More in the April 2007 issue of MBUK, due out on March 14th.
I still think raw is unquestionably the best way to extract the maximum possible quality from your camera, but if you're unable or unwilling to devote the time to process all your pictures there's a lot to be said for sticking with jpeg. Other reasons to consider jpeg over raw include:
- frees up more space on your card and hard drive
- in-camera noise reduction at high ISO settings can sometimes work better than third-party software
Ken's tips are specifically directed at Canon shooters, but equivalent adjustments can be made with any dSLR. His advice seems sound to me, so if you're disappointed with the results you're getting I'd suggest running some tests and experimenting with your camera's settings.
The light levels on this indoor climbing wall were terrible, but I didn't want to use flash. Because I had hundreds of images from this event I opted to shoot jpeg and turned the in-camera noise reduction setting to 'high' for these shots, which effectively killed the noise. It's one of the rare occasions when I chose to shoot jpeg rather than raw for quality reasons.
Magazine cover pictures don't get much attention from most readers, but that's kind of the way they're supposed to work. After all, the whole point of a great cover shot is to draw browsing eyes in at the newsstand, entice a quick flick through and - hopefully - encourage a trip to the till.
But that doesn't mean there isn't a lot of thought put into them. Quite the opposite, in fact, and mountain bike mags - for many years a bastion of the 'pick a nice shot from one of the features we're running and stick it on the cover' school of thought - have become more switched on to the impact a well-planned cover shot can have on sales. In a tough consumer magazine market, it's time and money well spent.
Here's an example from last year. MBUK wanted a shot to illustrate a technique feature on fast descending. So far, so much like any number of previous MBUK covers. But there were a number of specific criteria the picture had to meet:
- it had to be downhill (natch)
- it would preferably be rocky
- there'd be blue sky in the background
- there'd be plenty of motion blur
- the rider needed to look aggressive and confident
- the bike would probably not be airborne, but if it was it needed to be low to the ground
Just to add to the interest, a cover mount flap meant that most of the image had to be concentrated on the right hand side of the cover (which would also mean the rider travelling from left to right). MBUK's art ed, James Blackwell, even took the unusual step of sketching an image of how they wanted the cover to look for me.
So, armed with one of the most detailed briefs I've ever been given for a cover shoot, it was time to do some brainstorming.
Planning is everything for a shoot like this, so time spent at this stage can save a lot of hassle later on. Searching for a location can easily eat up a whole day, and experience has taught me that it's better to go straight to a specific place that I know will work (or at least will probably work) and that's close to at least one alternative. The fast, rocky and downhill requirements were easy to meet; the blue sky less so.
But I could think of one location that I'd used before that allowed a nice low angle and had all the other elements we needed. The only downside was that we'd need to be there early in the morning. And, assuming it was sunny, we'd have a window of just an hour to an hour and a half to get the shot before the rider would be completely backlit.
The weather on the day turned out gloomy and overcast, which was both a blessing and a curse. On the one hand, it potentially gave us a longer window of opportunity. On the other, it meant a lot more Photoshop work to insert a believable-looking blue sky behind the rider. Not my problem, ultimately, but I prefer the look of a 'real' blue sky on the whole.
Still, you've got to work with what you have. After a quick check to make sure that Oli Beckingsale, our rider for the day, was happy with the location (he was - a rider who isn't happy, incidentally, will never ride at their best, so this stage isn't just a formality) it was time to set up some lighting to lift the dull day. Setting my main flash to the right and below Oli, I positioned a second flash to the left and behind as a rim light. It's not quite the same as sunlight, but it adds contrast and punch in flat lighting.
And, um, action...
I quickly discovered that shooting from such a low angle - a choice forced on me by the need to include as much sky as possible - meant that I could only hear, rather than see, Oli approach. After a couple of missed shots I asked James to count me down... after which I had to rely on quick reflexes and a certain degree of good luck in getting the shot.
As usual in a situation like this, the only thing to do is to keep taking pictures. With this kind of setup, involving a high speed descent and a single shot per pass, it's hard work for the rider. It's also tricky to keep riding the same section of trail ad infinitum without feeling self-conscious, so it's good to get a useable shot or two in the bag early on.
This, as it turns out, was one of the best shots of the day - and also one of the first that I took. Although I was having difficulty with the need to crop aggressively to the right of the frame, I knew that a combination of some tight cropping and a bit of Photoshop - plus the covermount flap - would effectively shift the rider over to the right for me.
What's interesting about this cover is that it matches the original brief so closely, right down to the angle and positioning of the bike and rider. That doesn't happen all the time, of course - sometimes reality intervenes to rule out a specific angle or location. But on this occasion all the planning paid off, and MBUK got precisely the cover they wanted.
The Mac screensaver is finally ready for download (from the 'downloads' section, natch). Apologies to all Mac users for the delay, and thanks for your patience.
There are, unfortunately, a couple of known issues. First, the screensaver will install but won't run on Intel Macs. Second, the fade between images seems to be jerky on some Macs (it certainly is on both my G4s). The first issue will be solved when the software used to build the screensaver is updated. And we're working on the second, though I'm told it may be caused by an inherent problem with the way Macs deal with Flash.
If you download and install it, let me know how you get on.
Canon's designers aren't the only ones who've been busy - here's something I've been quietly working away on. Four vertical inches of image stabilisation (with adjustable compression damping) and a built-in flash.
Not, perhaps, as versatile as the new EOS. It's arguably a bit too specialist for the average sports photographer, too. And perhaps a bit bulky to take as a carry-on on a flight. Hum. Back to the drawing board?
* readers outside the UK and/or under the age of 30 can find enlightenment here
//Warning: long post with a complete lack of pretty pictures of bikes...//
Update: since originally posting this I've added a few sentences (to clarify a couple of points), generally tidied things up a bit and rearranged the seven bullet points at the end into something approaching an order of priority.
The question of whether to shoot jpeg or raw is one that crops up a lot. If you're new to dSLRs, chances are you've been sticking with jpeg - it's seen as the safe option, and there's no immediate need to fiddle around with extra software to tweak the results. But if you want to get the best out of your camera, raw has a lot of advantages over jpeg (and it's not as tricky to use as you might think).
There's a great deal of information out there on the web around this topic, some of it downright inaccurate and misleading. So let's keep things simple.
If you shoot jpeg the camera does the raw processing for you, using all the parameters in your camera's image menus (sharpening, tone, saturation and so on) to tweak the result. If you shoot raw the camera records the data straight off the sensor and you import it into a software package of your choice, where you can fiddle around to your heart's content before exporting as a jpeg or tiff.
Here's how the two file formats stack up:
- gives a ready-to-go result straight out of the camera
- doesn't need extra software or computer time
- reduces storage space
- contains less colour and tone information than a raw file (8 bits, compressed, for the technically-minded)
- reduces options for tweaking colour, contrast and brightness
- uses 'lossy' compression, which throws data out. Once it's gone, it's gone
- pulls the original data straight off your camera's sensor, unaltered
- acts as a 'digital negative' which you can come back to and re-process any number of times
- contains far more colour and tone information than jpeg (12 bits, uncompressed, typically), giving scope for more editing without losing quality
- allows post-capture tweaking of white balance, exposure and other variables not possible with jpeg
- needs extra software for processing and output to a useable file format for printing and viewing
- involves spending extra time at the computer
- takes up more storage space
(There are one or two myths surrounding raw floating around on the internet which are worth dispelling. One is that it increases a camera's dynamic range (the range of brightness values in a scene that can be recorded by the sensor). It doesn't do that, in fact, but it is certainly possible to make far bigger adjustments in software to a raw file than is the case with jpeg.
Another is that, armed with a raw file, the photographer can effectively make whatever white balance and exposure changes they like without affecting the quality of the outcome. In fact there's a visible difference between, say, a 2-stop underexposed raw file that's been recovered in software and one that was correctly exposed in the first place. Raw is no substitute for getting things right in-camera.)
The point, as far as I'm concerned, is this: if I shoot jpeg, I have limited options after I've pressed the button. If I shoot raw, I have far more flexibility at the expense of a little extra time spent at the computer. But that, in the end, is the crux for many people - is the time spent in front of a monitor worth the effort?
I think, on balance, that it is.
In the interests of making life a bit easier, here - in rough order of importance - are my tips for making the transition from jpeg to raw a bit smoother (including a few things I found out the hard way):
1. Accurate colour matters
If you're going to the trouble of shooting and processing raw, it's pretty important that your monitor is accurately profiled with a proprietary colorimeter and software package. If you don't take this step, you can't be certain that the colours you see on your monitor accurately reflect those contained in the file you're working on (chances are, they won't). And that means you'll never be able to print accurately. It's another expense, but it's worth doing for anyone interested in getting the best out of their dSLR.
2. Software design and performance makes a difference
Common problem areas include rendering speed (the novelty of watching a spinning wheel or eggtimer icon soon wears off when you've got dozens or even hundreds of files to process) and over-complicated screen layouts and tool palettes. How many different ways do you need to be able to tweak contrast, anyway? (the software I use has at least four...)
I've settled on Bibble because I'm used to the interface, it's blazingly fast at processing dozens of files at a time and the quality is pretty good. But apps like Apple's Aperture and Adobe's Lightroom set a new standard by making the entire process of downloading, organising and processing raw files far more intuitive. Well worth a look, if you can justify the price.
3. Getting it right first time saves time
Although there's lots more flexibility for making post-capture adjustments when you shoot raw, there's no substitute for getting things right in-camera. Not only does this mean your pictures will benefit from better quality, but you'll save a lot of time at the computer by not having to fiddle with all the basic adjustments. Nail the exposure and white balance to the best of your ability; you'll be grateful you did later.
4. Curves rule
Raw processors include powerful tools like white balance and exposure adjustments, but the most versatile tool in the box is also one of the simplest. Curves allow you to adjust the relationship between input and output pixels to alter image brightness and contrast. It takes some practice to achieve the effect you're after, but it's worth persevering with: curves allow greater control than any number of brightness, contrast or colour sliders.
5. Set your own 'look'
Don't be surprised if your software's settings give your raw files a disappointing look. Compared to out-of-camera jpegs, most raw conversion software defaults to relatively low contrast and saturation. Trial and error will allow you to find standard settings that work as a starting point, and all good software will allow you to save those settings.
6. In-camera image adjustments (mostly) have no effect
Aside from their effect on the embedded preview jpeg (see point 7., below), any in-camera image parameters you've set - including noise reduction and sharpening, for example - will only be recognised by your camera manufacturer's own raw conversion software. If you use third-party software it'll make a stab at setting the white balance (though the numbers may not match what you've set on the camera), but will otherwise use its own default settings.
7. The histogram sometimes lies
If you're using the histogram on your camera to tweak exposure (a subject I'll come back to another time), it's worth knowing that it's not displaying the data from your raw file. Instead, it shows a representation of pixel values in the preview jpeg embedded with the raw file (which is also the image that you're seeing on the camera's LCD). But the jpeg has been processed with white balance, sharpening and other adjustments set in the camera's menus, all of which can affect the histogram.
Bottom line: pay attention to what the histogram's telling you, but remember that it doesn't relate directly to the raw data you're recording. And set image adjustment parameters in your camera's menus (like sharpening, colour and tone) to 'off' or 'low' wherever possible.
Still here? If you haven't wandered off to another part of the web in sheer boredom by now, you should probably give raw a go. Once you've experienced the degree of control you have over your results you'll find it hard to go back to jpeg. It's rather like having your own darkroom. Except with far more control, a complete absence of noxious chemicals and no need to monopolise the bathroom for hours on end...
One of the liberating effects of my switch from film to digital has been the way I can use flash. Although I'd been using remote flash - triggered via a radio slave - for several years with film, it wasn't a technique I'd use unless I had to. Not being able to preview the results made it a bit of a hit and miss affair, although years of practice meant I could be confident of more hits than misses.
Digital widened the goalposts by allowing me to quickly test the position and strength of a remote flash before commiting to the shot. The result is that I carry at least one flash (with a stand or tripod and the radio slave) almost everywhere I go. I can punch a bit of directional light into a picture almost anywhere I want, using it to brighten shadows and / or freeze movement as appropriate. The effect can be fairly subtle, but an educated eye can usually spot the extra light source. Here's a shot with extra flash:
And here's the same setup without flash:
Remote flash - and lots of it, to the point where some shoots take on the look of a Hollywood film set - is currently a very fashionable look in photography. There's no doubt in my mind which is the better of these two shots, particularly since this was taken with a magazine cover in mind. Despite the tell-tale second shadows in the flash-lit shot, the extra contrast and punch helps lift the rider from the background and puts some light into areas that are otherwise verging on murky.
But tastes vary, and the question of whether or not to use flash can sometimes come down to personal preference. There are certainly situations where I simply wouldn't be able to get the shot without a remote flash, which makes the decision a no-brainer for me. Here's an example, shot in deeply shaded woodland:
Without the flash throwing a beam of light on the rider the picture would lack a focal point. It would also have been extremely difficult to keep both bike and rider sharp enough with just available light.
In good light and with no need for the flash's motion-freezing ability, the decision's not always so easy. Keep it real, or give the natural light a helping hand? Sometimes it comes down to the very prosaic consideration of how much time I've got to get the shot. If I've only got a minute or two, the flash will likely stay in the bag. No sense in setting up a tripod, flash and radio slave if the riders have already disappeared over the next hill.
When is a pre-9:30 courier delivery not a pre-9:30 delivery? Most of the time, in my experience. Chewing my nails off in my office yesterday waiting for a bike to be delivered to Marin rider Paul Lasenby for a cover shoot while the sky cleared to reveal a gorgeous dome of blue, the phone finally rang at three minutes to 12. 'It's here', said Paul. Game on. But only just.
By the time we'd both arrived at Cwm Carn in south Wales the sun was descending rapidly towards the horizon. Gear sorted, we elected not to wait for the uplift and headed off up the road climb under our own steam (thanks to Darrell from CwwDown for picking us up on his way past anyway). Up at the top with wall to wall blue sky, we had less than an hour and a half to finish the job. Hardly ideal, but then getting the shot in less than ideal circumstances is what I'm paid for.
And there are some advantages to being last one out. In the dying rays of the sun, at just before five o'clock, I snapped a few natural light shots of Paul enjoying himself on the singletrack. Ignore the growing chill and it could almost have been a late summer evening.
Technical aside: this is the kind of shot that would be very difficult to achieve in quite the same way on film. Faced with extremely high contrast, the camera simply can't record eveything that you can see. In a situation like this you have to make a call on where you're going to lose detail, and by how much.
For this shot it was important to keep some shadow detail, but I also didn't want to lose all the colour and tone in the sky. The solution was to expose so that the left hand edge of the histogram was steep but not clipped. The resulting image was dark and a bit flat on my monitor, but with some judicious use of curves and Bibble's cool fill light feature I was able to get the shot looking far closer to the way it appeared when I was standing there.
You simply couldn't do that with film...
Standing up for the humble prime lens in the face of ubiquitous zoomage might seem as anachronistic - not to say pointless - as arguing that the cart and horse work fine, thank you very much, and what's with all this new-fangled internal combustion nonsense anyway?
But bear with me on this. Many people, having stumped up for a new dSLR with kit lens, plump for a 70-300mm zoom (or thereabouts) as a second lens. The reasoning is sound enough: more reach and more versatility in a compact(ish) and light(ish) package. What's not to like?
There are at least three reasons why I think that the 70-300mm zoom is a poor second lens for mountain bike photography:
1. It's too long. Now be honest, when was the last time you shot at 300mm? The 70-300mm range harks back to 35mm days. With most modern dSLRs it gives the 35mm equivalent field of view of (roughly) a 105-450mm lens. You just don't need that length, most of the time. Really.
2. It's too slow. Long lenses are hard to handhold at the best of times, but the f/5.6 aperture of most zooms in this class makes life even more difficult. You'll need a shutter speed of at least 1/500sec to keep things blur-free at 300mm, and that's only going to happen in bright sunshine at ISO100. Boost the ISO? Well, sure, but that impacts on image quality. Got image stabilisation? Don't feel too smug - it doesn't help freeze a moving subject.
3. It doesn't work well wide open. If you're at all concerned about getting sharp pictures (and you should be), you'll soon find the limitations of a 70-300mm that's semi-permanently stuck wide open. Which it will be, to get the shutter speeds you need to do the job.
OK, so what's the alternative? Well, there's the humble 85mm prime, for a start. Consider this:
1. It's a field of view sweet spot. Now I realise this is partly down to personal preference, but on any dSLR that isn't full frame (that's most of them, by the way), the 85mm gives an approximate equivalent 35mm field of view of 135mm. In pre-zoom times, this was a popular lens for a good reason: it's long enough to give that compressed perspective look of a telephoto and short enough to be easy to use.
2. It's fast, fast, fast. The f/1.8 aperture of most 85mm primes is over two stops brighter than most 70-300mm zooms at the short end... and three stops brighter at the long end. That's the difference between 1/250 sec and 1/30 sec. Or, to put it another way, getting the shot... or not.
3. It's sharp, sharp, sharp. Most of the time you won't be shooting wide open, but an 85mm will perform brilliantly at f/2.8 - which is still one or two stops brighter than the zoom. And it'll be noticeably sharper, too. Much sharper.
4. It's small and light. For portability on a bike, size and weight is definitely an issue. An 85mm will be smaller than most 70-300mm zooms and will weigh about the same.
5. It's cheap. Primes just don't sell in the quantities that they used to. An 85mm f/1.8 can be had for a song from ebay or a secondhand camera store.
So what are the downsides? Well, you can't stand there and twiddle a ring to bring the subject closer to you, so you'll have to be prepared to use your legs occasionally. And it obviously lacks the reach of a 300mm lens, so there are a very few shots that you won't be able to do in quite the same way. But it's surprisingly easy to get used to seeing with 85mm eyes, and with practice you'll find yourself framing up shots as you go.
And if you don't like it (and you bought yours on ebay), you can sell it on for about the same price you paid for it. Bargain? I think so.
Photography forums are currently buzzing with rumours of impending upgrades to Canon's top-end dSLRs (and some muttering about the fact that Nikon won't be too far behind). More pixels, more clever technology... more expense.
Make it stop, please.
Don't get me wrong, I have the same mild gear fetish as many photographers. But from a self-employed pro's perspective, the pace (and cost) of updates is threatening to spiral out of control. Film cameras used to have a major update every five to eight years (and cost £1000-£2000 each). Their digi equivalents are now on a two to three year update cycle (and cost £2500-£6000 each).
Throw in the dire state of the commercial photography market (downward price pressure, hugely increased competition), and it's hard for many pros to make the sums add up.
Do I want more pixels, better high ISO performance, better wireless flash, better AF, a live histogram, built-in wifi? Yes. Do I NEED any of that stuff right now? No. Could I afford it, even if it were available? No.
Three years ago I took a look at my bulging camera bag containing a brace of F5s, a collection of fast lenses, several flash guns and a bunch of other bits and pieces, breathed a sigh of relief and gave myself a congratulatory pat on the back. After seven years, I told myself, I had everything I needed. And it was all paid for.
Three years later I've spent the same amount again... and it's already pretty much obsolete. It's progress, Jim, but not as we know it.