|
The Truth about Digital Image Quality... There are 3 primary things that affect IQ: 1) the lens, 2) the sensor, & 3) the image processing. Well, there is a forth: The nut
behind the lens, but I'm restricting this article to hardware and software.
[begin aside]Virtually all digital cameras can correctly be termed dSLR types, in the sense that they have one lens that both allows scenes to be visualized prior to capture and to be imaged on a sensor in order to actually take a photograph. The only ones that are not dSLR types are the very small, cheap pocket types that do not have an LCD on the back and the viewfinder is a separate optical system (with a small window on the front) and the few retro-rangefinder models. The reason this semantic issue is germane to IQ, is that the configuration variables of a camera, ie, fixed or removable lens, mirror or no mirror, pentaprism or pentamirror, optical or electronic viewfinder have (mostly) no impact on IQ. So, don't fall for this myth! These different configurations DO affect the way you, the photographer use the camera. NOTE: One of the above configuration options does have a measurable effect on IQ: A fixed-lens camera CAN have a better lens...but manufacturers rarely take advantage of this possibility.[end aside] Prologue… Before I start discussing The 3 Things that affect IQ, let me comment on mirrors. Some people equate a camera with a moving mirror as a "reflex" camera. But this is not completely accurate. There have been Twin Lens Reflex (TLR) cameras, as one example. A TLR camera doesn't move the mirror. There have also been a few SLRs with a fixed (pellicle) mirror. But what I want to discuss is SLR cameras w/o ANY mirror. The term "reflex" means "the reflection or image of an object as exhibited by a mirror or the like". In other words, anything that functions LIKE a mirror can be described as "reflexive". Modern fully electronic cameras that have an electronic viewfinder or LCD that displays the image seen through the "single lens" of a camera, work LIKE the original SLR concept. They project the image electronically, instead of opto-mechanically. The advantages are many: smaller, quieter, less vibration, more reliable, faster, less expensive, etc. There are also BIG advantages in the lens designs, which, in the absence of a mirror, can sit very close to the sensor (more on the impact of that later). When SLR cameras were invented in the 1930's, a mirror was the only option! But in 2007, there are better ways! There are cameras that "bridged" the gap between modern fixed-lens cameras and legacy dSLR designs. For example, Olympus has produced a few; the E-330, the E-410, E-510, and the E-3 (to mention just a few). And Sony did the R1. The Olympus designs still have a mirror and porro-mirror, but also have modes that lock the mirror up and "reflect" an electronic image to the LCD on the back. They have a conventional optical viewfinder. In contrast, the Sony offering is a fully electronic SLR (or eSLR). It has no mirror or penta-prism, penta-mirror, or porro-mirror. The Olympus designs employ a 4/3" (18mm x 13.5mm) sensor, where as the Sony design uses a slightly larger APS-C (21.5mm x 14.4mm) sensor. Panasonic has a 4/3" design, the Lumix DMC-L1 and Leica has the almost identical Digilux 3. Recently, every manufacturer has revised their venerable dSLR camera designs; they now have live preview in a similar way to the Olympus models. I think these cameras signal a shift to a different paradigm. Some people call these designs, "EVIL", which stands for "Electronic Viewfinder…Interchangeable Lens". But none of them are true "EVIL" cameras. The Sony R1 doesn't have a removable lens and the Olympus, Canon, Panasonic, and Leica cameras don't have electronic viewfinders. But they are very similar and seem to signal that a true "EVIL" camera is pending, down the road. In fact, the first "EVIL" camera is now fact. The Panasonic G1 is that camera. It has been wildly praised. Now if someone could just devise a better name than "EVIL"! While most dSLR cameras have removable lenses, there is nothing that says an SLR design must also have removable lenses. There have been a few mirror-SLR cameras that had fixed lenses. I make this point to forestall the idea that a "fixed- lens, single lens reflex" might be an oxymoron, but it's not! All this dSLR discussion leads up to a major point about lenses. In a camera system that employs a mirror to reflect the image to a viewfinder, the lens must be located well in front of the sensor. 
The above diagram shows clearly why the lens can't get close to the sensor. It has to be spaced forward to allow room for the mirror, room for the mirror to rotate out of the way, and room for the shutter. This distance is problematic for the lens designer. Given a free hand, the optical engineer would locate the lens much closer to the sensor than a mirror-SLR design permits. Even with mirror-type dSLR cameras, the engineers do everything to minimize that distance. Canon calls their special lenses "EF-S", where the "S" stands for "short-back-focus". It's most important on wide angle lenses, but other lenses also benefit. "The proximity of the rear element to the image sensor greatly enhances the possibilities for wide angle and very wide angle lenses, enabling them to be made smaller, lighter (containing less glass), faster (larger aperture), and less expensive." http://www.answers.com/topic/short-back-focus By eliminating the mirror, Sony opened up possibilities for Carl Zeiss engineers to design a very short-back-focus lens for the electronic-SLR R1 camera. Some cynical observers didn't understand the financial advantages of this design and claimed that Sony was losing money on every R1 sold, because the lens was so excellent. It was equivalent to (and they were comparing it to) conventional long-back-focus dSLR lenses costing several thousand dollars! 
The Lens… The lens is the most important of The 3 Things that affect IQ. It affects resolution, as much as the number of megapixels the sensor is subdivided into. But it also has other important contributions, such as contrast. Lenses also can have bad contributions, such as chromatic aberration, geometric distortion, curvature of field, and vignetting. A good lens minimizes these. Interestingly, having a very short-back-focus distance helps lens designers cope with these issues…economically. All this leads up to an interesting and somewhat startling conclusion: Digital cameras with no mirror can more easily control lens problems and deliver high IQ! This flies in the face of common knowledge: Everyone knows that to have a really good camera, it has to have a mirror…a mirror that moves…and produces a loud "whap-whap" sound…and shakes the camera. Extremely large zoom ranges are not good for IQ. This is especially true for zoom lenses mounted on cameras with large sensors and a mirror. Lately, there has been a rash of 18mm to 200+mm zooms designed for APS-C sized dSLR cameras. These are very popular, but do NOT have good IQ. For smaller sensor cameras, it's easier to get a big-zoom range and not sacrifice as much IQ. Part of this is because a small sensor doesn't have as much IQ to begin with; a small sensor just can't capture as much light. But, it's also easier/cheaper to build a big-zoom lens if it can be mounted close to a little sensor. Extremely "fast" lenses don't normally have high IQ. But often the "softness" of a f/1.4 50mm prime lens (for example) is a better tradeoff than the noise inherent in an ISO 1600 image. And in many cameras, that ISO 1600 is coupled with heavy NR, so the fast lens often produces images with more resolution! While depth-of-field (DOF) is not strictly an IQ issue, it's a variable…and one that experienced photographers care about. Having big DOF can render all of a scene in focus. Having small DOF can allow the photographer to selectively choose which part of a scene is in focus, making the rest of the picture blurry. Both these can be good or bad. It's hard to say which is better! An often heard comment is that, "…dSLRs give the photographer more control of the DOF." This is not strictly true, as DOF is, to a great extent dependent on the size of the sensor. I'm unaware of any camera, dSLR or otherwise, that has a variable size sensor! The correct statement about dSLRs is that they (having big sensors), allow the photographer to easily place background elements out-of-focus (OOF). In contrast, small sensor cameras tend to have everything in focus. This is especially valuable when taking macro pictures, as most people would prefer to have all of the subject in focus. Another point related to DOF is that contrast AF systems for small sensor cameras are typically slow, but if everything is in focus anyway, because of big DOF, what's the need for fast AF? [that's a joke, BTW]. The recent Panasonic G1 however, shows that it IS possible to make a contrast AF system fast. All these years and now we find out that ALL the camera manufacturers were intentionally making the cheap cameras slow… If you have OOF elements, then you may be interested in the subtle quality of these OOF parts. The term "bokeh" is a Japanese word that is destined to be a part of photography lore for years. It is used to describe the "feeling" you get when you experience blurry pictures. If you "feel" good, the camera that produced the image is said to have "good bokeh". I suspect it's tough to know how to design a lens with "good bokeh" as it's controlled not by engineering equations, but rather paranormal flows of energy. However, it may have something to do with the shape of the iris and/or the number of non-spherical elements used in the lens? Or not… Some people feel good all the time and for them the whole world has "good bokeh". The Sensor… There are 3 major types of sensors: CCD, CMOS, and MOS…and many variations on each of these major schemes. While advertisements may tout one of these sensor technologies over the others, they rarely have much to do with the performance of the sensor. The little details are more important…and ads never tell you about the little details! Another way to classify sensors is Bayer and non-Bayer. 99.9% of sensors are Bayer sensors. An article on Dr. Bayer's patent: http://en.wikipedia.org/wiki/Bayer_filter The sole non-Bayer sensor at the moment is the Foveon X3 sensor. It is quite expensive and in some ways better. It has large photosites that are stacked vertically. It has no color filters, but instead uses the Silicon to "filter" the photons. High energy (short wavelength) photons are captured by the top sensor. Low energy (long wavelength) photons are captured by the bottom sensor. An article on the Foveon X3 sensor: http://en.wikipedia.org/wiki/Foveon_X3_sensor One of the most important sensor-related factors is noise. All sensors produce some noise…it's inescapable. There are MANY causes of this noise, but to simplify the subject, noise is minimized by making the sensor large and keeping it cool. The temperature of a sensor is mostly ignored, but it can be quite important. There are a few generally true statements about temperature: o CMOS sensors operate cooler than the other types. o Sensors that are operating all the time, heat up. o Sensors that clock data at high rates get warmer. At this juncture, nobody (other than astronomers) mount thermoelectric coolers on sensors to keep them cool, but this might be a decent strategy for small sensor cameras in the future. There are several issues associated with using TEDs: they require LOTS of power and if the sensor is not sealed in a dry atmosphere, ice will form on the face (not conducive to good IQ)! A good discussion of sensor noise is at: http://www.luminous-landscape.com/essays/sensor-design.shtml For many reasons, a simplified rule is, "Buy the biggest sensor you can afford and one with the fewest photosites you can put up with." This is tough to do, because there is a marketing-driven megapixel race. We are bombarded with messages that run counter to this rule. Marketeers simply don't know how to explain complex technology to their average customers and revert to over-simplifications. Note that they don't say "10 megapixels is better than 8 megapixels", they just put "10 megapixels" in big, bold, bright letters on the box and their average customers jump to that conclusion (like "Why would a manufacturer advertise megapixels if it wasn't important?"). Unless you are and want to remain an average customer, don't fall for this! More megapixels can be good, but ONLY when the sensor has BIG photosites! Putting more pixels on a smaller sensor is a recipe for very bad IQ. Notice that marketeers don't mention how small the sensor is on the outside of the box in big letters! Many don't even tell you the sensor size in small letters. [begin aside] Small sensors are described as something like 1/2.7". This is called a complex fraction. Marketeers like it because as the sensor described by that complex fraction gets smaller (that's bad), the denominator gets bigger (that looks good to the average customer, who flunked"new" math in 3rd grade). Instead of using a complex fraction, they could express it as a common fraction. For example a 1/1.5" sensor can also be described as a 2/3" sensor. However, a 1/2.7" sensor would be described as a 10/27" sensor. Ugg! Another approach is to simply do the division. Divide 1 by 2.7 to get 0.370". If that number looks too small, they could express it as a metric value: 9.41mm. However, the plot is even more obscure than using a complex fraction to express sensor size. The truth is that a 1/2.7" sensor is MUCH smaller than 0.370" would imply! The original television sensor was a vacuum tube (remember them?). It was called a Vidicon (among other, earlier names) and was a round glass tube with a photosensitive end. The photosensitive area was a rectangle coated with a salt that released electrons when hit by photons. The rectangle was, of course, smaller than the round tube. Vidicons were described by the outside diameter of the tube, not the inside diameter or the dimensions of the photosensitive rectangle. The exact relationship of the tube diameter to the height and width of the rectangle is variable. Can you BELIEVE that digital camera manufacturers adopted the arcane Vidicon description for modern semiconductor sensor sizes? I can't! They obviously want the customer to be confused. So how big IS a 1/2.7" CCD sensor? Well, when you go looking for that information it's expressed in metric terms…something like 5.37mm x 4.04mm. The diagonal, from Pythagoras is, 6.71mm or 0.265". So why exaggerate the size by implying that it's 0.370" in diameter? Why not call it a 6.7mm sensor? After all, we got used to 8mm movie film…[end aside] The Image Processing… There are several kinds of processing inside digital cameras that can affect IQ: noise reduction, data compression, chromatic aberration removal, and automatic white balance setting. As the sensors have gotten smaller, NR has been employed in stronger doses to keep noise under control. And as megapixels have increased, image file sizes have risen, which has made compression more important. There are two benefits to compression: 1) it does the obvious of allowing more pictures to be stored in memory devices, and 2) do it faster. Writing less data to memory can help to keep the time between pictures down. With big zoom ratio lenses, CA becomes more of an issue. Some camera manufacturers remove these colored edges with internal processing, as it's predictable and easy to do. The processors used in digital cameras don't have a lot of power (ie, speed), because 1) fast processors are expensive and 2) more speed means more electrical power is required and thus less battery life. There ARE good NR algorithms that can BOTH reduce noise and retain detail, but these are very slow, even with really fast processors and lots of memory. Therefore, digital cameras do a quick and dirty NR. This can and does smear details in parts of the picture. Some cameras do a selective NR, by only applying heavy NR to areas determined to be less affected by smearing, such as regions of sky and shadows. There seems to be a point where sensor size, megapixels, and NR combine to really ruin picture IQ. At the moment it looks like when the physical sensor size falls to 1/2.5" (or below) and the logical size rises to 8 megapixels (or above), manufacturers can't find a way to unobtrusively apply NR 100% of the time, given the limited processing resources available in a battery-powered camera. I suspect manufacturers will get better at these algorithms and processing power w/ low energy usage will get better with time, but not much. Adding chromatic aberration correction to the processor load further slows things down. Given a specification from marketing, some designers choose to spend more on lenses than on processors. Their thinking seems to be that if a better lens can minimize CA, then it's better to go that way; going the other way, requires a more expensive processor and a bigger battery. However, when zoom ratios get to 10:1 and beyond, it becomes quite difficult (thus expensive) to eliminate CA. Some cameras use sensors with a microlens sheet above the photosites. This can also apparently cause color fringing. Purple is the most common color seen, but if you look closely, the opposite side of the object with a purple fringe (PF) often has a green fringe. While the cause may be different, the cure is the same: CA removal. [begin aside] Given an image taken with an inexpensive camera, how can you determine if, how strong, and what type NR/compression/CA reduction has been used? It's tough, even for experts. With professional cameras, it's EASY! They all have a RAW format, so you just take two pictures…one RAW and one JPEG…then compare these two pictures. The difference is the errors introduced. Fine or Super Fine JPEG usually is almost perfect, thus the artifacts attributable to other sources than the JPEG algorithm can be easily exposed and understood. Lately, manufacturers want to keep details of their in-camera processing secret, as it's a competitive advantage to have more advanced algorithms. Thus some have removed the RAW file format. Now it's tough, even for the experts to analyze what is going on inside the latest crop of cameras. Previously, if a camera over-processed pictures, the photographer could, for important shots, resort to RAW and get high IQ…but not anymore. [end aside] While it's not strictly "image processing", but more accurately called "camera automation", white balance can affect IQ too. Most people use Auto WB except when the conditions are unusual and might tend to confuse the WB algorithm. Experience with a specific camera will teach you when to trust automatic WB mode and when not to. Often, when mixed lighting is present (daylight + artificial light or a mixture of several types of artificial lighting), cameras have difficulty. In these instances, it's best to use manual WB (if your camera has that feature). To do this, you present a white or neutral gray target to the camera and it adjusts the RGB amplifier gains to render it a neutral color. One of the biggest differences in cameras seems to be the accuracy of Auto WB. Some cameras are quite good; others less so. Given a group of digital cameras, the most obvious difference in the pictures they take is the color accuracy. Observers can see the color differences from a distance, whereas all the other issues (focus, resolution, noise, chromatic aberration, barrel/pin-cushion distortion, blurry NR) require the observer to get up close and/or look closely. Luckily, it's usually easy to fix WB errors in post-processing, especially if a white surface is located in the picture. Some photographers (especially ones with cameras that have unreliable Auto WB) always locate a small white object in every important picture. They then use that white object as a reference and correct the color balance with Photoshop, or other photo editor. Then they "clone" out the small white object. An alternate way is to take the first picture of a series with a white object. This wastes a picture, but saves time later, as cloning it out is not necessary. What Type Camera? A very common question is, "Should I buy a dSLR or a P&S?" You now know that ALL cameras are dSLRs and "P&S" is a lifestyle choice. This goes a long way to explaining why that question never seems to get answered satisfactorily! [begin aside] Not only is "P&S" insulting, it also fails to distinguish anything. Concentrate now…"Point & Shoot" is a lifestyle, not a camera description! Almost ALL cameras have the ability to take pictures fully automatically, if the photographer chooses that style of operation. [end aside] OK, so if "P&S" and "dSLR" are not comprehensive classifications, what are? The experts with photography websites don't agree, but I'm pleased to see that most have totally dropped the "P&S" designation. Unfortunately, most people still use "P&S". If we combine all these expert opinions and their take on camera classifications, we are left with two obvious questions. Are there only 7 types of cameras (Ultra-compact, Compact, Super-zoom, SLR-like, Entry-level SLR, SLR, & SLR/Pro)? And what are the distinguishing characteristics of these? Hmmm… Going back to the top of this section, the "P&S vs dSLR" question is wrong because it over-simplifies the subject! It implies that there are only 2 types of cameras. Knowledgeable people, capable of answering the broader question of "What type camera (should I get)?", are dumbfounded that this question is asked repetitively. I, like many people, have worked hours to try to devise a simple, yet comprehensive scheme for classifying cameras. This scheme has to ignore the obvious (like size or price) and the trivial (like color)…it has to include the important differences. I have decided (for what that's worth) that there are 4 parameters worth of inclusion in a digital camera classification scheme: 1) How the lens is attached, 2)What type of viewfinder is used, 3) Where the shutter is located, and 4) How focus is done. I eliminated whether it uses a mirror, because the viewfinder type includes that detail. I further selected words that were mnemonics for the different choices and which could be abbreviated with a single letter. This allows a 4-letter abbreviation to be used. Don't worry what the letters mean...this is just a way to be sure I don't forget anything. Unfortunately, none of these make decent acronyms… FOSC...Typical "P&S" cameras FESC...Bridges w/ eVFs (eg, R1) RTBP...Removable-lens dSLR cameras RTSP...Hybrid-shutter dSLRs (eg, D50-D70) REBC..."EVIL" cameras (eg, G1) FTBP...Fixed-lens dSLRs (there are none of these that I know of) FNSN...Cell phone cameras ROLH, ROBH, FOLH, FOBH...Rangefinder cameras The top list shows that there are 6 basic types of digital cameras, per my scheme. The bottom list are things that may exist but are not really interesting to most of us that want to take a good digital picture. I also ignored details like how many pixels, what type of flash media, what type of batteries, how the optical viewfinder works, RAW capability, how the camera might be used, where it is kept when not in use, and lots of other things that some people think are important. [begin aside] Of all the flashy things on the outside of camera boxes, "Megapixels" is the most worthless. Most online review sites key off this stupid advertising and invariably list "Megapixels" first when announcing a new camera model. So, why isn't it significant? Because it tells us NOTHING about the image quality! There are 4 MP cameras with stunning IQ and 12 MP cameras with terrible IQ… not even good enough for a decent snapshot. The thing that matters is the size of the photosites. Rather than use a number measured in microns-squared, I thought that a number that was based on the total photoactive area might be a good compromise. My belief is that given a sensor size, the IQ is roughly the same, regardless of the number of megapixels it's divided into. Occasionally, manufacturers fail to deliver on this. Specifically, as megapixels increase, they have to develop processing software to address and remove noise and still retain sharpness. When they do this properly, the IQ is comparable to the same size sensor, with fewer megapixels. However, when they fail to develop appropriate internal image processing, the IQ suffers…it either has too much noise or too little detail. It's interesting that I have never seen an instance when a sensor was subdivided into more pixels (without increasing the total size) and actually IMPROVED the IQ! I conclude that the only advantages of more megapixels is advertising and bragging rights. [end aside] These characteristics are not fixed; they change as the products change. But I think the above 11 groupings are OK in 2007. Note that, IMO, there are few Prosumer and Bridge cameras being sold in 2007. In the recent past, there were…and at least on camera forums, they are talked about A LOT! At the moment, there are very few products between a Canon G7 and an Olympus E-410. That's quite a void! The relevance of these 11 camera types to IQ is simple. The cameras at the top of the list have less IQ than the cameras at the bottom. Sure…there are variations in IQ within each type and there is also some overlap between adjacent groups, but don't expect a "consumer" class camera to have better IQ than a "small SLR" class camera. If you think I'm wrong, you are un-talented and non-discerning…and probably worship your small camera. [begin aside] In the past, there was an apparently BIG market for the Prosumer class of digital camera. Many of the big players sold cameras of this type. Most had 2/3" sensors and 8X zoom lenses. They had innovative ergonomics and things like laser hologram AF assist. Examples are the Nikon 5700/8700/8800, the Sony 707/717/828, the Olympus 7070/8080, The Konica-Minolta A1/A2, the Canon Pro1, the Samsung Pro 815, etc. Today, only the Pro 815 is sold new and it's a 2-year-old product. There are some OK Prosumers being sold by Fuji, such as the S9100, but all these offerings have 1/1.7" or smaller sensors. All the manufacturers decided to abandon this market segment and sell equivalently priced, entry-level dSLRs. The Bridge segment is also empty. There was only one Bridge camera IMO; the Sony R1. It went away when Sony bought Konica-Minolta, discarded everything but their dSLR and rebranded it as the Sony A100. Although facts are unavailable, it appears that the R1 was quite popular and sold well (even though in the US it was not advertised). Many of these Prosumer and Bridge products are already "cult" cameras. They sell in good condition for as much, and sometimes more than, their lowest 'new' price.[end aside] As everyone knows, there are many different sizes of sensors. But not many know how many sizes there are or how they compare. I've compiled a list of sensor sizes, but this list is not comprehensive. There are countless variations! Even seemingly identical sensors may actually have slightly different sizes. For example, all 1/1.7" sensors don't have the same size…when you get to the manufacturers data sheet, they often have slightly different dimensions. A better example is that all APS-C sensors are different. And they are vastly different, yet people insist on using "APS-C size sensor" as if that means something. I count at least 6 different APS-C sizes.
To help understand the above table, this diagram may help: 
The Vidicon is a storage-type vacuum tube in which a charge-density pattern is formed by the imaged scene on a photoconductive plate (coated with antimony trisulfide) which is then scanned by a beam of electrons. The scanning electron beam dissipates the charges on the target and generates small voltages which are amplified to capture the image. The target is quite a bit smaller than the end of the Vidicon tube. However, it was common to describe Vidicon sizes by the outer diameter of the vacuum tube, with no reference to the dimensions of the target. When the camcorder was first designed, it had a Vidicon imager. But it was not very popular (large, heavy, short battery life, etc). As technology improved, the Vidicon was quickly replaced by CCD sensors. However, the camcorder manufacturers continued to refer to the sensor sizes as though they were still Vidicon's. This made some sense, since the customer was accustomed to that nomenclature. However, the still camera market didn't know what a Vidicon was or understand why it was called a ½" size, what that ½" referred to, or the relationship between the tube diameter and the target dimensions. The still camera customer knew what film sizes meant: 8mm, 9.5mm, 16mm, 35mm, etc. When the first small digital cameras appeared and the sensor was called something like ½", the customer didn't have a clue what it meant. Well, actually, the customer probably thought it meant the sensor was ½" in size. [begin aside] Proponents of grand conspiracy theories contend that the confusion was intentional. As an example, they say that calling a sensor a 1/3" size obscures the fact that the sensor is REALLY only 0.141" x 0.189"! That's smaller than the nail on your smallest finger. Also, they say, as the sensor gets smaller (bad) the complex fraction appears to get bigger (ie, it's not obvious to people that 1/1.5" is better than 1/1.7"). They say the use of this obscure 20th century vacuum tube terminology is intended to confuse the customer, allowing the manufacturers to sell cheap products for high prices. I'm not sure whether this conspiracy theory has merit, but I think it's interesting that almost everything else related to cameras is measured in metric values! The only exception I could think of is the ¼-20 female thread on the bottom used to attach the camera to a tripod. At least the larger, more professional cameras have the sensor height and width typically given in millimeters. [end aside] So, what do you do with all this information? How does it help you choose a camera? Just as with the list of camera types, the sensors at the top of the list have less IQ than the sensors at the bottom of the list…generally. But there is another important question yet to be pose. Going back to our original premise, that there are 3 Things that affect IQ, we haven't asked or answered the most important question: How important is IQ to YOU? If you are typical, the answer to that question is, "Not very!" But if you have read all the way down to here, you are not a typical camera customer. Let's assume then that you want more than a few snapshots to send to relatives and friends as e-mail attachments. Let's further assume you can describe the types of pictures you are likely to take. OK, then, look at the following table and select the scenes that apply to you:
This is not ALL the scenes and the suggestions are at best generalizations. The message is that different scenes need different cameras, different lenses, and different settings. Some type scenes benefit greatly from adding flash (or even increasing the ambient level with photofloods). Plus, there are many photographic styles. Some people like blurry backgrounds; some don't. Also, add the variable IQ expectations of photographers and you have a WIDE gamut of photographic equipment. It's hard to find the perfect camera; some contend it doesn't exist. A camera is a compromise to some extent. Remember the old adage? The customer says he wants a car that is 1) fast, 2) cheap, and 3) reliable. The car expert replies that he can only get 2 of these. The same is true for cameras. Some common desires are mutually exclusive. As examples, a fast lens is not light in weight; a high IQ camera is not small; a big sensor is not cheap. A good way to be permanently unhappy is to have impossible expectations. Be happy! | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||