Slashdot Mirror

A filter removes a portion of something. You filter out things. Phone cameras don't use an RGB filter.

Actually, they do. The standard single-chip image color image sensor uses what is called a Bayer filter array of red, green, and blue color filters in a regular pattern, one filter per pixel. From this data, a color image is reconstructed by combining the color data from that pixel with the colors from the neighboring pixels. The dirty secret is that the resolution of a "megapixel" camera is thus much less because a nine by nine array of pixels is used to create the color data for each color output pixel.

Color sensitivity is often reported in the datasheet for the sensor. This is a datasheet for a typical CMOS sensor, and the color sensitivity is shown on page 11. The individual sensitivity is due to the filters, the MONO data is for the sensor sans filter. CCD spectral curves are quite similar. The Sony ICX285AQ is an example. If you compare data sheets, you'll notice that the ICX285AL is identical to the AQ with the exception of the color filters, and the AL version is the monochrome sensor.

It would be interesting and not that expensive to create a different color filter pattern on the sensor to include IR and UV filters, but the use would be limited. Those who want IR images typically use the monochrome sensor and external IR pass filters, and UV is usually excluded by normal users by the use of a skylight or other UV cut filter. The dyes in the RGB filters are sensitive to the UV and will decompose over time, changing the color capability of the camera.

An inconsequential amount. I suspect a 20mA @3.2V LED would probably look similar to the one in the photo.

I was amazed at how good the headphones that came with my BlackBerry Bold 9000 are, as well as the quality of the output of the phone.

I doubt that they use more than 12bit dacs anyway.

SoAnIs from hydrogenaudio also gives an interesting perspective on the 24-bit thingy.

I did some digging and it looks like the Bold 9000 uses the Texas Instruments TLV320AIC3106 stereo audio DAC rated at 102 dBa S/N, 16,20,24,32-bit data, and 8 kHz - 96 kHz sample rates. I don't pretend to know if the Bold actually outputs at the full capability of the chip, but why bother using such a chip if you're not going to make the features available?

There is no doubt in my mind that 24-bit and higher is better than anything analog, but I guess I'm old enough that someone will have to prove to me that CD's can be better than reel-to-reel. I'll also add that there is simply no comparison between reel-to-reel and vinyl. Have you ever listened to a good quality reel-to-reel? One roughly the size of a 24" CRT (ca. 1995)? They really do sound quite good,not that I'd ever buy one.

Stepping voltage down is cheap and easy, you use a switching power supply. You can build one trivially for low voltage applications using one three-legged IC and two capacitors. For 5V, for example, use a 7805.

You probably know all this and just mixed up your wording, but a 7805 is not normally used as a switching regulator. OTOH, if you do use a switching regulator, then you can turn voltage up as easy as down. The word you're looking for is "flyback"

our childish "uh oh" introduction, your completely un-cited "a lot of issues" comment, and your vague "I recall" interruption reveal the fact that you're spouting off some crap on a subject you have no direct, real-world experience with. Find another subject with which to stroke your ego kid, because you're looking like a pompous dumbass on this one.

The specs for a 256Mb NAND flash memory chip by Samsung (which is by far the biggest NAND flash manufacturer today) quotes 100k millon write/erase cycles, and this is for an IC commonly used in USB pendrives. The figure usually tends to get worse with increased memory sizes since the memory "element" (float gate) becomes smaller. For example, Modern 16Mb chips, which are the ones i have experience with, usualy quote 1 million W/E cycles endurance.

But, it felt good stroking my ego a bit more :) Thanks!

Your closer to the mark than you think.

http://www.alldatasheet.com/datasheet-pdf/pdf/5612 /MOTOROLA/2N222.html

You could then pull the chip, and 'flag' the associated pins (bend them out, so they no longer enter the socket)

This is called "flying" a pin and its used frequently in debugging/modifying hardware. Good basic idea, but if you're going to attempt it, you'll need to tie it high as floating it will result in write cycles whenever the ambient RF decides to push your new "antenna" low. Most new bios uses a standard 1Mbit eeprom configuration (SGS datasheet here) so for the DIP configuration, write enable would be pin 31. Std procedure would be to tie it high through about a 10k resistor (ttl inputs can be damaged by power rail spikes if tied directly). Cut one lead of the resistor to about 1/4 inch length, tack solder it to the top of the (for now unbent) pin 31, stand the resistor vertically and loop the other lead over the top and down to the supply, pin 32. "Fly" the pin before reinserting or bend it back in to do flashing. The resistor remains. Find your specific mobo chip's datasheet here