Check out the video below:

Pretty impressive.

Lightroom 3 and ACR will ship with a selection of lenses for automatic correction of distortions, chromatic aberrations and vignetting but the really powerful feature is the profile editor, which will allow the creation of profiles for any camera/lens combination, very useful when dealing with third-party lenses or old models.

My only complain is that the perspective correction is not good enough; having three sliders for X, Y and Z-axis rotation is cumbersome compared to DXO keytoning feature.

Overall, it’s a very important upgrade.

Setting it up

1. Open Edit –> Preferences –> Guides, Grid and Slices
2. Edit Gridline every… 100 percent
3. Subdivisions… 3
4. Click OK

That’s it!

Now to show or hide the grid, just press CTRL+’ or select View –> Show –> Grid

Image with visible grid

What is this Rule of Thirds anyway?

Just in case you never heard of this “rule”, let me give you some points:

Beginners in photography tend to frame everything dead in the center. Sometimes it works, sometimes it doesn’t. There are many rules that can be used to achieve a more pleasing composition and this is one of them.

The Rule of Thirds states that you can achieve better composition balance if you place the interesting features of your photos along horizontal and/or vertical lines that divide your picture in thirds, or at the intersection points.

To illustrate my point, have a look at the pictures below:

Horizon and tower aligned to the lines

Eye at the intersection

As with any photography “rule”, you shouldn’t get obsessed with it. Not everything has to be framed according to it. Understand it, learn it, try it – then break it!

Updated on January 4, 2010 with the final release of Bibble 5.

The lineup

For this test I selected:

• Adobe Photoshop Lightroom 3 beta;
• Bibble 5 Pro;
• Capture One 5 Pro;
• DXO Optics Pro 6.

Notable absents would be Apple Aperture and Silky Pix.

I did not select Aperture for this review because it’s older (apart from updates and patches it’s still a 2008 product) and it’s also the only Mac-only product (all the programs in this test are available for both PC and Mac; Bibble is also available for Linux). In the case of Silky Pix, I started testing it, but I felt it’s not in the same league as the others, so a direct comparison would not be fair.

Also, in the case of Lightroom 3, it is still beta software; I decided that it’s stable enough to be used in production, so it wouldn’t be unfair to it. It’s good to keep in mind however that it may add features and/or improve image quality before it is released.

Originally, this article presented the beta version of Bibble 5, once it was released I retested and included new results from it.

Test images

For this test I found myself in the difficult position of not having enough test images. I made a habit years ago to convert all my pics to DNG. Unfortunately, DXO and Bibble don’t support DNG, so for this test I had to find some RAWs. If you think that would be easy, well, I actually had to run a program to recover some pics from my cards.

Second of all, I initially had some nice architectural pictures from Milan and Paris that were taken with a compact camera (a perfect test for noise reduction, chromatic aberrations and so on). Again, some programs (DXO and Capture One) did not support it.

Third, I wanted flawed images. I believe that test images should not be perfect – how would I be able to check chromatic aberrations if the image was taken with a razor-sharp, $2000 lens? Or how could I evaluate highlight recovery for a perfectly-exposed scene? Or noise at low ISO? or… you get the picture.

So if you look at the images in the test and wonder why they’re not so good – it’s because I selected them to be like that.

Interface

All four programs have remarkably similar interfaces: dark/gray color schemes, tabs and sliders.

All interfaces look similar, but there are differences in organization and workflow

I absolutely love Capture One‘s interface. It’s polished, simple and logical. It’s very easy to find the tools and the tabs are arranged in the proper order – from capture to details.

Lightroom comes very close, but for it the keyword would be ‘customization’. You can easily customize the interface to fit your own workflow, turn panels off, even customize the “nameplate” to your company name – a nice touch for when you’re working with a client.

Bibble is more of a mixed bag. They changed the UI completely from version 4 and copied many concepts from their competitors, but it’s still a bit confusing as buttons are all over the place.

DXO suffers from the same problems as Bibble – a rather confusing interface, with options arranged not very intuitively; for example, why is White Balance (a basic setting) placed after DXO Lighting (an advanced feature)? Still, in terms of speed and stability, it’s a huge improvement over DXO5.

Import and export

Lightroom has probably the widest camera support – it can read all camera formats under then sun and it also supports DNG (obviously, they invented it). Lightroom also goes beyond the typical RAW processor, as it aims to be a full DAM (Digital Asset Manager). In two years, it still hasn’t convinced me to use it exclusively, but it’s simple enough to import photos in its database and process them. It can also handle variations, has an unlimited history, snapshots and more. New in version 3 is the export system – you can publish from it directly to a service like Flikr. For me, it would be incredible to output directly to iStock, so I can only hope a plugin will come.

Bibble also has a wide camera support, but they don’t support DNG (despite numerous customers requesting it, I might add). Import and export are very easy and I like how printing is also treated like an output option and how easy it’s to define your own presets.

Capture One has a very straightforward way for import and export, focusing on simplicity. I might add that tethered shooting is perfectly integrated in the interface, a bonus for studio photographers. Capture One is more picky about direct camera support, but it does support DNG, so you can always convert to DNG first and then import.

DXO is the most cumbersome in this regard. First you must drag-and-drop files from the file system to a project, edit them and then drag and drop files from the project to the output queue. It just feels tedious for me, without enabling any advanced behaviour. DXO is also very picky about camera support and adds lenses into equation as it provides automatic correction for known camera-lens combinations. It also doesn’t import DNG (why oh why); it can output linear DNG, a feature I never used.

Unique features

As mentioned above, Lightroom is not only a RAW converter but a complete DAM, which comes in handy when you have tens of thousands of images (I have about 16,000 DNGs since 2003). You can easily filter by multiple criteria – tip: it really helps to use keywords for all images. The unlimited history, snapshots, virtual copies help you play creatively with photos, trying different looks. On the other hand, it’s the only program in this test that does not posses any sort of lens distortion correction, which is very disappointing (I still have hopes for the final version). Since version 2, LR also has some localized corrections (color, sharpness, exposure) via a brush or a gradient.

DXO is the unchallenged master in optical corrections. If you have a supported camera and lens, the precise corrections are great, and even if the lens is not recognized, you can still apply most of the corrections – sharpness falloff, light falloff, distortion corrections, even fish-eye corrections. The most impressive feature would be the keystoning, which lets you easily correct perspective distortions – a must-have for architectural shots (not everyone has a tilt-shift lens). It also has a built-in color rendering for emulating different film looks and more.

Capture One has the best support for tethered shooting  and a very useful focus checker – it will overlay the areas that are in focus, a great way to quickly asses a collection of images without zooming in, making it almost perfect for studio work.

Bibble has an incredibly powerful feature: layers. You can make a selection on the image and make any and all settings apply only to it. You can have an image with two or more different WB settings if you want. Areas can be defined using some tools or brushes and you can easily set/change the feathering. The system goes way beyond what Lightroom can offer. Bibble also has some interesting features licensed from other companies –  Noise Ninja for image reduction and a lens distortion database from PT Lens.

Processing Quality

This is where the fun begins.

Color rendition – portraits

With each program, I did not rely on defaults, but tried to achieve the best look. Looking at the image below, you can still see that each one has a different philosophy:

I think B5 nails it best with Lr3 second; though many will prefer C1's more natural look or even DXO's vividness

White balance was set for the black+white dress. I did not enable any “creative” color settings or film emulations.

Capture One really likes to stay neutral and true to the ‘real’ colors. Besides the WB tool, it also has a ‘skin color tool’ for achieving a pleasant skin color.

Lightroom throws more contrast; at least for my camera it has the tendency for more reddish skin, which I dislike. I know I can make my own color profiles (and I have), but how many users will?

Even with the Portrait contrast setting, Bibble generated a more contrasty image, while DXO created a very punchy result – very contasty, more yellowish skin and lipstick.

DXO6 makes gets too much contrast and C1 too little.

Four programs – four interpretations of the same scene with WB set for the eye white. What’s not to love?

Capture One again favors a very bland/neutral look, while DXO goes “a little” overboard with contrast. I think I like Bibble best.

Color rendition – landscapes

I really like the way Lr3 renders colors; DXO is over the top

For this image I also used a little Fill Light to open the shadows in the flowers and fruits. WB was adjusted for the white wall. Both Lightroom and DXO produce beautiful colors – Lightroom with a great blue sky (not very realistic since it was overcast, but beautiful nevertheless); DXO has a bit unnatural grass, but I’ve seen the color in magazines and ads a thousand times. Lightroom also gets the purples right but gets the orange walls a bit wrong. But this is just nitpicking.

Capture One misses it – the grass seems dirty/muddy and the Fill Light doesn’t work nearly as well. Bibble fares better but the colors are off and the contrast and shadow/highlight details are not that great.

Dynamic Range

You could get a taste of how well Fill Light & Highlight recovery work from the previous test, but now lets try something extreme:

DXO6 obtains the best results, followed by Lr3

DXO is really amazing with this image. Beautiful colors and look how detailed the dark, shadowed part of the building looks. Lightroom comes in close.

Initially, Bibble generated a rather poor image in the Preview release 3, but in the final version I managed to get better results.

As for Capture One, it just can’t.

Detail extraction

Let’s see now how much detail vs. noise can our four contestants extract from an image (100% crop):

Excellent detail in Capture One and Lr3

DXO produces a clean but not very detailed image, with some speckles here and there. Capture One removed too much chroma and has a quite speckled look (but note how natural the skin looks). Bibble 5 puts Noise Ninja to good use, leaving a pleasing image in terms of noise and details. Lightroom 3 renders the best image in terms of details, even though it’s slightly noisier.

Noise reduction

I don’t shoot usually at high ISO, but I have for this test- a boring detail of my keyboard.

It's B5's turn to show what it can do

The image above was shot at ISO6400.

Thanks to Noise Ninja, Bibble gets the crown. Lightroom 3 has the luminance noise reduction disabled in the beta, so the final version may look different. Although very noisy, it’s not that bad – I prefer Lr3 over C1. DXO cleans the luminance nicely but leaves some coarse chroma noise. With Capture One I tried 2 different approaches but the results are still very poor. The one shown above has luminance noise reduction very low, chroma to max; also reduced sharpening to zero.

Optical corrections

Unfortunately I did not have any RAW picture to really have a need to correct the chromatic abberations or barrel/pincushion. As I mentioned in the beginning, almost all of my pictures are DNG.

Having said that, based on my past experience I can safely say that DXO is unmatched in optical corrections – especially with a supported lens, whereas Lightroom comes in last since it has no distortion correction.

Here’s something from a previous test in 2007:

DXO Keystoning in action

I was able to test for purple fringing:

DXO really saves this image

Bibble has the least effective fringing correction (I couldn’t actually see it making any difference, although the Fringing option was set to Standard). Lightroom is also largely ineffective in this area (Purple fringing was set to All Edges). Capture One corrects the purple fringing but leaves a glow. DXO corrects both the color and the luminance in the affected area, leaving a sharp image.

It’s also worth mentioning the noise – Lightroom image is noisy but detailed (look at the black lace), while Capture One’s looks clean but speckled.

Speed

Speed comparisons are not very fair, as we’re comparing apples and oranges – some programs offer more advanced corrections so it wouldn’t be fair to say that they are slow when they also offer superior quality. Also, speed is dependent on many factors and will vary greatly from image to image.

As a general statement, Bibble is the fastest, followed by Lightroom and Capture One and unsurprisingly DXO comes at the end.

Pricing

Quality is not the only factor in making a decision. Let’s see how competitive are the prices:

• Adobe Lightroom 2: $300 USD (I can only assume version 3 will be priced the same)
• Bibble 5 Pro: $200 USD
• Capture One 5 Pro: $400 USD ($130 for the standard version)
• DXO Optics Pro 6 – Elite Edition : $200 USD ($109 for Standard edition)

Some notes:

• Capture One 5 standard does not offer optical correction and tethered shooting.
• DXO differentiates Stardard and Elite based on the supported camera bodies. High-end (full frame) digital SLRs require the Elite edition.

Updates

I intend to update this article when the final versions of Lightroom 3 and Bibble 5 are released and to also add more tests – so please come back.

I will also gladly correct any factual errors that may have slipped in this review (I worked for a week on it).

Updated on January 4, 2010 with the final release of Bibble 5. I changed some three of the Bibble results and used the new sales price.

Conclusions

It should come to no surprise that there’s no “perfect” program – each one has its strengths and weaknesses and your decision on which one is better for you should be based on the type of photography you do.

Having said that, I have a hard time recommending Capture One. I really wanted to like it and I know it has its fans, but I just can’t justify its price tag vs the output quality or the workflow. The interface is nice and clean, but it has nothing worth 4 times the price of DXO or the quality and workflow offered by Lightroom. It may work well for studio photographers because of its tethered shooting features and focus checker but even then it might be worth testing it along with Lightroom.

In terms of features and quality over price, DXO 6 is an absolute steal. If your camera and lenses are supported, $110 is nothing, even if you also use a different RAW converter. If you shoot landscapes or architectural elements, HDR, ultra-wide and so on DXO is a must. DXO 5 used to be a nightmare, with frequent crashes and a very slow interface, but DXO 6 is nothing like that – it offers a solid experience and in many cases it produces beautiful results.

Bibble 5 has great potential – it’s very fast and its adjustement layers feature is truly powerful. It supports more camera models than DXO and Noise Ninja does wonders for high-ISO images (obviously, you can always purchase Noise Ninja or Neat Image separately).

Lightroom is a very good all-round program. Its image management and workflow features are very useful for professional photographers. On the other hand, it’s relatively more expensive and lacks more advanced optical corrections. Adobe has included optical corrections settings in the latest DNG standard I hopefully support will be added in Lightroom 3.

So which one is best for you?

If you’re an amateur with a compact camera that supports RAW and want to get creative, go with Bibble. It will correct noise problems and optical distortions and will let you get creative.

Landscape and architectural photographers will benefit from DXO advanced corrections.

Portrait and studio photographers who work closely with their clients and shoot thousands of photos will be best suited by Lightroom.

Capture One has its fans too. It has some very professional features but its scope is limited. I would recommend it for professional studio photographers who worry more about correct colors than high ISO performance.

Ethics? What ethics?

Without trying to justify the abuses of retouching, it does has its uses. Acne for example is something transitory, I’ve had models with a few spots now and a perfect skin a week later – what should have I done, postpone the shoot? Ask her to put a thicker layer of makeup? (have you seen how much makeup is used in film/tv?) Moreover, the camera and studio lights are unforgiving. Stuff you’ll never notice in real life, like small wrinkles, shine and so on, become painfully obvious at when captured in full 25 Mp resolution. And even before Photoshop, people were using makeup and airbrushing and favourable angles to get the most flattering look.

The moral dilemma is, obviously, where to draw the line?

Portrait Professional 9 from Anthropics is definitely not going to make the previous question any simpler because, if anything, it pushes the boundaries further, allowing one to enhance portraits so easily it’s almost ridiculous.

How it works

At a glance, it’s a stand-alone program (not a Photoshop plugin) that does face sculpting, eye recoloring, skin repair, hair repair, teeth whitening and more. What’s really cool about it is that it creates some sort of a 3D model of the face, so after you set the control points, most adjustments are fully automatic.

You start by loading a picture and choosing the gender. You then click to indicate the corners of the eyes, the tip of the nose and the mouth. After that the program will build something like a wireframe model of the head, which you can fine-tune. I found that this is the most important stage; misplacing control points can lead to poor results.

In this screen you align the wireframe model with the facial features

After the model is built, the fun begins. The program has a very simple interface based on sliders.

The main slider areas are:

• Face sculpt (head, jaw, nose, neck, eyes, mouth);
• Skin controls (spots, wrinkles, shadows, hue, shine, texture);
• Eyes (whiten &brighten, sharpen, eye color, darken pupil, remove reflections);
• Mouth (whiten & brighten teeth, lip saturation and colour);
• Hair (shine, tidy, vibrance);
• Skin lighting (shadows, relight, contrast, highlights).

Overall, there are enough sliders to keep you happy for hours.

Almost anything can be ajusted via sliders

There are some mask-painting features to restrict the skin and hair areas. The skin controls even feature a Texture selection that allows you to retexture the skin so it doesn’t appear unnaturally smooth. The Skin lighting controls seemed the least useful, at least for me, but I think it can help in bad lighting situations.

Enough theory, let’s see it in action

Scenario 1 – spots

This is an all-time favorite. So ephemeral yet so low-esteem-inducing, they appear exactly when you want them the least – like before a photo session.

Removing spots and wrinkles is a one-click process

In the original, the acne was visible even with the makeup. The image on right is the result of automatic processing in Portrait Professional. Perfect skin yet not plastic-like.

Scenario 2 – Skin shine, teeth and eyes

Here I had a beautiful model with a different – it was very hot and the lighting was pretty crappy (indoors, available light – tungsten – yuk!).

With one click, skin was cleaned up, shine removed, teeth whitened and just for fun I recolored the eyes too

Sweat on makeup is always an issue, but Portrait Professional managed to clean it nicely. Also note some subtle teeth whitening (not really needed). Just for kicks, I made the eyes wider and changed their color.

Scenario 3 – Playing God with Face Sculpting

This is THE feature that sets Portrait Professional apart from competition. In short, it uses some internal algorithms to determine the facial structure and “push” them towards an ideal. Obviously, not everyone is going to like it. I tried some face sculpting on a photo of my wife and she hated it; others have loved the results.

This is understandable because, as the program tries to alter the photo to perfection, it also destroys one’s uniqueness (as an example, look at a beauty pageant – the contestants tend to look all the same) and I mean it when I say “perfection is boring“.

There’s something disturbingly fascinating in reshaping someone’s body to fit an idealized fantasy. Perhaps Photoshop (and plastic surgery) has made us all into little Frankensteins or Doctors Moreau, always trying to undo God’s / Nature’s work. But I digress.

For the final example I chose someone with a great attitude and a small flaw. She had a rather big jaw compared with the rest of her hear. I did not have Portrait Professional when I took the photos, but decided to try it now for this review and see what would the program do in her case.

Since there are many subtle changes difficult to notice side-by-side, I included a flash animation for the Before and After.

Conclusions

It took me a while to realize it, but the makers of Portrait Professional want it to be seen as an alternative to Photoshop (one of the reasons it’s a standalone app and not a plugin) and at $69 it’s quite tempting. It can definitely turn a few hours of retouching into a 10 minute play time and if used sparingly it can do wonders.

The downsides? Well, if everybody get their hand on it, no Facebook pic will remain untouched. God help us all.

Below you can see the ‘before’ and ‘after’:

Day to night conversion: before and after

The original, straight from the camera image is mind-numbingly dull; it’s the very definition of dullness. It was shot in a overcast February afternoon. As boring as it is, it was perfect for my purposes. I wanted to convey a sense of ‘out of place’, of a person that doesn’t really belong there, forcing the viewer to create a story – who is this young and attractive woman and what is she doing in this miserable street?

The overcast day had the advantage of decreased contrast and a complete lack of shadows, allowing me a wide range of editing.

Creating the light/dark versions

First thing, I created two Virtual Copies in Lightroom:

Original, with default contrast. Slightly overexposed too.

The “lit” version, Contrast was set to 100, also Clarity was was to max, increased vibrance. White Balance was set to Cloudy

The “unlit” version. The image was underexposed by 2 stops. White Balance was set to Tungsten.

Same image with different WB settings

As you can see above, the altered versions are already better — especially the “night” one, dark and moody. However, the street lamp gave me the opportunity to go the extra mile in creating something realistic.

Mixing light and dark

Having exported the two versions as 16bit TIFF (for maximum quality), I opened both of them in Photoshop. I then copied the “dark” one and pasted as a layer over the “lit” version.

Creating a new layer mask

It was pretty clear already that quite a lot of trial-and-error was to be expected, so instead of simply using the Eraser to remove parts of the “dark” layer, I decided to use a mask instead.

I selected the Dark layer and from the Masks panel, I clicked on the Pixel Mask button to create a new mask.

A Pixel Mask works just like an alpha channel for the layer – it’s a grayscale bitmap where white is opaque and black is completely transparent. What’s cool about it is that you can alter the opacity of the layer by painting on the mask.

The effect of the layer mask in mixing the lit/unlit areas

You can see on the left how the mask affects the mix between lit and unlit areas.

I clicked on the mask in the Layers panel to select it and then I simply used a semi-transparent, highly feathered black brush to create transparent areas.

There are three main lit areas: the light on the wall, faing out, the light on the pavement the light spill on the roof. You can notice that I preferred to paint everything instead of using gradients, to avoid the artificial “perfect” look.

Adjusting the exposure

The light was still too even in the lit area; close the the light source I needed something much more powerful (remember that light falls off exponentially). To improve the realism, I added an Adjustment Layer just above the Lit layer.

The Exposure Adjustment Layer

The Adjustment Layer was set to Exposure (Layer –> New Adjustment Layer –> Exposure). I added a simple radial gradient as a mask for this layer the same way as previously and then I tweaked it to get a slightly overexposed look.

Adding the glow

By now I had the light on the wall and pavement pretty much in place; it was time to turn to the light bulb. For this, I simply used the Lasso tool to select the visible area of the bulb and then copied and pasted it as new layer. I then used the Curves to make the bulb much brighter. Finally, from Layer –> Layer Style –> Outer Glow I created a nice amber glow around the bulb.

Setting the glow on the light bulb

Below you can see the exact settings for the glow:

Light bulb glow settings

I still wanted to add some nice touches: the glow should not be that strong in the upper area. Can you guess what I used? Yep, another Layer Mask.

Please note: By default, layer masks do not affect the layer effects. To make a layer mask hide the effect, open the Layer Style window and go to Blending Options section. From there, check the “Layer mask hides effects” option.

In the layer mask, I painted in black the areas I wanted the glow to be weak. Below you can see the result:

Glowing behind the edges

As you can see, I did not eliminate the glow completely, but created a glow-behind-the-edges effect that is seen in high-contrast situations.

Lens flare

A simple lens flare

Most designers, upon hearing the words “lens flare” run away screaming. They are very often overused and cheesy. Even the new JJ Abrams’ Star Trek features some lens flares annoying as hell. Still, they an unavoidable part of photography and can add some realism if used wisely.

Subtle lens flare applied

Unfortunately, Photoshop’s Lens Flare effect is laughable. I think it’s the same effect as 15 years ago. I only know of two decent lens effects – one in the old plugin Kai Power Tools 6, the other is Corel Photo-Paint. Both of them allow you to control the size, glow, ring, stars, streaks and reflection trail. This is not a tutorial on lens flares, but modern, good lenses don’t create reflection trails, but only a nice star pattern with minimal interference or random streaks.

For this photo I created a very simple star pattern on black background and I put it as a layer with blending mode set to Linear Dodge (Add) at 33% opacity. You can see that the effect is barely there; its purpose is to enhance the scene in a minimal way, not to overpower it.

Shadows

Selecting the body shape

If you recall from the beginning of this tutorial, the original image had no shadows because of the overcast sky. Now, as I created a light source, I needed to create a shadow too.

Refining selection edges

With the background layer active, I started to make a simple selection on the girl using the Magnetic Lasso. I didn’t even needed to be very careful, but just to follow the contours.

Once this was done, I clicked on Refine Edge and increased Smoothness and Feathering. After that, I created a new layer and filled the inside of the selection with black.

Distorting the shadow

The next step was to convert the new layer into a Smart Object via Layer –> Smart Objects –> Convert to Smart Object. Why? It’s because a smart object’s original appearance is preserved so I could distort it any way I wanted without degrading its appearance on each step.

Finally, I set the layer’s blending to Multiply and its opacity down to 50%.

Final result

Here’s the final view with all the layers:

All layers

In case you are curious what I’m dealing with in this article, have a look:

HDR of Orthodox Cathedral, Sibiu, Romania

The cathedral is located here.

If you find it interesting, read on.

Introduction

Just in case you’re unfamiliar with the therm, HDRI stands for High Dynamic Range Imaging. As I’ve discussed before, a regular photo can’t capture the full range of lights and shadows that the human eye can see. Photographers have long learned to avoid high-contrast situations or to use them to create artistic effects. It all changed when Paul Devebec presented at SIGGRAPH 1997 a method for combining several images with different exposures into one single image with a much higher range of luminosity. To put things in perspective, a camera sensor can cope with contrast ranges on the order of 4000:1, while a scene of a room with an outside view in full sun (something our eyes see every day) has a dynamic range of 100,000:1.

Initially HDR was used in 3D graphics as environment maps to create realistic scenes. Because a HDR image contains 32bits per channel (in floating point, e.g. can take almost infinite values), it cannot be displayed directly on the screen or printed on paper, so a second operation is needed, called Tone Mapping. Tone Mapping consists in “compressing” the high dynamic range back into something we can see on screen. This is something easier said than done; a simple conversion results in a lifeless, murky image so there’s a whole research field in creating algorithms that present images that are realistic and pleasing for the eye.

Shooting the scene

Equipment used:

• Sony α700 DSLR;
• Sigma 10-20mm lens;
• tripod (Canon, in case you’re interested).

First, a straight JPEG of the cathedral interior is hopelessly bad. The stained glasses are blown out while many areas are pure black. Colors are muted and overall it’s not an incredibly attractive picture. Shooting with Dynamic Range Optimizer marginally improves the shadow detail, but not by much.

JPEG image lacks the needed dynamic range

So, I put the camera on a tripod and shot eight pictures, from under- to over-exposed, one f-stop apart each. I started with just enough light to see the windows and ended with an exposure long enough to see clearly in shadows. Here they are:

8 images with different exposures make the final HDR

Picturenaut

Picturenaut is a nice little FREE tool from a guy named Christian Bloch. The interface is pretty barren, but it has to be the fastest HDR tool I’ve tried. Did I mention it’s free? (some tools can cost $700)

Selecting images for HDR

Tone Mapping dialog

First step is selecting all images that will make the final exposure. Picturenaut has the option to automatically align the images, which is extremely important since even with the camera on tripod the image can be slightly different from shot to shot.

Aside from being very accurate, Picturenaut is really, really fast. It generates the HDR in no time. The HDR image displayed on screen usually looks rather flat, but that’s expected. You can save it as a 32bit TIFF if you want.

Second step is the “fun” part – tone mapping. Picturenaut offers two algorithms (Adaptive Logarithmic and Photoreceptor physiology). Which one is better depends on the actual scene.

The tone mapping is also very fast, almost real-time.

Photoshop

Image selection for HDR

Tone Mapping in Ps

You access the HDR option from File –> Automate –> Merge to HDR (rather unintuitive, but that’s me). The process is very slow, “Aligning layers based on content” message seems to take forever.

For Tone Mapping (achieved via Image –> Mode –> 8 bits/channel), Photoshop offers four options: “Exposure and Gamma“, “Highlight Compression“, “Equalize Histogram” and “Local Adaptation“. Again, depending on the scene, one mode may be more useful that the other, but I find Local Adaptation to be good, with Highlight Compression and Equalize Histogram essentially useless.

Image Quality: Photoshop vs. Picturenaut

With both programs, after tone mapping, I spent some time tweaking the image, playing with Curves to increase the contrast and Saturation to bring out the true colors I saw in the cathedral.

Here’s the side-by-side end result:

HDR result - Photoshop vs. Picturenaut

Apart from some color differences, the two images are remarkably similar. I did manage to preserve highlight detail better in Photoshop and its local contrast made everything “pop” more, but overall they are on par, so if you can’t afford Photoshop or some “pro” tool, Picturenaut will deliver good results (but you still need to tweak it, it’s not a “push the button and you’re done” kind of operation).

What is this raw thing, anyway?

Camera sensor

Diagram of a Bayer sensor array

RAW in itself is just a term denoting an unprocessed image. It’s not a file format; all camera manufacturers (Canon, Nikon, Sony, etc.) have their own file formats to store this unprocessed information.

You see, all digital cameras work in the same way: there’s a sensor that is made of 6 (or 8 or 24) million elements laid out in rows and columns that can capture light. Each element produces a pixel in the final image. That’s where the “megapixels” term comes from. Now, these tiny elements don’t capture color, only brightness (there’s an exception, the Foveon sensor). To capture color, a trick is used, in the form of a so-called Bayer filter array placed in front of the sensor. The Bayer filter makes so that each recorded pixel can be either red, green or blue.

So, when saving a RAW kind of file, the camera simply takes all values read from each element on the senor, adds some info (date, time, camera settings and so on) and just writes that on the card, with no further processing.

How does the camera process the image?

When you’re saving files as JPEG, the camera has to go through several steps to get an usable image

1. take raw sensor data;
2. for each color channel (red, green, blue), pixels are interpolated to fill the gaps;
3. color balance and gamma correction are applied;
4. noise reduction, contrast, saturation and sharpening are applied;
5. image is encoded as JPEG and written to card.

How the RAW information is processed

Is RAW needed?

OK, you can’t see a RAW directly and the camera does the processing in less than a second, so why should we bother with RAW?

The answer is that under perfect shooting conditions, RAW wouldn’t be needed – that means perfect white balance, perfect exposure, perfect camera effect settings. In real life however, you don’t always get to control all variables. Color may be off (do you always calibrate the white balance?), you may have got blow-out (white) areas with no detail, and so on. You could try to tweak the jpeg in Photoshop, but if the shadow or highlight details are simply not there, you’ll never recover it.

What are the advantages for saving in RAW format?

Short answer: it allows you to correct your mistakes and be more creative.

Sounds good? Read on for the long answer.

Like I said in the beginning, the RAW file contains unprocessed data. It also contain more nuanced information per-pixel too. JPEG files store can represent 256 levels of brightness (8 bits/channel). RAW on the other hand allocates 12 (sometimes 14) bits for each pixel, resulting in at least 4096 levels of brightness! This helps you bring out details from the shadows or from blown-out areas.

Also, because when the image is saved, white balance is not applied yet, it can be easily set to whatever you want. If you ever took a photo with the wrong WB setting and then tried to fix it in Photohop, you know how hard it is. When editing the RAW, this is simple and accurate.

This is not all. In-camera conversion is done by the camera’s small processor, that also needs to do the processing in a fraction of a second, so all algorithms are optimized for speed. Modern RAW processing software allow for much sophisticated tools, resulting in more detail, better sharpness and less noise. Some of them even offer automatic optics compensation for chromatic aberrations, purple fringing and lens distortion.

How about some pics?

There are many articles on the web discussing the pros and cons of usig RAW, but very few of them actually post some examples. Why? Well, it’s most likely because artists hate for people too see their unfinished work. Still, seeing is believing, so let me show you some examples:

The RAW image (right) allowed for the image appearance to be improved compared to the JPEG (left)

I took this picture at Sacré-Cœur, Paris. The sun was very low in the sky, beginning twilight. The white travertine stone it’s made of was shining against the blue sky. Still, as you can see from the photo on the left, the camera could not capture all this. To avoid the sky and clouds from being blown out I had to reduce the exposure, resulting in a dark building, while the white balance, set on Auto, made a reddish cast on the whole picture. The image on the right was obtained in about a minute by processing the RAW. I set the white balance to make the stone white, increased the exposure a bit to bring out th details and added some highlight recovery to still have details in the clouds.

The RAW image

The differences in the image above, while not striking, are clearly visible; the eye at the bottom contains more detail and has less artifacts. Sure, the level of detail may not be important if you only print 6″x4″ (15x10cm) but if you’re doing big prints, all small details become important.

What’s the catch?

RAW is not the holy grail of photography. It can’t turn any snapshot into a work of art. On top of that, it has some clear downsides:

• bigger file size. RAW files are uncompressed or compressed using a lossless (think ZIP) compression. This is good because nothing is lost when the file is saved, but bad because instead of a 1-2Mb JPEG, you get a 8-10 MB RAW file. So if you plan on shooting RAW, you’ll need bigger cards.
• because of the bigger file size, it takes more time to save the files. Normal dSLR cameras have a buffer large enough to store about 5 RAWs, so if you’re shooting continuously, you’ll have to pause when the buffer is full. This is particularly annoying for sport and fast action.
• downloading and processing takes time. If time is of the essence, RAWs are not the way to go. This is why photojournalists don’t usually use RAW.

Do I need any special software?

Lightroom 2

Cameras come with some proprietary software for processing RAWs, but they are usually quite limited. There are a number of software capable of processing RAW images, some more powerful than others (I made a review of the biggest of them, here).

For now, my favorite in terms on features, workflow  and user-friendliness is Adobe’s Lightroom 2; other popular software include ACDSee Pro, Bibble, DXO and SilkyPix.

Conclusion?

In a nutshell, RAW is not for everyone, but if you are a creative type and like to get the most out of you camera, I think you should try it.

General information about the cameras  I’ve used in this experiment:

The Sony Cyber-shot DSC-W50 is stylish in its metal ultra-compact body, and features 6.0MP resolution on a 1/2.5-in. image sensor, 3x optical zoom and an extra large 2.5-in. LCD monitor.

The Fuji S8000fd has a tempting specification. It has an 8-megapixel CCD (although not a SuperCCD), and an 18x optical zoom, f/2.8-4.5 lens with a focal length range equivalent to 27- 486 mm.

The story I like to tell you:

First of all I’m not talking about technical details, those details are for professional photographers, I’m just offering you my honest opinion about two different photo cameras in a view  that you can actually understand.

I went home in Hermannstadt and got together with an old friend of mine and decided to start to make her some photos, I had two cameras with me Sony Cyber-shot DSC-W50 and Fuji S8000fd both in a very good condition.  I’ve been using both cameras for more than 6 months so there isn’t any doubt that I don’t know their settings by heart. If you want to make your first baby step towards  digital art photography a Sony Cyber Shot is a very good choice although a Fuji might seem a wiser choice, being a semi professional camera. A small camera as the Sony I presented is more light and easy to handle, you can fallow your model pretty easy whereas the Fuji is a little bit heavier and takes a lot of time when to comes to settings (complicated at the first look but very simplistic). Although another bonus point for our little Sony is its magical auto focus which Fuji doesn’t quite have, if your hand trembles a little your beautiful photo is gone.  The quality of a picture made with small Sony stunned me the first time I’ve used it, being a photshop fan, I like my pictures to be as clear as possible at its best quality and there’s where I couldn’t believe my eyes, a photo made by Sony, all of them actually fitted better with my needs than what my old Fuji had to offer. I have to admit the fact that I am also a sepia fan,  little Sony helped me there too, Sony has the best sepia color that not even photoshop can match whereas Fuji only allows you two options:  Black and White and Normal – disappointing!

These are simple things that you can easily notice too, remember also that both cameras are at an accessible price, of course that the Fuji is a little more expensive but I suggest you to invest in a Sony because in this case the cost does not represent  quality.

Here are the photos:

Photos made with Fuji S8000fd

I had some troubles coping with the model, the camera didn’t permit me to move that freely.

Photos made with Sony Cyber-shot DSC-W50

The little Sony allowed me to move easily, crawl and even lift it above my head so I could capture the model from different angles.

Exposure and lighting affect the mood of the photo

To aid those who want to be in control of their cameras, to go beyond point-and-shoot, I decided to write a series of tutorials on the basics of photography. If you don’t know what the exposure or aperture are, this little tutorial is for you, so read on. I’ve done my best to avoid overcomplicated technical details and give lots of analogies and examples instead.

What is exposure

The single most important thing to learn about when making a photo is exposure. You’ll never go beyond point-and-shoot without understanding how this works. Any gorgeous scene can and will be ruined by bad exposure — and you can turn a mundane setting into a startling photo by mastering the same exposure.

Tricky lighting required careful control of the exposure

In this photo, lighting the model while keeping the details in the sky was tricky, because the setting sun was behind the model.

How do we capture the light in the scene in a photo? Whether we talk about film or digital, the principle is the same: we’re letting a piece of light-sensitive material to gather light from the scene.

How much light do we need to capture? Here’s the tricky part. Each scene has a certain amount of illumination that can be measured objectively using a light meter. The typical unit of measure for luminance is called “lux” but most light meters are set up to display results in photographic terms (more on that later). Most likely however, you won’t have a light meter, so you can’t rely on them. All photo cameras have internal light meters, except they can’t measure the actual light levels from the light source(s), but the light that enters the camera, which is the reflected light by the scene objects. Why is this important? It’s because the camera has to “guess”. The camera tries to find a good average of the light intensities – here’s how it works:

what you see - what the camera meter sees - averaged gray

The camera takes readings of light intensities across different areas of the image. The number of segment and pattern varies from model to model. The readings are averaged based on different algorithms (“evaluative”, “3D matrix”, “center-weight”, etc.) and the exposure is adjusted to obtain a medium gray.

This method works most of the time – but sometimes it misses. Do you have your camera nearby? Take a picture of a white sheet of paper. — just the white paper, without anything on it. Download the picture and look at it on the screen. I’m willing to bet it came out gray-ish. Take another shot of a black T-shirt. Have a look at it, it’ll be gray or dark gray. Why?

Dark backgrounds will fool the camera meter

White or dark backgrounds can fool the camera meter. The camera tries to average everything to gray. If a scene has bright areas and dark areas, the camera can determine a proper average and from that, it can determine the exposure. But, if you’re shooting tricky scenes, it’ll fail (even a sunset or a portrait against a white wall can be considered tricky, so don’t think you won’t encounter such cases).

Hopefully, by now you understand what exposure is and why it’s important to get it right. Next, it’s how do we get it.

The right exposure – juggling with shutter, aperture and ISO values

Obtaining the correct exposure is like filling a bucket of water. Too little water and you don’t have enough to use it; too much water and it spills, making a mess.

Too little - Just right - too much

To fill the bucket with water you turn the tap. The longer you keep it open, the more water pours. You can also increase or decrease the rate of flow. So in effect you can get the same result either by keeping the tap open for longer and reducing the flow, or with a larger rate of flow for a shorter time. This is exactly how exposure works, except we use light instead of water.

The amount of time you let water, err, light “pour in” the camera is called “exposure time” or “shutter speed” and it can vary anywhere from 1/10000 of a second to minutes or hours. The amount of flow has its equivalent in “aperture size”, which basically measures how big the opening that lets the light shine on sensor or film is.

These two values – shutter speed and aperture are intertwined. To get the correct exposure, both have to be set correctly – that is, if you increase the exposure time then you need to decrease the aperture.

You may ask yourself now – why should we deal with two variables instead of one? Why not keep the aperture wide open and change just the shutter speed? It’s because unlike the bucket of water analogy, you’ll get different results depending on how you set these variables. Cellphone cameras or cameras set on Auto will try to work out a compromise, some sort of average between these two. Modes like “Portrait” or “Sport” are simple presets, favoring one or the other variable.

Shutter speed

The shutter looks like a curtain and acts like the tap in our example.When you press the shutter release button, it is opened briefly, allowing light to expose the film or sensor for the specified amount of time and then it’s closed again.

Fast action

The shutter speed value is specified in seconds, for example “2s”, or “1/30s”. High shutter speeds (low exposure times) are needed to capture fast action. Capturing a bullet as it passes through a glass requires extremely short exposure times, like 1/10000s, which are usually obtainable only in controlled environments. Fast sports or dancing need exposures like 1/125s – 1/250s. At the other end, night scenes may require exposures of several seconds, even minutes, which lead to interesting effects.

Aperture

The aperture is iris-shaped, with metal blades that allow it to grow or shrink. The larger it is, the more light enters the camera. Just as the pupils get narrower in bright light and wider when it’s dark, a photographer will adjust the aperture size to gather the needed light.

The aperture size is measured using a rather unintuitive notation, such as f/2 or f/16. The larger the number, the smaller the aperture. An aperture of f/1.4 is very very wide, while f/32 is extremely narrow.

Changing the aperture has a very clear impact on the pictures, that is called Depth of Field. Depth of Field (or DOF for short) is an interesting topic on its own, that’s outside the purpose of this tutorial. Depending on the camera you have, altering the aperture may or may not be that apparent. Compact cameras have a lot more depth than (d)SLRs, which can be an advantage or disadvantage depending on situation.

Shallow DOF - Deep DOF

Above you can see pictures taken with different aperture values. Note how the image is blurred in front and behind of the subject when the aperture is wide open compared to the picture taken with a small aperture. A “shallow” DOF that has little detail in front and behind the subject is desirable especially for portraits, because a blurred background doesn’t distract from the main subject; on the other hand, landscapes benefit from a deep DOF that ensures that all close and distant features are sharp.

ISO Speed

Going back to our initial tap analogy, if you can alter the quantity of water that is collected in a bucket by altering the time you let the tap open or by increasing/reducing the flow, what happens if you change the density of the poured liquid? If you want a pound of liquid, then the denser it is, the quicker you’d get it because it’d require less volume – a liquid twice the density of water would occupy half the space, so all else being equal (time, flow) you’d get it twice as fast.

It’s the same with ISO speed. For film, ISO specifies the film sensitivity to light. A ISO200 film would be twice as sensitive than a ISO100 film, a ISO400 film is twice as sensitive than ISO200 and so on. The sensitivity ranges usually from ISO50 to ISO3200.

Without a tripod, this scene required high ISO

As with all things in life, by gaining sensitivity, you lose something else. More sensitive films (also called “fast” films) result in grainier pictures, with less vibrant colors (actually all ISO3200 films I know of are black-and-white only). With digital, things are relatively the same. The analog signal is amplified by software, resulting in more visible noise, especially in darker areas. The camera software tries to clean some of the noise with varying degrees of success. Noise levels are dependent on a number of factors, among which the sensor size and resolution are very important; digital SLRs with their bigger sensors behave much better than compact cameras.

Putting all together

To recap, a certain exposure can be achieved using a combination of three factors: ISO, shutter speed and aperture.

ISO, Shutter speed and aperture opening

ISO goes from ISO50 (clean, bright image) to 100, 200, 400, 800, 1600 and 3200 (grainy, noisy) – doubling the sensitivity.

Shutter speed goes from let’s say 1s (slower), 1/2s, 1/4s, 1/8s, 1/15s, 1/30s (faster) and so on, doubling the speed.

Aperture has a stranger progression (than can be explained mathematically), from f/32 (narrow), f/22, f/16, f/11, f/8, f/5.6, f/4, f/2.8, f/2 (wide), each time doubling the amount of light that hits the film or sensor.

Each step in doubling or halving the exposure is called a “stop”, so the difference between 1/30s and 1/60s is one stop. The difference between f/5.6 and f/8? One stop. ISO3200 to ISO800? Gotcha – two stops. So, an exposure taken at ISO400, 1/125s, f/5.6 would be equivalent to one taken at ISO100, 1/30s, f/5.6 or ISO200, 1/15s, f/16. Modern cameras can adjust these values in half- or third- increments, so you will see values like 1/20s or f/9 or even ISO80.

Conclusion

Exposure and all the technical terms used here are really the foundation of photography. You may think you don’t need it – after all the cameras are pretty smart nowadays, right? I beg to differ. You should know your tools, first understand their capabilities, strengths and weaknesses and then use your creativity to push them to the limit.