AVS › AVS Forum › Display Devices › Display Calibration › AVS HD 709 - Blu-ray & MP4 Calibration
New Posts  All Forums:Forum Nav:

AVS HD 709 - Blu-ray & MP4 Calibration - Page 113

post #3361 of 3625
7/1/12 at 7:22pm
Quote:
Originally Posted by alluringreality View Post

The latest version was completely reauthored. During the reauthor there was an error found in with the v1.3 AVCHD authoring software, which may explain a report of measurements not matching exactly on a Panasonic player using the v1.3 AVCHD. If you are trying to use the CalMAN automation you would have to use whatever they support, but generally the recommendation is to use the last version. More release notes are at:
http://www.avsforum.com/avs-vb/showthread.php?p=19413135#post19413135



So, the latest version is from Nov 2010, right?
Reply
Reply

AVS Top Picks

post #3362 of 3625
7/5/12 at 3:14pm
Ok, i downloaded AVCHD and burned with DeepBurner, burn iso.

I insert the dvd in my Blueray drive BD-D5100 and he dont work. Not recognize my dvd ...wtf.
When i burn video-dvd with this dvd,it's #1!

Same **** with HDMV.

My tv is LG 47LK520 , so plz, i live in montreal-quebec-canada.

I need this disc but why not functionnal with me?
Edited by jo1985 - 7/5/12 at 4:14pm
Reply
Reply
post #3363 of 3625
7/5/12 at 4:00pm
Quote:
Originally Posted by kjgarrison View Post

So, the latest version is from Nov 2010, right?

yes, though unofficial releases might be lost somewhere in this thread (for when special requests are made for minor tweaks/mods)
Reply
Reply
post #3364 of 3625
7/19/12 at 4:10pm
Thread Starter 
Quote:
Originally Posted by jo1985 View Post

burned with DeepBurner

Based on other replies, this may be the issue, and my recommendation is to use Imgburn or Nero.
Reply
Reply
post #3365 of 3625
7/19/12 at 4:26pm
Thread Starter 
Quote:
Originally Posted by ontherocks View Post

Is there a mp4 or mkv with H.264 encoding of the HDMV version? I need this since the mp4 version in the first post doesn't contain all the patterns.

I think the patterns that are not included on the mp4 version are all mpeg2 from dr1394. I didn't want to transcode or re-author his video, so I just included his mpeg2 video on the HDMV. Anything that is not on listed on w6rz.net will be available in the links he posted at:
http://www.avsforum.com/t/570937/hd-mpeg-2-test-patterns
Reply
Reply
post #3366 of 3625
7/23/12 at 6:02pm
Hi,
I recently found out about the AVS 709 Calibration charts and had great success using them for basic brightness contrast color and tint settings on a samsung LED TV EH6050

My question is regarding the additional color flashing charts B1-Flashing primary color and B2-Flashing Color Decoder.

If after adjusting global color and tint using 4-Flashing color bars, I find some discrepancy in the individual RGB with B1-Flashing primary color and B2-Flashing Color Decoder, can I use Individual RGB Gain and Offset setting of the TV to get proper levels on B1 and/or B2 flashing charts, or are the gain and offset settings only to be used with a color measuring tool?

Thank you for any advice
Reply
Reply
post #3367 of 3625
7/23/12 at 6:56pm
Yep, you need a meter to adjust those, they are after the decoder.
Reply
Reply
post #3368 of 3625
7/24/12 at 11:29am
Quote:
Originally Posted by zoyd View Post

Yep, you need a meter to adjust those, they are after the decoder.
Thanks !
I was afraid of that.
I was hoping there was some flashing chart of sorts allowing to do some individual RGB ajustments with gain and offset.
Reply
Reply
post #3369 of 3625
7/24/12 at 12:47pm
Quote:
Originally Posted by scrambler3 View Post

Thanks !
I was afraid of that.
I was hoping there was some flashing chart of sorts allowing to do some individual RGB ajustments with gain and offset.

unfortunately, it doesn't work that way... in order to be able to do any grayscale, gamma or CMS adjustment, you need a reliable meter (especially for CMS)
Reply
Reply
post #3370 of 3625
7/31/12 at 9:42pm
I've been trying out CalMan with USB-UIRT + AVS HD 709 and I've ran into two issues:

1. I can't seem to find the version that lines up with the USB-UIRT commands. It goes off into the weeds (eventually going into the brightness/contrast slides) after measuring the first 100% white value. This was even with the version I got linked from SpectraCal.

2. Even if I'm able to find an older version that matches CalMan, there is no way to tell CalMan to use the SMALL APL patterns, which I find work much better when calibrating plasma. Instead, it uses the standard ColorHCFR windows, which are not only too big, but are non-APL.

Is there any way we could have a version made that ONLY uses the SMALL APL greyscale and 75% color slides, while preserving the same menu structure as the older CalMan-compatible AVS HD 709 disc? In other words, over-write each individual slide data with only its SMALL-APL counterpart?
Reply
Reply
post #3371 of 3625
8/1/12 at 5:39am
Thread Starter 
Quote:
Originally Posted by Tempest261 View Post

1. I can't seem to find the version that lines up with the USB-UIRT commands. It goes off into the weeds (eventually going into the brightness/contrast slides) after measuring the first 100% white value. This was even with the version I got linked from SpectraCal.

If you have the version selected, the rest is up to the CalMAN software.
Quote:
Is there any way we could have a version made that ONLY uses the SMALL APL greyscale and 75% color slides, while preserving the same menu structure as the older CalMan-compatible AVS HD 709 disc? In other words, over-write each individual slide data with only its SMALL-APL counterpart?

It's technically possible, but I have no plans to do any further disc authoring.
Reply
Reply
post #3372 of 3625
8/5/12 at 11:09am
I am trying to encode some ramps but when converting from tga to yuv it create some artifact on the picture.
I is there some limitations in rgbtoyuv.exe?

attached is the tga and the tga file converted back from the yuv.
ramps.zip 1133k .zip file
Cyan-White-ramp.tga is the original
Cyan-White-ramp.tga_res is the restored one

any comments on what I could do, I have tried 2 computers both same result
Reply
Reply
post #3373 of 3625
8/5/12 at 1:55pm
has anyone tried to use imagesource as opoused to rawsource?

I am using imagesource now (just for the ramps) but not sure if its 100% acurate.
I will verify and report back - it semas that for now it have solved my problem from before. however for test patterns (all with the execption for the ramps) I will still use rawsource.
Reply
Reply
post #3374 of 3625
8/5/12 at 3:10pm
Quote:
Originally Posted by visca blaugrana View Post

I am trying to encode some ramps but when converting from tga to yuv it create some artifact on the picture.
I is there some limitations in rgbtoyuv.exe?
attached is the tga and the tga file converted back from the yuv.
ramps.zip 1133k .zip file
Cyan-White-ramp.tga is the original
Cyan-White-ramp.tga_res is the restored one
any comments on what I could do, I have tried 2 computers both same result

You've created a pattern that overflows 8 bits in YCbCr space. rgbtoyuv.exe just truncates the 8-bit values, so you get wildly different colors. This is a feature and not a bug. wink.gif
Reply
Reply
post #3375 of 3625
8/5/12 at 3:17pm
Quote:
Originally Posted by dr1394 View Post

You've created a pattern that overflows 8 bits in YCbCr space. rgbtoyuv.exe just truncates the 8-bit values, so you get wildly different colors. This is a feature and not a bug. wink.gif

thank you for explaining what the problem is.

since the pattern is created in photoshop as a 8 bit pattern not sure why it overflows on the 8 bit. any sugestion on what could cause this to happen?

Cheers
Reply
Reply
post #3376 of 3625
8/5/12 at 5:23pm
The conversion from RGB to YCbCr is overflowing. Here are the last few values for a red only ramp (showing the input, floating point values and final 8-bit output).
Code:
R = 240, G = 0, B = 0
Y = 51.524000, Cb = 100.372000, Cr = 251.236000
Y = 51, Cb = 100, Cr = 251

R = 241, G = 0, B = 0
Y = 51.736600, Cb = 100.254800, Cr = 251.747400
Y = 51, Cb = 100, Cr = 251

R = 242, G = 0, B = 0
Y = 51.949200, Cb = 100.137600, Cr = 252.258800
Y = 51, Cb = 100, Cr = 252

R = 243, G = 0, B = 0
Y = 52.161800, Cb = 100.020400, Cr = 252.770200
Y = 52, Cb = 100, Cr = 252

R = 244, G = 0, B = 0
Y = 52.374400, Cb = 99.903200, Cr = 253.281600
Y = 52, Cb = 99, Cr = 253

R = 245, G = 0, B = 0
Y = 52.587000, Cb = 99.786000, Cr = 253.793000
Y = 52, Cb = 99, Cr = 253

R = 246, G = 0, B = 0
Y = 52.799600, Cb = 99.668800, Cr = 254.304400
Y = 52, Cb = 99, Cr = 254

R = 247, G = 0, B = 0
Y = 53.012200, Cb = 99.551600, Cr = 254.815800
Y = 53, Cb = 99, Cr = 254

R = 248, G = 0, B = 0
Y = 53.224800, Cb = 99.434400, Cr = 255.327200
Y = 53, Cb = 99, Cr = 255

R = 249, G = 0, B = 0
Y = 53.437400, Cb = 99.317200, Cr = 255.838600
Y = 53, Cb = 99, Cr = 255

R = 250, G = 0, B = 0
Y = 53.650000, Cb = 99.200000, Cr = 256.350000
Y = 53, Cb = 99, Cr = 0

R = 251, G = 0, B = 0
Y = 53.862600, Cb = 99.082800, Cr = 256.861400
Y = 53, Cb = 99, Cr = 0

R = 252, G = 0, B = 0
Y = 54.075200, Cb = 98.965600, Cr = 257.372800
Y = 54, Cb = 98, Cr = 1

R = 253, G = 0, B = 0
Y = 54.287800, Cb = 98.848400, Cr = 257.884200
Y = 54, Cb = 98, Cr = 1

R = 254, G = 0, B = 0
Y = 54.500400, Cb = 98.731200, Cr = 258.395600
Y = 54, Cb = 98, Cr = 2

R = 255, G = 0, B = 0
Y = 54.713000, Cb = 98.614000, Cr = 258.907000
Y = 54, Cb = 98, Cr = 2

The conversion from RGB to YCbCr was never meant to handle normalized RGB values less than 0 or more than 1. In video level RGB, you're using normalized values from below 0 (anything below 16) and above 1 (anything over 235).

For video level RGB, it's impossible to represent the entire 0 to 255 ramp (for most colors) in YCbCr.

Ron
Reply
Reply
post #3377 of 3625
8/6/12 at 1:30am
Ahh okay.. Thank you. Does that mean that my clip pattern (containing information in the range 236-255 and 1-15 is not usable?
Reply
Reply
post #3378 of 3625
8/6/12 at 4:12am
Quote:
Originally Posted by visca blaugrana View Post

Ahh okay.. Thank you. Does that mean that my clip pattern (containing information in the range 236-255 and 1-15 is not usable?

You're going to have to make a concession. You can either:

1) Limit the RGB values in your original .tga image to ones that don't overflow. This could be a little tedious since it's a little different for each color and you have to convert to YCbCr and back to RGB to look for artifacts (like you did with the Cyan-White-ramp.tga pattern). Some colors (like green) can go all the way to 255.

2) I can modify rgbtouyvy.exe for you to clip the 8-bit YCbCr values automatically. You won't see any artifacts, but you won't end up with exactly what you started with.

I believe there are some RGB clipping patterns on Spears and Munsil. I wonder what they did to get around this issue.

Ron
Reply
Reply
post #3379 of 3625
8/6/12 at 6:58am
Spears and Munsil:

Build each pattern in the color space and format that made the most sense. For example, if the pattern is for evaluating YCbCr issues, we built the pattern in YCbCr space directly rather than creating it in RGB and converting. This necessitated building our own pattern generation code, as no existing image editing software works in YCbCr natively. Similarly, most of the patterns were generated directly in 4:2:0, which again can’t be created natively by image editors.

• Use the very best algorithms and the highest precision. In most cases, the patterns were generated using 64-bit floating point values for each channel of each pixel, and then converted to 8-bit using special dithering algorithms to preserve dynamic range. Where we needed to anti-alias lines and curves (for example in the sharpness pattern) we oversampled each pixel (again, with custom-written software) by 256x or more to get the smoothest possible edges and contours.

• Build patterns with digital in mind from the beginning. For a long time, all the standard video patterns have been designed for the limitations of analog broadcast. Since our patterns were built using an all-digital signal path and will never be broadcast, we don’t need to conform to FCC bandwidth limitations or roll off high frequencies to avoid causing problems for old CRT displays. All of our patterns are built to use the absolute highest bandwidth available to modern all-digital equipment.
Reply
Reply
post #3380 of 3625
8/6/12 at 9:00am
This may have been asked, but it's a huge post to look through...

Anyway, is it possible to get full rez (1920x1080) TGA files of the test patterns? I suppose I could do screen captures from the videos, but I just figured it would be nice to have the raws of these. I actually have a playback source that plays TGAs and this would be a great thing to store on there to set up displays.
Reply
Reply
post #3381 of 3625
8/6/12 at 1:33pm
Ron,


thank you for the reply. If its not alot of work I would like a modified copy of rgbtoyuv, would be interesting to see how it compare.

if there is any artifact, would it show up clearly as in the example above or could you imagine that the artifact would be so minor you would not notice it?
Reply
Reply
post #3382 of 3625
8/6/12 at 4:37pm
Quote:
Originally Posted by visca blaugrana View Post

Ron,
thank you for the reply. If its not alot of work I would like a modified copy of rgbtoyuv, would be interesting to see how it compare.
if there is any artifact, would it show up clearly as in the example above or could you imagine that the artifact would be so minor you would not notice it?

No problem. The clipping function was already in the code, I just chose not to use it so that folks would be aware of overflows.

http://www.w6rz.net/rgbtouyvyclip.zip

Ron
Reply
Reply
post #3383 of 3625
8/6/12 at 5:16pm
Just wanted to say thanks for the disc!!!! I was using contrast and brightness patterns on the video processors and other disc, but when I used the AVSHD disc the picture quality on my calibrations are sooooooo much better in terms of wow, pop and picture detail. I could figure out I was never very impressed with my plasma until now! Thank again!! I wish I could put this patterns in my signal generator and processor!!!!!!!!!
Reply
Reply
post #3384 of 3625
8/6/12 at 10:19pm
Quote:
Originally Posted by avsform1 View Post

Just wanted to say thanks for the disc!!!! I was using contrast and brightness patterns on the video processors and other disc, but when I used the AVSHD disc the picture quality on my calibrations are sooooooo much better in terms of wow, pop and picture detail. I could figure out I was never very impressed with my plasma until now! Thank again!! I wish I could put this patterns in my signal generator and processor!!!!!!!!!

Did you use any meter to do calibration or just AVSHD disc?
Reply
Reply
post #3385 of 3625
8/7/12 at 10:46pm
I too want to thank you for the AVSHD disk!! WONDERFUL utility! The included video was helpful as well. I did not have a blue filter so it took me hours and hours of tweaking to get it perfect but I was not able to get it half this good before I downloaded your disk. One day I will bite the bullet and spend the $3 for the blue filter, but I'm so happy with the way my new Sharp 70" 8470U looks now that I'm pretty much done.
Reply
Reply
post #3386 of 3625
8/8/12 at 11:28am
Gamma 2.2 - Spectracal Calman always says the 100IRE is 2.2, no matter what. ChromaPure doesn't show 100IRE Gamma - how can you check that 100IRE is actually 2.2?
(Ran into a problem where 90IRE X=0.313 and Y=0.329, but the Gamma 2.2 was way too high, so it makes me think, the 100IRE isn't correct either.)
Seems to be too big a difference in finished calibrated settings between 90 and 100IRE. So why the lack of 100IRE Gamma correct values?
Have the AHCD 709 DVD for the Colour, Tint, Brightness and Contrast, but this 100IRE Gamma is bugging me on how to calibrate it to the correct 2.2 value.
Reply
Reply
post #3387 of 3625
8/8/12 at 11:54am
Quote:
Originally Posted by p5browne View Post

Gamma 2.2 - Spectracal Calman always says the 100IRE is 2.2, no matter what. ChromaPure doesn't show 100IRE Gamma - how can you check that 100IRE is actually 2.2?
(Ran into a problem where 90IRE X=0.313 and Y=0.329, but the Gamma 2.2 was way too high, so it makes me think, the 100IRE isn't correct either.)
Seems to be too big a difference in finished calibrated settings between 90 and 100IRE. So why the lack of 100IRE Gamma correct values?
Have the AHCD 709 DVD for the Colour, Tint, Brightness and Contrast, but this 100IRE Gamma is bugging me on how to calibrate it to the correct 2.2 value.

The 100% point is always the target, it really doesn't have gamma which is why chromapure leaves it off their charts entirely.

So if the target for 100% is the measured value for 100% then you have a normalized value of 1.

1^2.2 = 1, 1^2.5 = 1, 1^1.8=1

regardless of the gamut exponent 1's target value is always 1, which is what it always calculates too.

100% sets everyone else's targets, and itself doesn't have a particular target.
Reply
Reply
post #3388 of 3625
8/8/12 at 7:58pm
OK, the calibrated 100IRE, as long as it meets it's x and y values, then it's Gamma is always correct?
But why the discrepancies between the 100, and 90?
ie: First set 100IRE is 3, 2, 1 : 90IRE is 16, 12, -1 Resulting in the Gamma of 2.2
Second set 100IRE is-3, -3, 1 : 90IRE is 10, 4, 1 Resulting in the Gamma 0f 2.2
These seem to me to be Big differences in RGB values? I would have thought they'd be closer like the rest of the IREs down to 10IRE?
Reply
Reply
post #3389 of 3625
8/8/12 at 9:55pm
Gamma is the relation between power going into your tv and ligt coming out of it. 0% input is always 0% output, and 100% input is always 100% output. But 30% in is not 30% output, and 60% input is not 60% output. And that curve of difference between input and output is "gamma", and the way you try and get that curve right is by getting your gamma to 2.2 from 0% to 100% in 10% or 5% steps.
So 100% does not need to be measured because it is always the same, 100% = always 100%, doesn't matter if it is 50 lft or 1000 lft, 100% is 100%.

At least, this is how I understood it.
Edited by Wouter73 - 8/9/12 at 12:20am
Reply
Reply
post #3390 of 3625
8/8/12 at 10:33pm
Quote:
Originally Posted by p5browne View Post

OK, the calibrated 100IRE, as long as it meets it's x and y values, then it's Gamma is always correct?
But why the discrepancies between the 100, and 90?
ie: First set 100IRE is 3, 2, 1 : 90IRE is 16, 12, -1 Resulting in the Gamma of 2.2
Second set 100IRE is-3, -3, 1 : 90IRE is 10, 4, 1 Resulting in the Gamma 0f 2.2
These seem to me to be Big differences in RGB values? I would have thought they'd be closer like the rest of the IREs down to 10IRE?

Obviously not.

It could be that you have your contrast a bit too high.

The display basically has a native gamma curve and you are moving the points around on it so that 90% outputs .9^2.2 of 100% output. so you're sliding the 100% around on the curve and then the 90% point needs to follow it in output, the input corrections may change.

Another thing to note is that gamma isn't really the best measurement of how far you're off for 90 or 95%, very small changes can make big swings in gamma values at those levels.
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration

AVS Top Picks

AVS › AVS Forum › Display Devices › Display Calibration › AVS HD 709 - Blu-ray & MP4 Calibration