ColorMunki - How Good or How Bad? - Page 3

Disposable income for calibration and I can't seem to get it right. Color looks really good to me however.

I got CalMAN to work with the ColorMunki on my desktop PC (instead of the laptop) using Windows XP Pro.

This is my calibration result for a BenQ W5000 DLP. I did CMS, color temp, then ran the auto calibration on my HTPC. After that I went back and did some touch up to magenta using the CMS again. My original measurements while not far off for primaries and secondaries, measured closer to 5000K and had cyan push. Comments are welcome and appreciated.

It appears that you didn't get a valid reading for 80% white. I think your next step should be to do an 11-step grayscale run so you know what's going on with the middle grays as well.

I measured 80% before I did anything. This is what I got after RGB gain adjustments to my color temperature.



So I should go back and do this again? I thought that would be something done for other than for a spectrometer which is supposed to have good greyscale?

This is done with dynamic black on btw.

Note that there is no data captured for the 80% point, which is why I mentioned it.

A grayscale run with more than two points is a good idea (the type of meter used doesn't have anything to do with it) in any case. Unless your display has a very flat grayscale response-most don't-you need to know how the display is responding all along the grayscale, and using only the 30% and 80% points simply doesn't give you enough information. Because cuts (bias) and drives overlap in the middle grays, increasing them at both ends usually results in a "hump" in their response there.

I see what your saying now. I looked at where to adjust for 80% but could not find this adjustment. I started with the temp and greyscale, then I measured the 80% for each primary and secondary which I thought gave more than 11-step greyscale? I also thought by 80% you meant my 75%W measurement which to my understanding is not that different from 80%. Calibration for black level for my HTPC I think does actually improve greyscale versus a meter that does not, but that's probably not very relevant, and again this is from my limited knowledge on the subject. Do these results look appropriate for a high power screen? I expected really to get gamut dialed in, then hope for the best with greyscale using the automatic feature on the HTPC.

By 11-step grayscale, I mean that you measure at 11 different points (0%, 10%, 20%, etc.). You have to tell CalMAN that you want to use that many points. Also, the layout you're using calls for 2 different sets of measurements: 1 set for grayscale (at 30% and 80%), and a separate set for gamut (that's your 75% W,R,G,B,C,M,Y and 100% White).

I believe you have it a little turned around, Steve. A properly used and accurate meter will help you improve grayscale and gamma as much as the display's controls will allow. Calibrating black level (and contrast) is a first step and by itself does not necessarily improve your grayscale.

The closer you can get your display to D65 and a good gamma curve, the easier it will be for your HTPC to build an accurate profile.

I find the option to select a white point for greyscale but that is all I see. I also have the DeltaE Correction Greyscale box checked.

You guys have completely hijacked this thread. It's more helpful to start a new one with a subject that reflects what you're talking about.

Sorry about that. One post just lead to another...

I have some data on the ColorMunki. It measured fairly well. I was especially impressed by the fact that it did quite well even at relatively low light levels.

There were some hiccups, though. Although it did quite well with color, it was less accurate with white. Also, on average it was slightly less accurate than a stock Chroma 5 I picked at random. However, I must say that this particular Chroma 5 performs a little better than most of the units I have looked at.

Reference	W	R	G	B	Y	C	M	20%
x	0.312	0.638	0.292	0.15	0.409	0.23	0.320	0.301
y	0.331	0.339	0.577	0.07	0.497	0.330	0.162	0.321
Y	1	0.248	0.734	0.085	0.965	0.802	0.300
cd/m2	130.5	32.42	95.77	11.11	125.9	104.7	39.2	4.28
Chroma 5
x	0.315	0.639	0.298	0.148	0.414	0.231	0.321	0.301
y	0.332	0.336	0.573	0.077	0.493	0.331	0.166	0.319
Y	1	0.25	0.735	0.089	0.968	0.812	0.312
cd/m2	128.9	32.2	94.75	11.5	124.8	104.7	40.2	4.52	Ave.
CIE94	1.3	0.8	1	2.7	1.2	0.5	1.1	1.3	1.2
Luminance	1.20%	0.70%	1.10%	3.50%	0.90%	0.00%	2.60%	5.60%	1.90%
ColorMunki
x	0.314	0.643	0.294	0.149	0.41	0.231	0.322	0.301
y	0.339	0.339	0.582	0.071	0.503	0.338	0.163	0.326
Y	1	0.239	0.739	0.082	0.954	0.8	0.284
cd/m2	136.8	32.7	101.15	11.22	130.5	109.44	38.86	4.05	Ave.
CIE94	4.5	1.3	0.8	1	0.9	2.2	1.5	3.2	1.9
Luminance	4.80%	0.90%	5.60%	1.00%	3.70%	4.50%	0.90%	5.40%	3.30%

Perhaps my biggest concern about this instrument has nothing to do with its basic performance. This device was not intended for home theater use, and it is awkward in that application. Its large size and peculiar shape makes using it on large displays typical in home theater applications difficult.

Although it performs well, it is not clear that it performs any better than other instruments of similar cost that are better suited to home theater applications.

Reference instrument: Orb SP-100
Display Tested: Samsung LN32B650 LCD

What do you mean by not intended for home theater use?
Here is my latest results.

results of recent calibration

screenshot

I watched three movies so far with only two cases of RBE witch once looked more than normal compared to a RPTV, but I guess that is fine. I doubt you will get that with a a spyder3 or eye1 that can do that. The best meter they had for my previous front projector was a spyder 2. I tried a spyder 3.

I've written elsewhere about dealing with the (relatively) short strap and potentially abrasive case but I don't find it any harder to use on my 50" panel than a C5. Both are annoying compared to the i1p hanger although you can use the CM zipper to compensate for bezel depth. I also found luminance reading high relative to other meters although it appears internally consistent (same resultant gamma curve).

On my Kuro (using CalMAN) I have slow read problems with 0, 10, 50, and 60 percent using a C5. I've seen similar complaints. The CM appears immune to this problem. It's also very fast.

I think it might be a good enthusiast's spectro (in my world professionals are all using at least a K-10).

Any info about colormunki create? Is it the same as i1 display lt?

Yes same hardware as the Display2 it's a Gamma5 just different software from X-Rite. BTW, CalMAN supports the ColorMunki Create, Design and Photo.

I think all he's saying is it's awkward to use ("Its large size and peculiar shape makes using it on large displays typical in home theater applications difficult") even though it is reasonably accurate.

The white accuracy is rather disappointing for what I assume is a (WCG) CCFL backlight LCD display type. Do you plan on testing the CM with a plasma or other display types?

a little awkward, must have the strap (and usb cable) just right.. Some may have concern about the material in contact with the display.

Hanging on my PRO-141FD

Size Comparison

Tom,
My understanding is that non of the meters that we use (i1pro, chroma5 or Color Munki) were intended for Home Theater use....but they still work very well for that.



With that being said...I just got my Color Munki Photo today and will try and get some testing done this weekend, and compair it to my i1pro.
If the unit is as good as my i1pro...then this would be hands down the new meter for the Enthusiast to get. Personally I like the fact that everything is built in the meter...no more calibration plate or light defuser to loss.

Later
RayJr

Although loath to speak for Tom I find the HT issue (which does not apply to the i1p or C5) is one of mechanics. The (relatively short) strap is permanently attached to the case and the meter has a "high" center of gravity (tall aspect ratio). Although it may not be obvious from Turbe's photo you have to have a separate counter-weight on any screen larger than 20-30 inches* and the meter wants very much to rotate 90 degrees from its operational position (largely doe to the usb cable which goes the wrong way). As a contact meter its ergonomics are notably inferior to any "short" aspect ratio puck style meter.

*Or the ability to put your patterns at the top of the screen.

Strap and usb cable placement is key..


I don't need anything additional for it to work on the 60" (and within the standard 'small size' pattern windows), just make sure the strap and usb cable are in the positions noted in my picture - the ColorMunki sits flat against the display with this arrangement - just need to find that 'right' balance.

Closeup

Okay, I reviewed my placement. I'd be willing to use to AVSHD or CalMAN large patterns (or equivalent) however when using CalMAN presized "small" patterns and the requisite strap position the strap slipped (after a couple minutes). I'd be willing to use some non-slip mat in lieu of a counter-weight but I still wouldn't trust it unaided.

But then I'm just a bit obsessive about some things.

Blue painters tape works very well. I always carry some with me just in case I need to get a little creative.

Tom, what software did you use to get the raw xyY ColorMunki data? Since ChromaPure does not support the ColorMunki Create, Design or Photo. I was not aware the X-Rite CM software would give you raw XYZ or xyY results. If it was X-Rite CM software how did you control the exposure time?


Interesting that you just happen to pick your best Chroma5 in what should have been a objective review. How many ColorMunki's did you have on hand just the one? When was it last calibrated?


That is why a innovative person would create a tripod mount for the ColorMunki just like we do for the Chroma5 and i1Pro. BTW we are working on just such a thing.


It is a Spectro so that alone should be enough even if you get the same results as your "golden" Chroma5.

Not sure what you mean here, the Quantum Data (Sencore)/Accupel (per research many years ago for CRT's and Plasmas done by Joe Kane & associates and Gary Rogers) standard small patterns (size also used by AVS HD 709) are what are referenced as "Small Windowed Patterns".. no smaller (or larger) window size is needed for Plasma/CRTs.

There are some that may try to argue there is no standard for the "small window" size, but there is reserach for this size used by the Quantum Data (Sencore) and Accupel pattern generators (again AVS HD 709 uses these sizes).... there is a reason why the Professional Pattern Generator's small Window Size are the same (though some have options for additional sizes)


Do you mean the CalMAN PC Pattern Generator?

Thought that the Calman software was only good for pc monitors, how about a LCD PJ and a rear DLP????