X-Rite's Revolutionary New Colorimeter - Page 44

I've not heard anything about calibrating the meter using a 30% stimulus pattern. Might be unique to your software. Maybe if you can give me a link, I can make sense of it.

I've heard different about plasma and contact versus non-contact mode.

You can still use small apl windows if that's the type windows you want to use. Just like the contact versus contact modes for the meter, you can find quite a few opinions on whether to use small versus large windows and APL versus non APL windows. You may have to try various ones until you find one that gives you a good looking calibration.

I think this only applies to HCFR, not Calman or Chromapure.


bob

I'm using HCFR+i1d3
You can read it as an answer to my initial post:
http://www.avsforum.com/t/1342422/x-rites-revolutionary-new-colorimeter/1260#post_22743484

Then, No black, No white, No 30% sitimulus pattern. How is the way to calibrate the meter?

1. The i1D2 has to be calibrated with a black surface. The i1D3 needs to be calibrated using a white patch, and it's recommended to use a 30% white pattern (window, field, APL.. whatever you're using). With plasmas, you should use windows or maybe APL. LED/LCD or projector, you can use fields.
2. Yes you need to calibrate the meter before you start each session. When you create a new calibration in HCFR, it will ask you this.
3. When I say 'periodically calibrate the meter' I mean the HCFR internal sensor calibration. The i1D2/LT needs to be re-calibrated every 15-20min to ensure accuracy, but AFAIK, you don't have to do this with the i1D3.
Actually getting your meter re-calibrated/serviced by comparing it to an accurate meter (like a spectro) and getting offsets to compensate for meter drift is a different topic. This doesn't need to be done that often (several months, or years) because the i1D3 has sealed filters. To prolong your meter, it's best to store it in a dark, cool and dry place. I store mine with incandescent packs in a ziplock bag.
4. Yes, if you use a correction matrix, you will have to redo the calibration. There's a checkbox on the top right of HCFR that says "XYZ adjustment". You can see the how the matrix/offsets affect the measurements by simply checking or unchecking this box.

IMO, it's fine to use the i1D3 in contact mode with a plasma, as long as you let the meter warm up a bit on the screen. In my experience, even if you started measuring right away without letting the meter warm up, the difference in measurements is minimal. 20min or 30min tops should be enough time for the meter to warmup IMO.

Some other changes you will have to make in HCFR. Go to Advanced -> Preferences:
-On the General tab, uncheck 'use measured colors (when available)'
-Click on the References tab and make sure you select REC 709 for HD. You can also set your target gamma here.
-Click on the Advanced tab and check 'do not use luminance in delta E formula'

Would definitely be nice if there was an updated guide for HCFR. It's a very effective free tool, but one of the problems is that it takes a lot of searching and reading to find out how to set it up properly.
This post has a good summary of the things that should be changed in HCFR.
Edited by rahzel - 12/26/12 at 11:19am

Thanks a lot for all the info, Rahzel.
2.- HCFR does not ask me for recalibrate before each session, but I'm aware and will do it.
4. I've used the correction matrix for plasma loading the first mhc file for panasonic compiled by Make73.

I'm trying with cinema and normal mode on the display to choose one.

I have a new question. Curtpalme Guide indicates in his guide to do the first basic contrast/bright calibration getting the Y value of 10% IRE the 0,65% of the Y value for 100 IRE. In this way my gamma is accurate near 2,2, but I need to down bright to get the Y value and when using pluge patterns for bright calibration blacks are crushed. Wich is the best method?

To set Brightness and Contrast, I would just use the black and white clipping patterns on the AVSHD709 disc. For Contrast, the white clipping pattern shows bars representing levels of white from ~230 to 254. You ideally want all of them to be visible without any color shifting. Because video uses a range of 16-235, you at the very least want 235 and below flashing without any color shift but you should leave a bit of headroom in case there is some whiter than white video information being displayed. If you can set your contrast higher than your eyes can handle for your viewing environment without any negative effects, then just set the contrast to a comfortable brightness.

Here's another mini guide by Tom Huffman that you might want to check out.

That is the point. When using this method my gamma is under 2,2. If I use the method reported before (0,65% Y value for 10% IRE over 100 % IRE) gamma is better but blacks are crushed.

Having a slightly lower gamma is better than crushing black detail IMO. Personally I think the pluge/clipping pattern method is better for setting Contrast/Brightness.
Edited by rahzel - 12/26/12 at 5:27pm

Hello. I tried the 0.65% of 10% of white and it does crush blacks . It is not an accurate method. I would use avs black clipping pattern instead. Tom needs to update some of the info on those articles. I think he does mention an alternate way using a 0% black pattern lower brightness to the point where the black doesn't get any darker. I forget exactly what it says. I don't want to mislead you . I took this out of the chromapure thread. There is a little discussion on what is one percent above black and would it better to focus on 17 barely visible and 16 invisible from the black clipping pattern . This is what Tom had to say. You can read the rest i think it's on page 124/125. He also mentions to focus on 18 instead of 17 but that is not in the quote below

1% above black is more like 18.

8-bit video spans 16-235. That is 219 code values, so there is not a one-to-one relationship between a code value and percent values. 17 is one code value above black, so close in fact that it may be hard to visually distinguish from 16. If you can see 18 but cannot see bars below 16, then that's good. If brightness is set too high, you'll be able to see bars below 16. If it is set too low, 18 will fade into the background and become invisible.
Edited by hungro - 12/26/12 at 7:55pm

Thank you, Rahzel and hungro.
Is what I think after setting contrast/bright using these 2 methods. To my eyes is better the pluge/clipping pattern.

A last question. I'm planning to calibrate an LCD display too.
When you Rahzel say I guess I must choose Refresh type for LCD too. is that right?
Edited by alamagar - 12/27/12 at 12:29am

No, LCD is a non-refresh type. Only use Refresh type on CRT or Plasma (and maybe DLP?).
edit: http://sourceforge.net/p/hcfr/wiki/Home/

I'm using an i1d3 in contact mode with an ST30 plasma and HCFR. I simply cannot get stable readings below 40% IRE. Red especially fluctuates as much as 10%, I have let it sit for a few minutes hoping that it would stabilize, but no luck. How do I fix this?

it's the ABL terror of the plasma... use ur meter off-screen and calibrate as good as u can with ABL...

I don't understand what ABL would have to do with it. I'm not changing IRE's, Im just doing a continuous measure on 30%, and it fluctuates all over the place. I don't have this problem with my i1D2.

ABL is always there, I did a very similar test recently and the luma value kept continuously dropping...

I don't think it has anything to do with the problem i'm experiencing though. It doesn't do it on higher IRE's. I also tried it non-contact with the same results. I also tried 2 different sources (PS3 and HTPC) so I think that there must be something else going on. I had no problems with this meter or the USB extension cable that I'm using when I did my dad's LCD the other day. I also don't see this problem when using my i1D2 on the same plasma i'm trying to use my i1D3 on.

ttnuagmada

It could be one of several things.

1. Be sure that your plasma has been turned on for at least 50 minutes to warm up.

2. Have your meter connected to your computer and turned on for at least 30 minutes.

3. XRite issued some new code that helps the i1d3 with plasmas. Its in Chromapure and Calman but don't know if it made it into HCFR.

4. It's very odd that it would show up at 40% unless you've got your brightness set too low. It's not that unusual to get bouncy reading when the screen brightness is below the operating range of a meter but this shouldn't be happening at 40%.

5. You might want to review the i1d3 setup in HCFR.

As far as i know HCFR uses the libraries of argyllcms.
Have you checked if this problem also exists in the original program?
If so you could post your observations in the argyll forum.