If you’ve been in the market for a digital camera, you know that choosing one camera from all the competing brands and models can be a challenge. Many people like to research before they make a major investment, and there are some great resources to help guide you with your purchase. This blog post is mostly about a tool called Snapsort, but to more fully appreciate that tool, it’s worth briefly mentioning a few other resources that are available.

Flickr Camera Finder

A camera’s specifications are important, but so are the photographs that it can produce.
The Flickr Camera Finder is a great tool for looking at photographs taken by a specific model of camera. Flickr makes it easy to filter by the content of the photograph (such as portraits, landscapes and night shots), as well as photographs that other Flickr users found interesting. Flickr also lets users post their photos into groups that are specific to a camera or even a camera and lens combination, and these groups are another way of finding sample photos for a particular camera.

Of course, product reviews are important, and while there are many camera great review sites, two in particular stand out as my personal favorites: DPreview.com and DCResource.com.

DPReview

DPReview.com is well known for their extensive laboratory testing of cameras. Their highly technical reviews may seem intimidating if you’re less familiar with the technical aspects of photography, but you’d be hard pressed to find a site that performs more comprehensive, consistent and knowledgeable reviews. Their RAW image and lens sharpness comparison tools are commonly used as benchmarks and points of comparison across the web. Their user forums are extensive -- It’s likely that any question you have has been addressed by knowledgeable users, making this forum a top resource, despite the somewhat clunky search tools and the bitter arguing that sometimes occurs with less mature members of the forum. Still, that doesn’t detract from the overall value provided, and let’s face it: there are some immature folks on the Internet. Still DPReview, which while independenry operated, was recently purchased by Amazon, and the ability to vote up and down comments (and hide ones marked as unhelpful) would be welcome additions to this excellent site.

DCResrouce

DCResource.com is a site where Jeff Keller performs detailed camera reviews that are more approachable for less savvy users, and he also tends to review a wider range of cameras (more entry level point and shoots) than DPReview does. I have high regard for his opinion, and he is one of my favorite camera reviewers.

Information Overload

Often, after hours of online research, consumers end up with a shortlist of two or three cameras to focus on for more extensive research. Of course Flickr, DPreview,DCResource are just a few of the multitude of resources for researching cameras. Making sense of the virtually unlimited amount of information available can be a daunting task, especially if you’re not a seasoned photography expert, and can’t make camera research your full-time job.

Luckily, there is an amazing online tool that provides you with expert guidance through this sea of information.

Snapsort: It’s the bomb diggity

Yes, I said "bomb diggity."

Snapsort.com is an automated tool that provides a detailed, itemized comparison between specific camera models, and can even recommend the best model within your budget. Snapsort covers virtually every model of camera from entry-evel point and shoots to professional level DSLRs.

At the heart of Snapsort is an automated and highly comprehensive scoring system that evaluates cameras uniformly, assigning points for many different aspects of a camera, such as the size of its imaging sensor (the part of the camera analogous to film), how quickly it focuses, how many shots it can take on a single battery charge, the camera’s size and weight, along with many other factors.

Snapsort uses this overall score, and a camera’s score in some of the individual areas it measures, to provide several amazing tools for consumers looking for the right camera.

While its within-category and within brand ranking tools are highly valuable, it’s the automated camera comparison that I find most useful.

Automated camera comparison

If you’ve narrowed your shortlist down to a few models of camera, you’re ready to do a comparison. If you need help narrowing down your short list, don’t worry, Snapsort has plenty of other tools to help you hone in on a recommendation.

To perform a comparison, browse to the compare page and enter the model of each camera you’d like to compare. Snapsort instantly pulls up a picture of each camera along with an itemized list of benefits that each camera has over the other along with the cameras’ common strengths, followed by an overall score and recommendation of one camera (unless it’s too close to call).

The list of advantages for each camera are very detailed and comprehensive.
For example, if you compare the Nikon D7000 vs that Canon 7D, you’ll see that while the D7000 has a greater number of focus points, the 7D has a greater number of cross type focus points.

So, what the heck does that mean, and why should you care? Just click on the “i” icon next to any part of the comparison, and Snapsort provides you with a helpful explanation of why that particular aspect of the camera is important. These explanations are a key part of what makes Snapsort such a great resrouce.

Snapsort doesn’t just make a recommendation, it explains every aspect of each camera, so that once you’ve read a comparison, you’ll have the knowledge you need to understand the recommendation and to better understand cameras in general.

In fact, if you read a Snapsort comparison, odds are that you’ll be more knowledgeable than many salespeople you’ll encounter, and less susceptible to camera hype or pushy sales tactics.

In addition to some aspects of a camera that you might normally consider, such as its size or the resolution of its sensor, the camera comparison also points out details that can be easily overlooked, like the viewfinder magnification level and coverage level. (Some cameras only show a portion of the image that will be captured while others show 100% of the image.) Some more expensive cameras have pentaprism viewfinders which are typically brighter than less expensive pentamirror ones. Snapsort’s “true resolution” rating takes into account a camera’s actual resolution, rather than just the number of megapixels it has. These are just a few of the many handy pointers you can get from reading a Snapsort comparison.

If you’re less interested in the specifics and just want a recommendation, Snapsort’s “just tell me” feature allows you to enter your budget and receive a recommendation on what to buy.

The service that Snapsort offers is unique and valuable enough that they could probably get away with obnoxious, in-your-face advertisements to raise revenue. However, the ads on Snapsort are very unobtrusive and highly relevant to the content you’re viewing (like purchase links with the best prices for a particular camera), so they’re actually useful, rather than like a necessary evil.

There’s so much to love about Snapsort. Is there any room for improvement? I’ve got a few suggestions for making this excellent resource even better.

Improving Snapsort

Given the relative complexity of the task, it’s ridiculous how easy Snapsort makes it to read (and understand) a thorough review of cameras. Of course, this automated system wouldn’t be very valuable if the scoring system wasn’t very accurate. Luckily, the Snapsort scoring system is impressively accurate - their recommendations generally make a ton of sense. As a result, it’s hard to find much to complain about.

Still, there are few ways that Snapsort could be even more awesome. Some of those improvements have to do with the look and feel of the site, but many have to do with the scoring algorithm. Since the algorithm is at the core of everything Snapsort does, I’ll start with some suggestions for improving this part of the site.

Weight of DxOMark Scores

For some of its data points, Snapsort uses ratings from DxOMark.com, a site that performs rigorous laboratory testing of camera imaging sensors and lenses, and assigns scores in several different areas of the sensor’s performance. While Snapsort mentions the aggregate results when comparing two cameras, it uses the low light performance, color depth and dynamic range ratings from DxOMark to calculate scores. This makes sense, because weighing both the aggregate DxOMark score along with the specific scores seems like it would be double weighting.

Dynamic range

A camera’s dynamic range, measured in EV or stops, refers to its ability to capture areas with varying amounts of light and shadows in a single shot. Snapsort correctly weighs the dynamic range abilities of a camera’s sensor heavily. However, the DxOMark dynamic range rating doesn’t provide enough information here. Take for example, the sensor in the Nikon D7000, a sensor made by Sony and used in other cameras like the Pentax K-5. The DxOMark scores gives cameras with this sensor a roughly 2 EV advantage over the 18 megapixel sensor used across Canon’s APS-C format DSLR line. However, upon closer examination of DxOMark’s dynamic range chart, we can see that extent of this advantage is maintained at lower ISO values, with the gap narrowing significantly at ISO 800 and above. Snapsort could provide a more complete sense of a camera’s dynamic range capabilities by providing a score for dynamic range as reported by DxOMark at a low ISO level like, ISO 100 and a higher one, like ISO 800.

Score additional elements from the camera comparison

If you look at the score explanation that Snapsort provides with each camera comparison, you’ll see that not all items that were mentioned in the comparison are taken into account in a camera’s score. While the scoring system is still useful, it misses these valuable data points. Number of cross-type imaging sensors, pixel pitch, lens availability, and weight all seem like key aspects of the camera comparisons that would be valuable if weighted as part of the score.

Improved popularity metrics

While popularity isn’t the only indicator of quality, it certainly is important to consider, and Snapsort assigns points for a camera's popularity with a weight of 50 -- twice the weight of screen resolution viewfinder size, and in-body image stabilization, and equal to the individual DxOMark ratings. So, how does Snapsort evaluate popularity? According to the site, they measure, “popularity of a cameras based on how often users at Snapsort.com interact with them.” Page views definitely indicate interest in and excitement about a camera, but I come to Snapsort to cut through the hype surrounding camera models, not to add to it!

It would be more valuable to know how many people actually own and are using a camera, than how many people are looking it up on Snapsort’s website. There are a few different ways that Snapsort could get data about a camera’s level of real-world usage and ownership.

Flickr ranking

As part of its Camera Finder, Flickr exposes the number of items posted, average daily users, an “Activity factor” and overall rank of a camera’s popularity within Flickr. What’s useful about this data is that it’s based on real-world usage. As more people shoot with a camera and post their photos to Flickr -- Flickr surpassed 5 billion photos uploaded as of late 2010 -- we can use this data to get a sense of a camera’s popularity with confidence in a large sample size and real-world validity.

Amazon ranking

Snapsort wants to retain its sense of independence and lack of bias towards any camera manufacturer or merchant, but it’s hard to deny that Amazon is one of the largest retailers, and their category specific sales ranking would be another valuable measure of a camera’s popularity.

Similarly, Amazon and other large retailers like B&H provide average customer ratings of products. I definitely consider this data when researching a purchase, and while I realize the potential conflict of interest of using data from a particular retailer, I’d love to see this integrated as well.

As an alternative that may seem more independent, there are market share reports by camera model for different international markets, though parsing and combining these reports may be more trouble than it’s worth.

Regardless of the specific method chosen, improving the real-world vaildity of Snapsort’s popularity ranking would be a welcome addition to the algoritm, and would cetainly justify this score’s relatively heavy weight of 50 within the Snapsort scoring system.

Consider camera buffer size

One area that Snapsort seems to ignore is camera buffer size. After images are captured, they are stored in the camera’s buffer memory before being saved to the memory card. The camera can write to its internal buffer much faster than the memory card, and having a larger buffer means the camera can shoot continuously for a longer period of time before having to slow down and empty its buffer. It would be great if Snapsort could weigh continuous shooting speed and camera buffer size (measured in both RAW and JPEG format) in its camera scores.

Rank video functionality

While video isn’t my personal priority in a camera, it is for some folks.
Cameras that shoot at 1080p resolution at 24 frames per second should receive a lower score than those that can shoot 1080p at 30 fps.

Color bit depth

Comparing the Canon 7D DSLR to the Nikon D7000, we can see that the Canon has a color bit depth rating of 22, while the Nikon has 23.5. Because this is a logarithmic scale, the Nikon can distinguish 3 times as many colors, and thus gets a weighted 22.5 points, compared to the Canon’s 7.5 points -- a significant difference that could tip Snapsort’s overall recommendation between the two models. Being able to discern color is definitely an important aspect of camera quality, but beyond a certain degree of color accuracy, it seems like additional accuracy has diminishing returns for many users. Depending on which source you find, the human eye can see somewhere between 7 and 17 million colors, and many screens and printers don't exceed 16 million colors.
Snapsort could differences in weigh color bit depth above 22 bits less heavily than than those below that level of depth. One way of doing this would be to assign ranges of bit depth that all receive a certain score, though this would sacrifice some precision.

Improvements to look and feel

Overall, Snapsort has a very sensible information architecture and pleasant user experience. I love how the key areas where cameras’ scores differ is highlighted on the score explanation page. I love how easy it is to go from a list of cameras in a category to an overview of a specific model, and from there to comparison against another model.

Still, for me, the comparison feature is the meat and potatoes of Snapsort.com, and I’d love to be able to have camera comparison text fields on the top of every page of the site, so I don’t have to click “Compare”.

Better yet, I’d love it if there were a browser plugin I could install that would detect camera models in the page content I'm viewing (on a list of approved websites), and offer me a direct link to a comparison of the cameras being discussed. That way, I could go straight from a forum discussing two cameras and see the Snapsort comparison.

One last request

My final feature request for Snapsort is that they expand into other areas beyond photography. If Snapsort could help be decide what to eat for lunch, I’d save hours of excruciating scrutiny a week. A burger has so much protein, but also so much cholestoral. Should I really get a salad (probably not)?

Either way, I look forward to taking a picture of my lunch with whatever DSLr I end up buying, and I’ll be able to rest assured that I’ve made an educated, thoroughly reviewed purchase.

Disclosure: I wrote this blog post out of pure love for Snapsort.com, and also to participate in their community feedback contest, where I stand to win some awesome prizes.