Confusion over digital camera megapixels

This same problem exists in sound. Sound systems used to be rated in Watts (RMS), and average home stereo being 2 * 15W channels, and if you check the consumption, you'll see that the unit draws approximately 35-40W. Then some clown introduced Watts (PMPO) and I have seen USB speakers claiming 50 Watts (PMPO) output. An USB socket can only supply 2.5 Watts of Power, so PMPO figures are just bigger numbers that are essentially meaningless but attract the consumer away from another unit that may actually be better.

Hard drives are another example 2MB of hard drive does NOT = 2MB of RAM. Hard drives calculate in decimal to give 2,000,000 BYTES, where as RAM is calculated in HEXIDECIMAL to give 2,097,152 BYTES, hence you will now see CLARIFICATION of this issue as Drives will be quoted as MB but RAM will now be quoted as MiB because as we now move into TeraByte sized drives vs a TeraByte of memory the numbers get further and further apart. Under Windows, a 1TB hard drive will usually be reported as 1.8 TiB.

The same is true of Net connections being quoted in MegaBITS per second because it gives a higher number. To get MegaBYTES per second don't divide by eight, divide by TEN.

The same problem exists with LCD Monitors--21" Wow! But what's the resolution? Oh, it's only 1024 * 768. I have smaller monitors with much higher resolution.

So it is with digital still and video cameras. Instead of quoting the exact resolution e.g., 1024 * 768, they multiply them to give the megapixels i.e., 786432 which is 0.7 megapixels. To make this number look even better, some manufactures will also multiply it by 3 (each color element) giving 2359296 or 2.3 Mega-alleged-pixels.

I was never so worried about still camera resolution as I had a wonderful old HEAVY (stable) 35mm SLR with a fantastic zoom lens that could go from about 1m to 400m with excellent results. As it was stolen, I've had to go into details regarding digital cameras. You have to do your research (usually via the net) to find out the real details of resolution.

If you can't do all the calculations in the shop, get the salesperson to take a MAXIMUM resolution photo on each of the cameras that interest you, get them copied to a USB stick and compare them on your PC at home which will reveal the true resolution allowing you to decide which camera you want based on the results that you get on screen.

I'm a professional photographer who started with film and then moved to digital.

The difference between a pro camera sensor and consumer camera sensor isn't megapixels - it's noise and exposure range. Small sensors like in compact cameras are extremely noisy and cannot shoot well in low light - a double whammy.

When the 3MP Canon D30 came out, lots of pros started using it for things like weddings. It offered more than enough resolution for normal images of one or two people (but not really enough for group shots or landscapes).

6MP provides the same effective resolution as 35mm pro color-negative film.

12MP is equal to or better than the highest resolution 35mm film made and equal to Medium Format color negative film.

If you need good quality images buy a camera with a bigger sensor - in order of size (not counting point-and-shoots), those would be Micro 4/3, 4/3, APS-C, and Full-Frame.

As a former pro photog who transitioned to digital happily: Unless you need lots of technical detail, this is the best general guide I've seen (and used successfully for both pro and home use):
1) The sensor size IS more important, but for home use, don't worry about it.
2) Anything over 4 megapixels can take a good picture. 7 or more gets you into good 8x10 print territory, otherwise best to stick with 4x6 or 5x7.
3) Speed of autofocus will help you more than just megapixels in capturing pix, especially moving subjects. Most point-and-shoots over $200 are similar in focus time, and a faster flash card (Class 4 or 6 such as a Sandisk Ultra II, e.g.) will do more to speed up picture capture, write to the card, and recovery for the next picture.
4) Learn to use the camera's ISO setting. I took Carlbad Cavern pix without flash (needed brightening with Photoshop Elements 7.0) using ISO 1600, and outdoor sun pix using ISO 200 or (more flexible in sun and shadow) 400. I leave my TZ5 "home" camera on ISO 400 about 95% of the time.
5) Autofocus slows down (and may even "wander") at magnifications over 5X, so while I still often use my 10X zoom carefully, 5X is enough for most home users. 3X (to me) is pitiful.
6) Most consumer point-and-shoots have anemic flashes, often where your finger has to hold the camera (blocking the flash). Because the flash sensor reads the nearest reflector of light and is only good up to about 15 feet (20 if you're lucky), prefer stronger flash. You'll appreciate it.
7) While there is supposed to be a standard for measuring megapixels, you'll likely never be able to tell the difference. Generally, more Mp is better, but puts larger files (may take longer to transfer and take up more hard drive space) on your computer. Can also take longer for your printer to print (direct or via computer). Look for the "Effective megapixel" specification for your camera, and that will be accurate enough for non-pro use. More Mp takes longer to write to your card, to transfer, and to print, but you can make larger prints and more detail.
I hope that gives you enough to go on. Nikon, Cannon, Fuji, others, all make excellent cameras. Anything under $200 is just a complex toy (my opinion), but over $400 you start to get into prosumer (advanced). My Panasonic FZ50 (10.1 Mp) does great for my occasional pro work, and the Panasonic TZ5 (7.1 Mp, soon to be replaced by newer ZS3 with 10 Mp and a 16 Mb Ultra II card) are highly rated for prosumer (my home use, vacations, etc.)and TCO (total cost of ownership), and like all cameras, usually do excellent, but occasionally blow a shot (really me or circumstances, but I'll blame the camera's limitations). Good luck.

You can have loads of megapizels but if your resolution isn't at the maximum available for your camera, you'll still get pixelation on your enlarged shots. Unless you have won the lottery, few can afford a digital camera that will take as good a picture as an average film SLR. Using film as the format and a fast ASA800 speed film with a 1/1000 second exposure, there's not a shot you'll miss. The manufacturer that makes a digital element that fits in a film camera will make billions because you'd have the best of both worlds, digital element with optical superiority.

While the respondents have so far provided a decent overview of what the pixel is, and does, there are additional factors to be taken into account.

First of all, if the camera outputs in a compressed format, like .jpg, or .jpe, you usually have an interpolated image - a 5 megapixel CCD delivers a 10 megapixel image. That's not unlike the difference between a 1080i and a 1080p HDTV. Only when the manufacturer specifically mentions this, can you expect the real number. The Nikon D90, for instance, delivers 12.3 megapixels, and you can verify this because you can have the camera output images in raw format, Nikon's .NEF. Those cameras can output to compressed (JPEG) formats as well, but you should remember that any compressed image format is "lossy", and the advertised resolution is made up using software, smoke and mirrors.

Then there is the memory issue. Solid state memory is inherently slower, the larger the memory element is, and so a camera with a lot of real pixels will be relatively slow getting ready for its first shot, you can't turn it on and immediately push the button. Cheaper cameras use cheaper memory, of course, and that is always slower.

In that same vein, the CCD needs to "unload" the image into storage memory, after the picture has been taken, and you therefore cannot take the next picture immediately after taking one. The more expensive SLRs use fast buffer memory to achieve this relatively quickly, and this will allow you to take 10 or 20 or 30 "rapid fire" shots, until the buffer is full, when it needs to write its contents to slower card memory, which can take a while, and you wait.

Similarly, the more pixels, the longer it takes to load a picture into an application on your PC, and the longer it takes to process a picture, make changes to it, adjust the hue, crop it a litttle, and write a processed copy away to disk. If you're an avid photographer, looking through the 300 12.3 megapixel images you took on your last trip can be a time consuming exercise, and that's just opening them up, one by one, I am not even talking about processing here. And those 300 wonderful raw format pictures, you're a perfectionist and want to have your work in as much detail as possible, your 300 holiday snaps will take up 3.7 gigabytes of disk space. Better go out and get a terabyte RAID device, I have two sitting on a network server, because one little disk crash and years of work goes "poof". All I am saying is that the choice of resolution has consequences for the type and speed of PC, your electronic darkroom, you need to use, and the type and size of storage device you use as your electronic shoebox.

As a frame of reference - a decent slide scanner produces 64 megapixels out of a 35mm (24x36mm) old style slide. You're best off having two devices - I have a tiny 5 megapixel Wal-Mart cheapie on my belt, for quick snaps, and use a mix of Nikon SLRs (including a 35mm body, still, if I need truly high resolution!!) for the serious stuff.

Thanks to the forum members for the wealth of information. Allow me to summarise what I have learnt. Correct me if my understanding is incorrect.

1. Megapixel numbers mentioned in relation to digital cameras are not (no of pixels per unit area of the sensor) but simply the number of actual pixels on the sensor of the camera.

2. The greater the number of pixels, generally, greater the resolution.

3. The greater the number of pixels per square unit of the sensor, for a given pixel quality, lower the quality of the picture.

4. For a given pixel quality, there will be an optimum number of pixels which can be ideally placed in each square unit of the sensor.

5. I think the physics of light operate whether the camera is a film camera or a digital camera. I, therefore, tend to agree with the view that depth of field is a function of aperture rather than anything else for a given lens-size/quality.

6. The greater the quantity of light collected the better the picture quality.

7. The quality of the printed picture is governed by different parameters and different mega-pixels/sensor sizes can combine to give the same/similar printed picture quality given the same printing system. Different printing systems, manipulation of the picture with software etc are not reckoned for this.

8. In film cameras one used to have exposure level numbers which would give the same quantity of light for a given fstop/shutter speed combination. The Depth of field (DOF) would vary in such cases. However, in digital cameras the quantity of light measured in this way would not only affect the DOF but will also be affected by the sensor (mega-pixels/pixel-quality/pixels per square unit etc)

9. There is, unfortunately no measure or technical specification which gives these details while advertising the camera or no such metric is available even in the manuals or techinical specifications on the we-sites of manufacturers.

10. Thank you once again for the wealth of information.

V.K.Subramanian

Hi,
1 Mega-pixel means there is one million pixels in your picture. Usually, a 2+Mp Camera is good for personal use. There are companies that advertise of 12 mp cameras. recently Sony launched a 24 mp camera. But that's for what you see on road. The Biggest hoardings in the city are shot with no more than 8mp. Unless you are going to blow up your image to such big sizes, you don't need much of them. What you really need to look for is the OPTICAL zoom. That's the real power of the camera lens. I would recommend a camera no more with 3-5 mp and 10x Optical zoom.
Finally, if you are an armature, what i said above is for you.
If you are a pro, you wouldn't be here.

The definitions of MegaPixels I have seen are generally correct! More MP means more picture data, but there has been some confusion over the effect of sensor size. The "Standard Lens" on a camera was usually defined as equal to "the length of the diagonal of the film frame". A 35mm film frame, at 24 X 36 millimetres, had a diagonal of about 43mm and the standard lens was generally 50mm focal length. A 6 X 6 CM (120) camera had a diagonal of about 85mm, and most standard lenses were 80mm. A 35mm lens on a 35mm camera yielded a slightly wide angle image, and an 85mm lens yielded a short telephoto effect. An APS-C frame on a consumer-model D-SLR has a sensor size of about 17 X 25 mm, with a diagonal of about 30mm. That's where the Crop Factor comes in. It's normal lens would be about 30 - 35mm, and a 50mm lens would be a 1.7X telephoto. Smaller sensors in pocket digital cameras have correspondingly shorter focal length "Standard Lenses" and shorter focal length lenses, with their smaller aperture openings, have greater depth-of-field (DOF). Think Pinhole Camera, with a tiny aperture, and virtually infinite depth-of-field....

Taking photos at the highest resolution supported by your camera allows you to crop the photo (essentially zooming in on a portion of the photo) and retain reasonable resolution.

On many photos cropping improves the results by highlighting the essential parts of the photo.

Many people don't understand the value of closeups, and try to get an entire body in the photo (for example) or capture a big scene. Then when looking at the photo, the most interesting parts may be a fraction of the original photo (like a closeup of someone's face, or a building). This means cropping the photo.

With high pixel density for the original photo, the cropped version may still have enough pixels to print with reasonable quality. A photo with fewer pixels will look pretty bad when cropped due to the lower pixel density.

I take photos with high resolution and use a camera with lots of pixels (Nikon and Panasonic, and many others) and cropping photos works great, and prints look great.

Suggestion: After editing a photo, if you have "edge preserving smooth" option in your editor, try it, it will improve the photo by interpolating color gradation across the photo making it look more natural and more like a photo taken with a film camera.

This article goes to anyone that likes to brag how many mega pixels their cameras have and to those of you that are out there trying to buy the best thing that suits your needs without selling a kidney in the process.

First and the most important thing to do before buy a camera (and everything else) is think whats the purpose of that camera and how much i am going to need/use the camera and if there is someone i know that can go with me, someone that dont get too thrilled with all the stupid facts that a salesman will throw at us just to make a sell.

Back in the late 90?s digital cameras had around 2mp so back then the pixel count did matter a lot because if you take the percentage of difference between a 1.2mp and 2mp camera its a lot rather than a 7.2mp and a 8mp camera. I remember one first digital cameras i used, was a Canon PowerShot A300 with 3.2mp and i took this picture with it, dont know you but that picture looks fantastic to me!, and its quality is far better than the last Casio Exilim i used which had 10mp.

And let?s face it, who doesn?t bring down the quality or size of the picture setting in the camera when we see that we are running out of space in the memory card in the middle of a party? We all have done this, and it?s because is better to have a lower quality picture than no picture. Those of you that just take pictures to upload to your profiles on facebook or myspace would have faster uploads if you do it in a smaller sizes than the original, facebook and the rest of the social networks shrink your pictures anyways! because is a smart way to save space, faster loads and to fit the template of that particular website. So don?t bother uploading all your pictures from the original size because facebook will shrink the biggest width or height to a top 604 pixels. Use software to resize your pictures in advance and you will see how fast you will upload the 60 pictures in that facebook photo album next time you try it.

How often do you print your pictures? and how big? because i have seen wonderful that came out from a 5mp camera printed pretty big like 12

Most manufacturers display the number of megapixels that their cameras have. What exactly is this number? Is it the number of pixels per square unit? If so what is this unit?

The megapixels are the total number of detector elements (pixels) on the sensor. For comparison, HDTV 16 by 9 format has 1960 by 1080 pixel resolution. That is approx 2000 by 1000 (for easier math). Thus multiplying gives approximately 2 megapixels in a full 1080 HDTV display. A typical 12 megapixel camera with a 3 by 2 format will have about 4272 by 2848 pixels.

According to Brad Templeton, the very best 35mm pro film with the very best lens, lighting, focus and tripod stability can capture the equivalent of about 20 megapixels on film. For auto focus, hand held shots with typical consumer film and upper-end consumer SLR camera, practical performance for your old 35 mm SLR was likely more like 2 to 4 megapixels equivalent. For most purposes, a good (not low end) digital camera with about 6 megapixels will produce printed images that are equivalent or better than those from a 35 mm SLR in the past. Whether more pixels than 4 to 6 megapixels has any practical value depends on how you use the camera and how you produce and view the image.

=========================
In film cameras, the size on the film had a bearing on the quality of the final print, particularly enlargements and the larger formats were preferred by professionals who wanted to print large sized pictures. Is there an equivalence in digital cameras and do manufacturers display this? Under what name is this displayed? How much of this is good?

All cameras and photographs involve a compromise in key performance characteristics. As you recall, "faster" films needed less light for a good exposure but had "grain" that would be seen in enlargements. To make a bigger enlargement, you needed either a slower film (that needed more light and/or a slower shutter) or a bigger format (bulkier, heavier) camera and lens. For hand held photos of still or posed scenes that were to be viewed in typical 4 by 6 inch prints, the 35 mm format with ASA 100 film was the choice for most consumers. If action was to be captured, a faster ASA 200 or 400 film might be used with a flash to "stop" the action.

Digital cameras have a similar trade between sensitivity and resolution but since digital sensors are very sensitive (fast) this tends to be a secondary performance issue. How the sensor and lens are "matched" is the more important issue.

Larger format sensors have larger sensor elements than a smaller format sensor with the same number of pixels. This can effect sensitivity of the sensor to light, noise level experienced at low light levels, sensitivity to dust and the match between the size of the blur spot of a lens and the sensor element. A larger format can tolerate a lens with a larger blur spot (lower quality of lens or less precise focus). However, larger format sensors require a larger lens and camera body and thus incur a weight and size penalty for equivalent performance.

Just as a 35 mm SLR was smaller than the 8 by 10 format camera used by Ansel Adams, smaller sensors produce smaller cameras for the consumer but often at the expense of other performance features. Sensitivity, dynamic range, optical viewfinders, interchangeable lenses and other features are often reduced as cameras become smaller.

Some manufactures use sensors that are the same size as the old 35 mm film (full format). This allows the use of lenses that were originally purchased for a film camera with the digital body. However, one can use old lenses with a smaller sensor (the sensor only captures the center part of the image on the focal plane). This effectively magnifies the image produced with a lens of a given focal length. Thus a 50 mm focal length lens used with a half format size sensor captures the equivalent magnification image of a 100 mm lens on a full format sensor or film.

For most consumer digital cameras, sensor size is of low importance and while it can usually be found, it is not highlighted in the camera specifications. The lens on the camera is designed for use with the particular sensor and the consumer does not intend to use other lenses.

However if you have a set of lenses for your old 35 mm SLR that you would like to use with your new digital SLR, sensor size may be a significant issue. Sony cameras will often use old Minolta lenses and old auto focus lenses for Canon or Nikon cameras can still be used on many of their modern digital camera backs.

For most purposes, I think sensitivity (ISO) and size (weight) are the characteristics most correlated with sensor size and highlighted in the specs. Those are the "names" that sensor size effects. What is good depends on how you plan to use the camera but generally any camera that is the size of a 35 mm SLR or smaller with sensitivity of at least ISO 1600 are considered "good" by most.

Enjoy IT ... whatever IT is.

Hope this helps, In digital imaging, a pixel (or picture element) is a single point in a raster image. Pixels are normally arranged in a 2-dimensional grid, and are often represented using dots or squares. Each pixel is a sample of an original image, where more samples typically provide more-accurate representations of the original. The intensity of each pixel is variable; in color systems, each pixel has typically three or four components such as red, green, and blue, or cyan, magenta, yellow, and black.

The word pixel is based on a contraction of pix ("pictures") and e (for "element");

The more pixels used to represent an image, the closer the result can resemble the original. The number of pixels in an image is sometimes called the resolution, though resolution has a more specific definition. Pixel counts can be expressed as a single number, as in a "three-megapixel" digital camera, which has a nominal three million pixels, or as a pair of numbers, as in a "640 by 480 display", which has 640 pixels from side to side and 480 from top to bottom (as in a VGA display), and therefore has a total number of 640

As I see it, the megapixels count is not related to picture quality, but to the possible uses of the picture. For example, if you take a picture at 2MP, you can use it to send by email or as a desktop background, but it's not enough for a regular-sized print.

As far as the hardware is concerned, the quality of a photo depends the sensor, lens, camera firmware and settings and file format.

To give you a better idea of what a megapizel mean to you is to open any jpeg or gif or any picture file you have in your hard drive with any picture viewer you have, whatever. Now, you should know where to find in the menu of your picture viewer app to zoom into your picture, you know what I mean? If you never ever zoom in any picture before, (groan!), ok, I suggest you use your Windows Photo Gallery in Vista or similiar application called Windows Picture and Fax Viewer in XP to open any picture file you have in your computer, because it is easier to find the zoom feature in them. They look like magnifying glass icons at the bottom of the screen. The XP has two magnnifying glass icons with plus and minus symbols while the Vista is more elegant with a scrollbar that will pop up when you click on the magnifying glass icon. Anyway, you click plus magnifying icon to the max and you will start seeing the tiny pixels growing bigger until you can see the squares all over. The square is a pixel. So if you buy a 8 megapixel camera and shoot pics then move the pictures into your computer and zoom them up , you will see a really lot of pixels in the 8 megapixel pictures. If you compare this to one megapixel camera and zoom pictures from it, you will see a lot less pixels. The more pixels your camera can put in a picture , the more you can zoom up without starting seeing the tell tale pixels. So, it would seem logical to you that the more pixels or megapixels in a camera the better the picture is. I would both say yes and no, why?, well you can also look at the multiples of optical zoom your camera can have. My experience tell me that it is most optimal to balance the numbers of megapixels and the optical zoom multiples for the money you can buy. Moreover, the more megapixels the camera has the more likely that there will be black pixels peppering all over the picture due to the increased inherent defects in the sensor. You have to know that your computer LCD (17 inch) monitor generally can contain roughly up to 1.3 megapixels worth or (1280x1024). Some LCD (21 inch) monitors can handle up to 1600x1200) or close to two megapixels. So if you have a 4 megapixel camera and view pictures on your 1280x1024 LCD which can handle up to 1.3 megapixel, your computer will compress your original picture and throw out about 2/3 of the pixels that your camera captured in order to fit into the monitor as a full screen. So you would ask yourself , why do we need more than 1.3 or 2 megapixels unless we have 50 inch LCD monitors? Well, good point, but your Windows Photo Gallery or Windows Picture and Fax Viewer also have another feature called "Best Fit" (XP) or "Actual Size" (Vista) that you can click to alternately view all the pixels the picture file has to show, you will suddenly see the picture explode much bigger than your LCD screen allows and you will be forced to use the scrollbar to pan around the picture or if you see "hand" on your mouse cursor , you can drag the picture around. This is the real original picture that you shoot with your camera with 4, 5, 8, or 12 megapixel. They are really big pictures. It is too bad that LCD monitor manufacturers has not come up with 4, 8 or 12 megapixels capabilities, yet. It is funny that digital camera manufacturers can squeeze 4, 8, 12 megapixels in the tiny sensors, but the LCD monitor manufactures cannot show all of the megapixels in the much bigger monitor screens. Anyway, I am writing this because I am pretty sure that many people never see what a single puny pixel looks like until they use the magnifying glass icon to zoom in the picture all way to the maximum. Now back to making decision on which digital camera you want to buy for the money, I can tell you that I just chose a 5 megapixel with 8X optical zoom (not digital zoom). Digital zoom is an artificial zoom capability to piggyback on the optical zoom by means of electronic manipulation. I always ignore the digital zoom feature, always! I value optical zoom highly because it can bring your 5 megapixel worth as close to the target. Any camera with more than 8X optical zoom is not practical because of the air pollution or haze that will show up more with more than 8x optical zoom and fuzz up the picture anyway. Also, buy a good tripod because steady shooting on a tripod can help with extra crisp picture performance. My 5 megapixel camera still show a few black pixels here and there and can be seen when zoomed in on the computer screen, but I dont mind as much as if I have 8 or 12 megapixel camera which is far worse! Two megapixel cameras is actually good enough and get much better with optical zoom at least 3X or 5X or even 8x if you can find it. The most common digital cameras I like is 4-5 megapixels with 4-8x optical zoom. Forget about digital zoom feature, it is artificailly meangingless! I can explain what digital zoom mean, but I have written long enough here and I am exhausted!

I may have missed it but I did not see an explanation of just what pixels are.

In early programming graphics were monochrome. The basic image was 8 characters or pixels tall and 4 pixels wide These were easy numbers for binary computing. An easy example would be the letter E. The box for it contained from the top down X being a black pixel, - being a white one. the code looked like this XXXX,X---, X--- ,XXX-,X---,X---,X---,XXXX. If you tipe them one over the other you will an upper case E. That was 32 pixels. Larger graphics were created by conecting 32 pixel boxes together. Later coding was added to put shades of gray into pixels. When color monitors and printers became available coding was added to put color into pixels. The need for speed and more information with less memory lead to the picture formats. bmp jpg tiff gif

The early digital cameras and web cameras were for a very short time 320 X 240, and then (and some still being made are) 640 X 480. Do the math 640 times 480 equals 307,200 or a little less than one third of a megapixel. A megapixel is 1,048,576 pixels.

A picture at 5 Megapixels might only use up about 1 Megabyte of memory because software has compressed and encoded repetitious areas of the photo. After you take the pictuire, your camera is busy processing that data for a few seconds.

I understand that more pixels make sharper pictures and that's good. I personally don't need 10Mpx because I NEVER make 8 x 10 prints and the more Mpx's make larger files that eat up my hard drive. So I wish the manufacturers wouldn't make ALL of their camera's so large. For my purposes and many others, I think 3.1 Mpx would be adequate for my needs. I have a Nikon Coolpix 775 that is several years old , with 2.1 Mpx's and my biggest complaint is it is so darn slow that if I photograph my young cat playing, by the time the camera takes the shot, it reacts so slowly, the cat has moved by the time the camera shoots the picture. This camera is 4 years old and hopefully they've improved the reaction speed of electronic camera's, but now I'm more or less locked into buying one of 10Mpx's or greater.