Originally Posted by ahigee
Just like I said.....
Here is some goodness. I sorta tweaked with the settings from what others have said in posts and did my own thing.
The set is the HL-T5087S....I was so impatient I chose to pick it up from the shipper
I'm posting thumbs to let you preview the goodness.
I have definately got my wow factor already from Happy Feet, now it's time to play some games! Ask away if you want....I'm not tech savvy on TV's yet just to warn you though.
The quote above was from several days ago. What interested me was the King Kong screen shot. This is a very high resolution photo (downloads as a 12MB bitmap file), that appears to be an excellent portrayal of the image on the screen.
But what really caught my attention is the presence of what could be some unfortunate color renderings. You may have to zoom in on the picture a lot to see this, but here is what I see--
In the clouds at the top, there is a lot of magenta shading. I believe this occurs within a particular narrow luminance range.
On the girl's cheeks, there are light magenta banding areas. (This is sometimes called "solarization", more correctly "posterization", I believe.) Below this are two smaller, darker, and more noticeable magenta bands, one of which follows her chin line.
On the girl's neck, there are cyan splotches. These looks kind of like veins until you zoom in the see that they go in all directions and criss cross each other. These neck splotches continue to the left of the picture onto the gilrl's shirt--where they are joined by magenta splotches. On the right side of the picture, the girl's shirt sleeve goes almost fully cyan.
I am concerned about this, because I have picked the Samsung LED set as a good replacement for a plasma unit which does this very same thing, only worse. Plasma sets made in the past two years have been plagued by this problem--which appears to result from the inability of the three colors of plasma cells to linearly track input luminosity changes, and the difficulty in engineering data decoding and plasma driving circuits to cope with the non-linearity of the plasma panel. See the Sharp and Panasonic flat panel threads on "purple snakes" and "clay face" for examples. And see my posting at my web site
for a look at my Philips plasma problems. Both Panasonic and Sharp have put out replacement picture processing boards to reduce--but not completely cure--their problems. Philips has sent me four replacements for my TV, all with the identical problem. Philips has given up (on me) and is now refunding my payment.
So, what does anyone think about this? For starters, I do not believe any of what I have described is due to MPEG decoding artifacts. I have seen too many instances of this on my own TV and on screen pics in the forums noted above.
If you owners want to see if your set is suscetible to a color problem, grab the RGB Ramp and the Star Gray Ramp test files from 1zelda.com/tv
and display them on your TV. The RGB ramp is a 256 step NTSC pattern. You can likely see the individual steps. There should be no banding. There should be no instances where the steps do not change to the next darker or lighter shade. There should be no obviously too light or too dark steps. And there should never be an instance where, for instance, the red ramp displays anything other than a solid shade of red. With the star gray ramp, there should be no hint of any color rings, and, of course, no gradient errors.
You can display these JPEG files on your TV directly from a USB stick, or from a computer with an image viewer. For a viewer, I prefer the one that comes with Nero which can display a full screen image with no borders or boxes. You may have to adjust the TV brightness and contrast to resolve all the steps on the ramps.