I have a 50" 800u Panasonic plasma and use it about 90% of the time for gaming and love how it looks in every regard with it set to thx. My only problem is alot of times with games, it looks a bit muddy.

I would really appreciate if someone could explain where the gamma adjust is in the service menu. I know how to access the SM and needless to say, understand the dangers of it, which is why I'm asking instead of going in blindly. I've already adjusted the red to make it a bit more red and less pink but just cannot find the gamma. Thanks!

bump, please anyone. this will make this tv near perfect

There is no gamma adjustment for these in the SM. You just have your choice of THX mode, which has good gamma, and the other modes, which do not.

There is no gamma adjustment for these in the SM. You just have your choice of THX mode, which has good gamma, and the other modes, which do not.

The gamma only seems "good" if you look at a single gamma number for the entire luminance range. If you look at the gamma for each step on the grayscale, you get low gamma at lower luminance levels and high gamma at high luminance levels. That contributes to the flat dull look of the THX mode compared to panels with a reasonable gamma for each grayscale step. Some, like the original poster, might say the THX mode is "muddy" because the darker steps are too light... the lighter steps are too dark which contributes to the dull look. Adding luminance to try to overcome the dull look doesn't make it go away either.

The gamma only seems "good" if you look at a single gamma number for the entire luminance range. If you look at the gamma for each step on the grayscale, you get low gamma at lower luminance levels and high gamma at high luminance levels. That contributes to the flat dull look of the THX mode compared to panels with a reasonable gamma for each grayscale step. Some, like the original poster, might say the THX mode is "muddy" because the darker steps are too light... the lighter steps are too dark which contributes to the dull look. Adding luminance to try to overcome the dull look doesn't make it go away either.
My calibration results on many 800u's disagree, not with the character you describe, but at least with the lack of "goodness".
I get, after proper white and black level calibration, a gamma of 2.05 at the low end to 2.3 in the upper middles. While this is not perfect, it is well above average. Better, in fact, than any plasma I can think of except the Pioneers.
Here is a calibration report of an 800u in THX mode.
Check the point gamma chart.

My calibration results on many 800u's disagree, not with the character you describe, but at least with the lack of "goodness".
I get, after proper white and black level calibration, a gamma of 2.05 at the low end to 2.3 in the upper middles. While this is not perfect, it is well above average. Better, in fact, than any plasma I can think of except the Pioneers.
Here is a calibration report of an 800u in THX mode.
Check the point gamma chart.

Same here. In fact, my 58" 800U (manufacture date = December 2008) has an even better gamma after the 100 hour break-in. It's pretty linear at 2.25-2.3 from 10-60IRE, then peaks to 2.4 around 70IRE and falls back to 2.3 at 80-100IRE. See here:
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=129420&stc=1&d=1231739123

My grayscale in WARM mode was initially too warm. So rather than bringing RDRV down in the SM, I brought GDRV and BDRV up. As a consequence, my THX mode light output increased from 31 ftL to 35 ftL with PICTURE at 100. Basically, by driving the RGB higher it effectively increases the light output. I found light output to be too low in THX mode even with PICTURE at 100. That might be the reason the OP is finding it "muddy".

Same here. In fact, my 58" 800U (manufacture date = December 2008) has an even better gamma after the 100 hour break-in. It's pretty linear at 2.25-2.3 from 10-60IRE, then peaks to 2.4 around 70IRE and falls back to 2.3 at 80-100IRE. See here:
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=129420&stc=1&d=1231739123

My grayscale in WARM mode was initially too warm. So rather than bringing RDRV down in the SM, I brought GDRV and BDRV up. As a consequence, my THX mode light output increased from 31 ftL to 35 ftL with PICTURE at 100. Basically, by driving the RGB higher it effectively increases the light output. I found light output to be too low in THX mode even with PICTURE at 100. That might be the reason the OP is finding it "muddy".
I agree; that has been my strategy as well. I like more than 30-31 ftL also. Though I have been averaging closer to 33 ftL after cal, that is more than explainable by sample to sample differences and differences in meters. I suspect the gamma difference can be attributed to those 2 things as well.

My calibration results on many 800u's disagree, not with the character you describe, but at least with the lack of "goodness".
I get, after proper white and black level calibration, a gamma of 2.05 at the low end to 2.3 in the upper middles. While this is not perfect, it is well above average. Better, in fact, than any plasma I can think of except the Pioneers.
Here is a calibration report of an 800u in THX mode.
Check the point gamma chart.


Samsung plasmas have 7 Gamma steps to choose from in the User Menu and the Service Menu will move the range of those 7 steps up and down. Furthermore, Samsung doesn't play games with lower gamma at low luminance and higher gamma at higher luminance levels. You can setup a Samsung panel to have a gamma as high as 2.7 if you want to... and it will measure very close to 2.7 at each grayscale step (perhaps not 10% though, crush may be creeping in at that point). I'm certainly not advocating DOING that, but it's quite possible. 2.3 or so seems to be a good setting for watching Blu-ray movies in a dark room. Side-by-side with an 800U in THX mode, with both panels calibrated, the Samsung will have more accurate color, lack the muddy/flat/dull appearance, and there won't be an obvious grayscale error at 80% that just about every 800U has. The A550 Samsung won't look quite as black as the 800U, but the A650 Samsung is a pretty close match for the black level of the 800U.

Last night I used a Lumagen Radiance XD to produce the same sort of gamma performance exhibited by the 800U in THX mode and found it upset the images in exactly the same way I object to in the 800U's THX mode. I used a Samsung plasma and a Toshiba LCD just to see what would happen with 2 different panels and 2 different technologies. I copied the results I've gotten from 800U calibrations.

There have been a lot of AVS Forum 800U owners who have the same complaint about THX mode... muddy and/or flat/dull. I've had calibration calls from at least 10 people wanting to know if I can remove the flat-dull look of the THX mode in their 800U and I had to be honest... no, there's nothing that can be done to eliminate it. Half of those calibrations were "lost" because of the inability to eliminate that gamma issue. You can futz with the cuts & gains to move the average gamma for the 800U up and down a little, but you can never remove the characteristic low gamma at low luminance, and high gamma at high luminance problem. The spread from 2.05 to 2.4 is HUGE - and to me it compomises the images far too much. When I see these panels I'm torn between using a 1.95-2.3 range or going a little higher (your 2.05-2.4 range for example). I'll take a Samsung A650 plasma any day over an 800U - with calibration, you can have nearly perfect primaries and secondaries, any gamma setting you want (without much variation from low to high luminance), and grayscale performance about on par with the 800U.

Samsung plasmas have 7 Gamma steps to choose from in the User Menu and the Service Menu will move the range of those 7 steps up and down. Furthermore, Samsung doesn't play games with lower gamma at low luminance and higher gamma at higher luminance levels. You can setup a Samsung panel to have a gamma as high as 2.7 if you want to... and it will measure very close to 2.7 at each grayscale step (perhaps not 10% though, crush may be creeping in at that point). I'm certainly not advocating DOING that, but it's quite possible. 2.3 or so seems to be a good setting for watching Blu-ray movies in a dark room. Side-by-side with an 800U in THX mode, with both panels calibrated, the Samsung will have more accurate color, lack the muddy/flat/dull appearance, and there won't be an obvious grayscale error at 80% that just about every 800U has. The A550 Samsung won't look quite as black as the 800U, but the A650 Samsung is a pretty close match for the black level of the 800U.
What grayscale error at 80%? My grayscale is linear and tracks excellent from 20-100IRE and there's no discrepancy or error at 80IRE. My gamma curve posted above is not perfectly linear, but it's damn close. I don't find the post-calibrated THX mode to be muddy/flat/dull at all. In fact, with an average gamma of 2.31 (as reported by ColorHCFR), the colors are so vibrant and pop out of the screen. Black levels are so deep it feels like looking through a clear glass window at live imagery. Don't know if this "issue" you're referring to was with early production 800Us, but mine has a Dec. 2008 manufacturing date and doesn't exhibit ANY of the issues you're describing.

And to compare the black levels of a Samsung 550 to the 800U is ridiculous. I've calibrated the 550 and the difference is very noticeable. The 800U comes close to Pioneer Kuro performance. The Samsung really doesn't. I haven't seen the 650, but Samsung would have to have made a great leap in black levels from the 550 to the 650 to come close to the 800U. By account of almost all the reviews out there, that doesn't seem to be the case, but I'll believe what you say when I see it (i.e. 800U vs 650).

Samsung plasmas have 7 Gamma steps to choose from in the User Menu and the Service Menu will move the range of those 7 steps up and down. Furthermore, Samsung doesn't play games with lower gamma at low luminance and higher gamma at higher luminance levels. You can setup a Samsung panel to have a gamma as high as 2.7 if you want to... and it will measure very close to 2.7 at each grayscale step (perhaps not 10% though, crush may be creeping in at that point). I'm certainly not advocating DOING that, but it's quite possible. 2.3 or so seems to be a good setting for watching Blu-ray movies in a dark room. Side-by-side with an 800U in THX mode, with both panels calibrated, the Samsung will have more accurate color, lack the muddy/flat/dull appearance, and there won't be an obvious grayscale error at 80% that just about every 800U has. The A550 Samsung won't look quite as black as the 800U, but the A650 Samsung is a pretty close match for the black level of the 800U.
The other point worth mentioning is if you're comparing the 50" versions of the 800U to the 50" versions of the Samsung 550/650. I have the 58" 800U. OP has the 50", and maybe that model has the muddy/flat/dull look.

I have a 50" 800u Panasonic plasma and use it about 90% of the time for gaming and love how it looks in every regard with it set to thx. My only problem is alot of times with games, it looks a bit muddy.

I would really appreciate if someone could explain where the gamma adjust is in the service menu. I know how to access the SM and needless to say, understand the dangers of it, which is why I'm asking instead of going in blindly. I've already adjusted the red to make it a bit more red and less pink but just cannot find the gamma. Thanks!

You may want to give my user menu and service menu settings a go as detailed in this post: http://www.avsforum.com/avs-vb/showpost.php?p=15559155&postcount=2049. Be aware though that I have a 58" version of the PZ800U.

The 800U comes close to Pioneer Kuro performance.Whoa, I was with you until here. The Panny has a better black level than the Samsung, but it is not even close to the current generation Kuro, which is close to being so low it is unmeasurable.

Whoa, I was with you until here. The Panny has a better black level than the Samsung, but it is not even close to the current generation Kuro, which is close to being so low it is unmeasurable.
You're right. I should rephrase that: the 800U is the next best thing after the Kuro for black level performance.

Unfortunately, I've only had the opportunity to calibrate and do dark-room testing of the 800U and Samsung 550. I'd like to have a go with the Samsung 650, Kuro (non-Elite), and Kuro Elite for the sake of comparison. But I'd be surprised if the Samsung 650 has the same black levels as the 800U, because that would be a significant leap from the 550.

I had my Panasonic TH-46PZ800U ISF calibrated in the Custom mode by Bill Hergonson(Coast Calibration) recently, here's a link to the calibration report(pre-calibration data is THX-warm):

http://www.flickr.com/photos/22138595@N06/sets/72157612063712960/

Click on each thumbnail, then click on the magnifying glass icon above each page to see a clearer/larger view...

Tom Huffman calibrated my 50" 800U several months ago and here is the gamma curve post-calibration in the THX (warm) setting:

10%--2.22
20%--2.24
30%--2.26
40%--2.29
50%--2.34
60%--2.39
70%--2.32
80%--2.30
90%--2.32
100%-2.30

Not perfectly linear, but pretty damn good, IMO. There is nothing "washed out" about my picture, something that I couldn't say before he calibrated it. Several months post-calibration, the colors remain accurate and vibrant, with plenty of "pop" and excellent picture depth.

I couldn't be happier with the 800U and the job that Tom did. The change from the pre-calibration THX setting was so dramatic that I went through a day or two of "what happened to my set?"....followed by "oh, yea!". My wife--a professional photographer--saw the improvement immediately and remains impressed by what Tom was able to do for the set. I have had no reason to deviate more than one or two ticks from the user's menu settings that Tom gave me after he worked his magic in the service menu.

Tom Huffman calibrated my 50" 800U several months ago and here is the gamma curve post-calibration in the THX (warm) setting:

10%--2.22
20%--2.24
30%--2.26
40%--2.29
50%--2.34
60%--2.39
70%--2.32
80%--2.30
90%--2.32
100%-2.30

Not perfectly linear, but pretty damn good, IMO. There is nothing "washed out" about my picture, something that I couldn't say before he calibrated it. Several months post-calibration, the colors remain accurate and vibrant, with plenty of "pop" and excellent picture depth.

I couldn't be happier with the 800U and the job that Tom did. The change from the pre-calibration THX setting was so dramatic that I went through a day or two of "what happened to my set?"....followed by "oh, yea!". My wife--a professional photographer--saw the improvement immediately and remains impressed by what Tom was able to do for the set. I have had no reason to deviate more than one or two ticks from the user's menu settings that Tom gave me after he worked his magic in the service menu.
Your gamma values confirm that the 50" version of the PZ800U like the 58" version has a nearly linear gamma. I have no idea what Doug Blackburn is talking about and none of his claims were backed with data/facts.

I re-tweaked my set a little bit and my average gamma is now 2.29. Here are my gamma values from 10-90IRE:
10%: 2.25
20%: 2.25
30%: 2.26
40%: 2.30
50%: 2.29
60%: 2.32
70%: 2.40
80%: 2.28
90%: 2.27

Here's my latest gamma chart from ColorHCFR:
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=129995&stc=1&d=1232120262

None of my claims were backed up with data/facts? Really? My claims were that gamma is "good" in THX mode, not very good in other modes; and I posted a calibration report that shows a decent gamma response in THX mode that varies between 2.05 and 2.3. Aren't you getting me mixed up with another poster?
Oops, I meant Doug Blackburn. I've edited my post above with the correction. Sorry.

None of my claims were backed up with data/facts? Really? My claims were that gamma is "good" in THX mode, not very good in other modes; and I posted a calibration report that shows a decent gamma response in THX mode that varies between 2.05 and 2.3. Aren't you getting me mixed up with another poster?

My ISF calibrator(Bill Hergonson-Coast Calibration) sent me a heads up that you guys are the authorities on calibration so I'll tread lightly, what about my Custom mode post calibration gamma-seems linear and high enough to me? Here's a direct link to the gamma numbers/graph:

http://www.flickr.com/photos/22138595@N06/3155574543/sizes/l/in/set-72157612063712960/

Also, Bill said my set's color matrix was close to being right on(especially the important blue point.)


Btw, 37.72ftL seems perfect to me-plenty bright during daylight hours, and just right at night(absolute match for BD 24p in a totally dark room.)

My ISF calibrator(Bill Hergonson-Coast Calibration) sent me a heads up that you guys are the authorities on calibration so I'll tread lightly, what about my Custom mode post calibration gamma-seems linear and high enough to me? Here's a direct link to the gamma numbers/graph:

http://www.flickr.com/photos/22138595@N06/3155574543/sizes/l/in/set-72157612063712960/

Also, Bill said my set's color matrix was close to being right on(especially the important blue point.)


Btw, 37.72ftL seems perfect to me-plenty bright during daylight hours, and just right at night(absolute match for BD 24p in a totally dark room.)

Hmm, that isn't bad at all. While not perfect, that gamma should look pretty good. I'm surprised you got that in Custom mode, as the 800u's I've worked with had poor gamma in any mode but THX. Do you know if your calibrator took that final reading in the service menu or out of the service menu?
I like light output in that range also (upper 30's).

Oops, I meant Doug Blackburn. I've edited my post above with the correction. Sorry.

Thanks!:cool:

My ISF calibrator(Bill Hergonson-Coast Calibration) sent me a heads up that you guys are the authorities on calibration so I'll tread lightly, what about my Custom mode post calibration gamma-seems linear and high enough to me? Here's a direct link to the gamma numbers/graph:

http://www.flickr.com/photos/22138595@N06/3155574543/sizes/l/in/set-72157612063712960/

Also, Bill said my set's color matrix was close to being right on(especially the important blue point.)


Btw, 37.72ftL seems perfect to me-plenty bright during daylight hours, and just right at night(absolute match for BD 24p in a totally dark room.)
I find 1.99-2.10 from 10-50 IRE a little low. 2.2-2.3 result in more punch and slightly darker grays.

I find 1.99-2.10 from 10-50 IRE a little low. 2.2-2.3 result in more punch and slightly darker grays.

It took a bit of finessing to get the gamma up to an average of 2.1_but my calibrator only had a few hours to work on my set-not several days as he said some of you guys spend... ;)

Here's some screenshots of my Panasonic TH-46PZ800U's ISF calibrated Custom-normal mode:

http://www.avsforum.com/avs-vb/showthread.php?p=15964983#post15964983
http://www.avsforum.com/avs-vb/showthread.php?p=15882468#post15882468
http://www.avsforum.com/avs-vb/showthread.php?p=15879106#post15879106
http://www.avsforum.com/avs-vb/showthread.php?p=15834571#post15834571
http://www.avsforum.com/avs-vb/showthread.php?p=15819245#post15819245
http://www.avsforum.com/avs-vb/showthread.php?p=15800821#post15800821
http://www.avsforum.com/avs-vb/showthread.php?p=15790813#post15790813

Imho, the 800U's(and most likely future sets with a THX mode) NEED to be calibrated. Whether calibration is done in THX mode or a custom mode I'll leave up to those more knowledgeable than myself...

I've posted my set's calibration report and answered questions about it over at the 'official' 800U thread for a couple of weeks, now I'm going back to lurking and enjoying my TV.:D

Best regards,
John Schuyler

Hmm, that isn't bad at all. While not perfect, that gamma should look pretty good. I'm surprised you got that in Custom mode, as the 800u's I've worked with had poor gamma in any mode but THX. Do you know if your calibrator took that final reading in the service menu or out of the service menu?
I like light output in that range also (upper 30's).

I confirmed with my ISF calibrator(Bill Hergonson-Coast Calibration) that the final readings were taken after backing out of the Service Menu.

I had my Panasonic TH-46PZ800U ISF calibrated in the Custom mode by Bill Hergonson(Coast Calibration) recently, here's a link to the calibration report(pre-calibration data is THX-warm):

http://www.flickr.com/photos/22138595@N06/sets/72157612063712960/

Click on each thumbnail, then click on the magnifying glass icon above each page to see a clearer/larger view...

Do you have the pre-calibration gamma? Thanks

there is no way to change gamma on this display.
It is merely a preset that is linked to the user mode selected.
the THX mode will give you a gamma around 2.2, the Custom mode will give you a convex curve which will make the image appear much brighter (and therefor much more suitable to a room with ambient light present).

Do you have the pre-calibration gamma? Thanks

Sorry, no. After the first preliminary run the gamma was at 1.8, he then got it up to 2.0, still not being satisified he tweaked some more and it ended up at 2.1.

Here's my Panasonic TH-46PZ800U's Custom Mode/Normal Color Temperature post calibration gamma(taken after backing out of the Service Menu):

http://farm4.static.flickr.com/3108/3155574543_4652b165cb_b.jpg

As long as we're in the ballpark of comparing the Panny's to Sammy's, is there anything new to be told of the PN-A760 line from Samsung? Better black levels than the A650 or A550? Gamma the same? Thanks for your input. I'm with Doug Blackburn in terms of being able to get the Sammy's primaries and secondaries spot on but the gamma and/or black levels have been a problem spot for me on my home's A450. Now I'm looking into an A550 or A760 for my office and a more professional use.

Ok so I actually purchased a 42G10 from a local store yesterday under the condition that I could take it back if I couldn't calibrate it properly. I'm almost certain I'll be taking it back.

The gamma spikes up at the high end and starts low on the low end. The picture to me is still good but that much inaccuracy for what I need it for is going to kill me.

As far as color, the saturations aren't even close even when I turn the "Color" control to 100%. Red should be 21% of your overall white value and it was less than 10% even with Color maxed out... This is with THX mode. In other modes the colors were much brighter and could reach spec.

If the THX mode were accurate I could live with only 30ftL light output but it's very far from accurate for me at least.

I'd love some advice to try to fix this set up as much as possible. I did try the service menu and the G-DRV maxed out with only a few clicks so there wasn't much hope of adding luminance. The gray scale balanced out at least. Anything else I can try?

Initially I was quite pleased with the (somewhat darker) image but I need it to be more accurate for the work I'll be doing with it. Thanks for your help.

THX mode
Contrast 100
Brightness 68
Color 76
Here are some visuals.

As far as color, the saturations aren't even close even when I turn the "Color" control to 100%. Red should be 21% of your overall white value and it was less than 10% even with Color maxed out... This is with THX mode. In other modes the colors were much brighter and could reach spec.


This doesnt make sense? This is not how you measure primary luminance.

This doesnt make sense? This is not how you measure primary luminance.You beat me to it :D

This doesnt make sense? This is not how you measure primary luminance.

:) Sorry, you guys would know better than me. I'm just going off of The Greyscale and Color Calibration for Dummies guide. http://www.curtpalme.com/forum/viewtopic.php?t=10457

Here is the quote:

Skip to the 100% Red window pattern by selecting "100% Saturated Colors" -> "100% Red window".
Adjust the colour control on your display until the Y reading is 21% of the 100% white window reading measured earlier. For example, if the Y value earlier was 39.012 as seen above, then 21% would be 39.012 x 0.21 = 8.193.


Though now that I look at it again I may have been looking at the wrong number (ftL instead of Y). :rolleyes:

I'll recheck that again. Any suggestions for the rest?

until the Y reading is 21% of the 100% white window reading measured earlier.
I dont think you read this ??

I guess I'm a bit confused on which values to be looking at. My ftL at 100% white is 30. The "Y" value at 100% white is 104.7. Do I want 21% of 30 or 104.7?

Here are the resulting 100% red numbers:
ftL: 1.94 (6% of 30ftL)
Y: 6.6 (22% of 30ftL OR 6% of initial Y, 104.7)

Then in "Standard" mode I set the white window to 30ftL and bring up the red primary:
ftL: 6.5 (22% of 30ftL)
Y: 22.2 (74% of 30ftL OR 22% of initial Y, 105)

This is with both 100% white windows at 30ftL.

How do you measure primary luminance? When I calibrated my Samsung last year I went with the ftL percentage of ftL and it looked good. I really appreciate your help!

I guess I'm a bit confused on which values to be looking at. My ftL at 100% white is 30. The "Y" value at 100% white is 104.7. Do I want 21% of 30 or 104.7?

It doesn't matter which units luminance is measured in - just that red is 21% (if HD709 is the target) of white assuming the same level of stimulus (i.e. measure red in a 100% red window and measure white in same sized 100% white window) and the same units of measure. So, if your ftL for 100% white is 30 then you want your red luminance measure to be 6.3 ftL.

hope this helps,


--tom

Thanks, Thomas. That's what I suspected. So what do you make of the G10's THX mode not reading close to those levels? All the colors fall short of their respective percentages, some by more than half.

. So what do you make of the G10's THX mode not reading close to those levels? All the colors fall short of their respective percentages, some by more than half.

I think your methodology is past the point of being suspect. The above comment is way too premature to be stated.

As an FYI, use 75% stimulus for doing primary luminance measurements. If you are using 100% you may be doing a soft clip of one of the primaries would would adversely affect your measurements.

As an FYI, use 75% stimulus for doing primary luminance measurements. If you are using 100% you may be doing a soft clip of one of the primaries would would adversely affect your measurements.

I see what you mean. So if reading 75% color window then an example would be 75% of 21% for Red (15.75%)? Thanks for that tip. It makes sense.

So is there anything else you can do to help me out? What is your experience with the G10 and color luminance? I would LOVE to get this screen on track so I don't have to return it. Nothing else compares to it's black levels at this price point and it really creates a great picture. I just need to straighten out the gamma and figure out this color luminance/saturation situation. The only visible flaw for me is that skintone comes out a bit on the redish side so if that is solved in the process then I'll be a happy man.

I see what you mean. So if reading 75% color window then an example would be 75% of 21% for Red (15.75%)? Thanks for that tip. It makes sense

you are making NO sense

There is nothing wrong with you panel. There are things grossly wrong in your methodology of measurement and process of logic.

You do not set the color (and tint) control by measuring luminance.

You can measure color decoder by measuring the fL reading on 75% white. Your red reading should be about 21% of your fL reading for white. (assuming you are also measuring a 75% red window box (which would have to be the same size as the 75% white window box).

Has anyone measured the black level of the Panasonic panels?

BTW, buyer beware about Panny blacks. They rise over time. When I measured my 85U's black shortly after purchase it was 0.007 fL. Now, after 1962 hours of use, it is 0.018 fL.

hi Tom
What instrument are you measuring with?
You are talking about a difference of 0.011fL. Within the realm of error ?

Let's get back to fundamentals for a moment...

You haven't got a prayer in hell of getting any color measurements right on a plasma if you are using anything but 75% window patterns. 100% patterns will be brightness/luminance-limited by the thermal properties of the panel and the capabilities of the power supply. Using 100% windows with a plasma is problematic at best. Using 100% full fields is MUCH MUCH worse and won't give you accurate measurements of anything.

Why the Calibration for Dummies document would ignore this necessity for plasmas is beyond me.

So you are using 75% window patterns now... much better. But how do you ever know what the right settings are for the colors? Well, you measure 75% WHITE as well. This helps 2 ways... if you use 100% white, you are all the way up at the max output of the panel (ditto for 100% patterns for each color) so your adjustment applies to the highest-output the panel is ever capable of. With 75% windows, you are well down the luminance curve (probably close to 50% stimulus) and when you make an adjustment in the MIDDLE of the luminance range, you can very well end up with better results across the entire 0%-100% luminance range for each color than if you make your 1 adjustment all the way up at 100%.

Now... Y and fL are the same thing if you are using xyY. If you are using uvL... L is NOT the same thing as fL.

Your 100% white window should EASILY read 30 fL with Contrast set much lower than 100. You may be causing an above-white color shift on the Panasonics with Contrast set to 100. Obviously, when you are making color measurements with 75% patterns, you will read much less than 30 fL for the 75% white window.

Next... having calibration software like CalMAN that gives you all the target values for xyY or uvL (and others you can select if you prefer) makes this whole process much easier than using 21% for red, 72% for Green, etc. If your software doesn't have that level of support, you can use the free calculator from the Accupel website (www.accupel.com (http://www.accupel.com)), just download it from their firmware download page. Make the readings and type the numbers into the Accupel calculator and it will tell you what your targets are and how close your actual readings are. No software or the Accupel calculator requires you to use 100% patterns for the readings (because that's nothing but trouble with plasmas). So all you have to keep in mind with color measurements is that if you use 75% windows for the colors, you use a 75% window for white also... the 21%, 72%, 7% (approx) ratios don't change just because you are at 75% instead of 100%.

Hope this helps. Now... it may not fix the gamma completely... Panasonics tend to have the gamma issue you describe with great regularity, though they usually seem to have a relatively OK gamma in THX mode and you see the odd-shaped gamma curves in the non-THX modes. But keep in mind... luminance limiting WILL be a factor if you are using full-screen patterns instead of window patterns. And the gamma curve you get using full-screen patterns will look much worse than the gamma curve you get when using window patterns.

hi Tom
What instrument are you measuring with?
You are talking about a difference of 0.011fL. Within the realm of error ?Greg: I used a Chroma 5 colorimeter, whose luminance accuracy I checked against the Orb spectro and it was spot on.

This is the kind of thing one wouldn't notice without long-term exposure to the same panel. Being an owner I have had a chance to track this over time. I checked it several times when it was new and I got consistent readings in the 0.007-0.009 range. Now it tracks just as consistently at 0.017-0.019. The black level was also up a little after 200 hrs. of use and the trend has just continued, though I think it has now stabilized.

I thought I would go back and quickly review the black levels I had measured with some other Pannys of this generation (85U or 800U) that I have calibrated (and which were presumably fairly new), and I got the following:

0.011
0.015
0.013
0.016

By way of comparison, the latest dead-man-walking Kuros measure a generous 0.001 fL. I say "generous" because sometimes it is so low I can't measure it at all.

These Panny measurements are quite consistent, but they are also much closer to the 0.018 I measure now from my 85U than they are to the 0.007 I measured initially. So, either I somehow screwed up my initial reading (which I doubt), or my Panny's black level was unusually low out of the box and simply settled into a more normal value after a long period of use.

So, has no one measured the black level of one of the new G10s?

you are making NO sense


You have to admit I'm making SOME sense. I apologize if my terminology is not spot on but I do "get" what I'm trying to accomplish. I'm a DIY'er and do not want to give the impression that I'm a pro, so please have patience.

I read over Tom Huffman's Basic Guide to Color Calibration (http://www.avsforum.com/avs-vb/showthread.php?t=852536) just to get on the right page. Correct me if I'm wrong that the following terms are synonymous: Color decoding, Color brightness, Color luminance, Chroma. The exception is that color decoding also refers to the hue of the secondaries. Saturation and hue fall under Color Gamut and are plotted on the CIE chart, correct? I just want to make sure I understand these properties and feel free to correct me at any point as I'm here to learn.

You do not set the color (and tint) control by measuring luminance.

I agree you don't set tint by measuring luminance but from what I've read from Tom the color control effects color brightness. From the guide: Color Decoding: This is the display's ability to properly render the relative brightness of the primaries and hue of the secondaries. All commercial displays include a Color and Tint control. These are basically color decoding controls...
And brightness is...
Color Brightness - the luminance of color.

So from that I gather, you DO set the color control by measuring luminance. Again, if I'm wrong I don't mind being corrected.

There are things grossly wrong in your methodology of measurement and process of logic.

An example would be? I'm just trying to follow the only "textbooks" I have so if they fail to mention something important, don't blame me.


Doug, Thanks for your extensive post. I'll look into the Accupel calculator. I've also appreciated your patience in answering my PMs regularly and honestly. You've been a great resource.


Your 100% white window should EASILY read 30 fL with Contrast set much lower than 100.

That's what I thought as well but THX mode takes the brightness out of this display. I've read the same in other reviews. It actually came set at 100 contrast in THX mode. It's still very acceptable brightness but 35fL or the option of 40 would be nice to have. The grayscale seems to track well at this setting but I don't know about above-white color shift. There's not much you can do since lowering contrast just makes it too dark.

Haven't measured with 75% color windows yet but I'll let you know what I find. Thanks.

You have to admit I'm making SOME sense. I apologize if my terminology is not spot on but I do "get" what I'm trying to accomplish. I'm a DIY'er and do not want to give the impression that I'm a pro, so please have patience.

I read over Tom Huffman's Basic Guide to Color Calibration (http://www.avsforum.com/avs-vb/showthread.php?t=852536) just to get on the right page. Correct me if I'm wrong that the following terms are synonymous: Color decoding, Color brightness, Color luminance, Chroma. The exception is that color decoding also refers to the hue of the secondaries. Saturation and hue fall under Color Gamut and are plotted on the CIE chart, correct? I just want to make sure I understand these properties and feel free to correct me at any point as I'm here to learn.



I agree you don't set tint by measuring luminance but from what I've read from Tom the color control effects color brightness. From the guide:
And brightness is...


So from that I gather, you DO set the color control by measuring luminance. Again, if I'm wrong I don't mind being corrected.



An example would be? I'm just trying to follow the only "textbooks" I have so if they fail to mention something important, don't blame me.


Doug, Thanks for your extensive post. I'll look into the Accupel calculator. I've also appreciated your patience in answering my PMs regularly and honestly. You've been a great resource.



That's what I thought as well but THX mode takes the brightness out of this display. I've read the same in other reviews. It actually came set at 100 contrast in THX mode. It's still very acceptable brightness but 35fL or the option of 40 would be nice to have. The grayscale seems to track well at this setting but I don't know about above-white color shift. There's not much you can do since lowering contrast just makes it too dark.

Haven't measured with 75% color windows yet but I'll let you know what I find. Thanks.

I've calibrated a number of 800u models in THX mode... I never had to set contrast that high (100) to get 30 fL. They would typically get to 40 fL or a little more when set to 100. BUT... every time I've seen an 800u in THX mode it just plain looks DULL. Doesn't matter if it IS putting out 40 fL, it STILL looks DULL. I typically measure lower gammas at low luminance steps and higher gammas at higher luminance steps (as you mentioned you did, if I remember right). I believe THAT causes the DULL look the images have. The low gamma at low luminance makes shadows too light and the high gamma at high luminance steps makes 80% and 90% darker than they should be. Those 2 things combined WILL give images a dull, flat look.

You are having problems with terminology... not all of those things you mentioned are the same thing. Chroma is just a way of indicating you are talking about COLOR instead of the luminance portion of the image (the B&W image or grayscale). A "chroma error" is just a color error of any kind. Color Decoder is a very specific "thing"... it converts color data to different formats. When you send YCbCr data to a (current) video display, the data is processed in that format, but just before the data is displayed, it is converted to RGB - the Color Decoder or Color Decoding Matrix does that conversion. If you send RGB to almost all current video displays, the RGB goes to a color decoder matrix FIRST where it is converted to YCbCr, is processed, then put through a DIFFERENT color decoding matrix to convert the YCbCr back to RGB for display. The decoder matrix is no better than whoever enters the numbers/formulas for each "cell" in the matrix. You'd think by now that everybody would simply use accurate values in the decoding matrix - the values/formulas that give accurate results are certainly well-known by now. But manufacturers often force their engineers to make the matrix intentionally inaccurate to get what they think is a beneficial side-effect that will sell more video displays. This is where red-push came from years ago... TVs were so blue that fleshtones looked vampirish. Some big brain got the idea that you could add some red back in as YPbPr (it was mostly an analog thing when the red-push was so prominent, though you might still see some in a digital display) was converted to RGB in the matrix. A GOOD decoding matrix won't change anything.

You have to watch terminology a bit also... for example... uvL color space is hue, saturation, and Luminance. This is one of the more perceptually accurate color spaces. Some CMS controls work this way (i.e. 1 slider for hue, 1 slider for saturation, 1 slider for luminance). So clearly, saturation and luminance are not the same thing.

uvL space can be thought of as a cylinder that's tapered on the ends. One end is black, the other end is white... that's the luminance axis. The saturation axis is the radius of the cylinder... one end is 100% saturation, the other end is 0% saturation. Hue travels around the cylinder (circular). Hue tells you what the color is, saturation tells you how intense the color is, luminance tells you how bright the color is.

In xyY space xy defines a color and how saturated it is at the same time (the color is more saturated when the xy coordinates move the color closer to the limits/edges of the CIE color space). Y determines how bright or dark the color is.

It can be tricky to perform the mental gymnastics needed to convert the actions of CMS controls into moves in xyY space that get you where you want to be... that's why I recommended the AccuPel calculator if your software doesn't provide this information. CMS controls are still evolving, though Lumagen and Samsung are both at the same place now... take "green" for example... one slider moves green closer to blue, the second slider moves green closer to red, and the third slider controls how bright the color is. For me, using xyY is easier to keep straight because you can think of what you are doing in 2 dimensions (using the 2 color sliders to get you where you want to be) then simply setting luminance to be accurate. You're doing a 3-D adjustment, but your brain isn't overwhelmed with 3-D space while you are doing it.

Chroma is just a way of indicating you are talking about COLOR instead of the luminance portion of the image (the B&W image or grayscale). A "chroma error" is just a color error of any kind.In color science, "chroma" has a much more specific meaning than this. It is the colorfulness of an area relative to the brightness of a similarly illuminated area of white. This is distinct from, but closely related to, saturation, which is the colorfulness of an area relative to its own brightness.

From these two definitions it is clear that brightness plays a role in chroma, which is a mixture of brightness AND saturation. On the other hand, saturation is completely independent of brightness.

The vast majority of displays allow for an adjustment of chroma through the main Color control, which when adjusted, raises and lowers both the saturation of the color AND its brightness. In fact, modest adjustments of the Color control have a much larger impact on brightness than on saturation.

Only a few displays with a properly functioning CMS offer a pure saturation adjustment.

Regarding color decoding, it is important to distinguish between the meaning of the term, which is exactly as you describe, and the perceved effects of changes to or errors in color decoding, whch is exactly as the poster stated. Color decoding errors manifest themselves as errors in the brightness of the primaries and the hues of the secondaries. For this reason, the color and tint controls are basically color decoding controls, because their effect is to change the brightness of the primaries (Color control) and the hues of the secondaries (Tint control).

In color science, "chroma" has a much more specific meaning than this. It is the colorfulness of an area relative to the brightness of a similarly illuminated area of white. This is distinct from, but closely related to, saturation, which is the colorfulness of an area relative to its own brightness.

Mental gymnastics is right. :eek: I feel like I'm getting a PhD in this stuff by now.

Does anyone know about the Christie 2K projects used in theaters? They have a 2.6 gamma and XYZ colorspace. Our first project we tried to show in our local digital theater was much too dark and a bit over saturated and I suspect it was displaying a 2.2-mastered gamma at 2.6. Fascinating stuff but I never got a chance to really test out a few other methods. Just wondered if you had experience on that front.

I actually returned the G10 yesterday and ordered a Pioneer KRP-500M. Even if the G10 was spot on color-wise (which it wasn't and I had little hope of fixing it with the lack of CMS). I wanted to get the 50" which would have been only $350 less with tax than what I paid for the Kuro which was $1,974 shipped. It seemed like a no-brainer for someone in my position and the calibration reports people have posted look near perfect on all fronts. Looking forward to working with it. Thanks again for helping me understand the terminology and methodology better. You guys seem to have this down pretty well. :)

I've calibrated a number of 800u models in THX mode... I never had to set contrast that high (100) to get 30 fL. They would typically get to 40 fL or a little more when set to 100. BUT... every time I've seen an 800u in THX mode it just plain looks DULL. Doesn't matter if it IS putting out 40 fL, it STILL looks DULL. I typically measure lower gammas at low luminance steps and higher gammas at higher luminance steps (as you mentioned you did, if I remember right). I believe THAT causes the DULL look the images have. The low gamma at low luminance makes shadows too light and the high gamma at high luminance steps makes 80% and 90% darker than they should be. Those 2 things combined WILL give images a dull, flat look.

You are having problems with terminology... not all of those things you mentioned are the same thing. Chroma is just a way of indicating you are talking about COLOR instead of the luminance portion of the image (the B&W image or grayscale). A "chroma error" is just a color error of any kind. Color Decoder is a very specific "thing"... it converts color data to different formats. When you send YCbCr data to a (current) video display, the data is processed in that format, but just before the data is displayed, it is converted to RGB - the Color Decoder or Color Decoding Matrix does that conversion. If you send RGB to almost all current video displays, the RGB goes to a color decoder matrix FIRST where it is converted to YCbCr, is processed, then put through a DIFFERENT color decoding matrix to convert the YCbCr back to RGB for display. The decoder matrix is no better than whoever enters the numbers/formulas for each "cell" in the matrix. You'd think by now that everybody would simply use accurate values in the decoding matrix - the values/formulas that give accurate results are certainly well-known by now. But manufacturers often force their engineers to make the matrix intentionally inaccurate to get what they think is a beneficial side-effect that will sell more video displays. This is where red-push came from years ago... TVs were so blue that fleshtones looked vampirish. Some big brain got the idea that you could add some red back in as YPbPr (it was mostly an analog thing when the red-push was so prominent, though you might still see some in a digital display) was converted to RGB in the matrix. A GOOD decoding matrix won't change anything.

You have to watch terminology a bit also... for example... uvL color space is hue, saturation, and Luminance. This is one of the more perceptually accurate color spaces. Some CMS controls work this way (i.e. 1 slider for hue, 1 slider for saturation, 1 slider for luminance). So clearly, saturation and luminance are not the same thing.

uvL space can be thought of as a cylinder that's tapered on the ends. One end is black, the other end is white... that's the luminance axis. The saturation axis is the radius of the cylinder... one end is 100% saturation, the other end is 0% saturation. Hue travels around the cylinder (circular). Hue tells you what the color is, saturation tells you how intense the color is, luminance tells you how bright the color is.

In xyY space xy defines a color and how saturated it is at the same time (the color is more saturated when the xy coordinates move the color closer to the limits/edges of the CIE color space). Y determines how bright or dark the color is.

It can be tricky to perform the mental gymnastics needed to convert the actions of CMS controls into moves in xyY space that get you where you want to be... that's why I recommended the AccuPel calculator if your software doesn't provide this information. CMS controls are still evolving, though Lumagen and Samsung are both at the same place now... take "green" for example... one slider moves green closer to blue, the second slider moves green closer to red, and the third slider controls how bright the color is. For me, using xyY is easier to keep straight because you can think of what you are doing in 2 dimensions (using the 2 color sliders to get you where you want to be) then simply setting luminance to be accurate. You're doing a 3-D adjustment, but your brain isn't overwhelmed with 3-D space while you are doing it.

Ahhhh. Finally, a person who has technical credentials as a pro calibrator and reviewer stating what others and I have been trying to say in the face of all these EyeOne measurement folks. The bloody picture in THX-Warm mode is simply DULL and unacceptable to anyone spending the dollars on this beauty and expecting eye-popping imagery. THX will not give it to you and Mr. Blackburn substantiates that. Thank you sir!!

The bloody picture in THX-Warm mode is simply DULL and unacceptable to anyone spending the dollars on this beauty and expecting eye-popping imagery. THX will not give it to you and Mr. Blackburn substantiates that. Thank you sir!!


I dont agree at all with this. The THX mode looks just fine when viewed in a darkened environment.

The bloody picture in THX-Warm mode is simply DULL and unacceptable to anyone spending the dollars on this beauty and expecting eye-popping imagery.There is nothing to substantiate. "Dull and unacceptable" are purely subjective terms, so I would be careful about commenting on what "anyone" will think about this.

THX looks fine to me and it measures fine, offering around 35 fL, which is the brightness target I aim for in any case.

It will look "dull" compared to the other modes, which have outrageously inaccurate gamma curves that allow very high peak output.

TomHuffman

Quote:
Originally Posted by Gregg Loewen View Post
hi Tom
What instrument are you measuring with?
You are talking about a difference of 0.011fL. Within the realm of error ?
Greg: I used a Chroma 5 colorimeter, whose luminance accuracy I checked against the Orb spectro and it was spot on.

Tom,

Neither the C-5 or the Orb SP-100 are capable of measuring accurately to the light levels which you have quoted in your posts. The C-5 is accurate to approximately .06 fL and the SP-100 will not provide accurate results at this light level. You would need to measure the display with a much higher sensitivity instrument then you are using to get reliable data to quote from at this light level.

The fact that your instruments display repeatable values at a light level indicated in the software does not mean that the data is reliable only that it is repeatable! The accuracy of the C-5 will drop of rapidly as it proceeds past its low light level limit. At this light level there is very little energy available for the detectors to collect and the tracking accuracy will not be reliable!

To measure to .007 fL reliably requires an excellent instrument as well as a true "bat-cave" environment.

Neither the C-5 or the Orb SP-100 are capable of measuring accurately to the light levels which you have quoted in your posts. The C-5 is accurate to approximately .06 fL and the SP-100 will not provide accurate results at this light level. You would need to measure the display with a much higher sensitivity instrument then you are using to get reliable data to quote from at this light level.

The fact that your instruments display repeatable values at a light level indicated in the software does not mean that the data is reliable only that it is repeatable! The accuracy of the C-5 will drop of rapidly as it proceeds past its low light level limit. At this light level there is very little energy available for the detectors to collect and the tracking accuracy will not be reliable!

To measure to .007 fL reliably requires an excellent instrument as well as a true "bat-cave" environment.Nope.

I just measured it again. I got 0.063 nits/0.018 fL with the C5 and 0.058 nits/0.017 fL with the Minolta LS-100. The difference between 0.018 fL and 0.017 fL is negligible.

The claim that you need extremely sensitive/expensive equipment to reliably measure luminance in the 0.03-0.01 fL range is a myth. Below that, the Minolta or something similar would be advisable.

I didn't use the C5 directly for the 0.007 fL reading and didn't even own the Orb at the time. I measured that by placing the sensor of the AEMC CA813 illuminance meter directly aganst the screen and correlating the lux reading with a conversion offset I had established at a much higher light level with the C5. Using the same technique now, I get 0.019 fL, again a negligible difference.

The only displays whose black level is so low that I am unconfident of my readings are the JVC RS20 front projector and the 9th-generation Pioneer Kuro plasma, and even in these cases I get quite consistent results, though they are at the absolute lower limits of the equipment.

I just want to take a moment to state that some people understand what thing "really" look like. Having said that we strive to reproduce THAT. The THX mode is somewhat "close"on paper, but it doesnt look the way it should to those in the know. Once youve seen a true feed on a reference monitor you will understand. Until then, you have no idea what "correct" looks like.

but it doesnt look the way it should to those in the know. Once youve seen a true feed on a reference monitor you will understand. Until then, you have no idea what "correct" looks like.

A position that I do not agree with.

I apologize if this is the wrong thread for my question. But since you guys are talking about THX and color i will ask. I have a panny 50v10 which has about 75 hours on it and i calibrated it using DVE HD Basics(blu ray). When doing the color adjustments with the filters i was able to get all modes (custom, standard, game, etc) just about spot on. However the thx mode was unadjustable. After adjusting the blue the red and green were incredibly far off and i was not able to adjust even lcose to getting the colors to blend together. Any comments/suggestions would be extremely welcome. This is my first diy calibration and i thought i was doing a good job until i tried with no succes to do THX mode.

This is the wrong thread for V10 questions and it is also the wrong thread for questions about setting the basic user menu controls, you should post this in the V10 forum. This thread is for display calibration questions -- using filters and DVE is not calibration - that's just basic setup. Calibration requires a meter, software, and a known-good video source.

Filters are not necessarily accurate - THX mode has more accurate (measured) colors than the other modes. If the filter doesn't look right for THX mode SMPTE color bars, then it can't be trusted for the color bars in the other modes either.

Your disc player may be altering color also - not easy to determine if that is happening without measurements of your system.

Has anyone measured the black level of the Panasonic panels?

BTW, buyer beware about Panny blacks. They rise over time. When I measured my 85U's black shortly after purchase it was 0.007 fL. Now, after 1962 hours of use, it is 0.018 fL.

I'm seeing the same behavior on my 58" PZ800U after 540 hours, measuring with an EyeOne DisplayLT. When it had just about 150 hours on it it was measuring consistently at 0.008 ftL but now it's measuring 0.014 ftL.

With regards to the black level getting worse on some Panasonic sets, you may want to have a quick look at this short thread from one year ago : http://www.avsforum.com/avs-vb/showthread.php?t=1016048

After that, I've measured the same behavior on all three Panasonic sets (same model) that I had contact with. In the worst case, black level went up from 0.017ftL to 0.056ftL, 10 months later.

I'm not a professional calibrator, just an enthusiast with low cost, entry level equipment/software (DisplayLT/HCFR), but I'm convinced that black levels get worse over time on, at least, that particular Panasonic model. You can actually see it on the sets. Of course, that's just my own evidence and opinion.

Fernando

You haven't got a prayer in hell of getting any color measurements right on a plasma if you are using anything but 75% window patterns. 100% patterns will be brightness/luminance-limited by the thermal properties of the panel and the capabilities of the power supply. Using 100% windows with a plasma is problematic at best. Using 100% full fields is MUCH MUCH worse and won't give you accurate measurements of anything.

Why the Calibration for Dummies document would ignore this necessity for plasmas is beyond me.

This question is directed at pro calibrators... Is this commonly accepted belief or is it not significant whether one uses 75% or 100% color windows when setting color and tint on a plasma? Both Tom Huffman's Guide to Color Calibration and the Grayscale and Color Calibration for Dummies Guide show no preference for either 75% or 100%, only that one uses a matching % for the white window.

This question is directed at pro calibrators... Is this commonly accepted belief or is it not significant whether one uses 75% or 100% color windows when setting color and tint on a plasma? Both Tom Huffman's Guide to Color Calibration and the Grayscale and Color Calibration for Dummies Guide show no preference for either 75% or 100%, only that one uses a matching % for the white window.

It's simple logic... if you have only ONE adjustment point, would you be more likely to get better results at either end of the luminance scale (black or white) or in the middle of the luminance scale (a 75% window is about 50% stimulus). And yes, if you use 75% color windows you MUST use 75% white for the reference also.

That said, there may be cases where you want to use 100% or 50% windows because of a response problem or a very specific issue that needs to be fixed at one part of the luminance scale (but the display is fine at other steps along the luminance scale.

You have to keep your wits about you. But most of the time for most displays, especially plasmas, using the 75% windows will produce the best overall results. But there will always be the exception to the rule here or there - you just don't want to miss those because you get into a habit of doing it one way to the point that you forget to check what's going on with the display at several luminance levels.