THE SONY SERVICE CODES - Articles, Comments, Discoveries - Page 4 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #91 of 2962 Old 06-09-2005, 09:33 PM
 
CrocHunter's Avatar
 
Join Date: Jun 2004
Location: Massachusetts
Posts: 2,570
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Yes i know SHOF is not completely off but it looks better at 0 since there is less noise.
CrocHunter is offline  
Sponsored Links
Advertisement
 
post #92 of 2962 Old 06-10-2005, 12:41 AM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
This is a great thread and thanks for the efforts in supplying this information. I will need to spend more time reading and studying this.

Quote:
Originally Posted by KenTech View Post

I have determined to my satisfaction that some of the ISF standards are not required to such precision (6500K for example; 6350K would be fine) and the tendency to calibrate a "brightness" level (meaning Picture) is complete nonsense. There is no correct "brightness" of white; what is required depends on taste, the environment, and the particular TV's tendency to bloom at high brightness. These limits are easy to spot.

I would say you are only partially right here. The ISF standards are meaningful targets and rarely can all be achieved. The closer we can adjust all of the parameters, the better the picture will be. Any good ISF calibrator should calibrate to the CIE x/y coordinates where the true D65 = (x = .313/ y = .329), not to 6500K. True, D65 cannot always be achieved, however in calibrating, it is critical that no part of the gray scale goes plus green.

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #93 of 2962 Old 06-10-2005, 11:17 AM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
Quote:


Any good ISF calibrator should calibrate to the CIE x/y coordinates where the true D65 = (x = .313/ y = .329), not to 6500K.

Any chance of translating this into english for those of us who don't speak calibratese?

ADU
ADU is offline  
post #94 of 2962 Old 06-10-2005, 12:56 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by GlenC View Post

The closer we can adjust all of the parameters, the better the picture will be. Any good ISF calibrator should calibrate to the CIE x/y coordinates where the true D65 = (x = .313/ y = .329), not to 6500K.

Those precise decimals are very impressive, and I do know exactly what they mean, but . . . how does the *eye* see it? As ADU implies, you are speaking about a TV working backwards from your instruments. Even color temperature in degrees-Kelvin is sort of obscure, except it's the best way to make simple comparisons. Like pounds or inches.

In other words, better according to whom? Closer to what? If you have read what I have written, you know that I agree that there are some parameters that need to be nearly perfect -- good grayscale is one. But I believe that there are common-sense standards that can be achieved fairly easily without paying for "official" ISF calibration. I balk at the notion that there are people with "special knowledge" who need to be hired to make your TV "perfect" and that this is treated occasionally as incontestable religion or at least with considerable snobbery! Maybe -- if you are not technically inclined, are well-heeled, or are undisciplined or impatient with fussy detail. Out of the box, an ISF calibration can transform one of these sets, no question!

But for everyone? I am simply making the case that, for many folks, there are skills that you can learn that will give you long-term control over these excellent sets and how they behave *without* paying for help every time you get it in your head that something's not quite right. You get to pick how far you want to go at this, and you have the personal satisfaction of having done it yourself. My obvious bias is toward self-education rather than "call an expert," and I walk the talk.

Consider: Some ISF techs charge $300 per *input*? Give me a break! The most important setups that determine the overall appearance of the set are *common* to all the inputs. A fairer way to charge would be $ for basic setup and one input, plus $ for each additional input, no?

Further, I read in the AV groups that seem dominated by ISF technicians (not this one) that there is *one* standard for, say, white color temp, and anything that deviates a bit from that is hopelessly off. And no one speaks of correct gamma when, in fact there is an acknowledged standard for that (2.20). Is gamma adjustment part of the ISF calibration? Has anyone tried to watch ABC's dark "Lost" in HD/Pro mode with its gamma of about 2.45?

Back to color. My contention is that there is a range of whites that is satisfactorily perceived as pure white if that illuminant is dominating the vision field, and that there are certain colors that are "contaminants" of that scale that must be minimized (green and not-green/pink). My eyes say loud and clear: front-lighted mid-day (dense) clouds in clean air are really white. (Experts say this is 6100-6500K.) Sunlight, absent any clouds, is warmer. Thin overcast is cool. Yes, I can judge under all of these conditions what the real colors of objects are -- we are evolved to do this after all. But I have had the long-term task of calibrating computer monitors for graphic artists, and I have had to confront head-on "What is white?" This has been is on monitors that "emit" light (CTRs), and so the gray scale is seamlessly continuous with white, true also of CRT TVs and Plasma sets. My mantra has been: "In dim ambient light, eyes adjusted to screen, if it looks white, it *is* white." The other colors displayed are derived by the eye-brain combo as *relative* to this white. And so critical small color tints are correctly perceived.

So now comes my excellent TV and its precise calibrations. I calibrated my 36XS955's "Neutral" to a white point exactly matching front-lighted white clouds, and I made the gray scale as perfect as it can be. I then calculated what a correct color-temp deviation from that would be for Warm and Cool, using high-precision glass photographic filters as a standard, and set Warm to be equivalent to an 85A filter, or a bit less, and Cool to be its opposite. That's a weak salmon or blue-cyan color.

Result? In normal evening viewing conditions, all whites (graphics, blown highlights, pure-white test patterns) appear pure white. Really white. If I switch to Warm, I get used to it with out much trouble; but after an hour or so I can still look at the whites and say, "Y'know, those look a bit warm." Same with Cool: I accept the change without much complaint, but it never looks quite right to me, and I revert to that middle Neutral setting. (If I had preferred either Warm or Cool, I would have altered Neutral to match.)

That confirms more than any measurement that my color-temp settings are correct. Is my white 6500K? Maybe it is; actually I wish I knew someone with a colorimeter so we could measure it. But if it turned out to be 6250K, I wouldn't change it for anything! In viewing a variety of program material, there are *much* bigger problems to solve: black-level anarchy in broadcast SD, color casts and gross oversharpening in HD broadcasts, horrible excess color in a lot of PBS stuff (Jeez!). So after all this trouble, you are still lunging for the Menu button on the remote to make adjustments. But at least the whites are *white.*

KenTech
"We all get smart slowly."
KenTech is offline  
post #95 of 2962 Old 06-10-2005, 01:30 PM
Advanced Member
 
jmccorm's Avatar
 
Join Date: Oct 2000
Location: Tulsa, OK
Posts: 562
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ADU View Post

Any chance of translating this into english for those of us who don't speak calibratese?

"Show me the money."
jmccorm is offline  
post #96 of 2962 Old 06-10-2005, 01:33 PM
Newbie
 
Katmann's Avatar
 
Join Date: Dec 2004
Location: Portland, OR
Posts: 2
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
So these SONY codes will work as well for my 40" SONY direct view TV ?

If not do such codes exist for this unit ?

Thanks in advance.
Katmann is offline  
post #97 of 2962 Old 06-10-2005, 04:44 PM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ADU View Post

Any chance of translating this into english for those of us who don't speak calibratese?

Sure, most everyone has seen this in TV reviews that contain the color space triangle. The triangle, where Green is at the top, Red on the right and Blue in the lower left. This triangle represents the colors that phosphors were able to achieve for RGB. Within that triangle exist all of the colors a TV can produce. The triangle is plotted within the CIE color chart by their x/y coordinates. Within the triangle lies a range of values for white, from cool to warm. Movies and film base their whites on D65, therefore we try to achieve a uniform gray scale at D65 for correct reproduction of movies and film. (see attachment)

Quote:
Originally Posted by KenTech View Post

Those precise decimals are very impressive, and I do know exactly what they mean, but . . . how does the *eye* see it? As ADU implies, you are speaking about a TV working backwards from your instruments. Even color temperature in degrees-Kelvin is sort of obscure, except it's the best way to make simple comparisons. Like pounds or inches.

Well, when you have a colorimeter, you can actually measure the changes and that the eye can detect changes as small as .002, .004 can be very noticeable at times, especially when it is towards green. Without getting into too much detail, 6500K is within a range and can lie anywhere from .29 to .36 on the y axis, however D65 is a specific x/y point, .313/.329. If you have a white at .313/.340, it is still 6500K, however to the eye in comparison to D65, it will be GREEN.


Quote:
Originally Posted by KenTech View Post

In other words, better according to whom? Closer to what? If you have read what I have written, you know that I agree that there are some parameters that need to be nearly perfect -- good grayscale is one. But I believe that there are common-sense standards that can be achieved fairly easily without paying for "official" ISF calibration. I balk at the notion that there are people with "special knowledge" who need to be hired to make your TV "perfect" and that this is treated occasionally as incontestable religion or at least with considerable snobbery! Maybe -- if you are not technically inclined, are well-heeled, or are undisciplined or impatient with fussy detail. Out of the box, an ISF calibration can transform one of these sets, no question!

But for everyone? I am simply making the case that, for many folks, there are skills that you can learn that will give you long-term control over these excellent sets and how they behave *without* paying for help every time you get it in your head that something's not quite right. You get to pick how far you want to go at this, and you have the personal satisfaction of having done it yourself. My obvious bias is toward self-education rather than "call an expert," and I walk the talk.

Consider: Some ISF techs charge $300 per *input*? Give me a break! The most important setups that determine the overall appearance of the set are *common* to all the inputs. A fairer way to charge would be $ for basic setup and one input, plus $ for each additional input, no?

You should be getting what you pay for. An ISF calibrator should have $10K - $15K of equipment to do a proper calibration. Calibrators can charge what they want. The ISF recommended fees start at $225 - $325 for the first input and $125 for additional inputs. I even have this posted on my website.

Quote:
Originally Posted by KenTech View Post

Further, I read in the AV groups that seem dominated by ISF technicians (not this one) that there is *one* standard for, say, white color temp, and anything that deviates a bit from that is hopelessly off. And no one speaks of correct gamma when, in fact there is an acknowledged standard for that (2.20). Is gamma adjustment part of the ISF calibration? Has anyone tried to watch ABC's dark "Lost" in HD/Pro mode with its gamma of about 2.45?

Yes, for movies and film based material, D65 is correct for accurate reproduction. If it was produced at D65 it should be displayed at D65 to be correct. However for old B&W movies, an additional calibration at D54 may be appropriate.

Quote:
Originally Posted by KenTech View Post

Back to color. My contention is that there is a range of whites that is satisfactorily perceived as pure white if that illuminant is dominating the vision field, and that there are certain colors that are "contaminants" of that scale that must be minimized (green and not-green/pink). My eyes say loud and clear: front-lighted mid-day (dense) clouds in clean air are really white. (Experts say this is 6100-6500K.) Sunlight, absent any clouds, is warmer. Thin overcast is cool. Yes, I can judge under all of these conditions what the real colors of objects are -- we are evolved to do this after all. But I have had the long-term task of calibrating computer monitors for graphic artists, and I have had to confront head-on "What is white?" This has been is on monitors that "emit" light (CTRs), and so the gray scale is seamlessly continuous with white, true also of CRT TVs and Plasma sets. My mantra has been: "In dim ambient light, eyes adjusted to screen, if it looks white, it *is* white." The other colors displayed are derived by the eye-brain combo as *relative* to this white. And so critical small color tints are correctly perceived.

What you are saying is, you can go outside, look at the clouds, then go inside and look at your TV and say they are the same color, by memory? If you are looking at the clouds through a window, you have a problem. Just put a piece of white paper in front of a mirror and compare the color of the paper to the color of the image of the paper in the mirror.

Quote:
Originally Posted by KenTech View Post

So now comes my excellent TV and its precise calibrations. I calibrated my 36XS955's "Neutral" to a white point exactly matching front-lighted white clouds, and I made the gray scale as perfect as it can be. I then calculated what a correct color-temp deviation from that would be for Warm and Cool, using high-precision glass photographic filters as a standard, and set Warm to be equivalent to an 85A filter, or a bit less, and Cool to be its opposite. That's a weak salmon or blue-cyan color.

Result? In normal evening viewing conditions, all whites (graphics, blown highlights, pure-white test patterns) appear pure white. Really white. If I switch to Warm, I get used to it with out much trouble; but after an hour or so I can still look at the whites and say, "Y'know, those look a bit warm." Same with Cool: I accept the change without much complaint, but it never looks quite right to me, and I revert to that middle Neutral setting. (If I had preferred either Warm or Cool, I would have altered Neutral to match.)

That confirms more than any measurement that my color-temp settings are correct. Is my white 6500K? Maybe it is; actually I wish I knew someone with a colorimeter so we could measure it. But if it turned out to be 6250K, I wouldn't change it for anything! In viewing a variety of program material, there are *much* bigger problems to solve: black-level anarchy in broadcast SD, color casts and gross oversharpening in HD broadcasts, horrible excess color in a lot of PBS stuff (Jeez!). So after all this trouble, you are still lunging for the Menu button on the remote to make adjustments. But at least the whites are *white.*

If you are happy with the way you have white calibrated, then you are set. How uniform is your gray scale tracking from 10IRE, 20IRE.100IRE? what about color decoding, is Red the proper Red? You may be able to get a local ISF calibrator to come evaluate your calibration for a minimal fee or trade for your knowledge in your Sony and the SM adjustments, if he is not experienced with it.

I once read that of all of the recreational golfers, less than 10% can score under 100. it would be great if 10% of TV owners could properly set their TV. I have seen many try to set/calibrate their TV with Avia or DVE and just can't do it, most won't even buy the disc.

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #98 of 2962 Old 06-10-2005, 11:10 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by GlenC View Post

You should be getting what you pay for. An ISF calibrator should have $10K - $15K of equipment to do a proper calibration. Calibrators can charge what they want.

Yes, but I don't want their having to buy all this equipment to calibrate any TV they encounter under any lighting conditions to be my $$ problem, and anyone who learns to do *their own particular* TV doesn't have to make it their $$ problem, either. Having to buy all that equipment inflates the cost for setting up any one set considerably: any one customer is subsidizing your ability to set up other sets. *You* have to work with many different sets. But a typical owner has concern with only *one* set -- theirs. They can potentially learn a few tricks and buy *no* equipment at all -- except a good alignment DVD and maybe a memory stick.
Quote:


Yes, for movies and film based material, D65 is correct for accurate reproduction. If it was produced at D65 it should be displayed at D65 to be correct.

What does "produced" mean? I see a large proportion of movies and video that have obvious color casts, both accidental and for artistic effect. Are documentaries that show bluish snow or pinkish (real gray) rock "produced" at D65? And what do you think happens to pure white on a DVD or broadcast? You should be satisfying the *eye,* not an arbitrary standard. We are not matching fabric color swatches or printing inks, here, but attempting to reproduce what the real world of movie and video production throws at us.

BTW, you didn't address gamma. You *do* calibrate gamma, right?
Quote:


However for old B&W movies, an additional calibration at D54 may be appropriate.

Sheer nonsense! Who told you this? The makers of B/W films had no control over the color of white. Their film didn't record it. The projection systems were whatever color the arc lamps produced, and now are generally xenon-arc. What are you trying to accomplish with D54? The experience of average Joe in a typical theater of 1030-1945? Why would one want to do that? If the whites and all grays are uncontaminated by color, you have done your best, and approx 6500K is perfect! Want it warmer, turn the color-temp control to Warm.
Quote:


What you are saying is, you can go outside, look at the clouds, then go inside and look at your TV and say they are the same color, by memory?

You're misrepresenting my method on this entirely! Read my Article #03. I find this procedure very easy to do in a home environment. But you couldn't do this professionally in, say, an industrial environment. Hence the need for colorimetry instruments for *your* purposes.
Quote:


If you are happy with the way you have white calibrated, then you are set.

Yes, indeed, and I'm suggesting ways others might get there, too.
Quote:


How uniform is your gray scale tracking from 10IRE, 20IRE.100IRE? what about color decoding, is Red the proper Red?

My God, you really haven't read any of the articles, have you? Please do so before you assume I'm a complete novice. "Is red the proper red?" Jeez, this has been beat to death here and in other threads. On this late XS/XBR series of Sony sets, one has only so many parameters that can be adjusted, and you are stuck with the CRT phosphors Sony gave us. One already has to fool around with the three-color gamma settings to get a linear gray scale in the brightness range of relevance. Then you run out of options, and you move on. There are 16 increments in the four color-matrix parameters (RYR, etc.), not 128. You pick the one setting for each that gets the closest, and move on. The *encoding* matrix for HD isn't consistent from broadcast station to station, and theory says that the current CRT phosphors aren't really correct for the HD standard for perfect color anyway. Why are we agonizing over three-decimal-place coordinates in perceptual color space?

A television set is an *entertainment* system, not a precious religious icon! Not an antique violin! Source materials available for consumption vary widely in their perfection, no matter the format. When the TV is correctly adjusted according to common-sense rules that take *vision* into account, what you see can bring great pleasure. That's the point! Garbage will still be garbage. An expensive ISF calibration has to be evaluated in terms of benefit/cost ratio for the *result.* No amount of hand-waving statements about IRE levels (Of course you mean image- *brightness* levels, right? How many people do you think understand IRE levels?), the difference between D65 and 6500K, etc. means squat if the picture does not please the person watching.

My goal has been to suggest ways to achieve the best picture possible with these late XS and XBR CRT sets (and no others!), based on solid principles and sound engineering experience, and then publish what I have done so others can try it. Folks should judge based on the results.
Quote:


it would be great if 10% of TV owners could properly set their TV.

The ones who are truly interested had to at least get to this forum, and a subset of those actually dig in and try stuff. Three cheers! Others have two choices: (1) endure what they have purchased, out of the box, having only the User Menu at their disposal; and (2) hire someone to do the work for them at whatever cost. I'm addressing only the folks in this forum; I don't sell any TV-setup services.

KenTech
"We all get smart slowly."
KenTech is offline  
post #99 of 2962 Old 06-10-2005, 11:25 PM
 
CrocHunter's Avatar
 
Join Date: Jun 2004
Location: Massachusetts
Posts: 2,570
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks to Kentech, my tv set looks 70% better out of the box after doing some service menu changes and using DVE tp calibrate my set... Thanks Kentech

I probably saved myself some money doing the calibration myself too.
CrocHunter is offline  
post #100 of 2962 Old 06-11-2005, 01:46 AM
Senior Member
 
nathan118's Avatar
 
Join Date: Aug 2004
Posts: 222
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I gotta ask how you guys play with the crosshatch screens. I start staring at this stuff and nothing looks good! Do you actually get out a level to see if you've got straight lines? do you measure each box to see if they're all the same? Start from the top and work down??

My only real problem is that 480p material looks like crap compared to 480i. I fooled with different tweaking today, but can't say I came up with anything great. I get a lot of jagged edges. Maybe my 20" samsung using s-video acts as a great anti-aliaser, but I was expecting a lot more from gamecube on component cables in progressive mode.

The same can be said for my dvd player though. 480i output by the dvd player is better than 480p (which is just jaggy city).
nathan118 is offline  
post #101 of 2962 Old 06-11-2005, 07:54 AM
Member
 
loadams's Avatar
 
Join Date: Apr 2003
Location: Mount Pleasant,SC
Posts: 158
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Katmann View Post

So these SONY codes will work as well for my 40" SONY direct view TV ?

If not do such codes exist for this unit ?

Thanks in advance.


Sure do.
loadams is offline  
post #102 of 2962 Old 06-11-2005, 11:03 AM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
One last comment. Many calibrators, including ISF, have been calibrating with the aide of optical comparators, and it works. I have a Sony PVM-96 B&W D65 reference monitor to use as such. It's a great backup especially for difficult situations like dark CRT home theaters.

We have no control over poor camera calibrations, poor mastering and post production errors as well as transmission errors, however most movies that are produced and have been produced on film use D65 as a white reference.

Yes, I try to adjust Gamma, when I can. Some TVs just don't have much to work with. Then there is the issue of trying for 2.5 in a dark theater environment or 2.2 with ambient light conditions or the TV won't get any better than 1.75.
Quote:


An expensive ISF calibration has to be evaluated in terms of benefit/cost ratio for the *result.*

Exactly, some may want to go play a round of golf or go out to dinner or something other than spending hours trying to learn all of factors needed to do their own calibration, then spend the time doing it. Some people repair their own car, others take it to a mechanic...... I choose to repair my own cars.
Quote:


No amount of hand-waving statements about IRE levels (Of course you mean image- *brightness* levels, right? How many people do you think understand IRE levels?), the difference between D65 and 6500K, etc. means squat if the picture does not please the person watching.

IRE levels are simply a percentage white level. 100IRE is full white 0IRE is black (7.5IRE is black for NTSC). 100IRE may be 50 foot lamberts on your TV and only 25 on another. Kind of like full throttle on a Porsche may get 180MPH and on a mini-van 100MPH, but both are full throttle. For whatever reason, TVs have a control that is labeled Brightness who's function is to adjust Black Level. As for various people not wanting to watch a movie with white set to or as close to the D65 reference point as possible, it is their choice. All I am saying here is if white is at or close to the D65 point, then all other colors, assuming no major decoder errors, will fairly accurately represent/reproduce the original film image. As you move away from this point, you are introducing a color shift, usually plus blue.

This forum is for those who want more information, DIY and or modify/enhance their own devices. We are all free to choose how we want to watch what we watch. The same goes for calibrations, some can't do it, some can do it and some just don't want to spend the time to do it.

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #103 of 2962 Old 06-11-2005, 11:19 AM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by nathan118 View Post

I gotta ask how you guys play with the crosshatch screens. I start staring at this stuff and nothing looks good! Do you actually get out a level to see if you've got straight lines? do you measure each box to see if they're all the same? Start from the top and work down??

The first rule of crosshatch patterns: Don't obsess! A well-adjusted TV that pleases the viewer will show geometry defects on crosshatch patterns that are *not* apparent in the usual video material. They will, indeed, confirm something you suspect already (major bowing of horizontal lines, barrel distortion left and right, etc.). But little squiggles are best ignored given the imperfect state of all consumer CRT and deflection technology. Try to get the overall pattern square to the screen's edges with the TV oriented in its normal viewing position, and you can try the various internal/service-mode adjustments to minimize any overall trapezoidal or pincushion/barrel distortions. Then go back to watching real video.

For the technically-inclined, the patterns are especially useful for examining the focus of horizontal vs. vertical lines, revealing color-convergence issues, and seeing nonlinearity or wrong aspect ratio in the overall scan. The squares should ideally be *square* most everywhere on the tube. Circles work well, too. But the crosshatch pattern will reveal if the scan lines are being selectively squeezed somewhere on the screen by a nonlinearity problem.

Another particular example: For precisely focusing the newer Sony CRTs, you first should completely defocus the beam (internal adjustment), turn off all but the green gun, and *balance* the width of vertical lines to the left and right of center with one or two service-mode adjustments. Then you move on to precise, fine focus. White crosshatch on black is perfect for this first step. An upcoming article will detail this process.

(Your other questions should probably be in another, more general, thread.)

KenTech
"We all get smart slowly."
KenTech is offline  
post #104 of 2962 Old 06-11-2005, 11:49 AM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by GlenC View Post

IRE levels are simply a percentage white level. 100IRE is full white 0IRE is black (7.5IRE is black for NTSC). 100IRE may be 50 foot lamberts on your TV and only 25 on another. Kind of like full throttle on a Porsche may get 180MPH and on a mini-van 100MPH, but both are full throttle.

I understand this, and what is most important is that IRE is a way of stating how a certain brightness level is represented in the video signal by a specific voltage. Pity that broadcasters don't pay more attention -- they with their waveform monitors and vector scopes! Why does PBS's Friday-evening SD programming require cranking down the Brightness (black level) and Color by about 8-10 points each just to match other stations' appearance?

Given that, different sets with their different CRT and dynamic-focusing technologies are capable of very different maximum brightnesses for 100IRE white. In some sets, as you crank up the brightness ("picture"), the scanning spot bloats (blooms) so much that black detail against white is swallowed up and white text loses all form. Critical viewers must then be satisfied with a less-bright picture.

On the other hand, the newer Sony sets with the fine-pitch tubes have a great combination of electron gun and dynamic focus that allows a superbly-focused bright picture, whites and colors, making for a very satisfying viewing experience. Thus one can enjoy sports on a bright afternoon or a well-produced movie in the evening without taxing the poor tube's abilities.
Quote:


As for various people not wanting to watch a movie with white set to or as close to the D65 reference point as possible, it is their choice. All I am saying here is if white is at or close to the D65 point, then all other colors, assuming no major decoder errors, will fairly accurately represent/reproduce the original film image. As you move away from this point, you are introducing a color shift, usually plus blue.

I agree completely! Most sets come adjusted for some outrageously bluish white point approaching 8000-9000K because it makes the colors look super-bright when examined superficially, say in a store. Maybe most buyers simply don't care. But I'll bet they're not in this forum looking for answers and help.

KenTech
"We all get smart slowly."
KenTech is offline  
post #105 of 2962 Old 06-11-2005, 12:15 PM
AVS Club Gold
 
GlenC's Avatar
 
Join Date: Jun 2003
Location: Los Angeles - Whittier, CA
Posts: 2,555
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by KenTech View Post

I understand this, and what is most important is that IRE is a way of stating how a certain brightness level is represented in the video signal by a specific voltage. Pity that broadcasters don't pay more attention -- they with their waveform monitors and vector scopes! Why does PBS's Friday-evening SD programming require cranking down the Brightness (black level) and Color by about 8-10 points each just to match other stations' appearance?

The main issues here: HD with a black threshold of 0IRE and SD/NTSC with a 7.5IRE black level. This issue will probably never go away, in our lifetime, as long as older NTSC programming exists.

I usually set up an additional mode for the NTSC programs. This results in my setting/calibrating 4 modes for a TV input on the Sony, HD night and day and NTSC night and day. DVD gets two modes, night and day.

Glen Carter
Home Theater Calibration
www.ISFHT.com
GlenC is offline  
post #106 of 2962 Old 06-11-2005, 03:31 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by GlenC View Post

The main issues here: HD with a black threshold of 0IRE and SD/NTSC with a 7.5IRE black level.

Out of the box, these Sonys have mostly compensated for this. I originally found far greater differences between cable channels than between HD and average SD. In any case, if that needs adjusting, one only has to go (for example) to service mode 2170P-3 #13 UBOF to equalize black level among the basic input types and their scan modes. The service-data chart shows what's possible. Also useful: 2170P-1 #2, YOF. An additional compensation for near-black color contamination solves the problem for 1080i on these sets (yellowish on mine).
Quote:


I usually set up an additional mode for the NTSC programs. This results in my setting/calibrating 4 modes for a TV input on the Sony, HD night and day and NTSC night and day. DVD gets two modes, night and day.

A bit of overkill for these Sony sets, in my opinion, and not worth the expense. Having set up the HD/memory-stick signal path with a memory stick color-calibration pattern or just the component-video path (V5-V6) with an alignment DVD, SD video of all kinds is quite adequate, as there are more errors and problems with the material than nit-picking the TV can resolve in advance. If the user's high proprity is DVD movies, then DVD alignment disks take into account their own particular DVD player, as well. The memory-stick calibration really handles the HD signal-path very well, to my delight. I regret that the HS420 and HS510 folks don't have memory-stick slots!

Night and day? Why? If the room lighting is bright, turn up Picture about 10 notches and raise black level ("Brightness") a little, until shadow detail is resolved to your liking. How can "calibration" possibly help, here? Wouldn't it be better to teach the customer how to use the user-menu settings to their advantage? I doubt whether anyone following these forums wants their TV set up so as to keep them stupid: "Just push button A for day, and push button B for night."

These settings should be done at the time of viewing, not in advance. The criteria for correct color and gamma don't change from day to night; and they, too, are determined by the program material. The purpose for calibration should be to establish a *baseline* that the user can return to when appropriate (or confused), then make setting changes to suit preferences for the viewing conditions and video material. How does that baseline change from night to day?

I have set up my set so that the midpoint of all of the user-menu sliders ("31") is the best-calibrated position for color, hue (center), and black level (especially BL for HD broadcast). 31 is also my average Picture setting for evening HD and DVD-960i/anamorphic viewing. Progressive modes, some SD broadcasts, and bright afternoons require a higher Picture setting. I have set up three of the four picture modes so that they *only* vary gamma: maximum gamma (as in Pro mode, about 2.45), gamma = 2.2, and gamma = about 2.0 (which I never really use). Gamma = 2.45 makes a lot of SD channels come alive, since the lighting is so flat much of the time. For fine productions on HD and DVD, however, 2.2 seems much more correct to me.

KenTech
"We all get smart slowly."
KenTech is offline  
post #107 of 2962 Old 06-12-2005, 01:58 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
Quote:


Gamma = 2.45 makes a lot of SD channels come alive, since the lighting is so flat much of the time.

You may already know this Ken, but some of the variation you see in the "brightness" of SD content may be due to different levels of correction (or "undercorrection") used on the source. The basic idea is that the correction used on most high quality content these days is different than it was in the past, to produce more dynamic results and insure that the picture isn't too bright overall in darker home theater like conditions. This is one reason you probably hear alot of complaints about new DVDs or some TV shows looking too dark on some TVs. This is also why a PC interface (the desktop and applications other than the video overlay) will often look rather bright when plugged into a TV-- because it doesn't have the same level of correction as video (unless you make the appropriate adjustment in display properties yourself).

[Edited to reflect the newer gamma info in Post #343.]

ADU
ADU is offline  
post #108 of 2962 Old 06-12-2005, 03:07 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
I wonder if you'd be willing to indulge me in a little experiment. I'm curious to know how gamma works on the memory stick.

Attached are two patterns. One has RGB=127 greys, and the other RGB=106 greys. If my theory above holds true and the memory stick works essentially like a computer input with no undercorrection, the RGB=127 greys will look too bright on your TV and more like a light grey or even an off-white than a "neutral grey". While the RGB=106 greys may perhaps look more neutral.

[Edit: Based on new info in Post #343, RGB=106 may not be the right color of grey to use for this test. If there is in fact a specific color of grey that should look "neutral" in standard SMPTE-170M-corrected content (which is probably open to debate), it may perhaps be something a bit higher than 106 (but still less than 127). Bottom line-- in retrospect, this approach to using a particularly color of grey as a benchmark for gamma was probably rather flawed.]
LL

ADU
ADU is offline  
post #109 of 2962 Old 06-12-2005, 03:09 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
Pattern #2
LL

ADU
ADU is offline  
post #110 of 2962 Old 06-12-2005, 07:46 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

You may already know this Ken, but some of the variances you see in the "brightness" of SD content is likely due to different levels of undercorrection used on the source.

I have no perspective on this "undercorrection" issue at all, and your comments are very interesting and welcome!.

Help me with a definition, though. I think in terms of computer-screen gamma, which relates the actual signal voltage to brightness on-screen. CRTs inherently respond nonlinearly to video voltages, and this curvature (deviation from a straight line, 1:1 relationship) has typically been a bit over 2.0. So computer monitors for the PC world have for a long time simply standardized on gamma approximating 2.2, and the sRGB color space, used widely in Windows graphics programs and digital cameras, specifies a gamma of 2.2.

(On the other hand, I exist in the Mac world, and for some reason the Macintosh display standard is gamma = 1.8. Go figure. I can twiddle all of this in the Displays driver software part of the O/S.)

But I don't understand your use of the term "net gamma." Do you mean the ratio of video material's *design* or intended gamma and the display? All through the production chain, replete with precision monitors and vectorscopes and waveform monitors, the monitor gamma = 2.2, and I would imagine producers make judgements based on what they see on those monitors. Based on the whole notion of having some "calibrated" base to use as a touchstone, I would imagine having 2.2 as an available choice is a really good idea. How does this relate to what you said?

[Later Addnedum: I just discovered your gamma-etc. thread and links. (Why haven't I seen this before?) I see (1) my understanding of gamma needs to be more sophisticated, and (2) I have some reading to do tomorrow!]

KenTech
"We all get smart slowly."
KenTech is offline  
post #111 of 2962 Old 06-12-2005, 07:53 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

I wonder if you'd be willing to indulge me in a little experiment. I'm curious to know how gamma works on the memory stick. Attached are two patterns.

With pleasure! And if you haven't already gotten the tool that lets you easily ascertain your own screen's gamma, the patterns I posted in Post #29 are the key. (Use Pro mode only -- NO vertical sharpening!)

Forgive me if you've already said this -- Am I right that you don't have a MS slot? Or just haven't sprung for a stick yet? (I'm sure these patterns can be adapted to JPEG-on-CD or stills on DVD. I just haven't done it yet.)

KenTech
"We all get smart slowly."
KenTech is offline  
post #112 of 2962 Old 06-13-2005, 01:03 AM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
My TV supports memory stick, but I don't use it because I have a computer connected via the DVI port.

"Net gamma" or "viewing gamma" is my term for the end result that you actually see on the screen after file/source, system, and monitor gamma are all combined. It's the same thing this guy describes as "total gamma" at the end of this webpage:

http://exviking.net/back/monitor/gamma_2.htm

[Edit: And the same thing this other page refers to as the "end-to-end exponent".]

He probably gives a better explanation of it than I can, and the reasons video is undercorrected at the source to insure the picture isn't too bright on your TV.

Math has never been my forte, so its taken me quite awhile to get my mind around some of this stuff, especially the difference between file/source, system, monitor and perceived gamma (and I may still have some of it wrong.) When people start talking power curves, and 1.8 vs 2.2 vs. 2.5, I start to get confused, so it's often easier for me think in terms of the end result being displayed on the screen, and how that relates to the brightness of midtones in the image.

[Edited to reflect the newer gamma info in Post #343.]

ADU
ADU is offline  
post #113 of 2962 Old 06-13-2005, 01:46 AM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
Before I sign off I'll be adventurous and take one crack at explaining undercorrection as best I understand it.

Most CRTs have an exponent (or "display gamma") of about 2.2 to 2.5. When a high contrast CRT (like a TV) is viewed in very dim surroundings though, its midtones tend to appear much brighter (due to a phenomena known as either "simultaneous contrast" or "the surround effect"), and the perceived gamma of the picture is brighter by comparison to the surroundings. So instead of using a 1/2.2 or 1/2.5 correction to perfectly compliment the CRT gamma, modern TV cameras "shortchange" the correction a little, and use only about 1/2.0 correction to keep the end viewing gamma on the screen more in the 1.1 to 1.25 range (2.2/2.0=1.1 and 2.5/2.0= 1.25). This makes the picture slightly darker which is more comfortable to your eyes in the dim surroundings typical of nighttime home viewing, and it also increases the sense of depth and contrast in the picture, giving highlights and explosions more "pop".

The 1/2.0 (or ~.5) correction is relatively new though (probably a few decades old). In the past, the correction was more in the 1/2.2 (.45) or 1/2.5 (.4) range which meant that older NTSC video was a little "brighter" looking than most video now (2.2 CRT gamma x .4 or .45 correction = ~.9 or 1.0 viewing gamma). The gamma correction is different now I guess because more people are watching hi-contrast screens in darker home theater-like settings than in the 1950's when this stuff was originally worked out.

This unfortunately also means that the picture is a bit darker and less watchable during the day, when there tends to be more ambient light. Hence the addition of "Vivid" modes to TVs and DVD players, which boost white level and midtones to compensate. Glen's daytime/nighttime calibration scheme probably works in much the same way. The daytime calibration and Vivid modes on the TV/player are designed to compensate for the undercorrection applied in the video source for better nighttime viewing. At least I think that's the way all this jazz works.

[Edited to reflect the newer gamma info in Post #343.]

ADU
ADU is offline  
post #114 of 2962 Old 06-13-2005, 08:44 AM
Member
 
loadams's Avatar
 
Join Date: Apr 2003
Location: Mount Pleasant,SC
Posts: 158
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Man, I need some Excedrin now.
loadams is offline  
post #115 of 2962 Old 06-13-2005, 09:59 AM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
Yeah, maybe I shouldn't even have brought this up. Ignorance is bliss and Gamma can be a rather hairy topic.

If the gamma patterns in Post #29 are really telling you your monitor has a gamma of 2.45, that may not be so bad for certain types of content. If you change this near 2.5 exponent of the monitor to 2.2 using the gamma controls in the service menu, you may be brightening the picture on the TV more than needed for some video material (especially older NTSC content with brighter gamma), and sacrificing some of the contrast and depth on the display.

I'm not at all sure of this though. Maybe the patterns in Post #29 aren't telling you the monitor's gamma, but rather the "system" gamma of the memory stick (eek!).

Another thing to consider is whether you really want precisely the industry standard gamma on these TVs, or not. It's possible that you might want the midtones dialed up a bit (as you've already done) to compensate for some of the weaker points in the tube's design, especially black "purity" issues like phosphor lag, poor grey scale calibration, and internal reflections which tend to drown out and muddy up the "shadow detail" on these TVs.

As mentioned earlier, brightening the midtones/gamma is one way of compensating for a dimmer screen in a bright surround. And its possible you might want a little brighter than usual surround with these Sonys to hide the effects of phosphor lag on the screen. So that may be where the gamma adjustment is helping as well-- to compensate for slightly brighter than usual surrounding illumination for a home theater setting.

I guess this is a long way of saying that you may have made the right adjustment, but possibly for the wrong reasons(?) By boosting the midtones on the Pro mode though, you're kind of doing the exact same thing the Sony engineers already did in their Standard and/or Vivid picture mode on the TV. So there may be a bit of reinventing the wheel going on with these particular adjustments. Sony is obviously aware of all the issues above, and that may be precisely why they provided various "enhancements" to gamma/etc. that they did in their Standard and Vivid modes*.

Brightening the midtones does mean sacrificing some contrast/depth on the display though.

(*Footnote: Perhaps that may also be why they're biasing their gamma curve a little with GAMS, to pull more detail up out of the darker areas of the picture... if GAMS is indeed a bias control.)

[Edited to reflect the newer gamma info in Post #343.]

ADU
ADU is offline  
post #116 of 2962 Old 06-13-2005, 11:29 AM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

Attached are two patterns. One has RGB=127 greys, and the other RGB=106 greys. If my theory above holds true and the memory stick works essentially like a computer input with no undercorrection, the RGB=127 greys will look too bright on your TV and more like an off-white than a neutral grey. While the RGB=106 greys will look more neutral.

I just converted these to JPGs and viewed them on my TV at gamma = 2.2 as I have calibrated it. Also checked them out on my computer screen set to gamma = 2.2.

They look the same on computer and TV. The gray 106 (where 0 = black and 255 = white) is darker than the gray I associate with "50%" gray, and 127 looks about what I think is mid-gray. Neither is even close to white or off-white. If that were true, I would suspect "very" inappropriately low gamma or brightness-limiting in the video circuit!

I'm sorry some folks may be put off by this technical spinoff on gamma, but knowledge is power, here -- meaning one can take control of the TV and make what you choose to watch more attractively displayed. I think the time has arrived for a gamma control, and I regret having to take over two or three of the picture modes to do it. But this *does* do it. Having this control means, say, when sitting down to enjoy an episode of "Lost" on ABC:

(1) Reset black ("Brightness") to 31, then tweak it until there are no "black holes" and I can see shadow detail. This varies from one HD station to another, but not by much.
(2) Set "Picture" so blown-out whites and beach scenes aren't blindingly bright to my "evening" eyes.
(3) Choose the gamma the makes everything look okay to me. That's generally my current "middle" position of 2.2, which I have assigned to "Movie."

"Lost" is generally balanced sort of dark, and these settings work well. Works for CSI/Miami, too. But not for a couple of documentaries I have seen on PBS-HD, where the maxinum gamma (2.45) made the picture appear most realistic.

I *really* believe in the concept of a great baseline calibration, and then one should have the controls available to "tune" the display for the particular program or DVD you're watching. Your eyes don't lie, and they're *all* you have to please! Sharp focus, correct color decoding (no color push), available image enhancement if appropriate -- all are part of this baseline.

KenTech
"We all get smart slowly."
KenTech is offline  
post #117 of 2962 Old 06-13-2005, 11:55 AM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU
As I think about it, this statement I made eariler may be slightly off the mark:If the gamma patterns in Post #29 are really telling you your monitor has a power curve of 2.45, that may be close to what you want for standard video with an undercorrection of 1.25.
I get what you mean by "net gamma" and "undercorrection" now. If the "1.25" hypothesis is true, it is certainly a current issue, since gamma = 2.45 really improves almost all SD video of whatever source. So I use it when it looks better, which is the point of having it as a choice.
Quote:
I'm not at all sure of this though. Maybe the patterns in Post #29 aren't telling you the monitor's gamma, but rather the "system" gamma of the memory stick (eek!).
Well, yes: we have to treat the TV as a "black box" having inputs (the video connectors and tuners) and an output (the screen). The published technical block diagram of the set (attached) strongly implies that the first part of this signal-handling chain from input to screen is *common* for memory stick and the HD tuner. Rather late in the chain, the common CXA2170 signal-processing chip handles the gamma parameters that are adjustable in service mode, and it does this *just* before passing the video to the CRT video-interface board. I am hard-put to believe that the engineers built in "special" gamma processing for different inputs *early* in the chain (your worry, right?), as it seems to make no sense, especially economically; and provision has been made to store a unique gamma-parameter set for each of 17 different input/scan/picture-mode combinations. Implication: one gamma-processing step that we can adjust in code group 2170P-4.

I can't simulate an SD cable signal, and it's a huge pain to build test patters for perfect playing on a DVD player. But my viewing experience confirms that there are no surprises when switching between an excellent DVD and good HD broadcasts. So I am trusting what the test patterns show and the MS slot for calibration until I have substantial evidence to the contrary. (In fact, I currently find that the color-matrix calibration through the MS slot pleases me more than the one on the DVE DVD. Go figure.)

 

XS955|XBR960 Block Diagram.pdf 139.0234375k . file

KenTech
"We all get smart slowly."
KenTech is offline  
post #118 of 2962 Old 06-13-2005, 12:12 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

You may actually want the midtones dialed up a bit (as you've already done) to compensate for some of the weaker points in the tube's design, especially black "purity" issues like phosphor lag, poor grey scale calibration, and internal reflections which tend to drown out and muddy up the "shadow detail" on these TVs. If I had to take a guess I'd say that the reason your slight gamma adjustment in the SM looks good on some content is because it compensates somewhat for these other problems with the blacks on the Sonys, rather than "correcting the gamma on your monitor to conform with the '2.2 standard'. . . . boosting the midtones is also a way of compensating for a dimmer screen in a bright surround. And in my experience you really need a brighter than usual surround with these Sonys to hide the effects of phosphor lag on the screen. So that may be where the gamma adjustment is helping you as well-- to compensate for slightly brighter than usual surrounding illumination for a home theater setting.

This is very likely the case. I have a critical eye, and the final result is all that matters to me. It may well be that the "2.2" standard is of no real consequence and I am really just tweaking gamma for the viewing conditions. At least the gamma measurement lets me know where the TV stands relative to production monitors and my computer screen (Mitsubishi Diamontron 21" CRT).
Quote:


I guess this is a long way of saying that you may have made the right adjustment, but for the wrong reasons. By boosting the midtones on the Pro mode though, you're kind of doing the exact same thing the Sony engineers already did in their Standard picture mode on the TV.

Oh, I agree! Except Sony "polluted" those other modes with so many other "enhancements" (e.g. dynamic picture contrast, obnoxious and coarse VM). I guess I am trying to establish what effects I want and what I don't. Gamma control seems essential all by itself.
Quote:


Footnote: That may also be why they're biasing their gamma curve a little with GAMS, to pull more detail up out of the darker areas of the picture... if GAMS is indeed a bias control.

Haven't quite figured out exactly what GAMS does, but it seems to barely affect black level and raise the top brightness; and so it appears to add a linear slope to the gamma curve, doing what a little boost of the Picture slider would also do. For now I am leaving 2170P-4/GAMS at 0 for all combinations.

KenTech
"We all get smart slowly."
KenTech is offline  
post #119 of 2962 Old 06-13-2005, 12:13 PM
ADU
AVS Special Member
 
ADU's Avatar
 
Join Date: Apr 2003
Posts: 6,138
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 128
Quote:


I just converted these to JPGs and viewed them on my TV at gamma = 2.2 as I have calibrated it. Also checked them out on my computer screen set to gamma = 2.2.

They look the same on computer and TV. The gray 106 (where 0 = black and 255 = white) is darker than the gray I associate with "50%" gray, and 127 looks about what I think is mid-gray. Neither is even close to white or off-white.

Hmm... if that's really the case, then perhaps the memory stick may have some built-in correction, to conform the computer-created JPEGs more closely to corrected video. [Edit: Or not... still not quite sure about this.]

The DVI port doesn't seem to work quite like that, probably because it's designed to input high-def video which has already been pre-corrected before it arrives, rather than inputting directly from a computer. This is supported by the fact that DVDs in a computer video overlay look "correct" via DVI. Whereas the computer interface itself and other apps look much too bright on the TV (without some adjustment to the desktop gamma).

[Edited to reflect the newer gamma info in Post #343.]

ADU
ADU is offline  
post #120 of 2962 Old 06-13-2005, 12:56 PM - Thread Starter
Advanced Member
 
KenTech's Avatar
 
Join Date: Feb 2004
Location: Portland, OR
Posts: 714
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Quote:
Originally Posted by ADU View Post

If you want some more links to explore on the subject though, there are some here:

http://www.avsforum.com/avs-vb/showt...&&#post3555581

Thank you! Highly recommended reading for all (until your head aches, of course). I am happy to be prompted to re-read Poynton, whose articles I read a couple of years ago when I was trying to wrap my mind around the issues of digital-image manipulation and monitor calibration, important to my consulting business. Some of these links seem to be broken or deeply archived, and more digging is necessary.

Please note row #7 in the table here, regarding "gamma correction":
http://www.poynton.com/notes/color/GammaFQA.html

Maybe the "linear segment" referred to here is the purpose of 2170P-4/GAMS! Available if we want it. Very informative.

KenTech
"We all get smart slowly."
KenTech is offline  
Reply Direct View (single tube) CRT Displays

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off