| Forums | Register | Polls | Search | Statistics | File Bank | 
 (?)  
You must be logged in to post content on this forum.
Display Calibration Light Illusion Forums / Display Calibration /  
 

matching probes

 
 
Page  Page 1 of 2:  1  2  Next »

Author blasetheodore
ZRO
#1 | Posted: 17 Sep 2016 01:00 
Setup:
I have an LG E6 4k OLED. An i1d3, and an i1pro with full CMS suite.
Doing a "primary only" quick profile with either probe shows the OLED clearly exceeds rec709 gamut.

Problem:
I believe I'm following the documentation correctly, but when I do a matching probe profile, the results are wrong. They show a smaller gamut than rec709.

I created a probe matrix profile for the i1d3, and then one for the i1Pro. I loaded the i1d3 into the top "active" dropdown, and the i1pro into the bottom "reference" dropdown. Then I did a quick profile (primary only) with the i1d3. And the result showed a much contracted gamut compared to the clean profiles from either of the probes.

Thinking that maybe I'd got the combination wrong. I tried all possible variations:
i1d3 top > i1pro bottom > profile with i1d3
i1pro top > i1d3 bottom > profile with i1d3
i1d3 top > i1pro bottom > profile with i1pro
i1pro top > i1d3 bottom > profile with i1pro

None of them seem correct. They either shrink the gamut to something that seems wrong, or expand to something that seems impossible.

Obviously this is likely user error. But I'm banging my head through this to get somewhere and I'm not sure what to do next.

Thoughts?

(Happy to attach screen grabs as well.)

Author Steve

INF
Male
#2 | Posted: 17 Sep 2016 06:17 
Sounds like the actual reading taken when doing the probe matching are wrong, or you have also selected a preset matrix within the i1 Display Pro, so doubling the offsets?

Either way, the error suggest the probe value for the offset are wrong.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author blasetheodore
ZRO
#3 | Posted: 19 Sep 2016 15:47 
Hey Steve,

Yes thats definitely my hunch as well. Though can I ask you to explain in more detail what you meant by:
Steve:
or you have also selected a preset matrix within the i1 Display Pro

My process was based on your documentation.. roughly:

With the i1d3, I used the "measure all" button to generate RGBW points, and saved it as "i1d3". I used "switch probes" unplugged i1d3, plugged in i1pro, then pressed "measure all" again, and saved that as "i1pro".
Did I miss something?

Is the update luminance button (on right) a part of this process I'm missing? If so, do I do it for both probes before the "measure all" step?

And just to be clear, I'm using the i1d3 as my primary probe, and only using the i1pro to generate the offsets, correct?

Thanks for reading,
Blase

Author Steve

INF
Male
#4 | Posted: 19 Sep 2016 16:45 
What 'matrix' preset are you trying to use with the i1D3?

You need "General', or you will probably be duplicating offsets.

And as per rhe User Guide, the 'Luminance' min/max has nothing to do with probes matching. That is just for manual measure mode, when performing manual display calibration.

Hope that helps.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author blasetheodore
ZRO
#5 | Posted: 19 Sep 2016 17:39 
Hi Steve,

I mentioned this in an email earlier, but this "matrix preset" is where I keep losing you. There is no UI element I can find that uses the phrase "preset", or offers a "general" option. I'm obviously missing something, though not sure what. And you're not referring to the Calibration settings dropdown, right?

Yes I read the luminance units description, but I asked anyway while grasping for straws.

I appreciate the help and your patience, and I promise I've read the profiling manual, (and about 10 of your other guides). which are thorough and well written.

Author Steve

INF
Male
#6 | Posted: 19 Sep 2016 18:27 
Different Tristimulus probes have different inbuilt presets...

When using probe matching you need to select 'none', or the most basic, or you can get values that are out of range.

With the i1D3 the basic setting is (something like) general, and yes, it is in the 'Calibration' drop-down, as those are the inbuilt pre-sets.

Any help?

Steve
Steve Shaw
Mob Boss at Light Illusion

Author blasetheodore
ZRO
#7 | Posted: 19 Sep 2016 18:46 
Ok thanks Steve.

So when I build the preset for the i1d3, as part of the probe matching process, I use "general".

But once I've gathered those main and offset presets, am I using "general" with the i1d3 when I do the actual volumetric profiling? Or should used the i1d3's OLED setting instead?

Author blasetheodore
ZRO
#8 | Posted: 20 Sep 2016 16:49 
Thoughts on that last question? Thanks!

Author Steve

INF
Male
#9 | Posted: 20 Sep 2016 16:54 
You MUST stick with the same inbuilt matrix you used when making the probe match data, or all will be invalid.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author blasetheodore
ZRO
#10 | Posted: 20 Sep 2016 16:55 
Understood, thanks.

Author blasetheodore
ZRO
#11 | Posted: 13 Oct 2016 17:53 
Hi Steve,

I've been trying to get clean LUTs for a few weeks straight now, and they're still just a little too sloppy to use.

Just to be absolutely clear here, I am using my i1d3 for the actual display profiles, and the only time I'm using the i1pro is to generate the offsets. So in other words I'm using the cheaper colourimeter to do all the work.

Please confirm thats correct?

Author Steve

INF
Male
#12 | Posted: 13 Oct 2016 18:15 
Yes - that is correct.
The i1 Pro2 is used just to generate an offset, as explained in the User Guides.

But, what other settings are you using, and what are you using to generate the calibration patches, what patch size, etc?
All the different setting, patch generation, etc., have to be set correctly too, or you will get poor results.

LightSpace can only work with the profile data it is provided, and if the prove is reading poorly the LUT generated will always be inaccurate.

The Light Illusion website has a lot of information on this.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author blasetheodore
ZRO
#13 | Posted: 13 Oct 2016 18:41 
Thanks Steve,

Same settings as before, except now I'm cranking up the integration time to 2 sec.

The problems I'm having are:
bad greyscale (a 10 step is inconsistent)
strong color banding in shadows up to roughly 30%
general banding, especially noticeable in skin tones and defocused edges. Looks like using the plastic wrap filter in photoshop.

If I do a 0-30% grad, all my other monitors (including my Panasonic Pro plasma with Davio and cinespace) look smooth, but the LS-CMS is a mess of green and magenta stripes.

Thanks for reading,
Blase

Author Steve

INF
Male
#14 | Posted: 13 Oct 2016 18:49 
Initial comments...

Any LS result that is not 'correct' shows the profile data is incorrect in some way.
Looking at the info you have provided in the link there are a number of possible issues.

1 - a 33% patch will be too large for an LG OLED
Go with a patch size that is as small as possisble!
2 - a 15 min time difference between when the probe measurements were made for the probe matching to too long. They need to be done back to back.
Warm up the probe with a different USB port if needed.
3 - What do you mean 'Calibration settings: manual', and why 'Burst' - can you not use AIO?
(If you cannot use AIO you have the earlier i1D3 probe, and that 'could' be an issue?)
4 - if the 'Relax' filter is causing issue that suggests you really do have an underlying problem.

I'm downloading the zip file now. Will comment further when I have look through the profiles and LUTs.

Steve
Steve Shaw
Mob Boss at Light Illusion

Author Steve

INF
Male
#15 | Posted: 13 Oct 2016 19:30 
Another comment, based on the screen grabs only.

Using a value of '10' for drift, so a drift patch is inserted every 10 frames, is way to short a duration!
A value of around 100, or even higher for a display such as the LG OLED that suffers high-frequency instability would be a lot better.

The 'Matrix' screen grab also shows you are using an integration tike of just 0.25 - the smallest setting.
As has been said before, that needs to be a lot longer for OLEDs when using the i1D3!.

You may also need a small amount of Extra Delay Time as those displays can take a short while to 'stabilise' in-between patch changes.

Steve
Steve Shaw
Mob Boss at Light Illusion

Page  Page 1 of 2:  1  2  Next » 
You must be logged in to post content on this forum.
Display Calibration Light Illusion Forums / Display Calibration /
 matching probes

 

 
 
Online now: Guests - 5
Members - 0
Max. ever online: 22 [12 Jul 2022 20:37]
Guests - 21 / Members - 1