THE SONY SERVICE CODES - Articles, Comments, Discoveries - Page 12

You're very, very welcome! I'm quite boggled that what has been posted here in the way of *specific* instructions were of value to you. Who knew they would apply to your set, too? Power to the Paying Customer!

Trouble is, the factory settings for Standard and Vivid (and a bit with Movie) invoke a *dynamic-contrast* feature in the set, controlled by the settings for 2170P-4/BLK and the following three codes: DCTR (dynamic contrast ?), APED, and DSBO (dynamic sub-brightness offset?) BLK is a combo preset for those next three. Any picture mode that sets BLK at 0 turns OFF any dynamic-contrast activity: Pro on all inputs; Movie on all but a few 480i inputs plus ATSC (tuner) 1080i and 720p; and Standard and Vivid on all inputs. So in Pro mode, your black-level settings (SBRT) and xDRV and xCUT settings are seen as you have set them, and gamma is as you have set. On the other picture modes, the wildcard of dynamic-contrast affects both contrast/brightness and apparent gamma based on the video content.

Gamma settings for GAMM=0 (GAMx = 0) *are* max; on my 36XS955 it is about 2.45 in Pro mode. When I change the three GAMx settings to 3, I get a measured display gamma of about 2.2. I have made sure that dynamic-contrast effects are not working *anywhere,* as I don't want the TV making decisions like that based on the video material. I know "Dynamic Picture Control" serves some purposes for unsophisticated viewers and store demo, but it's not to my liking at all.

Bottom line: Do not trust the results of color, brightness, or gamma tests in anything but Pro mode out-of-the-box. (And caution: your black-level settings, SBRT or the Brightness slider, greatly affect the measurement of gamma. Make sure you have set it correctly in low room lighting.)

Of course, as I have, you can re-program the three other modes to be anything you wish. It takes a long time to go to all of the relevant settings for each mode and change them, but I believe it's worth it. (See article #05 - Customizing Picture Modes, in this thread.) I supposed one could leave one of the modes as-is for uncritical bright-afternoon viewing of kids programs or football, which is what may be intended by Sony. But I have never regretted making Pro and Movie *exactly* alike, then altering gamma for Movie to 2.2 (GAMM=1 and GAMx=3). I use that reduced gamma a lot for dark HD programming and some SD material where it's warranted. (CSI/Miami, Lost come to mind.) If it looks better in Movie mode, that's what I use instead of squinting at the shadows!

Glad you're being successful in improving your set, Hef! It will leave you feeling like you have more control, and you will get it to work like *you* wish it to.

Right. Your *service* warranty lasts for 90 days. As justsc said, you have every right to expect reasonably minimal defects out of the box (although *none* is hoping for too much). You could wait a little longer for the set to "settle in" before calling a service tech, maybe a month or two, but I have no proof that significant changes occur during that time -- especially with geometry and convergence.

Tip: Instead of calling any Sony national 800-number, you might mosey over to a high-endish dealership for Sony, be friendly, and ask them for a service-shop referral. Then call the shop directly. You are likely going to get a more-sympathetic response to your wishes from a local person than from a Sony-service call center! I had *no* problem with this 7 days before the warranty expired, and he was able to correct a global up/down red shift. Then he explained the nightmare of doing small local corrections that might throw off the otherwise-excellent convergence elsewhere. (It all interacts; I know this from experience.) Having his superior wisdom in the house for a while was a definite benefit!

You are confusing the concept of black level as input-standard (a voltage representing 0 or 7.5 IRE) with black level as observed (set, say, with a PLUGE pattern). As observed, black is black if properly adjusted no matter what the input is that represents black. In my case, I see a slight screen glow for *absolute* digital black, which is an orthodox result from a DVD PLUGE calibration using *my* player. One can confirm with the "digital black" screens on, say, the DVE DVD.

I did PLUGE only once for my DVD input, since there is little variation among high-quality DVDs, and that was with Brightness at 31. That is generally satisfactory.

For broadcast video, tapes, etc., black level is empirical: I adjust Brightness until the observed black level appears correct -- relatively easy. Brightness = 31 is only a starting point! I don't care if the source-black is 0, 7.5, or any other bizarre IRE level (some cable channels are way off!), it ends up *looking* black.

It would be best to refer to screen brightness as percent-white, not IRE. My measurement of gamma occurred after confirming a perfect black level in low-level evening room lighting.

How are these two statements related? The slight glow is generated by the tube, easy to see, since I have a 4:3 tube, and the boundary of a 16:9 display is clearly visible if there is a glow.

The thick, dark glass reduced the brightness of all light emission evenly, bright and dim. Picture and Brightness settings compensate for it completely. That the glass is dyed dark reduces the effect of room light on the tube face, making the shadow detail much clearer. (Room light has to pass thru the glass *twice.*)

If you set "black" to absolute black (no light emitted), the darkest shadow colors are too saturated and shadow detail is lost. The "toe" of the emission curve is also polluted by room light. I have experimentally adjusted black levels up and down with great sources (usually DVDs), and the best results are with video-black for the source chosen set *slightly* above absolute black for the CRT. The resulting picture qiality proves the point: anything else doesn't look right.

This illustrates the difficulty of measuring gamma with any type of screen pattern. The usual pattern juxtaposes two areas: (1) a specific composite "gray" that is composed of black and white in some pattern, say alternating black and white scan lines. (If black is truly black, that is a (linear) 50% brightness relative to the white level). And (2) a real solid gray of calculated density such that for a given gamma, say 2.2, the brightness of the solid gray appears exactly the same (average) brightness as the b/w composite gray. You blur your eyes, if that helps. Raising the black level with the Brightness control screws up the relationship between those two "gray" areas, giving a false reading.

In actual fact, Brightness (black level) setting has *no* effect on real gamma: In practice, the measured brightness response of any CRT display is composed of a fixed-brightness pedestal plus a curve relating video input to additional screen brightness. Changing the pedestal height (adding a screen glow or eliminating it or making it negative) does not change the power factor of the curve: the gamma. But measuring this would require a photometer aimed at the screen and a variable, calibrated video-level generator as input. Then you could draw the graph, separate the curve from the pedestal, and determine the exponent of the curve. *That* would be measureing gamma! But we're stuck with using patterns, and so one has to be careful to calibrate screen-black to a digital-black signal, and the whole measurement must be done in a very dark room. That is how I estimated the inherent max gamma of my set for GAMM=0 at 2.45. It amounts to an interesting tidbit, and one can set other gammas with the GAMx codes (3 yields about 2.2). But I yearn for a true Gamma Control, and *that* would be correctly labeled "Contrast." I set gamma to suit the program material, and the max setting is usually the best, especially with DVDs and SD broadcast. But then there are those few super-contrasty broadcasts that lose a lot in the shadows. Lower gamma to the rescue!

Or raises the bigger problem of broadcast quality control. The ideal is that correct shadows and correct dark-color should occur with the same black-level adjustment.

But I regularly encounter video that doesn't fit that description: Colors are oversaturated in shadows even though the shadow brightness (black level) is fairly high! Everybody in the scene has orange-red colors under their chins and noses and behind their ears, but there is no real black in the picture -- what to do? I diddle Brightness for a good compromise, then enjoy the show.

I regularly see meticulously set up video from the big three networks on their HD programming -- set black level to a real black, and the results are just about perfect. But there's a lot that doesn't conform.

My TV is calibrated for Brightness = 31 and Color = 31 to be a best-possible average. For DVDs it's perfect. Then I watch an ABC program called Brat Camp and can't get black levels right until I jack Brightness to 36. Color is fine. Later, the local ABC news requires me to set Brightness back to 31.

On a Friday evening I watch some PBS half-hour news programs, and I have to crank Brightness down to 25 and Color to (get this) 21 for a reasonable picture! Why is PBS transmitting 150% chroma?? Who is minding the store? Am I worried about gamma? Hah!

Oh, agreed. My black-level measurements are not made with bright material present. Program fade-outs give me an opportunity to see the edge of the 16:9 area relative to the unscanned area of the tube. Absent that, with a little practice, one can make decent judgements based on the appearance of the video. Either tow low or too high a black level will appear inferior to a satisfactory setting.