Quote
jornada720 - 2006-03-23 6:19 PM
What? There were once a replacement? Wow.
Could someone tell me how a TFT differs from a DSTN? Is it just the refresh rate?
DSTN - Dual-layer Super-Twist Nematic
It is a development of standard single-layer LCD devices. On an LCD each layer of Liquid Crystal can "twist"
(polarize
) light only to a certain degree angle. DSTN has 2 layers to double the angle the light rays going through are twisted on axis. By Twisting twice as far, as there are twice as many layers doing the twisting, more contrast is available due to the front/rear polarizers being able to do a more efficient, effective job.
TFT Thin Film Transistor
This is a radically different technology at the basis, versus DSTN or standard displays.
DSTN and "standard" single layer displays are
passive. The LCD controllers are external and
address the display - draw images on the display - in a row/column format. Wires go horizonatally and make columns, wires go vertically and make rows, and where the wires cross is a
pixel of LCD. The external controller charges, and can only charge, 1 row or column at a time, then it cycles through the cross opposing wires to draw an entire column or row. It then charges the next column/row and repeats the process.
This means the external controller
refreshes the screen in "swipes" - 1 row/column at a time until the entire screen is completed. This has the side effect of the fact that the row/columns are only turned on - active - during their instantaneous refresh. The rest of the time the pixel is actually turned "off" and the LCD twist decays slightly until the next refresh. This limits the total amount of light transmitted through the screen, as the Liquid Crystal only
(fully
) transmits light during the refresh addressing. Other times it is "flickering" as the LCD twist slightly decays.
TFT is different - it is "active". Each pixel has the transistor that controls it - turns it on, or off - directly behind it. The screen is a huge matrix of transistors with an LCD panel attached. Once the external controller tells a pixel to activate the transistor turns on and stays on until turned off - no decay. While the transistor takes up a certain amount of space behind the pixel, thereby blocking that portion of the light from going through the pixel, the net increase in light transmitted due to the pixel not "flickering" more than offsets the amount of light blocked by the switching transistor blocking a certain portion of that light. Therefore a TFT display is much brighter than a DSTN.
Also, since the controller only sends "turn on" or "turn off" signals to the panel and each pixel, in essence, controls the power to turn itself on or off the transistor / pixel pair operates much faster than a DSTN display. A DSTN display controller sends power to the pixels and that takes more time, as the power level and duration must be high and long enough to activate the Crystal to the level desired until the next refresh. A TFT sends a much lower-level "on" or "off" signal and the TFT transistor is responsible for keeping the Crystal supplied with the proper signal until told otherwise.
The transistors in a TFT are responsible for the "yield" and "dead pixels" you hear about regarding modern LCD displays - a large display can have millions of TFT transistors and each one must work otherwise you get a dead zone. Sometimes the transistor locks at "off" and you get a "dead pixel", other times the transistors locks at "on" and you get a "hot" pixel. Too many dead pixels found during the test after production gets the TFT panel
(rear substrate, actually
) scrapped and lowers the facility's total yield.
Late DSTN displays were being split into 2 "fields" to increase refresh rates - each half of the display would be refreshed separately and therefore the equivalent refresh rate doubled
(even though, actually, the clock frequency was the same
).
It is
technically possible for a DSTN to have the same color reproduction abilities as a TFT but at a lower brightness output. Most times a DSTN display is reduced in color depth, versus a TFT, to increase brightness - color depth in LCD's is created by "time dithering", switching the pixel on and off in order to control the amount of light passing through the Crystal. A DSTN "technically" could accomplish this but only with a fast, external controller, multiple fields and
(much
) lower light outputs; the controller would be spending more time dithering versus refreshing the entire panel
(remember that the Crystal would revert back to it's normal state, blocking light, as the refresh is slowed down
).