Calibrating HDR on Epson 5040/6040 projectors - Page 39 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 138Likes
Reply
 
Thread Tools
post #1141 of 1150 Old 02-14-2020, 10:05 PM
Newbie
 
Join Date: Jan 2020
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 1
Quote:
Originally Posted by ordivergens View Post
Hi All,

I am trying to calibrate my 5040UB for HDR. I get about 63 nits peak @ medium lamp on digital cinema. I tried high lamp which gets me a little more than 10 nits over medium, but then the projector, which is right over my head, is obnoxiously loud... I tried Natural and Bright Cinema which get me closer to ~110 nits with medium lamp, but I found the highlights to be a little too bright to be honest, so I figured I would stick with Digital Cinema and get the wider color gamut.

I had horrible errors on the low end for grayscale (way too much blue and red, way too little green). The only way I could get these under control at the low end was messing with the color uniformity.

The problem is that at the end of my calibration for 1000 nits, 27 nits diffuse white, I am having trouble dialing in the CMS. I tried using the 50% saturation targets but the other saturation levels are not really falling in line. If I try to get those closer to target then the 50% goes haywire... Is there anything I should try to get all the levels to hit their target more closely, or am I just out of luck with my display and screen?

I also see weird color gradients on skin tones with this calibration as well. It seems to be from the custom gamma settings I had to use to get the EOTF under control.

I've got a silver ticket grey screen. I am using an X-Rite i1Display Pro to measure and a CCast Ultra for getting HCFR internal patterns. I chose 'Projector' as the display type and left everything else alone. I manually set my projector to BT2020 color space and HDR 2.

My HCFR calibration is attached if that helps.

Thanks!
Gave it another go with better luck. This time I moved the probe closer to the screen and took more time to properly position it. I think I just had poor positioning previously. I also used HDR 1 (Auto bright) this time. Could have probably tried to get a better roll off at the peak, but good enough for now. Along with the LLDV hack, everything looks great! No more dark HDR scenes, weird color gradients or red push in the image!
Attached Files
File Type: zip DigitalCinema_HDR1_Medium_1000nits.zip (44.4 KB, 33 views)
ordivergens is offline  
Sponsored Links
Advertisement
 
post #1142 of 1150 Old 02-15-2020, 10:32 AM
Member
 
MMoser's Avatar
 
Join Date: Jul 2008
Location: Los Angeles, CA
Posts: 126
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 92 Post(s)
Liked: 30
Quote:
Originally Posted by ordivergens View Post
Gave it another go with better luck. This time I moved the probe closer to the screen and took more time to properly position it. I think I just had poor positioning previously. I also used HDR 1 (Auto bright) this time. Could have probably tried to get a better roll off at the peak, but good enough for now. Along with the LLDV hack, everything looks great! No more dark HDR scenes, weird color gradients or red push in the image!
Wondering if you have any other tips on dialing in the calibration you just achived? I have the same setup and have been struggling with skin gradients and some colors appearing off after measuring for HDR and getting the gamma curve lined up in HCFR. SDR calibration was easy but scratching my head with HDR.

Sent from my SM-N975U using Tapatalk
MMoser is offline  
post #1143 of 1150 Old 02-15-2020, 11:33 AM
Newbie
 
Join Date: Jan 2020
Posts: 12
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 1
Quote:
Originally Posted by MMoser View Post
Wondering if you have any other tips on dialing in the calibration you just achived? I have the same setup and have been struggling with skin gradients and some colors appearing off after measuring for HDR and getting the gamma curve lined up in HCFR. SDR calibration was easy but scratching my head with HDR.

Sent from my SM-N975U using Tapatalk
I only really looked at Digital Cinema, because it was bright enough for me. This is generally how I went about the calibration:

1) Using CCast to get internal patterns from HCFR. Manually set projector to HDR1 (could probably have used HDR2 also), BT2020 color space. Turn off all the image enhancements, dynamic iris, etc.

2) HCFR: UHDTV - Rec2020/P3 standard, SMPTE 2084 HDR EOTF and Delta E to CIE2000, absolute y w/ gamma. I used integration time of 0.5s for my probe.

3) I measured grayscale accuracy at various settings until I found the one that gave me the best response in terms of rgb balance and thus color temp closest to 6500. For me that was with color temp. set to 3 rather than 5. There was way too much blue at default settings. Basically all the default Epson color modes had too much blue and color temps closer to 8000k. Maybe it is due to my grey screen, I don't know.

4) Overide targets in HCFR, turn on BT2390 tonemapping and set diffuse white I wanted and the maximum luminance I measured. I adjusted rgb gain and offset to get the rgb balance errors lower as best I could (using 100% and 20%). This generally got everything in line except for below like 20 to 25% where everything was just out of whack...). The only way I could fix the low end was by tweaking color uniformity. I ended up adjust levels 1, 2 and 3 (only a minor change in 3, most of the changes were in levels 1 and 2) for color uniformity to get the low end correct. Once I did that I went back and checked the whole greyscale again to make slight adjustments as necessary.

5) I then tackled getting gamma correct. I used the custom gamma setting and just tweaked things (mostly C2 through C9, I left C1 alone) until I got the gamma in line with the target. I ended up having to push up the contrast a bit to hit the target. I then went back and checked the greyscale again to make sure rgb levels were still ok.

6) Next I used the 50% saturation color targets to calibrate the colors. I found the best results by getting hue correct first, followed by brightness and then saturation. Once 50% looked ok, I did the same at 75% and 25%. Going back and forth between 0%, 25%, 50% and 75% to get the colors close as I can with the targets and trying to make sure luminance errors didn't get too bad. Usually would have the worst errors at 100%, but I tried to prioritize getting the lower end of the scale more accurate. I did this for red, green and blue. Once I got blue dialed in I went back and checked red and green and made further adjustments. I then did the secondary colors the same way. Once those were dialed in, I once again went back and checked the primaries and the secondaries and tweaked as necessary.

7) Once the CMS was dialed in, I went through the whole process again. Checking greyscale accuracy, gamma, CMS, etc. and making changes where needed. I didn't have to make any drastic changes, just slight adjustments for gamma and rgb balance.

8) When I was satisfied with that, I ran the color checker patterns. As long as most of them were around 2.0 dE or below I called that good enough.

9) Run the full tilt boogie as a final check on everything. If something looked out of place, I went back and made whatever adjustments I needed.

10) Turn back on the image enhancments, dynamic iris and play some movies, etc. to make sure nothing looks off. When it all looked good, I saved my settings to memory. Hopefully won't have to touch it again until I have to replace the lamp or something...

The biggest problem I had (I think) was not properly positioning the probe. The first few times I did the calibration, I would have decent measurements at the end of the calibration, but then the image would look off with weird skin tones, color gradiants, etc. So this last time, I made sure to position the probe more carefully. I put it closer this time (about 4 to 6in away). I then put on the 100% white and continuous measurement and adjusted the probe (angle, direction, etc.) until I found the orientation that gave me the maximum luminance. Then I proceeded with the calibration.

Not sure if any of this helps, but this is what I have arrived at after many failed attempts at trying to calibrate this thing...
ordivergens is offline  
Sponsored Links
Advertisement
 
post #1144 of 1150 Old 02-15-2020, 12:16 PM
Member
 
MMoser's Avatar
 
Join Date: Jul 2008
Location: Los Angeles, CA
Posts: 126
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 92 Post(s)
Liked: 30
Quote:
Originally Posted by ordivergens View Post
I only really looked at Digital Cinema, because it was bright enough for me. This is generally how I went about the calibration:



1) Using CCast to get internal patterns from HCFR. Manually set projector to HDR1 (could probably have used HDR2 also), BT2020 color space. Turn off all the image enhancements, dynamic iris, etc.



2) HCFR: UHDTV - Rec2020/P3 standard, SMPTE 2084 HDR EOTF and Delta E to CIE2000, absolute y w/ gamma. I used integration time of 0.5s for my probe.



3) I measured grayscale accuracy at various settings until I found the one that gave me the best response in terms of rgb balance and thus color temp closest to 6500. For me that was with color temp. set to 3 rather than 5. There was way too much blue at default settings. Basically all the default Epson color modes had too much blue and color temps closer to 8000k. Maybe it is due to my grey screen, I don't know.



4) Overide targets in HCFR, turn on BT2390 tonemapping and set diffuse white I wanted and the maximum luminance I measured. I adjusted rgb gain and offset to get the rgb balance errors lower as best I could (using 100% and 20%). This generally got everything in line except for below like 20 to 25% where everything was just out of whack...). The only way I could fix the low end was by tweaking color uniformity. I ended up adjust levels 1, 2 and 3 (only a minor change in 3, most of the changes were in levels 1 and 2) for color uniformity to get the low end correct. Once I did that I went back and checked the whole greyscale again to make slight adjustments as necessary.



5) I then tackled getting gamma correct. I used the custom gamma setting and just tweaked things (mostly C2 through C9, I left C1 alone) until I got the gamma in line with the target. I ended up having to push up the contrast a bit to hit the target. I then went back and checked the greyscale again to make sure rgb levels were still ok.



6) Next I used the 50% saturation color targets to calibrate the colors. I found the best results by getting hue correct first, followed by brightness and then saturation. Once 50% looked ok, I did the same at 75% and 25%. Going back and forth between 0%, 25%, 50% and 75% to get the colors close as I can with the targets and trying to make sure luminance errors didn't get too bad. Usually would have the worst errors at 100%, but I tried to prioritize getting the lower end of the scale more accurate. I did this for red, green and blue. Once I got blue dialed in I went back and checked red and green and made further adjustments. I then did the secondary colors the same way. Once those were dialed in, I once again went back and checked the primaries and the secondaries and tweaked as necessary.



7) Once the CMS was dialed in, I went through the whole process again. Checking greyscale accuracy, gamma, CMS, etc. and making changes where needed. I didn't have to make any drastic changes, just slight adjustments for gamma and rgb balance.



8) When I was satisfied with that, I ran the color checker patterns. As long as most of them were around 2.0 dE or below I called that good enough.



9) Run the full tilt boogie as a final check on everything. If something looked out of place, I went back and made whatever adjustments I needed.



10) Turn back on the image enhancments, dynamic iris and play some movies, etc. to make sure nothing looks off. When it all looked good, I saved my settings to memory. Hopefully won't have to touch it again until I have to replace the lamp or something...



The biggest problem I had (I think) was not properly positioning the probe. The first few times I did the calibration, I would have decent measurements at the end of the calibration, but then the image would look off with weird skin tones, color gradiants, etc. So this last time, I made sure to position the probe more carefully. I put it closer this time (about 4 to 6in away). I then put on the 100% white and continuous measurement and adjusted the probe (angle, direction, etc.) until I found the orientation that gave me the maximum luminance. Then I proceeded with the calibration.



Not sure if any of this helps, but this is what I have arrived at after many failed attempts at trying to calibrate this thing...
Thank you for listing your steps in detail! I'll give it another go this weekend and report back.

Sent from my SM-N975U using Tapatalk
MMoser is offline  
post #1145 of 1150 Old 02-17-2020, 12:10 PM
Senior Member
 
Jameshtx's Avatar
 
Join Date: Dec 2011
Location: Texas, USA
Posts: 352
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 274 Post(s)
Liked: 27
Quote:
Originally Posted by viperlogic View Post
Here goes first attempt on coming up with a process to do HDR on the Epson. Feedback / corrections most welcome.

Pre-steps
- Using Firmware 1.11 and HCFR 3.4.6.6 (Note: as per Zoyd in post 484, RGBdE bars, NB/NW/satshift/lum graphs, and CIE diagram tooltip data have not been checked or updated for HDR) with internal patterns chromecasted (HCFR: Measures -> Generator -> Configure -> Display -> CCAST)
- Use Natural colour mode or colour mode of your choice (for my env I can get 160 nits for Natural and 70 for Digital Cinema, so you have to make the call of more specular highlights vs WCG)
- Reset chosen mode to defaults (Epson: Menu -> Image)
- Turn Iris off (Epson: Menu -> Image)
- Set Power Consumption to High (Epson: Menu -> Image), medium can be used also but at the sacrifice of nits which are important for the "HDR" affect but then less fan noise.
- Set projector Colour Space to BT.2020 and Dynamic Range to HDR Mode 2 (This was HDR Mode 1 pre 1.11 firmware) (Epson: Menu -> Signal -> Advanced)
- Turn on superwhite (Epson: Menu -> Image -> Advanced). Note, this is optional, jwhn can get the tone mapping curve to match in his setup without it and just using gamma controls, but I couldn't in my setup. For guidance as to use it or not, jwhn gets around 100nits max with his setup and I get around 160nits max.
- Unlike SDR, no need to change Brightness and Contrast (TBC)
- Set HCFR to 20 point (HCFR: Measures -> Parameters -> Number of grayscale levels)
- Set HCFR Colour Space to “UHDTV - Rec2020/P3” (HCFR: Advanced -> Preferences -> References)
- Set HCFR to SMPTE 2084 HDR (HCFR: Advanced -> Preferences -> References)
Grayscale
- Do 21 point grayscale sweep (HCFR: F2)
- Dial in using 30% and 80% points using continuous measure (HCFR: F8) and adjusting Red and Blue offset for 30% and Red and Blue gain for 80% so as to get a dE of 2 or lower, though lower than 1 should be easily achievable (Epson: Menu -> Image -> Colour Temp -> Customised).
- Note: 30% and 80% are typically used but you may get better results finding the best balance between 10/20/30 and 80/90/100.
- Note: There is a noticeable skew at the lower end (0 to 30%) on these Epson projectors. Some have attempted to reduce that skew using Colour Uniformity. See post 455
- May take a few attempts as offset and gain affect each other
Tone Mapping:
- Enable tone mapping in HCFR by checking "BT.2390 Tone Mapping" and "Override Targets" (HCFR: Advanced -> Preferences -> References)
- Set "Diffuse White[nits]" to what you personally would like, one school of thought is set it to what you use for 100% for SDR movies eg somewhere between 20 or 40 nits, note the higher you go the less room you leave for specular highlights
- Set “Target MaxL” to your 100% measure from the grayscale run
- Set “Master MaxL” and “Content MaxL” to 1000 or 4000 depending on what majority of your content is or even do two separate calibrations
- View Luminance graph. Grey dashed line is the target, yellow solid line is measured from grayscale sweep.
- Using continuous measure (HCFR: F8) adjust using customized gamma controls C1 to C9 (Epson: Menu -> Image -> Advanced -> Gamma -> Customized)
- Note: If can't get the curve to match by maxing the gamma controls then turn on superwhite and start over (Epson: Menu -> Image -> Advanced)
- Note: If can't get the curve to match by maxing the gamma controls then use Contrast (Epson: Menu -> Image)
- <Need to add more tips here as this is a black art!>
- Do a grayscale run after changes
CMS
- In HCFR, in the Information window, change the dropdown from "Comments" to "CIE Chart"
- In HCFR, change view to "Primaries and Secondaries", select white and run continuous measure so we get a reading for 50% white aka Diffuse White (HCFR: F8)
- Change view to “Red Saturation Scale", select 50% saturation and run continuous measure (HCFR: F8)
- Using Epson CMS (Epson: Menu -> Image -> Advanced -> RGBCMY -> R) dial in 50% saturation by using Hue (rotates the point around the center) and Saturation (moves it in and out from the center) so the yellow dot is in the second red box from the center of the chart.
- The third dimension, Brightness, cannot be done from the CIE Chart so adjust the CMS Brightness so "delta luminance" in the "Saturation colour" window is as close to 0% as possible for 50% saturation.
- Note, after adjusting brightness you may need to re-adjust CMS Hue and Saturation. This may take several attempts.
- Repeat for 50% Green, Blue, Cyan, Magenta, Yellow
- Run saturation sweep for all Primaries and Secondaries (HCFR: Measures -> Saturations -> Primary and Secondary colours)
- From the saturation sweep, 0%-25%-50%-75% saturations all should have a DE of less than 2. 100% saturations typically have a much higher DE as this projector, especially not having the cinema filter engaged, cant get the wider colour gamut. Green is the worst offender.
Post steps
- Check Grayscale, Tone Mapping and CMS again as adjustment to one can impact others
- Run a Full Tilt Boogie (HCFR: Measures -> Full Tilt Boogie)
- Save the HCFR chc file for future reference
- Turn Iris On (Epson: Menu -> Image) (TBC)
- Set projector Colour Space to Auto (Epson: Menu -> Signal -> Advanced)
- Set projector Dynamic Range to Auto if used HDR Mode 2 or Auto Bright if used HDR Mode 1 (Epson: Menu -> Signal -> Advanced)
- Set Image Preset to personal taste. Natural mode default is 1. Personally I prefer 4.
- Save to memory!
When following this method, is there a certain setting in the blu ray player that must be required? (in my case, Oppo 203)

Epson 5040ub Projector l Oppo 203 l Denon AVR-X4300h l Monolith 5x200w Amp l Def Tech BP9080 L/C/R/SL/SR/SBL/SBR l SVS PB-2000 x2 l Panamax M5400
Jameshtx is offline  
post #1146 of 1150 Old 04-22-2020, 06:37 PM
Senior Member
 
Join Date: Jan 2019
Posts: 325
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 252 Post(s)
Liked: 49
Hey guys,

I’m trying to create a custom gamma curve for my JVC projector, and I’m hoping you guys can help me with some HCFR Questions. First off, If I’m only measuring gamma, should I be using my spyder meter, or my i1D3?

My understanding on how to accomplish my goal here is that I need to create my ideal curve in the HCFR reference preferences, then adjust my JVC’s gamma profile curve until it matches what hcfr is giving me.

The issue I’m running into is that even with my measured peak nits and correctly set diffuse white, my curve still doesn’t match, and I’m not sure if the issue is my projector’s internal gamma curve is out of wack, or if I’m not setting correctly in HCFR.

I’m attaching my measured curve (using the i1d3)
Attached Files
File Type: zip HDR 85 NIT GAMMA.zip (1.3 KB, 5 views)
filmgeek47 is online now  
post #1147 of 1150 Old 04-22-2020, 07:25 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 8,747
Mentioned: 189 Post(s)
Tagged: 0 Thread(s)
Quoted: 6289 Post(s)
Liked: 2648
Reply posted in the HCFR thread as it has little to do with Epson projectors.
https://www.avsforum.com/forum/139-d...l#post59535772
Attached Thumbnails
Click image for larger version

Name:	HDR 85 Nits Match.jpg
Views:	52
Size:	1.03 MB
ID:	2714498  
Attached Files
File Type: zip HDR 85 NIT GAMMA_Match.zip (1.3 KB, 7 views)

JVC DLA-X550R; Denon AVR-X3400H; HD Fury Linker / Vertex 2; ATV4K; Sony UBP-X700
JVC Curves; Optimizing HDR;
Creating a basic HDR curve using Arve’s Tool;
Replacing JVC Bare Bulbs

Last edited by Dominic Chan; 04-22-2020 at 07:51 PM.
Dominic Chan is online now  
post #1148 of 1150 Old 06-17-2020, 03:02 PM
AVS Forum Special Member
 
Grayson73's Avatar
 
Join Date: Feb 2004
Location: Potomac, MD
Posts: 1,162
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 19
Just got a 5040UB refurbished. Whose HDR settings should I start with?
Grayson73 is offline  
post #1149 of 1150 Old 06-17-2020, 03:39 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 8,747
Mentioned: 189 Post(s)
Tagged: 0 Thread(s)
Quoted: 6289 Post(s)
Liked: 2648
Quote:
Originally Posted by Grayson73 View Post
Just got a 5040UB refurbished. Whose HDR settings should I start with?
This thread is about doing your own calibration.

You can find other people’s settings in other threads such as this one
https://www.avsforum.com/forum/139-d...cond-look.html

JVC DLA-X550R; Denon AVR-X3400H; HD Fury Linker / Vertex 2; ATV4K; Sony UBP-X700
JVC Curves; Optimizing HDR;
Creating a basic HDR curve using Arve’s Tool;
Replacing JVC Bare Bulbs

Last edited by Dominic Chan; 06-17-2020 at 04:28 PM.
Dominic Chan is online now  
post #1150 of 1150 Old 06-17-2020, 03:58 PM
AVS Forum Special Member
 
Grayson73's Avatar
 
Join Date: Feb 2004
Location: Potomac, MD
Posts: 1,162
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 19
Quote:
Originally Posted by Dominic Chan View Post
Thus thread is about doing your own calibration.

You can find other people’s settings in other threads such as this one
https://www.avsforum.com/forum/139-d...cond-look.html
Thanks! I'm not up to the level of you guys yet to belong in this thread
Grayson73 is offline  
Sponsored Links
Advertisement
 
Reply Display Calibration

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


Forum Jump: 

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

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