you can come up with much more economical combinations with the same philosophy – an ultra-wide zoom, a mid-range zoom, and a long-zoom.  Here’s what I’ve tried or considered in this range:

• 10-24 DX + 24-70 + 70-300 VR.  This combination costs a little over half of what the above would cost, and compromises with variable aperture zooms on the wide and long ends, but actually covers a much wider 10-300 range (compared to 14-200 above).  This is the combination I own and use, but I can’t recommend it for everyone due to the size & cost of the 24-70.
• 10-24 DX + 18-55 VR + 70-300 VR.  Swapping out the expensive pro 24-70 for a kit lens like the 18-55 takes this same approach but chops the price and size of the combination down enormously, which still retaining the 10-300 range.  You could also use the 55-300 DX VR zoom instead of the bigger full-frame 70-300 VR, but opinions of the latter are much better than of the former.
• 10-24 DX + 35/1.8 + 55-200 VR.  Even better than the above is breaking the rules a little and using the 35/1.8 prime to cover the middle of the range.  I’d definitely do  this over the above, but it does make the 55-200 (or 55-300) options more appealing to get a bit better mid-range coverage.

In the latter two combinations, the 10-24 DX contributes the largest part of the cost.  There are alternatives here, like the Tamron 10-24 (much cheaper, though reportedly lower image quality) or others – and you may not really need ultra-wide coverage below 18mm.

Of course, nothing says you need to go with three zoom lenses.  Notable two-lens combinations are:

• The classic 18-55 DX + 55-200 DX pair that provides great coverage, at half the price of an 18-200mm with a little less convenience.
• The 16-85 DX + 70-300 VR pair is something I think would work really well for a lot of people; it’s about three times as expensive at $1200, but both of those lenses are considered to be top notch from an image quality perspective.  The 16-300mm range is very decent unless you really want ultra-wide coverage.

Approach 4: DX Now, Full Frame Later

Should I favor (or exclusively purchase) full frame lenses over DX counterparts that only work on crop sensors?  This question comes up because almost every non-photographer is going to have a crop camera (anything below the D700 on Nikon, or the 5D on Canon), but will nonetheless wonder if they might upgrade someday – especially if full frame camera prices drop far enough.

In most cases, the answer to this question is: no, you won’t.  Why?

• Crop sensors today are really good.  When the D3 (Nikon’s first full frame camera) showed up, you’d be hard pressed to go beyond ISO 800 on a crop sensor.  Today, you can do ISO 3200 with the same approximate image quality on a crop camera.  Resolution (short of the $8,000 D3X) isn’t notable different between crop and full frame sensors, nor are other parameters like dynamic range.  Do you really need to get to ISO 6,400 or ISO 12,800?
• Full frame won’t be mass market, and will thus continue to be expensive.  If Nikon (or Canon) thinks they can get you for $2,000 a lens for full frame lenses (and they do), you can bet they’re not planning for a $1,000 full frame camera.
• You’ll still wind up with some DX gear that will inhibit switching in the interim.

Despite this reasoning, I still think I’ll go FX someday, and thus I now favor full-frame lenses.  This is because:

• I’m too foolish to follow my own advice.
• When I’m old, I think I’d rather have nice pictures to look at (with the 0.5MP eyes I will have), than memories of a nice car or a bigger house, so I’m prepared to spend proportionally more on this, mitigating the cost issue somewhat.
• I already have a decent collection of full frame lenses, and tend to like primes which are mostly designed for full frame anyways.
• I found the images from the D700 I used for a while to be really nice, notably better than my D7000 even with everything else held constant.  I don’t know if full frame is a requisite ingredient to achieve this, but despite a 2-3 year technology gap it’s still the best camera I’ve used – and I’m really wondering what it’s replacement is going to do.

Once again, I think the above are all factors that are sort of unique to me, and not true of most non-photographers.  Buying for DX is significantly cheaper, and if you go used, it’s not expensive to make the switch if you were wrong!

Approach 5: I Want It All

If you bumble around without a strategy for what approach works best for you, you may wind up staggering around through each of the above, picking up a little gear along each step of the way.  That’s exactly what happened to me:

• I started with the 18-200 all-in-one mentality;
• Then I warmed to prime lenses, using the 50/1.4 overwhelmingly, followed by the 35/1.8 + 85/1.8 combination;
• I discovered the DX wide prime hole, and plugged it with the 10-24;
• Around this point in time, my daughter started running – sometimes towards me – and I doubt even a pro could change lenses faster than she runs.  I got the 24-70 which quickly became the most popular lens in my collection.  Now running kids was slightly better handled – a shot seconds before this was at a very different focal length:

• With the above, I got a D700 (which wasn’t ultimately intended for me), which made clear that full frame was a distinctly attractive possibility to me despite all the drawbacks.
• With the 24-70 as my new favorite, coverage on the long end became more important as I was no longer bringing the 18-200 out; but 70mm maximum is short on the long side.  Hence the 80-200/2.8 and 70-300 VR (both of which I have to thank my stepfather for).

At least I can credibly claim that I understand the different approaches to putting a set of lenses together, having tried pretty much all options!  That said, this was certainly not the most economic route to get to the set of lenses I most actively use today.

Perhaps I do this to convince myself I’m not a fool for accumulating all this stuff, but a real benefit of having gone through the above is that I now have a choice – every time I head out and bring the camera, I can choose what makes the most sense for that outing. Sometimes, just the 35/1.8 is nice and small, and good enough for the situation.  Often, I just bring the 24-70. On a one-day, no-checked-bags trip to New Orleans, it was still the 18-200. Today at the park, it was the 10-24, 35, and 105. Earlier in the weekend, the zoo called for something like the 70-300, which was helpful in getting this shot that I sort of like – he (or she) seems to be wondering what lenses he needs too:

It’s definitely a luxury to have the choices I do – good thing I’m not into cars or boats or something else costly!

The moment you get sucked into the DSLR racket – or even as you consider it – you start the never-ending journey of asking yourself what lenses you need.  There’s really no way out of this; even if you bought every lens available for your camera (recommendation: go with something like the Sony NEX-5 that only has a handful of lenses!), you’d still probably check nikonrumors.com to see if anything new came along that you now need.  Almost exactly three years ago, I bought my first DSLR (Nikon D60), with a “do all” 18-200mm zoom lens that I never intended to take off the camera.  10 lenses later, I can’t say that this was a particularly accurate prediction.

Still, about half of photography forum chatter seems to be either “what lens should I buy” or “what lenses should I take on a trip to… my backyard”.  Amazingly, this question comes up a lot offline too; in one day, I wound up E-mailing my brother about lenses for a wedding his friend asked him to shoot (major pressure for a non-photographer!), and my friend Herman who is getting further along in pondering his switch to Nikon.  A little planning here is important, so you don’t wind up like Greg from my work, with a 16-85, 24-120, and 28-75 in a collection of just five or so lenses total (due to the significant overlap).  Or like me, with 10 lenses instead of a planned 1.

Lens selection is a 7-way trade-off between focal length coverage, maximum aperture, build quality, price, size, weight, and features (AF, VR, macro, etc).  Some of those (e.g. size & weight) are linked, others less so.  If the best lens wasn’t highly situational, there wouldn’t be so many of them.  This creates the usual photographer vs. non-photographer dilemma; what’s very good advice for an aspiring photographer can be very different from the right choices for a non-photographer.

With that said, how and what should a non-photographer choose?

Part 1: Learn From Yourself

Especially if you’re just starting out, there’s two things hugely in your favor when it comes to choosing lenses:

• A highly competent kit lens is just $150-200 over buying the camera body only and getting alternative lenses separately. On Nikon, you’d get a 18-105mm zoom with VR, which will already far outperform any compact you may have owned.  This will be sufficient to learn what you really need, and gives you a good initial range.
• Lenses actually hold their value quite nicely, and as a result, if you ever decide that you want to go a different route, you can often recover close to the full value if your kit lens if you didn’t get the very entry level 18-55mm that wouldn’t be an upgrade for anyone.

The nice thing about digital photography is that every choice you ever made is recorded somewhere, just waiting to be analyzed – what lens you used, at what focal length, aperture, shutter speed, whether you used flash, ISO, etc.  If you use Lightroom, you can filter by meta-information in the library view to get  some insight – like what focal lengths you use most often with a given camera & lens:

Now, Lightroom isn’t free (especially relative to a $150 kit lens!), and even if you have it, the above presentation isn’t great. Fortunately, there’s free tools that can extract EXIF information from your JPG files (or directly from raw files).  Several are available, but one I’ve used is ExifPro, available at http://www.exifpro.com/; it’s free though if you use it regularly you can buy a license.

Using ExifPro, I just select all my photos, add columns for camera and lens type, export the EXIF data to a text file, and then use something like Excel to graph your actual usage.  In a few minutes, you can have a simple chart like this:

This is my actual usage of my first DSLR, the D60, with “only lens I’ll ever need” – the 18-200 VR.  Looking at the graph above, what do you think I bought first – an ultra-wide lens (covering less than 18mm), or a longer telephoto (more than 200mm)? Not hard to guess the answer to that question; I was stunned at how often I was just zoomed out as wide as things would go. Another graph similarly tells the tale on aperture:

As you can see, I was pretty much shooting at maximum aperture (which ranges from f/3.5 to f/5.6 on the 18-200, depending on focal length) except on a few rare instances – with anything above f/16 being an outright error on my part.  Not surprising given how often the kids are indoor in poor light; the D60 was a great camera but not one you’re going to push to ISO 1600 if you can help it.  The next lenses I got after the 18-200 were thus fast primes (for other reasons besides aperture), followed by an ultrawide to address the first graph.

Of course, you don’t need to do data analysis to have a pretty good sense of what you’re wishing for.  And the most important attributes might not be things you can measure as easily – better sharpness, fast lenses for less motion blur or lower ISO, improved contrast in harsh conditions, more control over depth of field, nicer background blurring (bokeh), or any number of other things.  You’ll know what you want.  And all you’ll need is a dozen lenses to get it :).

I’ll thus leave it at this simple suggestion – start with a reasonable kit lens like the 18-105, see what you’re really doing, then decide on the more expensive stuff.  You can always sell the kit lens later, but you’ll learn what you need because nothing you read on the web – including what I’ll post next – is going to be exactly what YOU need.

Looking at some pictures my friend Wen shot at Olivia’s birthday party, indoors at a gymnastics-themed place, I realized that I should have opted for a higher minimum shutter speed – even if it came at the cost of having a higher ISO setting, and thus more noise, in the pictures.  After all, you can add some noise reduction after the fact, but there’s no blur reduction!  However, this made me realize that despite the existence of a large number of shooting modes, none of them actually does what I want.

Basically, every Nikon DSLR (and many film SLRs before it; same story for other brands) has the same core set of shooting modes that you can choose from:

Almost regardless of which camera you pick in the line-up, you’ll essentially get the equivalent of the dial you see above:

• Professional-oriented bodies – the D300, D700, and D3 models and variants – only give you the M/A/S/P  shooting modes.  These modes are the ones that it really makes sense to use anyways once you get the hang of things; more on this below.
• The D7000 pictured above adds a couple of programmable user  modes (U1/U2), an Auto mode where the camera picks everything, a useless flash-off mode (just push down the pop up flash or turn off the attached flash!), and access to scene modes (with pre-defined settings for portraits, landscapes, etc).
• The D90, D5100/5000, D3100/3000, and other cameras even back to my film-based F75 have the same M/A/S/P options, Auto, plus individual icons for selecting different scene types directly from the mode dial.

So basically, you always get M/A/S/P, and depending on what model you have, you might also get some automatic modes which are probably not why you decided to buy a DSLR anyways (but can be handy for learning).  The above is Nikon’s terminology, but Canon and other manufacturers all basically seem to take the same approach.  Each of the M/A/S/P modes (and most scene modes) control how aperture and shutter speed are chosen when taking pictures.

If you’re not familiar with how these affect what your pictures look like, it’s well worth understanding.  I can’t do the topic justice here, but I’ll provide a one-liner for each anyways.  Aperture controls how “open” your lens is – the wider the aperture (lower f-stop number), the more light it lets in, but the less depth of field you have (not as much will be in focus).  Shutter speed controls how long the sensor takes a picture for; longer (slower) shutter speeds let in more light, but if the object you’re taking a picture of moves during this time, you’ll get a blurry shot.  It’s all a trade-off – you generally want more light and it’s a question of what you’re willing to pay for it.

With that in mind, the M/A/S/P modes are very simple to understand:

• M – Manual.  You set both aperture and shutter speed to what you want.  Set them wrong, and your pictures will be too dark or blown out.
• A – Aperture Priority.  You set aperture, and the camera picks the matching shutter speed to get the right exposure (brightness).
• S – Shutter Priority.  You set the shutter speed, and the camera picks the aperture.
• P – Program Auto.  You set neither, the camera chooses for you – but allows you to adjust between options that allow an equal amount of light.  For instance, 1/200th @ f/4.0 is equivalent to 1/100th @ f/5.6, in terms of how bright your picture will be.

I usually use Aperture Priority, because aperture usually has the most impact on what you pictures look like.  Sometimes you want everything sharp, sometimes you deliberately want the background to be blurry to focus on the subject.  Many lenses aren’t quite as good wide open, but you get less light if you use a smaller aperture. Indeed, the difference between an inexpensive lens and a ridiculously expensive one – unless you look really close – is often in how wide an aperture is supported.  Stop down to f/8.0, and many lenses are hard to tell apart.

By contrast, shutter speed matters but only really makes a difference in one of three scenarios:

• If you are holding your camera in your hand (usually the case, especially for non-photographers like me), then a slower shutter speed increases the chance of a blurry picture because YOUR hands are shaking.  The rule of thumb here seems to be to shoot at a shutter speed that’s the inverse of your effective focal length.  So if you’re zoomed to 100mm, then use a shutter speed of at least 1/100th. Technologies like vibration reduction [VR] (called image stabilization [IS] on Canon) minimize the effect of camera shake and allow you to get away with longer shutter speeds without inducing blur.  You almost never want camera shake in your pictures.  For this reason, there’s a minimum shutter speed you typically want.
• If you are taking pictures of a fast moving object, like your kid playing sports, your kid jumping around in a gym, or your kid just being your kid, and you want to freeze the action, you may need an even higher minimum shutter speed.  Much depends on how fast things are moving, but 1/200th to 1/500th of a second is around what you may need.  Getting this fast is a challenge indoors in low light, though.
• If you want deliberate blur, you may want a lower shutter speed.  Non-photographers will rarely use this, but often in pictures where there’s flowing water, you’ll see this done on purpose.  Sometimes, some amount of blur of this kind is also desirable to highlight motion.

Okay, so the above is a bad explanation of photography basics that was much longer than intended, though necessary to make the point I opened with. The problem that I see is that each of the M/A/S/P shooting modes is a legacy from the days when shutter and aperture were the only things you could adjust. Every digital camera, even the one in the iPhone, now gives you at least one if not two added options:

• ISO – this controls the sensitivity (gain) of the sensor in your camera.  It used to be impossible to change this in the film days, except by changing the roll of film in your camera, which few people did at least until they finished the roll.  Now, it’s as easy as changing anything else.  ISO is also the easiest choice to make – higher ISO means more noise (grain) in your images, so you always want the lowest ISO necessary for the aperture & shutter speed you’re using.
• Flash output – the camera can actually control how much power to use when activating the flash, which is pretty handy both for not blinding your subject, not draining your batteries faster than necessary, and getting the picture you want.  How this happens is a whole other complex topic, which we can skip for now.

So, with all that in mind, here’s how I wish my camera worked:

• First I pick the aperture I want, knowing that wider apertures means more light, less depth-of-field (i.e. blurry backgrounds), and perhaps less sharpness (depending on the lens), with narrower apertures doing the opposite.
• Then I pick the minimum shutter speed I want (unless I want to blur some flowing water, in which case I will just want to set the shutter speed to a fixed value). This the faster of:
  • The camera shake minimum speed, which in turn is determined by the focal length (longer focal lengths = higher required shutter speed), whether or not the lens has VR (and how good it is), and how steady the camera users hands are.
  • The subject movement speed, which can vary from totally still, to a speeding race car, all the way up to the average movement speed of your kids.  (Relative to the camera, the kids indeed can move faster than a racecar!)
• Then I want the minimum ISO necessary to get the right exposure given the above.

This seems pretty logical (to me, anyways!).  Flash complicates this otherwise simple formula, and the way the D7000 handles this is one of my few complaints about the camera, but that’s a topic for another day.

Can the camera work this way?  No.  Despite all those shooting modes and lots of menu options, it actually can’t.  There are two options that come close, and both depend on a feature that exists on Nikon cameras and most others called Auto-ISO, which automatically raises ISO if shutter speed falls below a specific value that you can set. The two options, both of which have their own failings, are as follows:

• Use Aperture Priority mode, with Auto-ISO on, and the minimum shutter speed set to whatever you need.  This mostly works,  it’s how I have my camera configured by default, and it’s how I take most non-flash pictures.  What doesn’t work?  Ergonomics.  While you can normally change shutter speed with one touch by rotating the rear command dial (which is easily accessible), you can’t change the auto ISO minimum shutter speed this way.  The latter has to be changed via a deeply nested set of menus.  Even if you assign one of very few programmable buttons to take you directly to this menu option (as I do), you still need to navigate through things on the back LCD and scroll through all available shutter speeds to find the one you want.  Did you zoom in or out?  Different required shutter speed.  Did your kid switch from the balance beam to the trampoline – or did they stop for a second allowing a portrait shot?  Vastly different shutter speed requirements.  It’s just impossible to use the menus to adequately adjust for this on the fly.  At Olivia’s birthday, I stuck with a default of 1/100th – 1/125th for most shots as it was impossible to switch shot-by-shot, but this was more than necessary for some and not nearly fast enough for others.
• Use Manual mode, with Auto-ISO on.  In manual mode, the minimum shutter speed you set for Auto-ISO doesn’t do anything, so shutter will be whatever you tell the camera to use, and you can set the shutter speed via the rear command dial (which means you can adjust very quickly without menus).  What’s the problem?  Now you’re setting the absolute shutter speed, not the minimum shutter speed.  Since Auto-ISO can’t go lower than the base ISO of your camera (ISO 100 in the case of the D7000), very bad things happen once you get to base ISO – any additional light means your picture will be overexposed and blown out, potentially quite severely so.  But ironically, you want the lowest possible ISO for the best image quality – so if it is bright enough for you to get to base ISO in at least some shots, you are constantly trying to adjust your shutter speed for the lowest speed that is both higher than your minimum requirements and doesn’t cause overexposure.  This is near impossible, and you will lose pictures if you try (at least I did, in dynamic situations).

What I’m asking for is actually trivial to implement – in Aperture Priority mode with Auto-ISO enabled, allow the rear command dial (which sets shutter speed in manual mode) to set the minimum Auto-ISO shutter speed (and show this in the viewfinder).  Sadly, I have no friends or family who work at Nikon, so they’ll never see this, and even if they did, updating firmware is sadly not Nikon’s thing.  If the firmware was open source, I’d add this myself, but unfortunately it isn’t.

If I wasn’t just tweaking the current interface, then this whole mess of M/A/S/P + ISO vs. Auto-ISO + aperture/shutter settings could be simplified much further:

• One setting for aperture, which can be a specific value like f/2.8, f/5.6, or AUTO.
• One setting for shutter speed, which can be any valid shutter speed or AUTO, plus a qualifier of whether the shutter speed is a fixed value or a minimum requirement.
• One setting for ISO, which can be a specific value like ISO 100, ISO 400, or AUTO.  I’d probably just replace the mode dial with this.

The above would take fewer controls and less menu items than exist now, and is essentially about just treating ISO as an adjustment like any other setting – which it has been, ever since digital cameras were introduced.  It’s kind of amazing that with so many knobs and options, the simple one I want isn’t somewhere to be found amongst them!  Just as amazing is that even compact camera designers, like Canon with my PowerShot S90, has the same idea about how things should be (Av = A and Tv = S in Canon’s terminology):

 

Well, here’s to hoping that some future iteration of cameras simplifies things down and just does what seems to make the most sense, at least for how I use the camera!

One page I spent a good amount of time writing up earlier was this one on processing, in which I really just wanted to recommend Lightroom, but wound up rambling at some length about how processing is especially important for the non-photographer because of the latitude it provides to fix basic mistakes that a professional would never make.

I didn’t talk at all about one important thing that processing helps you do, though, which is black & white conversions.  At least one reason for the omission is perhaps that I’m pretty bad at doing such conversions! However, I do think some black & white pictures look better than their color counterparts, so when I saw this blog post from David duChemin for a $4 (regularly $5!) eBook on better B&W conversions using Lightroom, I didn’t waste much time before buying it.  I haven’t gotten through everything yet, but have already learned several things I didn’t previously know – I’d highly recommend it.  I don’t know why it’s called a “Masterclass” but the book itself is written assuming you know nothing about B&W, so don’t let the title fool you.

At first it might be counter-intuitive as to why getting a black & white image isn’t trivial; after all, you can just set your camera to black & white, and even simpler tools like Picasa have one-button B&W conversions. But in reality, there’s a huge amount of choice in how you do the conversion, including on how each color is converted to some shade of grey. I personally find this more difficult than color processing; you have to make many more choices about what you want the end result to look like. Whatever you do, don’t set your camera to B&W mode!

Here’s an example image from Olivia’s 3rd birthday, which we just celebrated a couple of days ago with a gymnastics-themed event.  The original color photo, with fairly limited processing, is as follows:

Yes, I wish I could get rid of that hand in the upper right too, but as typical for kids you don’t get to set things up or get a 2nd chance at a shot!  Still, I kind of liked the picture, and as I was reading the book I thought it would be handy to pick a reference image to follow along with and decided to go with this one.

Lightroom’s default B&W conversion of the original photo looked like this:

It looks OK, but not too encouraging, I personally prefer the original picture over this B&W conversion.  Now, I’m still not going to win any awards or even pass B&W 101 with the 30 minutes of added knowledge I gained today, and you may think that this is even worse than the default conversion – but it’s still a bit closer to what I had in mind:

While there are a few non-B&W related adjustments, like a little cropping and a decent amount of post-crop vignetting (which is especially visible when looking at the small version of the image above – click for a larger version where it’s less noticeable), a good amount of the difference comes from how specific colors were blended to create the final result. Most notably, I brought the blues down a lot to try and separate Olivia a bit more from the environment (and note how the default B&W conversion makes the mat on the right look monotone, even though from the color photo you can see it’s high contrast red vs. blue).

In any case, this is just scratching the surface of things (as am I); I just hope it reinforces how much flexibility you can get out of processing your photos in different ways!

[Picture of the lens was taken by Nikon]

I’ve wondered a few times whether a non-photographer like me can make use of a macro lens as part of their toolkit. I think the answer is still “no”; since my definition of non-photographers is people who shoot to remember things vs. shooting to create art, it seems fairly unlikely that what you’d want to remember is the hair on a bug that turned up.

Still, when an AF-S VR Micro-Nikkor 105mm f/2.8G IF-ED showed up on Craigslist, in new condition, at what looked like a pretty reasonable price (asking $750), I finally decided to take the plunge.  With under 10 pictures since I picked it up last night, and no natural light to work with, I can’t say I know anything about it yet, but hopefully that will change.  If not, the nice thing about buying used is that you can always sell a lens for roughly what you paid for it!

Now, in defense to my credentials as a non-photographer, I thought it’d be nice to have one macro lens for rare occasions, but most of my interest in the lens was that it’s supposed to be stellar for portraits too.  All data seemed to indicate that the 105VR is as sharp as my 85/1.8 at f/2.8 (which is about as wide as I go with the 85/1.8).  But at 85mm+, with my shaky hands, the addition of vibration reduction on the 105/2.8 is very attractive, and while it’s not as acclaimed as the 85/1.4 (which I’ve never touched) for portraiture, the consensus seemed to be that it would outdo the 85/1.8 for this purpose.

Finally, since the amazing 24-70/2.8 is by far my most used (and thus carried) lens, the 105mm is a better complement at the long end than the 85/1.8, which I don’t carry or use if I’ve got the 24-70 with me; it’s not worth it for 15mm extra.  So if I’m out and about with 3 lenses in a situation where I can’t predict in advance what I’ll use, I’ll likely bring the 10-24, 24-70, and either this 105 or the 70-300.  Big difference vs. the 70-300 at the long end, but I rarely go that long.

On macro – it didn’t take too much playing around last night to realize both how little I knew and how much there is to learn if you want to get half decent macro shots.  I have a reasonable sense of depth of field on “normal” shots, but I was just stunned how little there was when operating at very small distances.  The tiniest back-and-forth motion of the camera completely knocked any handheld shots out of focus.  Finally I stopped down a little, got out the flash and tripod, and found an innocent strawberry to throw up on the chopping block for the first real test shot for the lens:

That’s not even close to the minimum focusing distance of the lens, and already just the surface of the strawberry is in focus (it’s stem is somewhat blurred).  Always hearing that diffraction kicks in after f/11 (not to mention my flash power kicking out), I bumped it up to f/11; things were a little clearer:

The edges of the strawberry are still not too sharp if you look at the bigger version, though at least you can see some texture in the stem now.  Still, it sure does highlight that depth of field kicks out much faster than diffraction kicks in!  At least know I understand why anyone would bother with techniques like focus stacking (where you digitally combine multiple shots of the same object that each have a different focus point)!

The other thing I learned from the exercise is that the exact positioning of the light source matters that much more at macro distances, because the surface texture and details look so remarkably different.  The two shots above look quite different in this regard, even though the change to the position of the flash was relatively minor.  I like the top better for primarily this reason.

I don’t think I’ll be taking many macro shots, but I have to admit it was pretty cool looking at some of the shots that were much closer than the two above, because there’s so much you don’t see with your naked eye.  The 105VR seems well suited to the macro task, though, and I’m really looking forward to trying it out in good light with the family on some definitively non-photographer shots!