Display calibration - Hardware question

[János, Tóth F.]

Hy,

During the last few years, I have been using different colori- and
spectrophoto-meters (mostly together) for display calibration as
consumer grade products (both displays and measuring instruments) and
my knowledge changed (hopefully all improved but I wouldn't dare to
say it with confidence about either :D).

After I upgraded my small toolset with an i1 Display Pro (i1d3)
colorimeter (replacing an old i1d2), I started to question the benefit
of creating correction matrices with my older ColorMunki Photo spectro
(the old i1d2 obviously needed that but I wasn't sure if the
ColorMunki Photo is any more accurate --- if not slightly but
unequivocally less accurate ??? --- on a random display than an i1d3
alone). I tried to read the opinions and look at test results of as
many individuals as possible but it was very puzzling (some said the
i1d3 is very accurate without corrections while the CM is NOT and some
said the i1d3 still needs corrections and a CM grade spectro is still
much more reliable), so I couldn't decide.

Since I wasn't sure, I kept using the CM Photo to create correction
matrices for the i1d3 and I thought I might upgrade to an i1 Pro 2 in
time (hopefully when it becomes cheaper - for example, an OEM version
without the extra tools for cheap...).
But now, my CM stopped working completely. It must be some kind of
contact problem inside it's case: the device is accessible in the
default measuring position but becomes unavailable in any other
"dialer" positions, so self-calibration is no longer possible (and
thus the device is useless). But I don't even care to try and possibly
get it fixed because it's ~3 years old already, so it would be time to
dispose of and get a fresh one anyway (especially when I consider the
"i1d3: better of worse" question).



But now I wonder if somebody got some more solid informations by now.

Should I even care to create profiles for the i1d3 with a CM or i1 Pro
2 grade spectro? Is the i1d3 accurate enough (or even more accurate)
alone on a random (*) display?

* I mostly use it on HDTV (mostly PDP, sometimes through 3D LCD
glasses) and projector (mostly DLP) screens, so the supplied EDR files
don't really cover my usage cases. However, the difference between the
EDRs are not that big and I don't really need "spot-on", just a "good
enough" accuracy in general (consumer grade HDTVs are never "spot-on"
regardless how you calibrate them, just better than out of box...).

So, what do you think? Should I go for an i1 Pro 2 and use it to
profile my i1d3 or use the i1d3 alone in any case (even if I get a
spectro for other reasons anyway)?

[Graeme Gill]

Post by János, Tóth F.
After I upgraded my small toolset with an i1 Display Pro (i1d3)
colorimeter (replacing an old i1d2), I started to question the benefit
of creating correction matrices with my older ColorMunki Photo spectro
(the old i1d2 obviously needed that but I wasn't sure if the
ColorMunki Photo is any more accurate --- if not slightly but
unequivocally less accurate ??? --- on a random display than an i1d3
alone). I tried to read the opinions and look at test results of as
many individuals as possible but it was very puzzling (some said the
i1d3 is very accurate without corrections while the CM is NOT and some
said the i1d3 still needs corrections and a CM grade spectro is still
much more reliable), so I couldn't decide.

Hi,

The impression I get is "it depends". If your display is close to
one of the preset .edr/.ccss samples, then the instrument is probably
about as accurate as the ColorMunki Photo.

But if your display is not close to the preset samples, and you
happen to have an i1d3 that is at an unfortunate end of the filter accuracy
distribution (i.e. it's spectral sensitivities are not so like the standard
observer), then you may be better off creating a .ccss or .ccmx for it
using the ColorMunki. [ And some of the people with access to a lot
of i1d3's have commented that a few of them have not so good filter
shapes as the average i1d3. ]

Post by János, Tóth F.
But now, my CM stopped working completely. It must be some kind of
contact problem inside it's case: the device is accessible in the
default measuring position but becomes unavailable in any other
"dialer" positions, so self-calibration is no longer possible (and
thus the device is useless). But I don't even care to try and possibly
get it fixed because it's ~3 years old already, so it would be time to
dispose of and get a fresh one anyway (especially when I consider the
"i1d3: better of worse" question).

A few people have fixed this problem by carefully disassembling it
and repairing it. It's not too hard, you start by opening the
measurement guide and popping the clips out, then the plastic
covers, working around until they are all out. The only tricky
bit in the re-assembly is sorting out where the flexible PCB
that connects to the USB socket and spring go, as you put it
together. I'm not exactly sure what goes wrong with the position
detector though - I think it is optical, so should be reasonably robust.

Post by János, Tóth F.
enough" accuracy in general (consumer grade HDTVs are never "spot-on"
regardless how you calibrate them, just better than out of box...).

Some people have got very good results using hardware or software
than can apply a cLUT transform:
<http://www.avsforum.com/forum/139-display-calibration/1517849-comparison-3dlut-solutions-eecolor-box.html>

Post by János, Tóth F.
So, what do you think? Should I go for an i1 Pro 2 and use it to
profile my i1d3 or use the i1d3 alone in any case (even if I get a
spectro for other reasons anyway)?

The i1pro2 is a nice instrument (apart from it's fussiness about
USB cables!), but it's a big expense if you just want to do
display profiling. Fixing your ColorMunki is probably a cheaper
solution.

Graeme Gill.

[János, Tóth F.]

Thank you for the reply.


I disassembled it.

The position detector is indeed optical. I don't see that easily
repaired without some replacement parts or at all (at least easily and
economically at home, unless the problem is dust on the sensors and/or
reflective surfaces - there are two sensors and three reflective
sticks [two in a group] to unambiguously distinguishes the four dialer
positions, it's a clever an indeed a robust design).

But I guess the problem is the USB ribbon cable. Before the
disassembling I heard a small clicking-like noise and saw the
notification lights rapidly light up for a short moment as I moved the
dialer away from the reflective/display position and the USB
connection got lost halfway through to the self-calibration position.
Now I that I see this ribbon cable in place and I heard what kind of
noise it makes when it moves, I guess it was that. And not there is no
USB connection anymore, no matter how I move that cable.

This should be easily to fix (by finding a similar cable and replacing it).

ccss/ccmx

Ah, now I remember. That's where I got confused.
The difference was significantly smaller with a ccss file than with a
ccmx. The ccmx essentially aligned the two sensors but the ccss didn't
really bring the i1d3 readings too close to the CM readings (I even
saw some cases when it moved further away). - However, we talk about
dE differenced of about 1-2 or so, thus nothing extreme.
This made me assume that either the ccss based correction is too
"experimental" and can't be relayed on (the i1d3 was a new thing back
then) or the CM isn't really that accurate (probably less accurate
than the CM).
Since this ccss implementation was a new (not a widely and/or long
tested) thing, I chose to go with the first assumption and thus
continued to use matrices (created with the CM).
I think this CM malfunction was a good reminder that I should reevaluate this.

[Graeme Gill]

Post by János, Tóth F.
The position detector is indeed optical. I don't see that easily
repaired without some replacement parts or at all (at least easily and
economically at home, unless the problem is dust on the sensors and/or
reflective surfaces - there are two sensors and three reflective
sticks [two in a group] to unambiguously distinguishes the four dialer
positions, it's a clever an indeed a robust design).

It's worth just cleaning up the sensors and optical target, and
seeing if it works any better.

Post by János, Tóth F.
But I guess the problem is the USB ribbon cable. Before the
disassembling I heard a small clicking-like noise and saw the
notification lights rapidly light up for a short moment as I moved the
dialer away from the reflective/display position and the USB
connection got lost halfway through to the self-calibration position.
Now I that I see this ribbon cable in place and I heard what kind of
noise it makes when it moves, I guess it was that. And not there is no
USB connection anymore, no matter how I move that cable.

Yes, conceivably the flexible PCB might go open circuit with a lot of flexing.
Finding a replacement is probably not easy though...

Post by János, Tóth F.
saw some cases when it moved further away). - However, we talk about
dE differenced of about 1-2 or so, thus nothing extreme.

The Munki spectral sensitivity curves derive their reference indirectly
to a CS2000, but that "indirect" could conceivably add a delta E.

Whereas creating a .ccmx ties the i1d3 directly to the Munki, so
one would expect better agreement.

Graeme Gill.

[János, Tóth F.]

I think I can use virtually any kind of cable (ribbon or not) which
fits inside the case and isn't too prone to flexing. A cable used for
digital data transfer shouldn't effect the measurement
accuracy/precision. And I can solder it on, so it doesn't need to fit
into the connectors.


The i1Pro2 comes for 3 times the price of a CMP.
Do you think it's worth the price difference?
I ask both in terms of general use (including reflective measurements
as well) and for display measuring (colorimeter profiling) only.
I liked to experiment with printer profiling and measurement based
camera profiling, even though I never really achieved great results
(especial with cameras) and thus I rarely done similar things (only a
few printer profiles here and there over the years).

I ask since you got to play with a Jeti spectro as well as the iPro
1/2 and the i1d3.

1: The better self-calibration/self-check capabilities sound nice.
Knowing that the device would report if it drifted too much due to
aging or dust/damage would make me more comfortable on an everyday
basis. However, I am not sure if this actually works as intended
and/or how I imagine it. And there are cheaper ways to check this,
like comparing the readings with other similar instruments from other
people you know (not every days but after every few months or so...).

2: Are iPro2's tend to be (I know they are not all the same) any more
accurate (in an absolute manner) in general and/or on displays than
the CMP's?
I saw some reports about "not so accurate" CMPs. Most of the experts
always preferred the old iPro.
And for that matter, unlike the i1Pro2, the CMP doesn't come with any
kind of factory certificate (and it also can't be reevaluated...).

3: The CMP is strictly UV-cut.
I am not sure if some (current or near-future generation) display
devices might emit some UV light (I know some of them emit significant
amount or IR light). In any case, I never liked the idea of having
this filter in general.

[r***@public.gmane.org]

<<< The i1Pro2 comes for 3 times the price of a CMP.
Do you think it's worth the price difference?
I ask both in terms of general use (including reflective measurements
as well) and for display measuring (colorimeter profiling) only.
I liked to experiment with printer profiling and measurement based
camera profiling, even though I never really achieved great results
(especial with cameras) and thus I rarely done similar things (only a
few printer profiles here and there over the years).>>>

Here are my 2-cents worth, strictly as a user. I have both an i1D3 and an
i1Pro2. The i1D3 calibrates an Eizo CG277 I have extremely well, using
Eizo's ColorNavigator (it gets a better black point than the i1Pro2).

I also have a Dell U2713H and the i1D3 is almost unusable for it (or was,
the last time I tried) using Dell's (X-Rite's) Ultrasharp Color Calibration
Solution or X-Rite's i1Profiler (basically the same software), giving a
calibration around 800K below what it should be.

With a ccmx file created using the i1Pro2, the i1D3 calibrates/characterises
the Dell fine (with Argyll) ... but the problem is that this can only be
used for the Custom Color setting, which does not use the display's hardware
calibration.

Both monitors have basically the same wide-gamut panel from LG, so the
problem is not with the i1D3, but with the X-Rite EDF file (effectively same
as the ccmx). So it's a software problem rather than a fundamental problem
with the i1D3.

It's very nice having an i1Pro2 (and you can always buy the i1Basic package
if you don't need to do print profiles using i1Profiler ... you can use
Argyll for that with the i1Basic anyway) ... but I don't see how it would be
worth the money if you only really want it for display calibration. Better
to use the i1D3 and if necessary borrow an i1Pro/2 to generate the ccmx file
for it.

Robert

[János, Tóth F.]

Turns out it isn't really a feasible option to solder on some random
cable at home, however I can order a piece from the exact same cable
(for a ridiculous price but it's still a lot cheaper than a new or
even a used but working CMP, so I guess will do this regardless of any
other instrument purchases).

Today I could borrow (for free but only for a limited time) both a CMP
and an i1Pro2.
I measured a Panasoinc 42G30E with these two and my own i1d3 (without
using any EDR/ccss or ccmx corrections ; I repeated all measurements 3
times and they all looked reasonably stable).
These are not pure primary colors but an emulated ~Rec709 gamut (it
was calibrated with my old CMP)..


i1Pro2
----------------
Red
0.640937
0.331168
18.736442
Green
0.301860
0.599112
58.608018
Blue
0.150282
0.064774
6.836370
White
0.314922
0.330515
83.470609
----------------

CMP
----------------
Red
0.641842
0.329491
19.427972
Green
0.304845
0.597412
60.711867
Blue
0.149922
0.065672
7.185770
White
0.315066
0.330667
85.876157
----------------

i1d3
----------------
Red
0.642188
0.332041
19.768431
Green
0.301456
0.598586
61.233681
Blue
0.150078
0.063216
7.087048
White
0.311883
0.327727
86.325481
----------------

Post by r***@public.gmane.org
<<< The i1Pro2 comes for 3 times the price of a CMP.
Do you think it's worth the price difference?
I ask both in terms of general use (including reflective measurements
as well) and for display measuring (colorimeter profiling) only.
I liked to experiment with printer profiling and measurement based
camera profiling, even though I never really achieved great results
(especial with cameras) and thus I rarely done similar things (only a
few printer profiles here and there over the years).>>>
Here are my 2-cents worth, strictly as a user. I have both an i1D3 and an
i1Pro2. The i1D3 calibrates an Eizo CG277 I have extremely well, using
Eizo's ColorNavigator (it gets a better black point than the i1Pro2).
I also have a Dell U2713H and the i1D3 is almost unusable for it (or was,
the last time I tried) using Dell's (X-Rite's) Ultrasharp Color Calibration
Solution or X-Rite's i1Profiler (basically the same software), giving a
calibration around 800K below what it should be.
With a ccmx file created using the i1Pro2, the i1D3 calibrates/characterises
the Dell fine (with Argyll) ... but the problem is that this can only be
used for the Custom Color setting, which does not use the display's hardware
calibration.
Both monitors have basically the same wide-gamut panel from LG, so the
problem is not with the i1D3, but with the X-Rite EDF file (effectively same
as the ccmx). So it's a software problem rather than a fundamental problem
with the i1D3.
It's very nice having an i1Pro2 (and you can always buy the i1Basic package
if you don't need to do print profiles using i1Profiler ... you can use
Argyll for that with the i1Basic anyway) ... but I don't see how it would be
worth the money if you only really want it for display calibration. Better
to use the i1D3 and if necessary borrow an i1Pro/2 to generate the ccmx file
for it.
Robert