or Connect
AVS › AVS Forum › Display Devices › Display Calibration › X-Rite's Revolutionary New Colorimeter
New Posts  All Forums:Forum Nav:

X-Rite's Revolutionary New Colorimeter - Page 45

post #1321 of 1402
What are the mechanics of a i1pro2?

Does it only read RGB or does it do more?
post #1322 of 1402
Quote:
Originally Posted by JimP View Post

What are the mechanics of a i1pro2?

Does it only read RGB or does it do more?

Like the original i1pro it samples the entire spectrum of visible light at 10nm intervals and then integrates the results into X, Y, Z (bar) "coordinates." Your software package(s) then takes that X,Y,Z output and coverts it into more familiar coordinate systems like the x,y,Y
post #1323 of 1402
How does that compare to how a colormeter such as the i1display3 works?
post #1324 of 1402
Thread Starter 
Quote:
Originally Posted by JimP View Post

How does that compare to how a colormeter such as the i1display3 works?
The D3 has photo-sensitive diodes that receive light through filters. These filters mimic the response of the standard observer. The disadvantage of this design is that the performance depends largely on the accuracy and stability of the filters. The advantage is that the diodes are MUCH larger than what's found in spectros, such as the i1Pro and more expensive reference instruments such as the Jeti, which means that they offer much greater sensitivity and speed.

What makes the D3 revolutionary, especially at this price point, is that the filters are exceptionally accurate and stable. Also, the lens focuses a lot of light on the diodes, which increases speed and sensitivity. Finally, the integrated diffuser and tripod attachment make it exceptionally well-suited for measuring front projectors. The D3 is no more accurate than the Chroma 5--an excellent previous generation colorimeter--but it is much faster, has much better low light sensitivity, requires recalibration less often, and is much easier to use with projectors.

X-Rite has a nice white paper explaining the difference between tristimulus filter-based colorimeters and spectroradiometers.
Edited by TomHuffman - 6/20/13 at 11:24pm
post #1325 of 1402
Has X-Rite in the mean time posted anywhere any specifications of the D3?
(measuring range (nits), accuracy (chroma, luminance), repeatability (chroma, luminance), ...)
post #1326 of 1402
Tom,

Since the i1pro2 can go fairly low for luminance, should there be any difference between using an i1pro2 or an i1display3pro(your tweaked version) when it comes to calibrating gamma? 10% reads .25 ftl. with a target gamma of 2.22
post #1327 of 1402
Quote:
Originally Posted by TomHuffman View Post

Nothing definitive. However, the fact that the white, red, and blue channels were almost exactly the same as the previous i1Pro 1 measurement relative to the reference certainly suggests something more than random results.

Also, I haven't noticed a meaningful difference between good color analyzers when measuring plasmas. The previous test showed that a D3 profiled with the reference on a Panny plasma seemed to perform just as well when measuring a Pioneer plasma. LCDs (LED and CCFL) and bulb-based projectors that use different illumination technologies is where you start to see real differences in color analyzer performance.

So, you're saying display types like CRT and Plasma (phosphor-based) are more consistent in terms of SPD than other display types that use LEDs, CCFL, or bulbs? Have you had a chance to compare more than 2 plasmas?
post #1328 of 1402
Correction:

I made a small error above. You actually can get the raw spectral data out of the i1Pro (and I assume v 2, as well,) so, if you were brave enough to write your own integration routines ...

I count at least 36 relatively discrete measurement points in the spectrum output from my i1Pro vs. ... 3 not so discrete points for a tristimulus colorimeter (profiled or not.) ... Just a sanity check here. wink.gif

I don't have gigabytes of comparison data, but if I were to approach a display with an unknown illumination source and no idea of how much affect the spectrum of that illumination source would have on the resulting whitepoint/greyscale, I think I know which meter I'd rather have. ... YMMV

Also, The spectrum of illumination "sources" have a tendency to age/drift (sometimes quite a lot) over time .... I'm probably missing something "obvious," but I don't see how fixed calibration tables can really deal with that effectively.
Edited by HDTVChallenged - 6/21/13 at 10:52am
post #1329 of 1402
What method would be recommended for comparing two D3's against each other? My first thought was to run each and then simply compare the reports generated by Chromapure. I thought I'd check to see if there might be other suggestions though. Thanks.
post #1330 of 1402
You might want to use the profiling module in Chromapure to take readings with each D3.
post #1331 of 1402
Quote:
Originally Posted by JimP View Post

You might want to use the profiling module in Chromapure to take readings with each D3.

Good idea. Thanks.
post #1332 of 1402
Tom,

It seems that there would be a good argument to use an enhanced i1display3 pro (from Chromapure, of course smile.gif ) over an off the shelf i1pro2.

A variable that I don't understand is how do they compare as to linearity? Is one more linear therefore more correct at different levels that the other?
post #1333 of 1402
Thread Starter 
Quote:
Originally Posted by PlasmaPZ80U View Post

So, you're saying display types like CRT and Plasma (phosphor-based) are more consistent in terms of SPD than other display types that use LEDs, CCFL, or bulbs? Have you had a chance to compare more than 2 plasmas?
That's been my experience, though I have a lot more data on various Pannys and Pioneers than Samsungs. I would be surprised if they were substantially different.
post #1334 of 1402
Quote:
Originally Posted by TomHuffman View Post

That's been my experience, though I have a lot more data on various Pannys and Pioneers than Samsungs. I would be surprised if they were substantially different.

So to clarify: We are saying that the supposed accuracy of the "Display 3 PRO" is not inherent but rather the result of "brute force" correction(s) to a much more accurate meter (specifically either a JETI or Minolta 5nm spectroradiometer.) Furthermore, the only real reason this method works is because the spectral distributions of CRT, Plasma and Front Projectors are relatively consistent across different makes and models of these types of displays.

Furthermore, the "system" begins to breakdown when confronted with display types that have 'less than consistent' spectral distributions across different makes and models, not to mention across time. Otherwise, I would assume that there would no longer be any reason to continue selling i1Pro2s ... Or the need for a dealer's choice smorgasbord of "LCD Correction tables."

?
post #1335 of 1402
Thread Starter 
Quote:
Originally Posted by JimP View Post

Tom,

It seems that there would be a good argument to use an enhanced i1display3 pro (from Chromapure, of course smile.gif ) over an off the shelf i1pro2.

A variable that I don't understand is how do they compare as to linearity? Is one more linear therefore more correct at different levels that the other?
I would only use the i1Pro 2 to correct a stock D3 or other colorimeter. Even the i1Pro 2 has not-very-good repeatability at 10% on flat panels, which is much worse on front projectors (this can be remedied by using it in illuminance mode).
post #1336 of 1402
Thread Starter 
Quote:
Originally Posted by HDTVChallenged View Post

So to clarify: We are saying that the supposed accuracy of the "Display 3 PRO" is not inherent but rather the result of "brute force" correction(s) to a much more accurate meter (specifically either a JETI or Minolta 5nm spectroradiometer.) Furthermore, the only real reason this method works is because the spectral distributions of CRT, Plasma and Front Projectors are relatively consistent across different makes and models of these types of displays.

Furthermore, the "system" begins to breakdown when confronted with display types that have 'less than consistent' spectral distributions across different makes and models, not to mention across time. Otherwise, I would assume that there would no longer be any reason to continue selling i1Pro2s ... Or the need for a dealer's choice smorgasbord of "LCD Correction tables."?
Actually, I don't think that I saying any of this. The largest source of the accuracy of the Display 3 PRO is the stock meter itself. Even uncorrected, it is quite good. The "system" is not limited to a single correction. Any reasonable approach to this includes corrections for a variety of sources. The Display 3 PRO, for example, includes 12 independent corrections for different operating modes/display types. My only point above is that I have seen no need for different corrections for different brands of plasma, nor different brands of front projectors with UHP bulbs. To accommodate changes throughout time, the meter should be periodically re-calibrated, though I have got to say that the D3 is much more stable in this regard than was the D2. I used to recommend annual recalibration. That now seems overly aggressive for the D3.

There are two competing performance issues here: acceptable colorimeteric accuracy across a wide variety of displays and acceptable sensitivity for low-light sources. I have noticed a tendency to exaggerate the former while neglecting the latter. Modern colorimeters are better than most people think in this regard--indeed quite good once corrected--and budget spectros are not as good as many people assume they are. However, the i1Pro 2 offers a significant improvement. Its accuracy seems improved and the user is not hobbled by the need to take dark readings every 10 minutes, which for me was unacceptable. Nonetheless, the sensitivity issue remains and this can be critical in getting reasonable accuracy and repeatability at low light levels.

Finally, some of this is just personal preference. I would much rather use a good colorimeter that has been corrected by a reference spectro than use a budget spectro. I even prefer the corrected colorimeter over a very sensitive reference spectro, such as the Jeti 1211. Even that instrument is painfully slow at 10% stimulus on front projectors.
post #1337 of 1402
So again ... Why do you bother to continue selling the i1Pro2 (at nearly twice the price of the D3Pro?)

Either what you're claiming about the D3Pro is "exaggerated" or you're taking advantage of people by selling a more expensive product that you "know" or at least feel is somehow "inferior" to the "D3Pro."

I'm sorry to be a pest about this, but as the saying goes 'if something sounds too good to be true ..."

PS: I have the same reservations about the claims made in reference to C6's, so I'm not singling you out ...

PPS: My personal philosophy is that its wishful thinking to rely on one and only one meter in the field, and furthermore it makes more sense to use an i1Pro/i1Pro2 for the range it good at, profile your colorimeter in the field on the target display, then use the colorimeter for the sub-30% range. If nothing else, It makes sense to have the spectro in the bag as a sanity check. Of course if I were doing this professionally, I'd probably pull a Micheal Chen and just haul a Jeti around with me. wink.gif

Finally, with seven(ish) (or more) different LCD corrections to pick from, how am I supposed to have *any* confidence that the one I use is actually going to work on the display in hand? How do I know that a correction factor made on a display with 100 Hours on the backlight will work on a display with 6500 hours on the backlight (or vice versa?)
Edited by HDTVChallenged - 6/23/13 at 11:12pm
post #1338 of 1402
"... The future's the past, the past is the present ... it all just gives me a headache! ..."

Discussions like these are why I've deliberately tried to avoid this section of the forum. wink.gifbiggrin.gif
post #1339 of 1402
Quote:
Originally Posted by HDTVChallenged View Post

PPS: My personal philosophy is that its wishful thinking to rely on one and only one meter in the field, and furthermore it makes more sense to use an i1Pro/i1Pro2 for the range it good at, profile your colorimeter in the field on the target display, then use the colorimeter for the sub-30% range. If nothing else, It makes sense to have the spectro in the bag as a sanity check. Of course if I were doing this professionally, I'd probably pull a Micheal Chen and just haul a Jeti around with me. wink.gif

Hello, This is not good, each meter reads the luminance differently, doing this you will mess your gamma at that 0-30% Range.

See these numbers that are taken during the profiling of i1D3 OEM with i1PRO2 using my KURO measuring 100% White:

i1PRO2

x white= 0.313365
y white= 0.328795
Y white= 133.798 cdm/2

i1D3 OEM

x white=0.302729
y white=0.3280
Y white= 138.311 cdm/2

This shows that the id3 OEM is reporting more light, so if you use your colorimeter to fix your 0-30% Range, your gamma will be higher at that range because of this difference in luminance levels that the 2 meters are reporting.

** BTW the way i used back in 2010 to calibrate using both meters the same time with 2 notebooks, one for each meter connected with ChromaPure, is to profile your colorimeter to the spectro and then use the Spectro from the one notebook for RGB Balance for each stimulus and the other laptop with the profiled colorimeter for Gamma/Luminance and to do the RGB Balance of 5-30% Stimulus only. But this is really an added extra calibration time for nothing....
post #1340 of 1402
I realise this is a thread that is about a meter but the majority of us will use it to calibrate a display.

With this use in mind will calibrating using any of its variants from an unprofiled D3 up to any of the enhanced versions show any variation in Display output from a PQ point of view?
If so, which Displays will demonstrate this difference?

I am asking really if the super accuracy that can be achieved by any meter is of anything other than academic use when the manufacturers, perhaps with good reasons, prefer to not calibrate 'out of the box'.?
post #1341 of 1402
Quote:
Originally Posted by PE06MCG View Post

I realise this is a thread that is about a meter but the majority of us will use it to calibrate a display.

With this use in mind will calibrating using any of its variants from an unprofiled D3 up to any of the enhanced versions show any variation in Display output from a PQ point of view?
If so, which Displays will demonstrate this difference?

I am asking really if the super accuracy that can be achieved by any meter is of anything other than academic use when the manufacturers, perhaps with good reasons, prefer to not calibrate 'out of the box'.?

Some display types it will be very accurate on, other not so much. The object of the calibration is to make the display as accurate as possible with the controls provided. So using a measuring tool that is inaccurate kind of defeats the purpose doesn't it? The other issue is, once you start with this hobby, you will drive yourself crazy trying to get prefect charts.. well just cause the dot is in the box does not make it correct.. especially if you don't know if you measuring device is accurate. Will you see a difference .. maybe .. maybe not you will never know for sure.

You may want to read these articles by Michael Chen as well as look over his site.
Why TVs are not Calibrated
About Calibration Tables
post #1342 of 1402
Quote:
Originally Posted by airscapes View Post

Some display types it will be very accurate on, other not so much. The object of the calibration is to make the display as accurate as possible with the controls provided. So using a measuring tool that is inaccurate kind of defeats the purpose doesn't it? The other issue is, once you start with this hobby, you will drive yourself crazy trying to get prefect charts.. well just cause the dot is in the box does not make it correct.. especially if you don't know if you measuring device is accurate. Will you see a difference .. maybe .. maybe not you will never know for sure.

You may want to read these articles by Michael Chen as well as look over his site.
Why TVs are not Calibrated
About Calibration Tables

I'm not disputing there are differences in accuracy I am merely asking whether this extra accuracy translates into a visible difference on its intended use.

I agree that chart chasing becomes obsessive but to what ends? I'm far from convinced that any display, irrespective of its make, type or price can maintain its super calibration for any length of time, the manufacturers have not made it to be that way..

However they probably have made it to maintain its visible calibrated characteristics over a much longer time because they use PQ as the final objective not absolute accuracy.
post #1343 of 1402
Greetings

The only sets that I have encountered that hold their calibration "forever" given the effective lifespan of a TV ... are the Pioneer Kuros ... I've returned after 3 years on a few of them to find no change.

Similar experience with some of the older Panasonic plasma sets ... from 2008-2009 time frame ... little change over 2 to 3 years.

Projectors are another matter ... given the bulb wear.

If you use dE as a simple guide ... Stock Cinema and Movie modes almost never read dE 3 or less. They are usually in the 5 to 15 range ... which goes anywhere from very good to worse than tolerable. Once you dial it in that first time, it will take the TV a long time to drift back to that original type of error again. Beyond 5 to 6 years ... and since most are changing out by that time ... then one calibration is really all that is needed for the useful life of any flat panel.

Regards
post #1344 of 1402
Quote:
Originally Posted by Michael TLV View Post

Greetings

The only sets that I have encountered that hold their calibration "forever" given the effective lifespan of a TV ... are the Pioneer Kuros ... I've returned after 3 years on a few of them to find no change.

Similar experience with some of the older Panasonic plasma sets ... from 2008-2009 time frame ... little change over 2 to 3 years.

Projectors are another matter ... given the bulb wear.

If you use dE as a simple guide ... Stock Cinema and Movie modes almost never read dE 3 or less. They are usually in the 5 to 15 range ... which goes anywhere from very good to worse than tolerable. Once you dial it in that first time, it will take the TV a long time to drift back to that original type of error again. Beyond 5 to 6 years ... and since most are changing out by that time ... then one calibration is really all that is needed for the useful life of any flat panel.

Regards

I've found the same with the Samsung B860 Plasma. Still holding the same settings after three years.
post #1345 of 1402
Quote:
Originally Posted by ConnecTEDDD View Post

Hello, This is not good, each meter reads the luminance differently, doing this you will mess your gamma at that 0-30% Range.

See these numbers that are taken during the profiling of i1D3 OEM with i1PRO2 using my KURO measuring 100% White:

i1PRO2

x white= 0.313365
y white= 0.328795
Y white= 133.798 cdm/2

i1D3 OEM

x white=0.302729
y white=0.3280
Y white= 138.311 cdm/2

This shows that the id3 OEM is reporting more light, so if you use your colorimeter to fix your 0-30% Range, your gamma will be higher at that range because of this difference in luminance levels that the 2 meters are reporting.

** BTW the way i used back in 2010 to calibrate using both meters the same time with 2 notebooks, one for each meter connected with ChromaPure, is to profile your colorimeter to the spectro and then use the Spectro from the one notebook for RGB Balance for each stimulus and the other laptop with the profiled colorimeter for Gamma/Luminance and to do the RGB Balance of 5-30% Stimulus only. But this is really an added extra calibration time for nothing....

Yeah ...but the profiling process should, in fact, fix the difference in the Y values ... your profiled D3 Y should read the same as your original i1Pro2 Y ... it's just a scalar factor ... For me, the difference in your small x is far more significant ... the small y diff I could probably live with. smile.gif I suppose if I trusted my (not a D3) colorimeter more, I'd just use it for everything once it had been profiled. But I don't trust it that much, so limiting the potential "damage" to the sub-30% range makes for sense ... for now.
Edited by HDTVChallenged - 6/24/13 at 10:23am
post #1346 of 1402
Quote:
Originally Posted by Michael TLV View Post

Greetings
... then one calibration is really all that is needed for the useful life of any flat panel.

Regards

I hope you really meant Plasma "flat-panel," because my (CCFL/EEFL) LCD's have been drifting enough to drive me a bit nuts ... In fairness probably only ~3.5dE worth at a time, but it's *just* enough to cause mild brain seizures for a week or so post recalibration. biggrin.gif The drift is consistent with what one would expect from an aging FL backlight (or even a UHP,) and the hours do tend to add up quickly on them. I plan to recheck more often in the future, I let the main one go for ~10 months the last time.

OTOH, after the first 3 or so years, my old CRT RPTV rarely changed more than +/-2 clicks on any of RGB Drive or Cuts ...
post #1347 of 1402
Greetings

You may be right there. The non-drift has been on plasma panels ... not the LCD ones. My own Samsung B860 has been rock solid ... another plasma again.

Regards
post #1348 of 1402
Quote:
Originally Posted by Michael TLV View Post

Greetings

The only sets that I have encountered that hold their calibration "forever" given the effective lifespan of a TV ... are the Pioneer Kuros ... I've returned after 3 years on a few of them to find no change.

Similar experience with some of the older Panasonic plasma sets ... from 2008-2009 time frame ... little change over 2 to 3 years.

Projectors are another matter ... given the bulb wear.

If you use dE as a simple guide ... Stock Cinema and Movie modes almost never read dE 3 or less. They are usually in the 5 to 15 range ... which goes anywhere from very good to worse than tolerable. Once you dial it in that first time, it will take the TV a long time to drift back to that original type of error again. Beyond 5 to 6 years ... and since most are changing out by that time ... then one calibration is really all that is needed for the useful life of any flat panel.

Regards

Hi Michael,

So if I use a standard none profiled D3 then an enhanced D3 will I see a difference in picture quality?
I am aware it is probably more accurate but is this reflected in a difference in PQ?
post #1349 of 1402
Greetings,

Peter,

If you use a stock D3 with the calman program, it uses tables built into the calman program. The stock d3, of course, won't work with CP ... from what I recall so that is moot. So on Calman, since tables are being used, you get the same result as the OEM d3 that you would buy from Spectracal really. Same ... and you'd certainly not be able to see a difference ... likely talking deviations of 1 dE at most and people need about 3 dE to actually see the difference on real material ... not test patterns.

Regards
post #1350 of 1402
Quote:
Originally Posted by Michael TLV View Post

The stock d3, of course, won't work with CP ... from what I recall so that is moot.

CP gained support for retail meters just recently.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › X-Rite's Revolutionary New Colorimeter