Confusion over megapixels vs. megabytes

Megapixel referances to a picture clarity or resoluation when displayed on a viewing format like a Tv,Monitor or such..
While megapixels are important usually the higher the better this is up for disscusion..

Megabite referances to the amount of space taken up while recording any information be it video or data material to any type of storage media into a file.

Please don't confuse the two just because they are in lots of 1,000 units..

The size of the picture is due to its Codec used, your old camera uses old codec's not used anymore. Codec's are the format used to save on device.

wiki

A codec is a device or computer program capable of encoding and/or decoding a digital data stream or signal.[1][2][3][4] The word codec is a portmanteau of 'compressor-decompressor' or, more commonly, 'coder-decoder'. A codec (the program) should not be confused with a coding or compression format or standard - a format is a document (the standard), a way of storing data, while a codec is a program (an implementation) which can read or write such files. In practice "codec" is sometimes used loosely to refer to formats, however.

A codec encodes a data stream or signal for transmission, storage or encryption, or decodes it for playback or editing. Codecs are used in videoconferencing, streaming media and video editing applications. A video camera's analog-to-digital converter (ADC) converts its analog signals into digital signals, which are then passed through a video compressor for digital transmission or storage. A receiving device then runs the signal through a video decompressor, then a digital-to-analog converter (DAC) for analog display. The term codec is also used as a generic name for a video conferencing unit.


With the new hardware comes new codec's - raw formats from more than 350 digital camera models from all top manufacturers and supports CR2, NEF, DNG, SR2 and more, as well as Photoshop PSD and specialists image formats like OpenEXR, Radiance HDR, TGA, DirectX DDS...

NEF Codec is a module that makes Nikon RAW (.NEF) image files as easy to work with as JPEG and TIFF images.

I believe codecs refer to video, not image compression.

That's the funniest response yet.

In an uncompressed image, 1 pixel equals aproximately 1 byte of data. Therefore, a 14 megapixel image, or 14 million pixels, would take up approximately 14 million bytes, or 14 megabytes (14 MB). JPEG image compression, and other forms of image compression, reduce the file size by removing redundant information. That is why the images are smaller than expected. Some types of compression are "lossless", meaning the compressed images are exactly the same as the original uncompressed images. Some types of compression, such as JPEG, are "lossy", meaning that some data has been lost in the process of compression. Compression is NOT a bad thing, however, even if it is lossy. Most of the time the human eye cannot see the difference, except at high levels of compression. The savings in storage space easily makes up for the "loss".

Excellent explanation, Lee! I thought I understood
it myself until I read your response in detail!

I also have a nikon digital camera (coolpix p2). The megapixel is 5. The image size setup option may need changing in the menu - image mode.

According to nikon:

5 megapixel = 2592 x 1944
3 mp = 2048 x 1536
2mp = 1600 x 1200
1mp = 1280 x 960
PC = 1024 x 768
TV = 640 x 480
3:2 = 2592 x 1728

14.2 megapixel means that the image has 14.2 million pixels... period. Not per inch, per centimeter, per anything.

You're right, it's 14.2 megapixels per photo.

Pixels are the squares (or triangles, i don't know for sure) in a picture, containing 1 color (a mixture of red, blue and green). The more pixels, the better the quality of the picture.
1 kilopixel = 10^3 = 1 000 pixels
1 megapixel = 10^6 = 1 000 000 pixels
1 gigapixel = 10^9 = 1 000 000 000 pixels
and so on...

The quality of the picture also depends on the steps of a color (from black to red/blue/green) The more steps, the more colors and a more realistic, less "pixelated" picture. This means a better quality of the picture.

Bytes (8 bits or "binairy digits", meaning there are 8 numbers that can have 2 values, 0 or 1) are used to store data. A photo can be stored in different ways, so 1 photo can be stored with 5 megabytes in the first photo and in 3 megabytes in the seconds. This depends on the way of storing the foto, like a PNG format, a BMP format, a JPG format, a GIF format, etc.

How much bytes depends also on how much pixels there are in the photo and how many colors are used (8 bits (256 colors), 16 bits (65536 colors) 24 bits (16777216 colors, 256 colors for green, for red and for blue), etc.).

(there are also bits used for transparency, like 8 bits in the 32 bits color, used if you use multiple layers of photo's on top of each other)

1 byte (B) = 8 bits (b) = 2^8 = 256 values
2 bytes = 16 bits = 65536 values

1 kilobyte (KB) = 2^10 = 1024 bytes
1 MB = 2^20 = 1048576 bytes
1 GB = 2^30 = 1073741824 bytes
1 Terabyte = 2^40 = 1099511627776 bytes

24bit colour, the RGB we are all used to on our computer monitors, by definition contains 3 bytes per pixel. 3 x 8 = 24
So, uncompressed, the file size is 3 times the physical image size.
A 14.2 megapixel uncompressed RGB image will be 42,600,000 bytes (or 40.63 megabytes).

CMYK (4 colour for reprographics) on the other hand is 32bit colour, and 4 bytes per pixel.

It's not per inch or per anything. It's exactly as you say, but given the folks who think it's per inch how much else don't they understand?
A 14.2 megapixel image contains 14.2 million pixels. Even high quality images intnded for reprographic reproduction only usually contain 300 pixels per inch (90,000 per square inch). What on earth would you do with 14.2 million per square inch?

A 14.2 megapixel sensor has that many photsites in total and depending on the dimensions of the sensor this could mean that the sites are larger or smaller than another camera sensor but the total number of photosites is the same...Yes is does relate of course to the final resolution of details of the image and it is also true that all things being equal two cameras ought to have the same details though a larger or smaller sensor will have differing signal to noise and different rendering of the colours etc. but the actual number of dots of colour ought to be roughly the same. What is confusing the writer is that the file size is not 14.2 MB. The reason it appears much smaller is that likely the writer is seeing a jpg file of the original and that uses methods to compress and eliminate repetition of identical pixel data in the saving of the file and thus, the computer rebuilds the data ( more or less but it is called lossy for a reason) into the final seen image. Only a RAW file or a TIFF writing camera has a file consistent with the number of photosites and even here there is some modest compression (non losssy) that makes the file smaller but not small. When one expands a jpg, you can see its real open size with most photo viewers and it is as you want it to be 14.2 megapixels.

An excellent answer Rick1025!

Rick1025's answer is a bit confusing, which shows you why the entire subject is confusing. The 14.2 megapixels means there are a TOTAL of 14.2 million dots (pixels) making up the picture image. It doesn't matter how many inches the camera sensor is or how big the eventual print will be. When you make a print, those 14.2 million dots will be spread out over the entire area of the print.

A small computer in your camera takes the color and brightness information in all 14.2 million dots and compresses it into a digital file that it stores in the camera's memory card. Better compression means a smaller file, as others have discussed.

Mega is a prefix that means a million of something, say, a million pixels. Megapixels is not a per inch number. It is the actual number of image sensors on the chip were the image is formed. In most point and shot cameras the chips are much smaller than one square inch making the density of sensors on the chip as high as 40-50 megapixels per square inch. For a variety of technical reasons smaller density numbers are better. This means that a lower megapixel camera, with those pixels on a larger chip, may take better pictures then a camera with a higher pixel count on a smaller chip. This information is available on the dpreview site.

I thought that 14.2 megapixels are for the whole picture not per inch of area. Even if it is per inch I think it must be per square inch and not per linear inch as stated. In fact they are the number of pixels for the whole chip.

Correction to "...the camera contain 14.2 megapixels per inch of area". More accurately, "the camera sensor contains 14.2 megapixels", period. Since sensor of the Nikon S6000 is not one square inch in size, your statement is incorrect. (The sensor is actually closer to three hundredth of a square inch! About 4mm by 5.3mm).
See http://www.kenrockwell.com/tech/angle-of-view.htm for information on the dimensions of various sensors.

Clarification of "...the image is converted to a file that contains 1.8 megabytes of data".
The file size can vary, depending on resolution and quality settings, so the 1.8 megabyte size is only true for this camera at one particular setting.

That's a good general explanation, but with a significant error. (Significant, at least, for those interested in photo display and printing.) The megapixel property of a camera indicates the total number of pixels (color dots) in the entire image, not "per inch of area." (For one thing, area would be measured in square inches; inches would just indicate the length or width.) Since each pixel is typically stored as three bytes, the entire 14.2 megapixel image would occupy around 42.6 megabytes of disk space, if stored as a byte-for-byte copy of the sensor image. Fortunately, digital pictures are normally stored in a compressed format, which is why the high-resolution image may only occupy a few megabytes.

The entire thing is layered and we need simple words to make it work in daytime. It goes like this.

A pixel is short for a picture element. Your PC does not know colors, it does know bits. It uses 8 bits to code a pixel, but 8 bits also form a byte. That was in the old days. To handle 16 million colors you need 24 (2^24) bits or 3 bytes. That, nicely, fits onto the RGB system. Open your color palette in MS Office to see it work. Computers can only count, that's it. To get the magic we tell them how to count what. The TFT screen was the last nail in the coffin of analogue computing in PC's.

Megapixels is the short word used in TFT displays only. Megabytes that for data storage. The PC uses the bytes to show you the pixels. There is a relationship, but that is determined by the compression algorithm and they won't tell you that! If your camera can get 14.2 MP into 1.8 MB, you have a compression ratio or 87%!! Excellent!!! Any setting will not improve picture quality at all. Dollars for a better camera will - when they decide the market is ripe for another harvest. So watch what you buy when and what do you get for the outlay.

In the old days they used bitmaps, but when you zoom in you get jagged edges. Then there is vector graphics - and a few other things like Mpeg and Jpeg and so on. All of this was caused by the various problems to make counting computers understood by seeing humans. This entire process of deciding which method to use for displays initially took some twenty years. That was when they had something like 400MHz as a top speed for a PC. Now it is in the region of 3.7 GHz, right? Literally about a 1000 times faster. This saw to it that you may now convert just about any file format into any other.

The cynicism on scamming devils is slightly out of place. You can understand and calculate everything once you have all the information.

Simple solution. Highlight any text you see anywhere, Ctrl-C and Ctrl-V that into your Google search box. Read everything you get. As suggested Wiki is an excellent start. Happy hunting.

One minor correction: your statement that 14.2 Megapixels means "the camera contains 14.2 megapixels per inch of area." is NOT correct. "Per inch" of a picture is meaningless until you print the picture and then the relevant measure is SQUARE inches. In fact, 14.2 Megapixels means that the picture you have taken contains 14.2 megapixels. To get pixels per square inch you would need to divide 14.2 megapixels by the area of the printed picture.

The megapixel rating of and sensor is the actual amount of pixels on that sensor. Not all sensors are the same size and some pixels can be much smaller then pixels on a more expensive camera. It is correct that the amount of pixels will help clarity and will benefit you when you make larger prints but since most people only print 4 x 6 prints it really doesn't matter. In fact most people use their images to go into Facebook or site like that so the images are reduced in there over all size by Facebook anyway. They need to make them smaller so that they can load faster and the resolution of a monitor is not that great to begin with.

Megabytes refer to the size of the file stored nothing to do with megapixels or cameras. Raw files are the best but if you don't know how to handle them or correct them or why you're even using them then don't. I teach photography and this is not an easy subject for students to understand. Worry less about this and start to look at composition and lighting and you will start to enjoy photography as it should be.

Remember the word photography came from the Greek and it means to "write with light". Enjoy.

14.2Mpx means 14.2 million pixels in the entire image, not in one square inch of the image. In chemical photography and printing, we used to break up an image into tiny dots (much like pixels in a digital image) by a process called screening. The screen could be characterized by the number of holes per inch-- therefore the number of dots per inch in the print. This legacy of DPI(Dots Per Inch) thinking has overlaid many peoples' thinking about digital images. We still think and talk about DPI in printing and video display screens (dot pitch) but in describing an image file, which has no physical dimensions, resolution is called out as the number of pixels wide by the number of pixelas high contained in the file. A 1920 x 1080 image can be printed an inch high or projected 50 feet high, and it will have the same number of pixels, but the pixels per inch will be radically different. Even on a CCD or CMOS sensor inside the camera, pixels per inch is not a useful measure. It takes typically three light-sensitive elements on a multi-color sensor, or 1 element in a monochrome sensor to generate 1 raw pixel, but interpolation and Beyer filtering and multiplexing of three-chip systems generate various pixel densities, even from the same chip architecture, so again pixels per inch is not used.
The greatest variation you will find in file size vs. pixel count is caused by compression. We use run-lenth and various other compression systems to reduce the file size and the bandwidth required to process the image. The effectiveness of those compression algorithms is radically affected by the complexity of the image. In the same camera, with identical resolutioin and quality settings, a solid white image with a single tiny black dot in the middle will create a relatively small file, whereas ocean waves will create a much larger one. The ratio can be hundreds to one.
SO.....
The resloution of an image, and the number of pixels in it, and the file size, are all distincct attributes which are generally in direct proportion, but can have very different values and even radically different ratios, depending on the image content, type of compression and method of processing.

Using the ratio's in your post, the 14.2MP camera will give a side count of about 4927x2886. You assume correctly on file sizes. Computers don't do voodoo. The three settings allow you to choose the level of coding. See my reply to rick1025 and timhood's post.

Tim, thanks for taking the time to post this excellent response. It definitely has helped my understanding.

All the best for your holiday season.

Jeff

One other not on the megapixels is that frequently seem to use higher numbers for advertising. To really grasp image quality and true megapixels you need to remember the top of the line digital cameras use sensors that are the same size a 35 mm negastive. Most point and shoot camera's like the Coolpix series use tiny sensors that go with the very small camera bodies. So go figure how a point and shoot is able to cram in 14 million pixels into a tiny area that might be 5% of the size you would find in an SLR sensor that also has 14 million pixels.

don't be fooled by advertising.