AVS HD 709 - Blu-ray & MP4 Calibration - Page 59

The APL window patterns were mainly created to measure the auto iris setting on my SXRD. The idea is that the average picture level doesn't change on the APL window patterns, so the iris (or any other dynamic setting) will not adjust during measurements. With a typical window or field the average picture level changes with each pattern, and on a dynamic iris that will cause the light output to change between different patterns. The problem with the iris adjustment is for example the measured gamma is nonsense and the Y measurements for color are not relative to each other. So the main point of the APL window patterns was to stop any dynamic display from adjusting so that relative Y measurements can be taken - primarily for gamma and color.

I've read about floating black levels on some plasmas (Panasonic?), and in such a case the APL windows would stop that sort of behavior for measurements. I haven't looked at how much blacks might change on some plasmas, but odds are that floating blacks will not impact measurements near as much as an adjusting projector iris or a dynamic LCD might. The only other possibly plasma-related use of the APL windows would be if power limiting affects the relative Y measurements for color. Again I haven't looked into this at all either, so I don't know if relative Y for color would vary much between typical color windows and APL color windows, but I'd expect the APL color window measurements to be directly relative to each other and generally in line with what you'll get from a movie scene.


The black clipping pattern has an extremely low average picture level, and the white clipping pattern is extremely high. The APL clipping pattern has an average picture level that is going to be closer to say a typical movie scene. The pattern is mainly included to offer an alternate pattern for setting white and black levels, sort of like how DVE has more than one pluge pattern.

Thanks for the reply. I'll update the list to reflect PATCHED working and AVCHD not working. Most likely if HDMV plays from DVD media then PATCHED should work.

If you're using a puck or a meter with a small measurement area, a step further would be to compare if the large and small sets of APL windows vary much in regards to relative Y for color measurements. Ideally the large and small APL windows would be within repeatability, but whether or not that would commonly be the case I couldn't say.


With a typical gamma measurement there are two things changing - the video level (% shown) and the average picture level (APL) of the displayed image. If a display varies brightness depending on APL, which some do, a typical window measurement ends up at least partially relating how the display changes between very dark and very bright video scenes. Generally in most gamma discussions it's never mentioned that some displays may vary light output depending on the APL of the displayed image, and typically the focus of such discussion is more on how you're measuring different video levels.

The APL window patterns eliminate the adjusting APL, so the only variable is the change in the center video level. Effectively the APL windows will measure how the display increases in brightness between black and white for a given image. I might call this "on-screen" gamma, because the APL window measurement is generally in line with what you'll see looking at an individual image. It is possible that a display may vary depending on APL, so the large and small APL windows may give slightly different on-screen gamma measurements.

For the HD standards percentage and IRE match, but I'll take it that you are talking about the IRE labels used with Avia. The Avia IRE labels do not correspond with the percent gray patterns here. For example the 7.5 IRE label is the same as 0%, and the Avia 10 IRE is not the same as 10%. Because the patterns don't match, you have to account for that with your measurement software. For example with ColorHCFR you have to select the "Measures" drop down menu, choose "Parameters", and then check "Grayscale in NTSC IRE Levels" to measure the Avia IRE labeled patterns. When measuring the AVS HD 709 patterns, you do not want "Grayscale in NTSC IRE Levels" checked in ColorHCFR. While percentage labeled patterns do not directly correspond with the Avia IRE labels, if your measurement software accounts for the difference the resulting graphs would generally be expected to look similar.

The encoding side was mainly a contribution from hwjohn, but I believe the flags are basically intended to describe the video. The flag change was primarily done to bring the video in line with the original authoring intent. It was more to avoid any possible problems, rather than to necessarily correct any specific known issues. Personally my players measure the same with either video type, and I'm not aware of any examples that measure differently between flagged and unflagged video. The Misc Patterns area of v1.3 contains sections with flagged and unflagged video for comparison if you want to test your electronics. There was a suggestion in this thread that it's possible that not all commercial Blu-rays are flagged, but I personally have not tried to look for flagged and unflagged Blu-rays considering my electronics operate the same with either.

Five playlists were changed on the v1.3b HDMV and PATCHED. Without going through the navigation I'm not sure which playlist goes with which title, but I think all the sections now have a chapter right before the out time to loop.

Thanks for the player feedback.

Avia might have one of the spot gamma patterns, but I personally find those hard to interpret and of limited usefulness. If you just want a general idea of what gamma does, the Grayscale Bars or Ramp in the Misc Patterns can be used to see how gamma affects the image. For example a higher gamma will reduce the brightness increase from one bar to the next near black, and a lower gamma will generally make the image come out of black quicker. The Bars pattern is probably the easiest way to quickly observe how brightness increases from one bar to the next between black and white, or the Dynamic and Center pattern can show if your display tends to vary gamma depending on APL. At the very least you should always be able make out all the bars between black and white, including the bar brighter than black and the one darker than white. Grayscale is meant to be approximately perceptually uniform, so generally the step from one bar to the next between black and white should seem about equal. If say the steps near black were hard to tell apart while the ones near white were very clear that might be an example of a situation to lower gamma. Anyway if you need more specific information about gamma there are a couple things you could do:

1 - The easiest way to get an idea what gamma is doing is to take measurements. If you hire a calibrator they can do this for you, or else you'll need some way to measure the display. The X-rite Display LT is an example of a device that works with the free ColorHCFR software and can be used to give an idea what gamma is doing. The links area of the first post will take you to more information about taking measurements.

2 - If you don't have access to a way to measure gamma, the next best thing to do would probably be to see if other owners of your display model have measured their displays. Sometimes display reviews will also list how gamma measured. When I measured my display it generally ended up within the same range as what others had measured on theirs using similar settings, but it's possible different display types might show more variation than my model seems to have.

Strictly speaking this isn't necessarily what is going on. You might be outputting the full range and the display may simply not be able to display it. Like you said, "using the same pattern he is able to get the full range out of the player" so it's not necessarily a given that the player isn't outputting the full range. To test if the source is outputting the full range you could hook the player up to another display that you know is able to display the entire range.

Personally I wouldn't touch the brightness control on the player, especially if I didn't test another display. I've read that some displays will clip the below-black grayscale, but for a video source it doesn't really matter because you're not displaying that range anyway. The only place where being unable to display 1-7 would necessarily be an issue would be if the source was something like a computer where the range was intended to be displayed. Without any more information to go on, I would leave the player controls at their defaults and just adjust brightness at the display so 16 and lower is all the same shade.

Assuming a player will work with any version, go with the HDMV. If there was just one version offered it would be the HDMV, but the other versions are necessary to address practical concerns with different players. For example many people don't have a way to burn to BD-RE or BD-R, but that might not matter with computer playback if you're using a virtual drive.

The quick answer is that I would just use a video pattern. There is a similar reply at http://www.avsforum.com/avs-vb/showt...8#post17115888