Armadillo wrote:
A Pixel is a tiny light sensitive transistor that converts Photon energy into electrical energy. AKA Picture Element. Michael G
Each pixel is derived from multiple sensor sites' data sensor sites are not pixels. The sensor sites send analog data to a Digital-to-Analog converter, then to a processor that turns a matrix of RGB filtered, monochrome energy representations into an output file in one of two forms either a raw file that is (mostly) just quantized sensor site information distilled to an intermediate pixel format, or a fully rendered JPEG file. The pixels are created through very complex algorithms, when the raw file is created.
Armadillo wrote:
PPI refers to how many photosensitive transistors can fit onto a specific sized photosensitive chip. If your photosensitive chip is 1 inch long and the transistors are small enough to fit 300 lengthwise into the chip you will have 300 PPI. Michael G
Incorrect. PPI is a header setting in an EXIF or TIFF file, created in software or firmware. It is irrelevant to the overall X by Y dimensions of the image in pixels. It merely tells SOME software how large to display or print the file... i.e.; how large to make each pixel before converting it to visible dots.
Armadillo wrote:
What this all boils down to, PPI can be used to gauge the resolution of a device to capture an image electronically, digital camera to computer. DPI can be used to gauge the resolution of an image to a printer. Fortunately, technology has evolved to the point where the transistors on the photosensitive chip are close enough to the ink jets of inkjet printers to be equal. Our modern consumer printers perform the required calculations to convert the PPI values to DPI values to obtain the best quality printed document within the standards of current resolutions. Michael G
br What this all boils down to, PPI can be used ... (
show quote)
INPUT DEVICE (camera) resolution is measured as the maximum number of derived pixels it can store. That is often smaller than the number of sensor sites on the sensor. It is expressed as X by Y, i.e.; 6000x4000 pixels or 24 Megapixels.
The relationship between input and output resolution is quite indirect. Different imaging technologies require and use vastly different numbers of "dots" to create essentially the same resolution on output substrates. A dye sublimation printer might use 300 dpi, while an inkjet printer might use 2880x1440 dpi, to achieve the same output resolution from the same input file.
You can say, roughly, that input file resolution (expressed as PPI) determines AVAILABLE IMAGE DETAIL, while output file resolution (printer dpi) determines DEVICE RESOLUTION. Even that varies with the printed substrate (paper type). It is best to deal with these as separate considerations.
Armadillo wrote:
When it comes down to providing image products in the 'Real World', we must supply the imagery products in the format and resolution values the customer requires. If the customer wants DPI because his professional level printers require that value, we must provide DPI in the image product. Failure to do that and the customer will most likely say, "Very nice picture, but we cannot use your content." Michael G
Huh? In the real world, photographic labs and service bureaus will either tell you to provide an image at a specific number of PPI, or will give you a chart showing file dimensions in pixels and the maximum size of print that can be made from each. A few über nerds will provide a complex mathematical formula from which to compute this.
Printers all PRINT in dpi. They take input files with specific PPI in the header. At the lab I worked in, we made ID cards at 300 dpi, with dye sublimation printers. We made portrait labels on RIP-driven color copiers at 400 dpi. We made portrait prints on Noritsu mini-labs at 600 dpi. We made larger portrait prints on Epson printers at 1440x2880 dpi. But we told our customers that our lab standard resolution was 250 PPI, which we always converted their images to, internally, if they had not submitted them to us in that resolution. Software printer drivers or hardware RIPs unique to each output device did the rest.
Armadillo wrote:
Another area of contention is images on the WWW.
If you take an image straight out of the camera, save that image as a .jpg @ 300PPI/DPI, then go back to the camera image and save this image as a .jpg @ 72PPI/DPI with a modified file-name. Now, using your computer navigation tool read the file details for file-size on both image file-names you will see the 300DPI is much larger than the 72DPI file.
Michael G
Not necessarily! Remember that Resolution is only a field value in the EXIF header of an image file. If you save a SOOC file, you have a choice to save it WITHOUT re-sizing it, in which case the header changes, but the pixel count and pixel dimensions stay exactly the same, while the pixels "get smaller" (not really they're just numbers but they're reproduced smaller by the output device's interpretation of the PPI in the Resolution field). Michael G[/quote]
Armadillo wrote:
In the USA, most ISPs charge for the bandwidth service provided (5Mbaud - 100Mbaud), not for the bandwidth consumed. Michael G
Baud is an appropriate term for gauging the speed of a telephone modem. Although they are called "modems", modern cable internet receivers use fully digital data, and the speed is measured in megabits per second. Old telephone modems transmitted data converted to analog audio signals, where the term baud or Kbaud made sense. There are various reasons why these terms bits per second and baud are different or the same, under various circumstances.
Some ISPs DO throttle your connection to a lower bandwidth after you download a certain quantity of data in a month, much as they do for wireless (4G or LTE) service. As usage grows, as competition gets stiffer, and as technology gets better, this will vary.