AE4000 Calibration help please - Page 2 - AVS Forum | Home Theater Discussions And Reviews
Baselworld is only a few weeks away. Getting the latest news is easy, Click Here for info on how to join the Watchuseek.com newsletter list. Follow our team for updates featuring event coverage, new product unveilings, watch industry news & more!


Forum Jump: 
Reply
 
Thread Tools
post #31 of 59 Old 06-16-2012, 04:50 AM
Member
 
dominickwok's Avatar
 
Join Date: Jul 2009
Posts: 188
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by jeansibelius24 View Post

Just wondered, do I set the colour temperature to -3 then set the colour and tint using a test pattern with the blue filter before moving onto setting the primaries, or do I leave the settings on zero?

If you're using a meter, it is not necessary to use a blue filter. You can leave the main Colour and Tint controls at 0 to start with. I set Colour = +1 and Tint = -1 simply because it can minimize the changes that I needed to make to the tint (for Hue), colour (for Satuation), and brightness (for Lightness) controls for the individual RGBCMY colours inside the AE4000's CMS (which is using the HSL model). If you discover that you need to make a big increase of Brightness to a, e.g. primary colour say Green, you can increase the main Colour control (which has the main affect on Lightness of all the primaries and secondaries) so that a less +ve Brightness value is required on Green (but of course, at the same time, a lower Brightness value is required on all the other primaries & secondaries too).
Quote:
Originally Posted by jeansibelius24 View Post

My contrast is currently on the default setting generating 10.6ftL with a 100% Ire pattern, is that good enough as a starting point?

That should be a good starting point (assuming you mean setting Contrast = 0, which is the default in Cinema 1); but you should check your gamma to see whether it is still a too high setting which gives you a gamma drop at the high end (typically when the unit is running out of Red in the high end). In my case, I set Contrast to -2 as a compromise.
Quote:
Originally Posted by jeansibelius24 View Post

When setting the CMS RGB levels with the 75% saturation windows do I just adjust the colour slider to attain the desired RGB percentages as I noticed there was also a tint and brightness control too?

No. In AE4000's CMS, the "Tint" control is for adjusting Hue, "Colour" control is for adjusting Satuation, and "Brightness" control is for adjusting Lightness. You can refer to the following post for some fundamentals:

http://www.avsforum.com/t/852536/basic-guide-to-color-calibration-using-a-cms-updated-and-enhanced
dominickwok is offline  
Sponsored Links
Advertisement
 
post #32 of 59 Old 06-23-2012, 07:52 PM - Thread Starter
Member
 
jeansibelius24's Avatar
 
Join Date: May 2012
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
After all these useful tips I thought that working my way through the steps wouldn't be too difficult, however trying just to set the correct 75% RGB saturation levels has proven to be a real headache.

I just wondered if I have set the sliders incorrectly because while trying to achieve the correct Red saturation level I had to reduce colour to -25, tint -3 amd brightness +10 (R 378% G 24% B 24%). For Green I had to reduce the colour down to -30 and increase the brightness to +20 which are the minimum and maximums of the sliders, even then I couldn't quite achieve the required R and B percentages which were both 2% under. The biggest problem however was Blue which I couldn't even get close to. According to the spreadsheet Blue needs to be 555% and reducing the colour slider completely reduced it only to 780%.

Also looking at the colours afterwards they looked really washed out.

These are the settings I used before attempting the CMS RGB saturation levels as recommended in a previous post.

Cinema1 mode
Contrast -2
Brightness +1
Colour +1
Tint -1
Colour Temperature -3
Dynamic Iris: Off

Any help would be really appreciated as the frustration levels are hitting a new high.

Thanks
jeansibelius24 is offline  
post #33 of 59 Old 06-24-2012, 06:09 AM
Member
 
dominickwok's Avatar
 
Join Date: Jul 2009
Posts: 188
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by jeansibelius24 View Post

After all these useful tips I thought that working my way through the steps wouldn't be too difficult, however trying just to set the correct 75% RGB saturation levels has proven to be a real headache.
I just wondered if I have set the sliders incorrectly because while trying to achieve the correct Red saturation level I had to reduce colour to -25, tint -3 amd brightness +10 (R 378% G 24% B 24%). For Green I had to reduce the colour down to -30 and increase the brightness to +20 which are the minimum and maximums of the sliders, even then I couldn't quite achieve the required R and B percentages which were both 2% under. The biggest problem however was Blue which I couldn't even get close to. According to the spreadsheet Blue needs to be 555% and reducing the colour slider completely reduced it only to 780%.
Also looking at the colours afterwards they looked really washed out.
These are the settings I used before attempting the CMS RGB saturation levels as recommended in a previous post.
Cinema1 mode
Contrast -2
Brightness +1
Colour +1
Tint -1
Colour Temperature -3
Dynamic Iris: Off
Any help would be really appreciated as the frustration levels are hitting a new high.
Thanks

What type of screen you're using? Besides your viewing environment (e.g. walls, ceiling, floor, furniture, etc.), the screen is one of the very significant factors that can change the colors seen by your eyes (and the meter of course). I'm using a matt white screen with gain ~0.8, which is famous to be best perserving color accuracy of the reflected light (vs. other types such as gray screen, etc.). I'm in no way required to use your extreme values in the AE4000 CMS. HItting 75% saturation targets is just like a piece of cake for me. Below are my values for your reference.

Colour / Tint / Brightness
R: -14 / 0 / -6
G: -11 / -6 / +10
B: +1 / +7 / 0
C: -12 / +4 / +11
M: -10 / -5 / -7
Y: -9 / +8 / +3
dominickwok is offline  
post #34 of 59 Old 06-24-2012, 02:22 PM - Thread Starter
Member
 
jeansibelius24's Avatar
 
Join Date: May 2012
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks for posting those settings, I will try them and see if they give me a better result.

My screen is a Grandview Cyber 92", here is the link to the product - http://www.cyberscreens.co.uk/cyberpulldownscreens.htm

The room is painted with flat matt, the ceiling dark grey and the walls dark green. I also use a blackout blind with black curtains. However I do have books on the floor which could possibly be reflecting light back onto the screen.

The meter I am using is the i1Display Pro III retail with the latest version of HCFR.

Think that I might have to give up on trying to calibrate the projector as it is really becoming a pain.
jeansibelius24 is offline  
post #35 of 59 Old 06-26-2012, 09:48 AM
AVS Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,316
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 264 Post(s)
Liked: 818
OK, I just calibrated a 4000U yesterday and although I didn't take notes of my procedure, there are a few things I wanted to post here.

It was light starved, as it was projecting a large image on a DIY gray screen ("Black Widow") painted wall. I don't know the gain of that paint, but I wouldn't be surprised if it was negative gain.

Before cal it was putting out around 9.9 fL, with a very low gamma and poor color. It was in Normal mode with minimal modifications. Lamp power was on normal (high).

My only option for calibration was Normal mode, since switching to the more accurate Cinema1 mode would cut light output to well below 5 fL. Probably end up around 3 fL after cal, which is unwatchable. Normal has very significant red drop off above 60-70% or so, but it's still a ton better than trying to watch 3-5 fL.

I calibrated to D75 rather than D65. That minimized the visibility of the unavoidable red dropoff.

With brightness (set with a low APL pluge pattern) at +4 and contrast at -3, and all color temp and white balance controls at default, the 10 point adjustment tracks well and is very useful for flattening out the gamma and grayscale throughout most of the range. If you want to use the multipoint, It's important to NOT adjust the R,G,B high and low white balance adjustments. Otherwise, the different colors will start getting displaced by differing amounts. In other words, if you adjust the white balance or the color temp adjustment much then when you try to adjust the multipoint you will find that the 80% red control may actually effect 95%, the 80% green may adjust 70%, and the 80% blue might adjust 85%, etc. I used the multipoint to flatten gamma over the entire range and flatten grayscale from 10-70%. Above 70%, the red ran out of steam, but as mentioned calibrating to D75 made that much less apparent. Dynamic Iris should be turned off for these adjustments.

With GS and gamma as good as could be, I moved on to CMS. The measurements, at least the conventional 75% levels/100% saturation measurements, indicated Red should be turned down in level and it's hue should be swung toward magenta. Don't do this! It results in pale and purplish people. I ended up taking the overall color control down a few clicks, but not as much as the measurements indicated, and then I actually gave the red hue just a tiny (2 click) nudge toward yellow. Flesh tones looked fantastic. The other primaries and the secondary's subjective looks followed matched up much better with the measurements, so calibrating the green CMS by the measurements eliminated the neon green grass in stadiums and ball fields. The magenta was also way off before cal and looked much better after CMS adjustment.

I ended up with just a tiny bit over 10 fL of light output and a dramatic transformation and startling improvement in PQ.

It's true, when projecting the 4000u on a low gain/large screen, the normal rules of calibration can not be followed. But the picture can still look subjectively very good, and it can be a huge improvement over stock Normal mode.

ISF/THX calibrator with Jeti 1211 reference spectro
Latest reviews:LG 65EG9600, Sony XBR75X940C, Samsung UN65JS9500
Copied settings measured, LG OLED calibration tips, Review index and rankings
Chad B is online now  
post #36 of 59 Old 06-27-2012, 03:43 PM - Thread Starter
Member
 
jeansibelius24's Avatar
 
Join Date: May 2012
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hi, I tried the RBCMY CMS settings in a previous post last night. The image looked fine to me but to be honest I don't know what to look for when viewing the image and whether or not it's closer to the correct saturation levels. I compared it to Color1 mode and they both looked very similar with the calibrated Cinema1 mode looking a bit darker with slightly less red in it. Is there a way I can tell if the primaries and secondaries saturation levels are improved, perhaps I could post a particular graph here from HCFR?

Also my RGB levels are now getting much closer with the exception of red dipping at 10% IRE and all three a bit untidy at 20% IRE. To try and correct this do I display a 10% IRE window and try and get the RGB bars to match up at 100% or will that throw off the other settings?

Thanks
jeansibelius24 is offline  
post #37 of 59 Old 06-28-2012, 08:20 AM
Member
 
dominickwok's Avatar
 
Join Date: Jul 2009
Posts: 188
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by jeansibelius24 View Post

Hi, I tried the RBCMY CMS settings in a previous post last night. The image looked fine to me but to be honest I don't know what to look for when viewing the image and whether or not it's closer to the correct saturation levels. I compared it to Color1 mode and they both looked very similar with the calibrated Cinema1 mode looking a bit darker with slightly less red in it. Is there a way I can tell if the primaries and secondaries saturation levels are improved, perhaps I could post a particular graph here from HCFR?
Also my RGB levels are now getting much closer with the exception of red dipping at 10% IRE and all three a bit untidy at 20% IRE. To try and correct this do I display a 10% IRE window and try and get the RGB bars to match up at 100% or will that throw off the other settings?
Thanks

You can do a full measurement of grayscale and a sweep of saturation levels (0% / 25% / 50% / 75% / 100%) for all the primaries and secondaries (using the saturation patterns in AVSHD709 disc) and then see how your particular AE4000 performs with respect to the gamut tracking. I'm sorry for not mentioning before but in fact you SHOULDN'T copy my settings because we do not have the same viewing environment, screen, bulb age, etc. etc. which can significantly affect your image's colors. By the way, I attach here some graphs that I extracted from the Stereomandan spreadsheet and the HCFR software (after I took the measurement of my calibrated Cinema 1 mode) in the PDF here, just to show you that in fact (it might be varies from unit to unit) how AE4000 is capable to have linear tracking of saturation, hue and luminance of colors up to 75% saturation level, together with a good tracking on the D65 white point and gamma. I'm using a i1Pro-profiled Spyder3 for calibration/measurement.

Even for colorimeter such as Spyder3, which can have much better light sensitivity vs. a spectro such as i1Pro, I will not trust any readings below 30% - trust your eyes instead.

Gamut.jpg 106k .jpg file

Color Luminance.jpg 93k .jpg file

Grayscale.jpg 84k .jpg file

Gamma.jpg 79k .jpg file
dominickwok is offline  
post #38 of 59 Old 06-28-2012, 11:37 AM
Member
 
morty343's Avatar
 
Join Date: Feb 2006
Posts: 50
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Chad B View Post

OK, I just calibrated a 4000U yesterday and although I didn't take notes of my procedure, there are a few things I wanted to post here.
It was light starved, as it was projecting a large image on a DIY gray screen ("Black Widow") painted wall. I don't know the gain of that paint, but I wouldn't be surprised if it was negative gain.
Before cal it was putting out around 9.9 fL, with a very low gamma and poor color. It was in Normal mode with minimal modifications. Lamp power was on normal (high).
My only option for calibration was Normal mode, since switching to the more accurate Cinema1 mode would cut light output to well below 5 fL. Probably end up around 3 fL after cal, which is unwatchable. Normal has very significant red drop off above 60-70% or so, but it's still a ton better than trying to watch 3-5 fL.
I calibrated to D75 rather than D65. That minimized the visibility of the unavoidable red dropoff.
With brightness (set with a low APL pluge pattern) at +4 and contrast at -3, and all color temp and white balance controls at default, the 10 point adjustment tracks well and is very useful for flattening out the gamma and grayscale throughout most of the range. If you want to use the multipoint, It's important to NOT adjust the R,G,B high and low white balance adjustments. Otherwise, the different colors will start getting displaced by differing amounts. In other words, if you adjust the white balance or the color temp adjustment much then when you try to adjust the multipoint you will find that the 80% red control may actually effect 95%, the 80% green may adjust 70%, and the 80% blue might adjust 85%, etc. I used the multipoint to flatten gamma over the entire range and flatten grayscale from 10-70%. Above 70%, the red ran out of steam, but as mentioned calibrating to D75 made that much less apparent. Dynamic Iris should be turned off for these adjustments.
With GS and gamma as good as could be, I moved on to CMS. The measurements, at least the conventional 75% levels/100% saturation measurements, indicated Red should be turned down in level and it's hue should be swung toward magenta. Don't do this! It results in pale and purplish people. I ended up taking the overall color control down a few clicks, but not as much as the measurements indicated, and then I actually gave the red hue just a tiny (2 click) nudge toward yellow. Flesh tones looked fantastic. The other primaries and the secondary's subjective looks followed matched up much better with the measurements, so calibrating the green CMS by the measurements eliminated the neon green grass in stadiums and ball fields. The magenta was also way off before cal and looked much better after CMS adjustment.
I ended up with just a tiny bit over 10 fL of light output and a dramatic transformation and startling improvement in PQ.
It's true, when projecting the 4000u on a low gain/large screen, the normal rules of calibration can not be followed. But the picture can still look subjectively very good, and it can be a huge improvement over stock Normal mode.

Thank you Chad,

I'm a little confused by your referral to the "RGB high and low white balance adjustments", and the "color temp adjustments", and what these are. Are you referring to the sliders for the individual primaries' brightness and contrast in the advanced menu, and the color temperature slider in the picture basic menu? Or are you referring to the 10-point gamma scale, saying that only the Y should be adjusted, and the gammas for RGB should not be individually adjusted. If it's the former, I don't understand how D75 can be attained without adjusting these sliders, because HCFR would report the RGB levels as not being relatively equal unless they are tweaked. Or is using the RGB gamma 10-point adjustments an alternative to playing with the RGB brightness/contrast, and you're saying that these can be left at 0 and greyscale balance can be achieved by solely adjusting RGB gammas?
morty343 is offline  
post #39 of 59 Old 06-28-2012, 12:03 PM
AVS Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,316
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 264 Post(s)
Liked: 818
Quote:
Originally Posted by morty343 View Post

Thank you Chad,
I'm a little confused by your referral to the "RGB high and low white balance adjustments", and the "color temp adjustments", and what these are. Are you referring to the sliders for the individual primaries' brightness and contrast in the advanced menu, and the color temperature slider in the picture basic menu? Or are you referring to the 10-point gamma scale, saying that only the Y should be adjusted, and the gammas for RGB should not be individually adjusted. If it's the former, I don't understand how D75 can be attained without adjusting these sliders, because HCFR would report the RGB levels as not being relatively equal unless they are tweaked. Or is using the RGB gamma 10-point adjustments an alternative to playing with the RGB brightness/contrast, and you're saying that these can be left at 0 and greyscale balance can be achieved by solely adjusting RGB gammas?

Yes, I am referring to the sliders for the R, G, and B brightness and contrast in the advanced menu and also the color temperature slider in the basic menu.
I am not at all familiar with HCFR, so any questions specific to that I will leave to others. However, yes, it's easy to get D75 (from 10-70%) by adjusting just the 10 pt adjustment. You will have to use each color's 10 pt adjustment. Also at the same time it can be used to flatten out the gamma, which was very important to getting the PQ improvement.
So to recap, if you want to try this method, leave the R/G/B brightness and contrast at or near 0 (I ended up turning blue brightness up by a click, but that was just enough to help smooth things out a bit and was not enough to cause problems). Leave the Color temp slider at default. Use all colors and the Y controls of the 10 point adjustment to achieve a flat gamma across the whole range and also D75 from 10-70%. Keep G and B tracking relatively close together from 80-100, but you won't be able to maintain red at those levels.

I am not recommending this practice for all calibrations. It's a deviation from the norm and should only be used on 4000us that would be struggling with weak light output in Cinema 1 mode.

ISF/THX calibrator with Jeti 1211 reference spectro
Latest reviews:LG 65EG9600, Sony XBR75X940C, Samsung UN65JS9500
Copied settings measured, LG OLED calibration tips, Review index and rankings
Chad B is online now  
post #40 of 59 Old 06-28-2012, 03:21 PM
Member
 
morty343's Avatar
 
Join Date: Feb 2006
Posts: 50
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Excellent, thanks very much for the suggestions! Very much appreciated.
morty343 is offline  
post #41 of 59 Old 06-28-2012, 03:24 PM - Thread Starter
Member
 
jeansibelius24's Avatar
 
Join Date: May 2012
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by dominickwok View Post

You can do a full measurement of grayscale and a sweep of saturation levels (0% / 25% / 50% / 75% / 100%) for all the primaries and secondaries (using the saturation patterns in AVSHD709 disc) and then see how your particular AE4000 performs with respect to the gamut tracking. I'm sorry for not mentioning before but in fact you SHOULDN'T copy my settings because we do not have the same viewing environment, screen, bulb age, etc. etc. which can significantly affect your image's colors. By the way, I attach here some graphs that I extracted from the Stereomandan spreadsheet and the HCFR software (after I took the measurement of my calibrated Cinema 1 mode) in the PDF here, just to show you that in fact (it might be varies from unit to unit) how AE4000 is capable to have linear tracking of saturation, hue and luminance of colors up to 75% saturation level, together with a good tracking on the D65 white point and gamma. I'm using a i1Pro-profiled Spyder3 for calibration/measurement.
Even for colorimeter such as Spyder3, which can have much better light sensitivity vs. a spectro such as i1Pro, I will not trust any readings below 30% - trust your eyes instead.
Gamut.jpg 106k .jpg file
Color Luminance.jpg 93k .jpg file
Grayscale.jpg 84k .jpg file
Gamma.jpg 79k .jpg file

I just wondered if I do a sweep of the primary and secondary saturation levels do I use the saturation luminance graph in HCFR and try and get each as flat as possible. If yes do I go through each saturation pattern from 10-100% and adjust the colour slider in the CMS until it sits on the reference line or will I have to use the tint and brightness slider as well to achiever the correct setting which could get somewhat confusing?

Thanks
jeansibelius24 is offline  
post #42 of 59 Old 06-29-2012, 08:15 AM
Member
 
dominickwok's Avatar
 
Join Date: Jul 2009
Posts: 188
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by jeansibelius24 View Post

I just wondered if I do a sweep of the primary and secondary saturation levels do I use the saturation luminance graph in HCFR and try and get each as flat as possible.

DON't use the Saturation-luminance graph in HCFR - it has serious bug there and it presents you a totally wrong graph. The problem lies in that it 'normalizes" all the color luminance curves in such a way that it 'assumes' the measured luminance at the 100% saturation level is perfectly correct and then plot the data points for 0%, 25%, 50%, and 75% as relative error to the 100% data point. So if your luminance of Red at 100% saturation level is +10% off from the standard, and you try to 'level' the luminance curve for the other 4 saturation points, you will end up with +10% luminance error for all the saturation levels (even though the HCFR graph will tell you "a perfect alignment with standard").
Quote:
Originally Posted by jeansibelius24 View Post

If yes do I go through each saturation pattern from 10-100% and adjust the colour slider in the CMS until it sits on the reference line or will I have to use the tint and brightness slider as well to achiever the correct setting which could get somewhat confusing?
Thanks

From my reading on your message above, it appears that you might get the concept wrong with respect to calibrating grayscale vs. color gamut using the controls in AE4000.

First of all, saturation patterns in the AVSHD709 disc consists of 30 patterns (no matter window or full field) - satruation level of 0%, 25%, 50%, 75%, 100% (5 patterns) x RGBCMY (6 colors) = 30 patterns. These are used for calibrating the color gamut using AE4000's Color Management System's "RGBCMY mode" (so that's where located the individual Color, Tint, Brightness sliders for R, G, B, C, M, Y). There is no saturation pattern namely "from 10-100%". The only set of patterns that you measure for 10-100% is the grayscale patterns (actually from 0% black to 100% white), in which you use Red / Green / Blue Contrast sliders to adjust the high end of your grayscale, and Red / Green / Blue Brightness sliders to adjust the low end (these slides are those located just at the top of the Advanced Menu, nothing to do with the controls in Color Management), so that your white point for 10%, 20%, ..., 100% stimilus level sits closely to the D65 point. The concept of "grayscale" (or white balance / white point) has nothing to do with the color gamut (although the controls for adjusing grayscale and the controls for adjusting gamut may, and likely will, interact / interfere with each other).

I suggest you first get familiar with the difference between grayscale and color gamut before you mess up your AE4000 settings. It seems that you use the adjustment controls for the wrong purpose. The following posts will help:

http://www.curtpalme.com/forum/viewtopic.php?t=10457

http://www.avsforum.com/t/852536/basic-guide-to-color-calibration-using-a-cms-updated-and-enhanced

http://www.avsforum.com/t/1134710/epson-calibration-guide-1080-1080ub-6100-6500ub-7500ub#post_16166537
(AE4000 behaves very similar to these Epson models with respect to CMS behaviour)
dominickwok is offline  
post #43 of 59 Old 06-29-2012, 10:18 AM
AVS Special Member
 
HDTVChallenged's Avatar
 
Join Date: Jan 2002
Posts: 8,889
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 195
Quote:
Originally Posted by dominickwok View Post

DON't use the Saturation-luminance graph in HCFR - it has serious bug there and it presents you a totally wrong graph. The problem lies in that it 'normalizes" all the color luminance curves in such a way that it 'assumes' the measured luminance at the 100% saturation level is perfectly correct and then plot the data points for 0%, 25%, 50%, and 75% as relative error to the 100% data point. So if your luminance of Red at 100% saturation level is +10% off from the standard, and you try to 'level' the luminance curve for the other 4 saturation points, you will end up with +10% luminance error for all the saturation levels (even though the HCFR graph will tell you "a perfect alignment with standard").

It's not a "bug" it's a feature. smile.gif Seriously, as far as I can tell it was just meant to show how much the saturations are drifting wrt to the 100% points. But you are correct in that it shouldn't be used for any other purpose.

Also, be wary of the primary/secondary Delta Luma's ... these *are* referenced to the "standard," but in most cases you'll probably want use targets "normalized" to the primaries you actually have, not the ones you wish you had. smile.gif
HDTVChallenged is offline  
post #44 of 59 Old 06-29-2012, 12:51 PM - Thread Starter
Member
 
jeansibelius24's Avatar
 
Join Date: May 2012
Posts: 16
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Sorry my previous post was a bit confused. I was aware that the saturation levels go 0% then 25% etc and the greyscale goes from 10-100IRE%. The problem I am faced with is that when trying to do the Stereomandan’s 75% saturation method I cannot get the RGB level bars to the correct percentages. I just wondered if there was an alternative method I could possibly try.

I have read and followed those guides that you recommended but I will re-read them as there is alot of information to take in.

Thanks for your help.
jeansibelius24 is offline  
post #45 of 59 Old 07-12-2015, 07:38 AM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Smile Thanks for the help!

I know this is an old thread (and an old projector) but this thread was a huge help to me in getting my PT-AE4000U calibrated. Before this I had only used the various calibration disks with color filters. I've had this projector a while but it was in a place where it got only sporadic use. Thus it only has about 300 hours on the original lamp (and I bought it when they were giving spare lamps for free so I have another lamp for it too). I always loved the image, though this one definitely isn't one of the best optically or in terms of panel alignment. Still it looked great for movies, but I wanted to see if it could do even better. I want this projector to last me until 4k is standard (and cheap!).

I started with the main guides (which are linked earlier in this thread). Most were close to what I needed, but there were a couple nuances I didn't understand. The main ones, which this thread explained to me, were the very low gamma of this display by default, running out of red up high, and the strong interconnection between controls (when setting greyscale/gamma).

I purchased a Colormunki Display for this calibration and used HCFR. The calibration took pretty much all day yesterday due to this being my first real calibration (so standard learning curve) and because for the three quirks of this projector listed above.

My final procedure (I won't detail all of the failed methods), from memory, was this:

First I set Cinema 1 to all defaults and then turned off dynamic iris, frame create, and detail clarity for the calibration. Lamp mode I set to normal as I knew I would be giving up lumens for red (and I have a free lamp sitting in the closet).

The only numbers I changed in basic were the overall brightness and contrast. Using the standard method (from any of the disks) I got a brightness of 0 and a contrast of 10 to get the full light output range of the projector. From some earlier testing I knew a contrast of 0 would limit lumens enough that I wouldn't run out of red much at all. So to compromise reds vs. lumens I chose a contrast of 5.

For this calibration I set to D75 rather than D65 to try to reduce how noticeable the red loss up high was.

I then did not touch any other color settings (in basic or advanced) and went right into the Advanced Gamma settings. In here I did two things at once using the 9 point settings curve. I set my gamma curve to the reference curve HCFR supplied using the Y gamma and then I tweaked the RGB gammas to set my greyscale. You should end up with fairly smooth curves - if there are any large outliers then something probably has gone wrong. It is also an iterative method - first get Y where it gets you the correct lumens output, then adjust R and B gammas to get your color percentages correct. Of course as you increase/decrease the RB gammas you will change your lumens output, so you will have to tweak Y, R and B until everything is balanced. For me I started running out of red around the 80% point. So at 80-90-100 I mainly focused on getting B to match G. For the red gamma I just set it to smooth my red gamma curve back to zero offset. The main goal here was to try to make the red dropoff as smooth as possible. I might have forgotten something here, so if anyone wants any details on when I did (if anyone out there is still trying to calibrate this projector ) let me know.

With that done I had what I felt was a very good greyscale and gamma (my gamma is a little below the reference curve, however I wanted that as my room has some lighter colored surfaces). Obviously the top two points aren't great due to the red loss, however I found that red cast whites seem to bother me more than blue cast whites, so the top end white still looks okay to me. I'll post all of the "FINAL" charts at the end of the post (I also have the "BEFORE" charts if anyone is interested, except for color saturation which I forgot to do).

After that I followed the forum post on color calibrating Epsons pretty much exactly. It was only a quick run through though I think the results were okay. Cyan is still off a bit, but the rest look okay on the plot to my untrained eye. The Epson guide said your 100% color saturations will be too high but 75% and lower should be pretty good. That I think I managed to do, though like I said it probably needs another pass to really lock things in.

After a long day of work I decided to use Interstellar as my "test disk". I have never seen it before (I wanted something "good" visually that I had not seen before, so that I would not have any preconceived "correct" colors). Probably a bad idea as I got pretty into the movie and stopped paying attention to what I thought of skin tone accuracy in every scene. However when I was paying attention to skin tone I thought it looked pretty good, so hopefully I have things set not too far off the mark (though I will tweak color a hair more - just not this weekend!). I will try these settings with a variety of movies to see if I am happy before I make any changes, but Interstellar definitely was a check-mark. I have no idea how much the changes to color will effect my movie watching, but with how far off Gamma was from preferred (I think default was way down around 1.5 or something like that) I know the image is VERY different from what I had before. And greyscale is drastically improved too.

As promised I've attached all of the "AFTER" charts to this post. Comments and critiques are welcome(ish)
Attached Thumbnails
Click image for larger version

Name:	Grey Scale Table.PNG
Views:	15
Size:	35.5 KB
ID:	825978   Click image for larger version

Name:	Luminance Plot.PNG
Views:	18
Size:	505.5 KB
ID:	825986   Click image for larger version

Name:	Gamma Plot.PNG
Views:	16
Size:	494.0 KB
ID:	825994   Click image for larger version

Name:	RGB Plot.PNG
Views:	16
Size:	501.7 KB
ID:	826002   Click image for larger version

Name:	Color Temperature Plot.PNG
Views:	17
Size:	501.8 KB
ID:	826010  

Click image for larger version

Name:	CIE Plot.PNG
Views:	15
Size:	698.3 KB
ID:	826018  

Last edited by EXT64; 07-12-2015 at 07:47 AM.
EXT64 is online now  
post #46 of 59 Old 07-12-2015, 09:49 AM
Senior Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 351
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 175 Post(s)
Liked: 24
Quote:
Originally Posted by EXT64 View Post
As promised I've attached all of the "AFTER" charts to this post. Comments and critiques are welcome(ish)
Thanks for the post. I had an AE4000 previously, before I got into calibration. I didn't realize that, in spite of Panasonic's claim regarding "red rich" lamp, the red colour is still deficient. Can you also post the "before" and "after" .chc files?

I wish HCFR could display the "unnormalized" R/G/B luminance graphs; the normalization kinda distorts the picture in cases like this (see 2nd attached picture).
[EDIT 2015-07-12 17:35 EDT] The latest version of 3.3.4 (posted today) normalizes the luminance graph to Y-Target instead of the measured Y. The luminance graph now shows the red deficiency much more clearly (see first attached picture)
Attached Thumbnails
Click image for larger version

Name:	Color Measures FINAL FINAL ALL.png
Views:	13
Size:	23.1 KB
ID:	826770   Click image for larger version

Name:	Luminance Plot_Normalized.PNG
Views:	7
Size:	505.5 KB
ID:	827258  

Last edited by Dominic Chan; 07-12-2015 at 07:13 PM.
Dominic Chan is online now  
post #47 of 59 Old 07-12-2015, 10:48 AM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Here they are. Unfortunately I forgot to save my "Before" Color Saturation. I'll grab that at some point in the future. I also can't remember exactly what my before was (Default Cinema 1 or Default Color 1). Sometime soon I'll grab a new one so I am certain.
Attached Files
File Type: zip PT-AE4000U_Before_and_After.zip (27.6 KB, 6 views)
EXT64 is online now  
post #48 of 59 Old 07-12-2015, 06:22 PM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Thanks for the note in the PM (unfortunately I can't respond to PMs as I only have 8 posts) - I now have the latest version.
EXT64 is online now  
post #49 of 59 Old 07-12-2015, 07:15 PM
Senior Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 351
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 175 Post(s)
Liked: 24
Quote:
Originally Posted by EXT64 View Post
Thanks for the note in the PM (unfortunately I can't respond to PMs as I only have 8 posts) - I now have the latest version.
I have updated my post to compare the new plot with the old plot.
Dominic Chan is online now  
post #50 of 59 Old 07-13-2015, 06:40 AM
Senior Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 351
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 175 Post(s)
Liked: 24
Quote:
Originally Posted by EXT64 View Post
For me I started running out of red around the 80% point. So at 80-90-100 I mainly focused on getting B to match G.
What exactly happens when red runs out? It not longer responds to the control? Also, instead of getting B to match G, have you also tried to minimize dE instead?
Dominic Chan is online now  
post #51 of 59 Old 07-13-2015, 03:27 PM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Basically that is it, though I'll try to get some details this weekend. I'll also see if I can get the high a little better. I was exhausted after this 'pretty good' cal and it seemed like the upper red controls were having minimal effect, so I didn't push my luck. But now I can write down/save the settings and give it another poke.
EXT64 is online now  
post #52 of 59 Old Today, 04:08 AM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
So quick update on my color calibration.

First to see the red limit: to recheck this I set all the RGB brightness/contrast to default (0) and the main contrast to +5 (my compromise point). HCFR then reported color balance of approx. 82%/102%/100% for the 100% greyscale. I then started increasing R Contrast. After a few clicks it got up to 88%, and then increasing up to max did nothing more. I also played with R gamma at 80% and 90% greyscale and could improve those a little. However messing with any of the high end red controls seemed to have subtle but weird effects on lower red content. I didn't have the time to mess with it so I went back to my old settings with only very minor tweaks.

I did tweak my color saturations to be a little better. I think they are fairly close to the limit of my projector's CMS (with my knowledge level at least).

I did do a quick look at the default CMS settings. Wow, they are pretty far off the mark, so at least I know I made an improvement in gamut, gamma and greyscale.

I also attached the latest .chc file, though it is not much different than before (I changed a lot but then changed a lot back). Still could use some small tweaks to gamma at least.

Now a couple questions:

How am I supposed to collect data for the Saturation - Luminance plot? When I look at it right now there is very little data on the plot and I am way off the targets. Am I that far off of the mark (if so, what mark?) or is there another measure I need to run (if so, what?)?

And wow, Dynamic Iris ON releases a lot more light on full white. It cracked 41 cd/m^2 whereas I am around 33 cd/m^2 with it off. It also really messed up my gamma and greyscale. Though I guess full screen grey scenes in 10% increments isn't the most realistic movie watching scenario. Those who have irises on your projector what do you do? Use it? Try to tweak calibrations for it? Do the best cal with it off then flip it on? I might try a few movies with it both ways and see what I think. It seems to be a decent iris (where you don't really notice it change much during a normal movie, not like some led backlights).
Attached Files
File Type: zip Color Measures FINALCUBEDALL.zip (12.8 KB, 3 views)
EXT64 is online now  
post #53 of 59 Old Today, 04:28 AM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Ah, ok. I figured out a little of my first question. In a right click menu I could turn on all of the colors. However it still looks like I am comparing a 0-100% reference to my 0-30 cd/m^2. Must be another setting somewhere?
EXT64 is online now  
post #54 of 59 Old Today, 06:32 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,777
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 929 Post(s)
Liked: 795
The reason the saturation plots are strange is because you haven't run a primaries/secondaries scan and the program is using the default 100 cd/m^2 white level as reference. You should do this run when doing CMS adjustments to get the primaries and secondaries aligned to reference (within adjustment capabilities). Then re-run saturations and color check to evaluate the entire gamut. Also looks like you should be able to improve grayscale quite a bit and this will help align the secondaries.

Regarding the iris, I would align everything as best I could with the iris off and then turn it on and run the full series of measurements using APL patterns to check how things behave. From there you can decide if calibrating with APL patterns/iris on makes any sense.

Last edited by zoyd; Today at 07:01 AM.
zoyd is online now  
post #55 of 59 Old Today, 07:31 AM
Senior Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 351
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 175 Post(s)
Liked: 24
Quote:
Originally Posted by zoyd View Post
Also looks like you should be able to improve grayscale quite a bit and this will help align the secondaries.
zoyd,

The runs were made with D75 as the white point, and the grey scale is actually quite close except at 90 and 100 IRE where the red "runs out".

Just wondering if there's some way in HCFR to identify the target color space used for the runs, so that that results can be evaluated accordingly.

Last edited by Dominic Chan; Today at 07:38 AM.
Dominic Chan is online now  
post #56 of 59 Old Today, 08:27 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,777
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 929 Post(s)
Liked: 795
Quote:
Originally Posted by Dominic Chan View Post
Just wondering if there's some way in HCFR to identify the target color space used for the runs, so that that results can be evaluated accordingly.
no but this is on my todo list.
zoyd is online now  
post #57 of 59 Old Today, 09:14 AM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Thanks both of you. The software told me I had to do a greyscale first before Saturation but I missed doing the primaries/secondaries scan. I will give that a try tonight. I think I missed doing them because I did the 75% Saturation method (as calibrating 100% will crush the lower levels of some colors on this projector) for setting the CMS.

And sorry about the confusion on the color space. I should've mentioned that in the most recent post.

I'll keep playing with the iris too, though probably not much tonight - I need to get another movie in
EXT64 is online now  
post #58 of 59 Old Today, 09:21 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,777
Mentioned: 61 Post(s)
Tagged: 0 Thread(s)
Quoted: 929 Post(s)
Liked: 795
Quote:
Originally Posted by EXT64 View Post
I think I missed doing them because I did the 75% Saturation method (as calibrating 100% will crush the lower levels of some colors on this projector) for setting the CMS.
You can also select the REC709(75%/75%) colorspace option to do the alignment using 75% saturation 75% luminance patterns.
zoyd is online now  
post #59 of 59 Old Today, 10:29 AM
Newbie
 
Join Date: Mar 2015
Location: Blacksburg, VA, USA
Posts: 13
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Oh great, I'll take a look at that too, thanks!
EXT64 is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration



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