Impact of Sensor Size on Performance — a Contemporary Survey

I am in the middle of a choice between interchangeable lens systems and one of my pressing needs is the selection of a system that lets me have a smaller, lighter camera bag. Now is not a bad time to be shopping for cameras, with so much choice being available across price brackets and IQ variance. As always, having a lot to choose from is another way of getting confused, so I decided to do some objective data analysis to figure out what’s what.

The choice is between 35mm (full frame or FX in Nikon parlance), APS-C, Micro Four-Thirds (MFT) and 1″ (CX in Nikon parlance) sensor platforms. It’s worth noting that there isn’t any FX mirrorless interchangeable lens camera on the market so far, so going FX means using a DSLR with a mirror and built-in OVF. Similarly, there’s only the Nikon 1 system available in the 1″ format.

Crop Factors

Crop factor is the ratio of the field of view between FX and the sensor in question. One of the things it tells is the focal length required on a 35mm sensor for a given focal length on the sensor in question. E.g. the crop factor of APS-C sensors is usually 1.5, which means that a 100mm lens on APS-C would give the same kind of frame as a 150 mm lens on FX.

Crop factor also determines the increase in Depth of Field (DoF) for a given aperture value between FX and the sensor in question. E.g. a 100mm lens at f/4 on APS-C would give the same frame and DoF as a 150mm lens at f/6 on an FX sensor. One thing the crop factor doesn’t change, though, is the metering. A scene that requires 1/100s shutter speed at f/5.6, ISO200 on FX would still require 1/100s at f/5.6, ISO 200 on a CX sensor. I.e., a smaller sensor allows you to shoot at wider apertures and lower ISO or higher shutter speeds for the same DoF.

Lastly, the square of the crop factor indicates how much larger the FX sensor is compared to the sensor in question in terms of surface area. E.g. FX is 1.52 = 2.25 times larger than APS-C sensors.

So, let’s take a look at the crop factors across the systems and see how they compare. Here’s a chart:

CropFactors

 

As you can see, the biggest gap is between FX and APS-C. The smallest is between APS-C and MFT. All of them are fairly close to 1 stop (1.414) so you can assume that you change a stop’s worth of equivalent aperture while going across the spectrum.

These crop factors have interesting consequences. They mean, for example, that a 50mm f/1.8 “normal prime” would behave like a 135mm f/4.9 telephoto lens on a CX camera, except for exposure metrics.

Tonality

Tonality, or colour depth, is the ability of a sensor to distinguish very similar looking colours from one another. This metric is what gives photos from larger sensor cameras that “3D look” while the pictures form cellphones and small cameras look very “flat”. I have pulled the best in class tonality data from DxOMark scores for each format. The chart below shows how they compare.

Tonality

This chart shows an almost linear progression across formats, except that the best CX format sensor (Sony CyberShot RX100) is nearly as good as the best MFT sensor. The good news here is that if you’re looking for colour depth in your photos, you are much better off in 2013 than just 5 years ago. The colour depth of the RX100 (22.6 bits) is about the same as that of the Nikon D90 (22.7 bits), despite the latter having more than 3 times the sensor area.

The not-so-good news, of course, is that an FX sensor is still quite far ahead in terms of tonality, and this becomes especially important if you are a landscape photographer shooting in snow or haze. Portrait photographers would also get much better skin tone gradation with FX sensors.

Dynamic Range

Dynamic Range (DR) is a measure of the maximum variation of light and dark areas in a scene before the sensor loses the ability to resolve colour or tone, rendering the dark areas as greyish black and bright areas as white. DR is especially important for shooting outdoors in daylight or indoors with overhead lighting or spotlights. People who shoot Christian weddings would also want to be careful about the DR of their sensor since they often need to shoot the groom wearing a black suit along with the bride wearing a white dress.

The DR variance, again based on the segment topping sensor scores on DxOMark across formats is shown below.

DynamicRange

This is an interesting chart because the variance in DR doesn’t have any correspondence with the crop factors. The best FX and APS-C sensors are quite close to each other, but there is a huge gap from APS-C to MFT. In fact the RX100 outperforms its MFT siblings in DR despite a sensor that’s almost half as large.

Noise

Now we come to the most important metric for a sensor, i.e. noise performance. We all recognise noise as the speckles and freckles that we see all over a photograph. That is the most visible effect of noise, but it’s not the only one. Noise also reduces a sensor’s tonality and DR quite a bit. What’s even worse, a noisier sensor would also perform worse at its base ISO for long exposures compared to a less noisy sensor.

So what does the Noise progression look like across formats? Well, here it is. No surprises.

MaxISO

The chart above is based on DxOMark‘s listing of ISO measurement at which the degradation in image quality becomes perceptible. I have converted those numbers into “stops” for a more easy comparison. E.g. according to the chart above, the best CX sensor (Nikon 1 V2, in this case) is 3 stops noisier than the best FX sensor (Nikon D3s).

Sensor Performance Conclusions

Based on the above charts, we see that there’s a significant gap between every segment, except MFT and CX, wherein the RX100 is acting as a bit of a leveler, albeit with 1.14 stops poorer noise performance.

If you are an outdoors shooter, it’s probably not a bad idea to downsize since the key metrics for outdoor shooting (Tonality and DR) are far more level across the formats, while the kit size reduction is significant. If you shoot indoors without studio lighting, though, you should strongly consider having a large sensor system.

Another important thing to remember, though, is that today’s sensors in any segment match up to or outperform the sensors of the next segment from 5 years ago across all metrics. If you are satisfied with your 5 year old equipment’s IQ, you are ready to downsize without consequence.

State of the Camera 2012: Part 2

In Part 1 of this two-part series we looked at some of the dominant technology trends over the last couple of years and the impact they are having on cameras and photography. In this part of the series, I write about my impressions of a few recent cameras that I have had the opportunity to use.

Sony CyberShot DSC-RX100

Voted by Time Magazine as one of the best inventions of 2012 and declared to be the best pocket camera ever made by New York Times reviewer David Pogue, this camera has changed the way people respond to small sensor sizes.

RX100 is a fixed zoom pocket camera with features and controls designed for enthusiast photographers. At 36mm thickness, it is just about small enough to fit into the front pocket of a pair of jeans without discomfort. Its claim to fame, though, comes from the fact that despite such a diminutive size, the camera’s 1 inch (13.2mm x 8.8mm) sensor delivers dynamic range and colour depth that matches that of the much bigger APS-C sensors, e.g. that of the Nikon D90. This excellent 20MP sensor is mated to an f/1.8-4.9 Carl Zeiss 3.6x zoom lens that is sharp enough to justify the super-high resolution.

I acquired this camera in September 2012 and since then, I have hardly used my DSLR (Nikon D90) and I sold off my Sony NEX-5 as well. Other than the 100mm+ telephoto range, I do not miss shooting with the DSLR. RX100, in fact, has a few smarts that make it nicer to shoot with than the D90:

  • Fast AF, accurate metering and white-balance
  • Astounding detail in the 20MP files captured with this camera
  • Horizon Indicator, to show camera tilt across two axes
  • High resolution, high brightness WhiteMagic™ LCD that makes daytime shooting a pleasant experience
  • Front and rear rotary dials offer direct control of shooting parameters
  • 7-option customisable “Fn” button for quick access to shooting options
  • 3 memory presets for shooting parameters that one can customise for specific conditions
  • Tiltable on-board flash for bounce-flash capability

Another thing that RX100 does much better than the D90 and its bigger cousin NEX-5 is video recording. The video stabilisation on RX100 is amazing and the results from 50 fps 1080i video are striking.

There isn’t much that this camera leaves to complain about. If at all, I sometimes find the minimum focus distance at tele end to be a bit too much.

Olympus OM-D E-M5

Voted by DPReview readers as the best camera of 2012, the Olympus OM-D E-M5 is indeed a remarkable camera. This camera is remarkable for 3 reasons:

  1. Exquisite styling based on the Olympus OM series film cameras, with solid construction and weather sealing
  2. Excellent new technology incorporated into the camera
  3. Large selection of small, high quality micro Four-Thirds (MFT) lenses

OM-D EM-5
Indeed, the styling of the camera is admirable. Its modern machine tooled cuts combined with classic styling cues give it an intriguing and likable shape from the front and top. The rear side of the camera is dominated by the large LCD swivel touch-screen that makes it look and feel like a modern camera. The looks of the camera are enough to raise expectations from its performance, handling and IQ.

Handling of this camera is quite good. Apart from the awkwardly placed on-off switch, everything is just about where it should be. The tilting LCD screen offers a better shooting experience than the built-in 1.4M dot EVF, though the latter may be used for extra stability. While using the LCD, you can also use the touch-to-shoot functionality to focus on a desired point in the scene and shoot.

The E-M5 features hybrid AF (on-sensor phase detect + contrast detect) that is very fast to lock on and quite accurate. The 16MP imager has great colour depth and high ISO performance. The JPEG engine from Olympus also retains its characteristic colour response to give pleasing images straight out of camera. It did have some issues with auto white-balance under fluorescent lighting.

Another first in the E-M5 is the 5-axis in-body image stabilisation. While most IS systems only compensate for yaw and pitch, the E-M5 compensates for yaw, pitch, roll and horizontal and sideways translation movements. This works very well for shooting stills, and also for shooting video. However, the RX100 still outclasses the E-M5 in video shooting for one basic reason — continuous tracking AF in E-M5 is very, very slow, at least with indoor lighting.

Overall, this is a very attractive camera and just by its looks and handling, it inspires one to shoot. However, it’s still a bit rough at the edges and things like iffy AWB and slow focus tracking can be a bit disappointing. Nevertheless, this is the first camera from the FourThirds stable since the Panasonic GF1 that has held my attention and is having me think of switching to the MFT stable instead of Nikon APS-C.

Nikon 1 J1

The Nikon 1 J1 is a strange camera in many ways. In photographs, it appears to be all cutesy and properly small, in line with other compacts. When you see it in real life (I was going to say, “flesh and blood”, but stopped for obvious reasons), it turns out to be rather big and very solidly built.

The most striking thing about the J1 is its build quality. It is crafted out of seemingly thick metal, with large, nicely tooled buttons on top. The back-plate has some more large and well constructed buttons, along with the LCD screen. It makes the very solidly built Sony RX100 feel like a flimsy little thing.

So, the J1 is uncharacteristically well built. What about the performance? It has the same sized sensor as RX100 but whereas RX100 only does contrast detect AF, the J1 has hybrid (phase detect + contrast detect) AF like the OM-D E-M5. This does result in super-fast focusing with a high level of accuracy. Raw IQ is excellent and matches that of the RX100 where low light is concerned. At base ISO, RX100 has higher colour depth and dynamic range, though. The camera is also capable of up to 60 fps frame burst, which enables it to offer interesting features like Motion Snapshot and Smart Photo Selector. That’s not all, the camera can also shoot 400 fps videos at 240p and 1200 fps videos at 120p. For all this performance, though, it doesn’t offer exposure/WB bracketing or in-camera HDR, which brings us to the next point.

The J1, unfortunately, is a bit of a strange camera for another reason. It’s designed for P&S upgraders and novices, but requires an enthusiast — a tinkerer — to get the most out of it. The metering and AWB of this camera are both unremarkable, making it necessary to shoot raw and use EV compensation more often. That, however, is a problem because the camera being designed for novices, doesn’t offer an up-front control to set EV compensation. The high speed videos look like an attractive prospect, until you discover the amount of ambient light the camera needs to be able to capture them.

Another trade-off between RX100 and J1 is that of size vs. versatility. The J1 is big, compared to RX100 (though smaller than Olympus E-PM2) and its lenses are bigger still. But for the added bulk, you do get the ability to choose between not just the interesting Nikon CX lenses like the 18.5mm f/1.8 or the announced 32mm f/1.2, but with the FT1 adapter, you can also get your Nikon DX and FX lenses to work with it.

Overall, it’s a very interesting camera and for its now discounted price, it offers a great value proposition. Indeed, it’s the best selling mirrorless ILC in Japan by a healthy margin.

Buying Recommendations

If you are looking for a pocket camera and are willing to spend the money, I can recommend the RX100 without reservations. It does what a pocket camera should do, and then some more, with uncompromising IQ on-the-go.

For a mirrorless ILC system, though, the choice is still a bit confusing, especially if you consider cameras I haven’t covered yet. There are 3 broad choices based on sensor size:

  1. APS-C: This includes Sony NEX, Fuji X and Samsung NX bodies. With a big sensor comes the burden of big lenses. None of the manufacturers in this segment have managed to address lens size issues satisfactorily, though Sony made some progress with its 35mm f/1.8 prime and 16-50mm collapsible zoom. If you’re going mirrorless to reduce size, you might want to consider the entry level DSLR bodies, which aren’t significantly larger but offer better value for money.
  2. MFT: Micro-Four-Thirds platform has plenty of choice among lenses and the MFT champions, Olympus and Panasonic, have managed to make some really small, really good lenses for this system. Unfortunately, these manufacturers lagged behind in the sensor department. All that looks set to change going by what we see in the E-M5 and E-PL5, and given the sensor supply deal that Sony has struck with Olympus. I would myself have considered this system, but…
  3. Nikon CX: being a Nikon guy, I am watching this mount with interest. The RX100 has shown the potential of a sensor of this size, and I can happily live with the IQ it returns. The CX lens selection is sparse at the moment, but at least they are sufficiently small, even with VR. The AF and continuous shooting performance of Nikon 1 bodies is beyond anything in its price bracket, or even higher. I would have gone for the Nikon 1 V2, but the lack of exposure bracketing  and/or multi-shot exposure modes a la Sony’s Superior Auto is a deal-breaker for me. Let’s hope Nikon addresses this concern soon enough.

State of the Camera 2012: Part 1

It’s been exactly two years since I last wrote about a camera on this blog.  A lot has changed in the camera scene over the last two years and I fell in and out of love with the NEX-5 during this time as well. In this post I shall begin with talking about the recent trends in the camera market followed by short reviews of a few new generation cameras that I have tried or bought.

Mirrorless/EVIL Systems

The hot new stuff these days undoubtedly is the rise of “mirrorless” or “EVIL” (Electronic Viewfinder Interchangeable Lens) cameras. “Mirrorless” is a broader term that just indicates the absence of a mirror that is traditionally used to operate the Auto-focus system in SLR cameras. Mirrorless cameras use their sensors directly to auto-focus either through contrast detection (CDAF) or phase detection AF sensors embedded directly on the main sensor. Mirrorless cameras may not have a viewfinder or allow changing of lenses. EVIL cameras include both.

By discarding the mirror, the bodies of these cameras can be considerably thin. By dropping out the optical viewfinder, these bodies also save space otherwise occupied by the pentaprism in SLRs. This opportunity in body size reduction for the same sensor size is the main selling point of the mirrorless cameras. They allow you to carry cameras capable of seriously good IQ in a smaller package.

Nikon vs Sony

Small Sensors

As the Sony NEX system demonstrates, while it’s possible to reduce the body size considerably by going the mirrorless route, the size of the lenses does not reduce since it’s largely related to sensor size. To reduce the overall size of the package, the sensor size must also be reduced.

Traditionally, small sensors have had a big gap up to the IQ delivered by bigger ones (APS-C/APS). However, smaller sensors have also now reached the IQ sweet spot where minor improvements in IQ are harder to perceive and more expensive to deliver. A striking example of this is the 1 inch sensor inside Sony Cybershot DSC-RX100. As per DXOMark, this sensor is better than that of the Nikon D80, despite being 3.24x smaller. It even matches up to the D90 in all except low light performance!

This isn’t to mean that the large sensors haven’t improved in the mean time. However, the IQ delivered by small sensor systems is now enough to make fairly large (8×12″ or 12×18″) prints without noticing defects. That really is more than most non-professional photographers would want out of their camera.

Summary

To summarize, digital cameras are now in a phase of miniaturisation as they have reached and exceeded the levels of “acceptable” performance. Miniaturisation is enabled mainly by getting rid of the mirror and pentaprism assembly from traditional SLRs and having their functions performed by the camera electronics (CDAF, Electronic Viewfinder). Systems with smaller than APS-C sensors also benefit from a significant reduction in the size and cost of lenses. The resulting cameras are not only physically smaller, but they are also cheaper than their large-sensor counterparts when you include the cost of lenses.

In the next part of this post, I shall write about a few recent cameras that are at the forefront of this miniaturisation and see how they compare to DSLRs.

NEX-5 Early Impressions

It’s been a week since I’ve had my baby camera, i.e. the Sony α NEX-5 and I’ve been enjoying it so far. Here’s a quick low down of the ups and downs I’ve encountered with the camera.

The Good

  • Outstanding IQ in any shooting conditions that can only be trumped by careful, deliberate shooting with a DSLR equipped with high quality optics
  • Small enough to fit in my laptop bag, if not the jeans pocket
  • Excellent iAuto mode for quick snapshots
  • Fairly easy to use controls, esp. with button customisation offered by Firmware v03
  • Auto HDR and Hand-held Twilight Modes really work!
  • Sweep Panorama works really well for hand-held shooting. You can’t do better with a DSLR until you mount it on a tripod.
  • Focuses almost as fast as an AF-D type Nikkor with D80/D90 body
  • Optical Steady Shot (OSS) on the 18-55mm f/3.5-5.6 Kit Lens is outstanding
  • Movie making is super easy and fun!
Top Left: ISO 12800 shot. Top Right: “Hand-held Twilight” shot @ ISO 6400. Bottom: Sweep Panorama

NEX-5 ISO 12800 Mega Performance
Studio-I

The Bad

  • JPEGs tend to be a little less sharp by default
  • Chromatic Aberration is visible in high-contrast scenes, though not unusually excessive and removed in one click with Nikon ViewNX
  • Zooming while composing is cumbersome due to missing support from the forehead as with an OVF equipped DSLR
  • Cumbersome MF adjustment despite on-screen MF assist. The electronic focus ring twists indefinitely so you’re not really sure if it’s locked at minimum/infinity focus, when you need it

The Ugly

  • Excessive battery drain if you take time while composing a shot due to OSS, continuous AF and LCD
  • Fingerprint -magnet of an LCD and screen protector not provided as part of the kit

I’ll probably add more to this post as I get more familiar with the camera, especially when I start shooting RAW (+JPEG, of course).

My Imaging Tools and Workflow

As my photos get backed up and burnt on to a DVD, I thought I might just do a quick cataloguing of the software I’ve found useful for developing my photographs and what each does. I’d also outline my workflow as I present each software in the order in which it appears in the workflow. Just to set the context, I use Microsoft Windows 7 for my imaging tasks, and my camera is a Nikon D90, so I do use a lot of Nikon software.

File Management and Archival

1. Nikon Transfer (Freeware)

This is the program that detects photos on external media and transfers them to a local disk or other external media. I’ve tried connecting the camera through its own USB interface and through the built-in card reader on my computer, but I find that the San Disk card reader that came with my Extreme III card gives the fastest throughput. Anyway, so Nikon Transfer is launched when the SD card is inserted and the nice thing about this software is that

  1. It allows simultaneously downloading the photos to a secondary location, which in my case happens to be just another disk partition but can also be an external USB drive (I’m contemplating doing the latter, going forward)
  2. It detects which photos have already been transferred from a batch and skips those (configurable)
  3. It allows manually selecting the photos to be transferred. I don’t normally use this option but I’ve needed it once or twice so I thought it might be worth mentioning.

My folder hierarchy for transferring these photos into is like this:

Camera Model/Name > Year > MonthNumber-MonthName or EventName

E.g. D90\2010\09-September, or Vesper\2010\MulaniWedding, etc.

The RAW files get dropped directly into those folders and the converted JPEGs go into a subfolder. I drop the camera-wise separation for permanent archival.

2. Nikon View NX2 (Freeware)

ViewNX 2 is a recent release from Nikon and Nikon Transfer now comes bundled with it. ViewNX 2 allows rating and tagging photos, while also allowing for conversion of RAW files into JPEGs or 8/16-bit TIFFs. It has also started offering the basic photo editing options that are available in Nikon cameras’ retouch menu. I primarily use it to select the photos that I want to develop. The way it allows files to be compared at 100% zoom and the way it can filter the film strip on rating ranges makes it extremely easy to do so. Here’s the rating system that I’ve evolved for myself over the months:

  • 5 Stars: Awesome snap! Convert, retouch and upload to Flickr. Send for a medium/large print too.
  • 4 Stars: Awesome snap! Convert, retouch and upload to Flickr. Send for a small/medium print too.
  • 3 Stars: Good snap. Convert and retouch. Likely upload to Flickr. Maybe send for a small print.
  • 2 Stars: Acceptable snap. Convert, maybe retouch. Unlikely to be uploaded to Flickr or printed.
  • 1 Star: Not that great. Just meant for keeping. May not even be converted to JPEG.
  • 0 Stars: Delete.

Please note that while I ruthlessly delete items based on the star rating assigned as above, I do it only on the primary working copy. The backup copy is always there, just in case I need to revert on my decisions. It doesn’t happen often but on the rare occasion that it does, it’s good to know that you’ve got all the shots with you.

When I’m done with all the processing, etc. for a month or few months, I copy the finished and shortlisted content to a more permanent backup solution (DVDs, in my case).

Image Development

Nikon Capture NX 2.2.4 ($180): This software is all that I use for developing my images. Why? Here are a few reasons:

  1. It’s arguably the best converter for Nikon’s RAW files, giving rich tones, smooth gradations and excellent detail
  2. It recognises all the camera settings used while taking a photograph and applies them as a starting point while developing the image
  3. It has in-built lens correction features like distortion control, CA control and vignetting. Really awesome to have this if you’re shooting with Nikon’s own lenses
  4. It’s got the U-Point controls that make selective retouching really really easy, without requiring you to drop into complex layer masks, etc., which I could never get the hang of
  5. It saves all the development and retouch settings right within the NEF file. The ability to retain the exact settings used to create a JPEG is amazing!
  6. It supports multiple edit-versions of the same file. You want a full colour as well as a monochrome version of the same snap? Sure, your NEF can carry multiple sets of editing steps all in itself
  7. I’ve come to love the very functional interface. Though a lot of people say it’s not as great as Aperture or Lightroom, I disagree. I’ve tried LR3 and I didn’t like it as much as CNX2. Yeah, batch conversion is a pain so if you do a lot of batch conversions, you might have to get something else or use View NX2 if you can live with its glacial speed of conversion and limited retouching support.

Specialised Development

While Capture NX 2 handles development and retouching of single images, some times you need multi-image solutions for panoramas and HDRs. Here are the tools I’ve found useful for these applications:

1. Picturenaut HDR Imaging (Freeware)

Picturenaut HDR is not the best HDR creation software out there, but it gets the job done for me. On the positive side, it’s got a very simple interface and it’s very fast. On the negative side, it doesn’t do image alignment too well (despite the option being present). If you want the super-saturated, artificial-looking, sick HDR shots, just select “Bilateral” tonemapper, set saturation to the max, set contrast to the min and you’re done. I don’t like that, so my workflow is a wee bit more involved.

My HDR workflow is to create 16-bit TIFFs from the bracketed shots and “Neutral” Picture Control setting via View NX2. I then load the TIFFs into Picturenaut and through some juggling with “Adaptive Logarithmic” global tonemapper. Flying by the histogram (I’m now pretty good at interpreting histograms cool), I get the HDR tonemapped as best as I can. The output is saved into another 16-bit TIFF. I load that TIFF into Capture NX2 for local enhacements and, thanks to the 16-bit depth, I get to recover amazing detail and colour from the shadows, if needed.

2. ArcSoft Panorama Maker 5 Pro ($80)

While trying to make this panorama out of 14 images, I spent hours breaking my head with Hugin. Hugin is Free Software, and it’s supposedly highly capable. I would’ve recommended it but unfortunately, it failed to work for me on MS Windows 7 (64-bit). ArcSoft’s Panorama Maker 5 Pro is a nifty little tool. All I had to do was feed it my 16-bit TIFFs and click one button to generate the panorama! Since panorama stitching doesn’t give a perfectly rectangular image, the tool offers a crop suggestion to crop out all the jagged edges and thereafter you can save your panorama.

I saved it as a 16-bit TIFF again, and retouched it in Capture NX2 to remove some issues caused by uneven exposure in the source images. Oh yeah, remember to always lock focus, white-balance, picture control mode and exposure (aperture, ISO, shutter speed) while taking multiple shots to make a panorama. This tool also includes a little tutorial on capturing individual shots for a panorama. Thoughtful.

3. HDR Alignment Tool v2.0 (Freeware)

Since Picturenaut doesn’t do a great job of aligning bracketed shots, I found this neat tool that allows you to put 2 control points on a pair of images and then re-aligns them. If you have a bunch of images, you need to set 2 control points for each pair in the sequence (1-2, 2-3, …) and this tool will finally yield an aligned stack or re-aligned individuals. Unfortunately, it doesn’t to TIFF output.

Honourable Mentions

While the above mentioned software is what I’ve used so far, here are a few honourable mentions to conclude the post:

  1. Flickr Uploadr for Windows/Mac OS X: Excellent tool for managing your Flickr uploads. Never upload a photo without the right title, description or permissions!
  2. UFRaw: Before moving to Capture NX 2 and Windows, I used Ufraw for my NEF conversion. It’s an excellent tool, but the underlying dcraw converter doesn’t do a great job handling colours and noise reduction. If GNU/Linux or F/OSS is your thing, UFraw is highly recommended.
  3. The Gimp: Also for F/OSS buffs, the Gimp is a highly capable image editor. I suppose it’s indispensable for any imaging workflow on Linux that involves non-trivial retouching.

Which Compact Interchangeable-Lens Camera?

Why I settled on the Sony α NEX-5 as the best mirrorless interchangeable lens camera for myself.

These are exciting times for DSLR enthusiasts. We’ve all marvelled at the creative and operational flexibility afforded by the large dial, switch and button infested DSLR bodies and interchangeable purpose-built lenses. We’ve been spoiled for the impeccable image quality afforded by the 8.5-15x larger APS-C sensors (upto 34x larger if you’re a 35mm shooter). It’s impossible to look back at compacts. Or is it?

Most DSLR shooters sooner or later realise that their beloved hunk can’t be their only camera. They can’t carry it all the time to family events. They can’t do anything about it if they happened to dine in a fancy restaurant on impulse. Carrying a DSLR has to be planned ahead, owing to its bulk. In the last couple of years, though, the B-level interchangeable lens system manufacturers (anyone other than Canon and Sony) had been pushing the boundaries of how small an interchangeable-lens system could be made. While I’ve followed this category (dubbed EVIL — Electronic Viewfinder, Interchangeable Lens), it’s right about now that I have finally settled on a system. It’s going to be none but the Sony α NEX-5 for me. If you follow me on twitter, you’d already know about this. Here’s a brief overview of stuff that I considered and what sold me on NEX-5.

Luxury Compacts

These are essentially non-interchangeable zoom compacts with puny sensors. Canon G10, G11 and the latest, G12, are leaders in this category, owing to their small size and controls that are suited to DSLR users. Heck, they have an ISO selection dial! Some people also choose the smaller Canon S90/S95 systems. Their undoing, however, is their image quality. While they’re considerably good in terms of IQ, sensor technology has been favouring large and small sensors equally and large sensor systems have leaped further ahead in terms of dynamic range and low-light capabilities. Nikon’s recent offering, the P7000, is a worthy addition to this lineup. It trumps the G12 in terms of specifications though deeper IQ analysis is pending.

Certainly, though, these systems aren’t going to offer the kind of IQ one gets from today’s DSLRs. Written off.

m4/3 EVILs

While Canon and Nikon were busy fighting it out in the 24mm and 35mm DSLR arena, a bunch of B-segment makers got together to create a new shared sensor format and lens-mount called Four-Thirds. This system boasts a 4:3 sensor (compared to 3:2 for 24 or 35mm ones) that has about 1/4 the area of a 35mm sensor. Much better than tiny compacts (4-9x larger) but still not up to the level of 24mm sensors. Early on, these systems focused more on the almost-similar IQ to larger DSLRs with a wide selection of lenses thanks to the shared lens-mount between manufacturers. Models like the Panasonic

Panasonic Lumix GF-1 and Olympus PEN EP-1, however, put the same sensor in rangefinder/EV bodies that were significantly smaller. What’s remarkable about them is that they are about the same size as the Luxury Compacts if you put a pancake prime lens on them. Yet, they offer visibly better IQ and flexibility to change lenses. GF-1 and EP-1 were instant hits with shooters of even Nikon and Canon DSLRs as they afforded DSLR-like IQ and operations, without the bulk. Around this time, people started speculating about Nikon and Canon jumping into the Mirrorless Interchangeable Lens (MIL) segment as well.

I was waiting for a GF-1 update, that fixed some of the auto-focus speed issues and various other shortcomings. That update is yet to come but something else happened along the way…

Enter Sony α NEX-5

This May, Sony announced two APS-C MIL cameras — α NEX-5 and α NEX-3. For the purpose of this post, I’m going to ignore NEX-3, which is just a cheaper version of NEX-5 designed to win price wars where needed.

Sony α NEX-5 with 16mm f/2.8 Lens
Sony α NEX-5 with 16mm f/2.8 Lens

What’s cool about NEX-5? Well, consider that despite being an all-metal body, it’s considerably lighter than anything in this category. Not only that, it’s also the smallest. It’s smaller and lighter than the Canon G10/11/12, it’s smaller and lighter than Nikon P7000, it’s smaller and lighter than Lumix GF-1, than PEN EP-1 or EP-2… heck, it’s smaller and lighter than anything at all!

In terms of IQ, this sensor beats all contemporary m4/3 offerings hands-down, blows the fixed lens, tiny-sensor compacts out of reckoning and stands all the way up to Nikon D5000’s image quality. Other notable and interesting features are:

  • Tilt-up/-down LCD screen with best-in-class display.
  • Hand-held HDR creation with upto 6-stop difference among bracketed shots. It works.
  • Multiple-shot High ISO blending for reduced noise low-light photography of still subjects.
  • 3-function programmable jog dial and 1-function programmable lower soft-button (firmware v3).
  • High-resolution 2D/3D sweep panoramas. It’s not magical, but it does usually work. A pan-head tripod like the <a href=”http://www.flickr.com/photos/code_martial/3757510707/”>Sony VCT R640</a> that I have will be of tremendous help.
  • 1080p/24 fps videos with continuous AF (Nikon D3100 is the first to offer continuous video-AF in DSLR form-factor).

No wonder then, that the Sony α NEX-5 is the best mirrorless camera available. In the absence of any significant moves from Canon or Nikon and Sony’s continued ground-breaking advancement into compact APS-C bodies, NEX-5 is also likely to hold its position this year.

Creating Hand-held HDRs

On my recent vacation, I took a bunch of bracketed exposures to turn into HDRs. Before this trip I only used a tripod for bracketing. This time around, however, I had to deal with hand-held bracketed shots. To make things a bit worse, these shots included foliage, which isn’t always stationary between shots. Picturenaut utterly failed to align these images. I tried Luminance HDR (qtpfsgui) for Windows 7 (64-bit), and it simply kept crashing. Then I tried Hugin, which too failed to do much. Besides, it was extremely confusing since it is a tool for stitching panoramas, with HDR and alignment being a part of the whole.

Finally, I found HDR Alignment Tool — a free Windows utility for aligning multiple images using manually selected control points. Following is a screenshot of what the tool looks like.

HDR Alignment Tool

There’s a selection of input files at the top, and the output preview in the bottom right corner. The “In”, “Out”, “1:1” and “Fit” buttons are zoom settings for either the source images all together, or for the output preview. “CP1” and “CP2” refer to the control points and allow centering on the chosen control point.

The default selection of control point locations and the resulting alignment aren’t great, so just relocate each of them manually and fine-tune the alignment by visually inspecting the sharpness of the output image near the chosen control point. It’s cumbersome, but it works pretty well. The tool also includes some lens correction features but I didn’t use them, instead making Capture NX2 apply lens correction while creating the source JPEGs. Oh, let me now get to the workflow, which is a bit involved since HDR Alignment Tool doesn’t work with TIFF or RAW files.

  1. Load each bracketed exposure in Capture NX2.
  2. Ensure that all exposure parameters (white-balance, etc.) are the same, apply any other pre-processing steps (lens distortion correction, in my case) and create JPEGs.
  3. Open the JPEGs in HDR Alignment Tool, carry out the alignments and save the aligned JPEGs. By default, the tool will append “Aligned” as a suffix to the file name for distinction.
  4. Open the “*Aligned.JPG” files in Picturenaut. Apply the desired Tonemapping. Save the tonemapped output as a 16-bit TIFF.
  5. Load the TIFF back in Capture NX2, apply post-processing (tilt, sharpening, colour adjustments, etc.) and finally save the HDR JPEG.

The following photo is one example of the manually aligned HDR that I created. Creating one HDR took me about 20 minutes, after I got all things sorted out. The bottomline, however, is that its best to avoid these acrobatics and directly shoot with a tripod if you’re intending to create HDRs. You never know when you might want to create an HDR, so the second thing to remember is to always carry a tripod where possible… or get a Sony NEX-5 😉

Terrace & Garden

Shooting in the Dark

Digital sensors have made a lot of progress on the light efficiency front. The Nikon D3s sensor, currently the most efficient sensor available, offers amazingly clean images at crazy high ISOs. Something that film shooters could only dream of. There is still some time, however, before a D3s calibre sensor makes it to consumer bodies. Meanwhile, it will help knowing the tips and tricks of shooting in the dark for the win.

Gear

Before proceeding further on the topic, I’d like to say that the best thing to do is simply not shoot in low light. Most low light situations can benefit hugely from an off-camera speedlight bounced from the ceiling or walls. This essentially involves placing a large external flash like the SB-600 somewhere out of your field of view (your camera’s hot shoe is also a good candidate) and pointing it toward the ceiling. You can now happily shoot at ISO 200 and enjoy good, clean images with soft pleasant lighting. The following shot of pizza toppings was made at f/5.6, ISO 400, 1/60s in ambient light provided by 2 11-watt CFLs. Without the flash, this shot would have had less tonality, more noise and perhaps motion blur (corresponding flash-less settings would be something like ISO 3200, 1/8s).

Pizzascape

The other very important piece would be a lens that has a large aperture, i.e. at least f/2.8 or larger. The AF 50mm f/1.8 D Nikkor aka Nifty-Fifty is a good, inexpensive purchase. Beware, though, shooting at very wide apertures places a big ask on focus accuracy because the DoF at large apertures might become too thin to be workable. Auto-focus systems also tend to perform poorly in low light. One useful tip for shooting people with wide apertures is to try and focus on the near eye — eyes are easier for AF systems to lock on, and it gives a natural guide for the viewer of the photograph.

Despite having the best gear, you still might end up in situations where you can’t do anything but bump up the ISO and face the lack of light. That’s where some of the following techniques will help.

Taking the Photographs

1. Exposure vs. ISO

The first time I was faced with shooting in low light, the camera metered some terribly low shutter speed like 1/20s at ISO 640, which is tough to shoot with even in low light. I decided to under-expose by a stop at the same ISO to get up to 1/8s because I didn’t want to hit the limit of the ISO range at 1600. Higher ISO = More Noise, right? Yes, provided the exposure is comparable in both. Under-exposure results in even more noisy photos. My shots from the above-mentioned sequence were all wasted. Here’s a noisy, under-exposed shot. Do you want to guess what ISO the following shot was made at?

Electro-forking-daemon

ISO 100! You can see the effect of -5 EV under-exposure on the noise even in this small version. It would print really horrible on 6″x4″ paper. When you have to choose between underexposing and bumping up the ISO to get to a desired shutter speed, always go for higher ISO. An underexposed, low ISO shot is almost always more noisy than a properly exposed high ISO shot. Here’s another shot where I chose to over-expose +2/3 EV even at ISO 1600, to allow good tonality and saturation. The result speaks for itself.

Tandoori Burra Champ

2. Auto-Focus

Another area that suffers badly in low-light is auto-focus. Most AF systems try to lock in on edges that they find under the focus points where the boundaries appear the sharpest. This turns out to be difficult in low light because the difference in light intensity between light and dark parts of an edge are reduced. Having a speedlight like the SB-600 helps here, because it projects a focusing grid on to the field of view (much like the red grid you might see in supermarket barcode readers). If you find your hardware failing, you could try some of these tricks:

  • Zoom-in; Focus; Zoom-out: This works if you have a zoom lens that’s not “varifocal“, i.e. its focus doesn’t change while zooming in and out. The trick is to zoom all the way in, acquire focus — either manually or automatically — and zoom out to the frame you want. Most of the modern SLR zoom lenses are parfocal. I know that the 18-200mm f/3.5-5.6 VR Nikkor works.
  • Focus and Recompose: Sometimes, we would like to focus on a subject that’s not contrasty enough for AF or manual focus to work. In this case, it might be helpful to focus on a more edgy object at roughly the same distance and recompose. The results are iffy, though. Also, if you are not able to verify exact focus through the camera view-finder, chances are that your photograph would pass on as fairly in focus, anyway.
  • Live View: Most new DSLRs now support Live View feature. Live View can help you a lot when trying to focus manually. Turn on Live View and zoom all the way in on the view. Since you’re zooming in on the image preview and not on the lens itself, this trick works even with non-parfocal lenses. I made the following shot using the same technique in fading daylight. The AF was being tricked too often by the bushes and rubble, and generally being unable to cope with the fast moving bird.

Li'l Birdie

3. People Shots with Slow Sync Flash

One of the most frustrating experiences in low-light shooting is blurry people shots. When your shutter speeds drop to levels below 1/20s, asking people to stand still also doesn’t help. You end up needing to try something else. You also can’t pop the on-board flash because

  1. it gets into people’s eyes (not a problem with ceiling bounce), and
  2. it doesn’t bring out the ambient light properly, leading to an unpleasant photograph.

Issue #2 can be solved by using the flash in “Slow Sync” mode. In this mode, the flash fires for a short time (1/60s) but the shutter stays open for as long as it would have opened without the flash. You can choose whether the flash fires in the beginning of the exposure (Front Curtain Sync) or toward the end (Rear Curtain Sync). Why would you use slow sync instead of just shooting without flash? Because it de-emphasises motion blur that occurs while the flash is not on and captures a sharp outline when the flash pops.

While shooting people in slow sync mode, Rear curtain sync works better than front curtain for two reasons. One, people don’t notice a snap until the flash goes off. In front curtain, they tend to think that the snap is done when the flash goes off and let go of their pose while the shutter is still open. This leading to blurry shots.

Secondly, rear curtain captures motion in the forward direction. You get to see the final position of a moving object most prominently, which is more natural than seeing the starting position of an object and seeing a blur in its later direction of movement.

While using slow-sync, just always remember to disable pre-flashes by using FV lock to determine exposure. You want your flash to fire exactly once, and toward the end of the exposure, to get people to instinctively comply with the slow shutter. You might also want to use an orange colour gel in front of your flash if you don’t want the people to appear blue while shooting in incandescent light. Here’s a 1/8s full-on action shot of my friend strumming on the guitar, taken using this technique. If you look carefully, you could reveal a lot more motion blur than what is apparent in first glance. That’s the rear sync advantage!

Ambar and the Guitar

4. Camera Stability

One of the easy ways of success in low light is to be able to use slow shutter speed, at least for relatively still subjects. The problem here is that slow speed results in camera shake. It is, therefore, important to learn how to have a stable grip on the camera. I myself, took a very long time getting this right. Part of the reason is that people talk about this aspect very infrequently on forums. Part of it is that you only get the information in bits and pieces. I’d invite you to go through this excellent blog post on how to make a “human tripod”. Supporting the camera: holding with your hands.

One of the techniques I often apply when faced with camera shake is to burst a series of shots. Chances are that one of those shots will be sharp. It frequently happens to be the first in the series and I find bursting for more than 2 seconds to be a waste. The following is a 1/4s shot made using a Canon PowerShot A630 P&S in burst mode.

More Nikon D80 Switches

That’s it for this article. It still talks about shooting at high ISO and it is inevitable that you’ll get noise in such shots. In the next post, I’ll talk about development techniques that would help you reduce the apparent noise, beyond just pushing the NR slider all the way up. Oh yeah, you’re shooting RAW, aren’t you? 😉