| Forums | Polls | Register | Reply | Search | Statistics |
Display Calibration Light Illusion Forums / Display Calibration /   
 

Want to pull my hair out

 
 
Page  Page 1 of 2:  1  2  Next »

Author toby5
ZRO
Male
#1 | Posted: 5 Oct 2020 22:30 
Ok, I'm sorry for asking like this, but could we please go through some kind of step by step procedure. I know there are the videos, I've watched, and manuals, I've looked through them. I even know how to manually calibrate for the most part which is why I thought this wouldn't be so hard, but I've just tried calibrating SDR and HDR 2 times in 2 long days and there is something wrong with my calibration. I keep seeing weird artifacts in things. Yesterday I could see pink in shimmering white during the football game. Today it's in faces. This is on regular television. I'm not sure if I'm doing something wrong in the program, which I probably am, because the workflow is so different from what I'm used to, or if it's something wrong with one of my meters, because I tried doing probe matching the 1st day, and could not get it matched, I don't know how to verify with MPVM and kept getting errors every single time with 4 color method. I am so frustrated because I was hoping I could just hop right in, to an extent, and work my way from there, but that is absolutely not the case here. So let's start with 1st question, when measuring luminance that I want, measure with 100% white to see the luminance. When I get it to where I want it, do I need to put that into Luma Target? I wanted to do 80 nits and 400 nits calibration, but I ran into trouble every step of the way. When doing 2 step and 21 step grayscale, should I do it with or without the chroma only option? I tried it without, but couldn't get my 2 point grayscale matched up without it on, 21 worked ok. I tried it with, can get 2 step, but luminance doesn't go where I want it, but 21 gave me trouble. I know the TVs have wonky controls because the 1st TV I ever tried to calibrate had it, and this Sony X950G seems to have wonky controls with 21 grayscale. I'm not even sure the right value for 0 black. I watch Steve put 0.05000, but I'm not sure if that was for OLED or what? I know this is exhausting, and I'm sorry, but again, I am so frustrated right now. Also, there's weird chroma in white going on with this new calibration. I'm not only frustrated, but kind of sad right now because that is so much time just gone. Please help!

Author toby5
ZRO
Male
#2 | Posted: 5 Oct 2020 22:44 | Edited by: toby5 
I'm not gonna let this get the better of me. It seems like when you get frustrated, nothing works right. I've tried slowing down and taking it step by step, but I'm still getting it wrong on my calibrations. When trying to do the 80 nits, the controls are wonky on the TV, meaning, just like the other TV I've calibrated on, the controls don't actually go 20 for 20, you have to move down to 19 and 18 for 20 to work, and there's a crossover somewhere in the controls of the tv. When I have tried to calibrate the grayscale, I'm getting stuck on the low end for 80 nits, and even high end. It's either not enough luminance for my targeted nits value, and I'm maxing out in the RGB. I feel really stupid right now. I know this is a lot, but if we could try to point me in the right direction in novice terms, please

Author Steve

INF
Male
#3 | Posted: 5 Oct 2020 23:39 
When doing manual calibration you are really only using ColourSpace to show the values being read.

To get the Target/Actual values to correctly define how close, or not, you are when making adjustments, you need to set the Luma Target Nits (Min/Max) in Settings to the desired values.
The variation of the Actual values to the Target values are defined by those, in combination with the Target ColourSpace.

In most instances the Min/Max values are the TVs measured Black/White values, after you set them to the approx. wanted values, and can be automatically updated every time black or white is measured, by selecting the '^' tick box in Settings.

Then, when measuring the patches, the patches must match the TV's expect input values.
If they do not match, the results will never be right.
This is critical.

It then a question of just getting the correct readings for each patch while taking manual measurements, and adjusting the TVs CMS controls.
I would not use probe matching initially, until you have things close - use it after to verify and make small corrections.

Using things like Chroma Only for the RGB Bars is a user preference, as they Luma bar always show that anyway.
Some users like to have Luma in the RGB bars calculation, some don't...
Your choice.

But yes, it is very likely that changing one level on the TVs CMS will affect other levels, as the controls tend to work over a 'range', and not a single level.

But all of this is why we focus on LUT based calibration.
Manual TV calibration really is extremely limited in the level of accuracy it can attain.
And many TVs have very poor controls.
An external LUT box, or home cinema system is a far better bet.
(Or a TV with internal LUT capability... if you can put up with LG TVs...)

Steve
Steve Shaw
Mob Boss at Light Illusion

Author toby5
ZRO
Male
#4 | Posted: 6 Oct 2020 03:32 | Edited by: toby5 
ok, I basically had a meltdown earlier, I'm sorry. I hope I am on the right track here. I just did a simple calibration for 80 nit SDE and 400 nit high SDR for HDR. I went back to basics with this one only fixing red and blue bars as needed, and green last, only as needed, always watching the DE to try and get the lowest acceptable reading for me at this time. I ignored the luminance on this go round, only making bare minimum adjustments to RGB. Once you start messing with green it seems to get a little messy. I calibrated with the widget in standard, then clicking on chroma only after getting RGB down to check further, and fixing those, then moving on to CMS.

I still didn't get my probe matched, but I'm going to try and keep adjusting until I see that both meters give me the same luminance, and hope that does the trick for me. Please let me know if I am going in the right direction.

Author Steve

INF
Male
#5 | Posted: 6 Oct 2020 09:09 
The initial approach seems to make sense.
(Always difficult to know when just following written information.)

toby5:
I still didn't get my probe matched...

But that I don't understand.
What can you not get to work with probe matching?
Just use FCVM matching - it really is very simple.

toby5:
...but I'm going to try and keep adjusting until I see that both meters give me the same luminance, and hope that does the trick for me.

And I do not understand that at all.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author toby5
ZRO
Male
#6 | Posted: 6 Oct 2020 15:01 
alright, so I'm using the last 2 examples as an instance. Every time I tried to match my probes, I can't get the xyY within nist requirements. They always come up in red. The thing that I noticed on the last 2 examples is that White Y was always off from each other. This is why I was thinking of trying to read 100% White Point until they match up in the program, and then trying to probe match from there. I have tried setting up my iPro2 with a laser pointer and measuring tape to make sure it's centered, but I still can't get a match, or I'm reading the verification wrong. I'm using displaycalibrationtool website to verify the meters. I could be reading it wrong, but I've been checking the xyY values, and they're always usually in the red in spots.

Author Steve

INF
Male
#7 | Posted: 6 Oct 2020 15:10 
Are you saying that the probes do not match each other, or that the readings taken with the Tristimulus when match to the Spectro do not match the Spectro?
You probably need to define exactly the steps you are taking.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author toby5
ZRO
Male
#8 | Posted: 6 Oct 2020 15:34 | Edited by: toby5 
I couldn't link the website. I'm setting up the tristimulus per the distance given on the website stated above for a 5% window, then doing the same thing with the spectro. I'm doing both of these at 120 nits to read lower colors, like blue. I am not able to get everything in the green on xyY. I was also having problems with Calman trying to do the same thing, but was able to get it done easier than I am with Colourspace so far. I still didn't get high nits exactly correct with this last calibration, but it did fix errors, like bringing out contrast for HDR. They actually look like the presentation that they're supposed to! I hope the contrast is repeatable and not just some fluke. There were always some testers scenes that had problems, but not look like they're supposed to, and that contrast was what it needed, or bringing the green from RGB up.

Author Steve

INF
Male
#9 | Posted: 6 Oct 2020 15:51 
Sorry, but that is not actually descriptive enough...
At what point are you checking the probe readings, and how are you checking them?
You really do need to define every step you are doing, in some detail.
At the moment we really have no idea of the steps you are taking.

(And to add a link to a post, use the 'Http' button above the message post window.)

Steve
Steve Shaw
Mob Boss at Light Illusion

Author toby5
ZRO
Male
#10 | Posted: 6 Oct 2020 15:56 
I'm verifying probe readings after taking both readings from Colourspace. The distance I have the iPro2 is 24 inches from the screen, and the distance for the i1D3 from the screen is 12 inches. is the website I'm using to verify meter readings. This is the same website I'm getting distance from screen as well. I had to take out the link again because of spam protection for under 25 posts.

Author Steve

INF
Male
#11 | Posted: 6 Oct 2020 16:19 | Edited by: Steve 
Sorry, what I mean is what settings are you using with ColourSpace?
What is the configuration for each menu when you are taking the readings?
It is key for us to understand what exactly you are doing.
Sorry to be so pedantic, but without such info any suggestions will be mute.

(And I'll remove the 20 post limit for links - forgot that was there.)

Steve
Steve Shaw
Mob Boss at Light Illusion

Author toby5
ZRO
Male
#12 | Posted: 6 Oct 2020 16:55 
For the i1D3 I'm calibrating at 120 nits rec709 1st and foremost to get the white point as correct as possible with 2 point grayscale, then I'm setting it up to do probe match. The settings in ColourSpace are integration time at 0.75 with delay of 0.50, I also have Intelligent Int. at 0.500. I do my read for probe matching, save the file. For the i1Pro 2 I'm not using any settings for that probe. I take its readings, save file. Then I check those saved files on that website. The xyY on that website never show that the probes are in green, or that they are in nist recs for xyY. They are always off. For xy, for example, they always come up as like 0.005, 0.009, but never 0.015, or a 1.5 tolerance for any of my files for some reason. I don't know if I'm reading something wrong there or what.

Author Steve

INF
Male
#13 | Posted: 6 Oct 2020 16:57 
That means you are just comparing the raw probe readings.
They will never match, as neither is being correct by the other.

Basically, you are not doing probe matching.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author toby5
ZRO
Male
#14 | Posted: 6 Oct 2020 17:01 | Edited by: toby5 
huh, then what should the correct procedure be to probe matching then? And I ask because I watched the video multiple times and I thought that was the procedure. I was checking the results before clicking on probe matching active.

Author Steve

INF
Male
#15 | Posted: 6 Oct 2020 17:12 
https://www.lightillusion.com/probe_matching.html

Look at the 'Application of Probe Matching' section.
You must apply the matching, so the Tristimulus readings are being 'corrected' by the Probe Matching, making the Tristimulus act in alignment with the Spectro.

As Probe Matching is a 'post-process' in ColourSpace, you can simple take measurement with the Tristimulus (without any probe matching), and then enable/disable Probe Matching after the event, so you can directly see the difference.

This is very unique, and very powerful.

Steve
Steve Shaw
Mob Boss at Light Illusion

Page  Page 1 of 2:  1  2  Next » 
Display Calibration Light Illusion Forums / Display Calibration /
 Want to pull my hair out

Your Reply Click this icon to move up to the quoted message

 

 (?)
Only registered users are allowed to post here. Please, enter your username/password details when posting a message, or register first.

 

 
 
Online now: Guests - 1
Members - 0
Max. ever online: 264 [6 Jun 2020 12:02]
Guests - 264 / Members - 0