I decided to do a quick comparison of the 4 meters I have available.
Display was a Sony 60XBR1 SXRD rear projector.
Room lighting was subdued; curtains drawn and the room was dim overall, but not black.
Test pattern source was consistent throughout, and each meter was contact mounted at the exact center of the screen. All readings were done in about 1/2 hour.
Times taken to run through the gamut and grayscale readings are posted next to the meter name.
Software was CalMAN Pro v3.1 (I'm a big CalMAN fan!)
Low light handler was on.
Units standard is Imperial.

Meters:
1. Milori Trichromat-1, AKA Sencore ColorPro III, AKA Chroma 4. Approximately 5 years old and never factory recalibrated.
2. Spyder2. About 7 or 8 months old.
3. GretagMacbeth i1 monitor, older version of the X-rite i1Pro. Approximately 3-4 years old and factory recalibrated 2 years ago.
4. X-Rite i1Pro. Around 6 months old.

Results are attached in a Word document. Zoom to 150 or 200% to view clearly.
I welcome everyone's comments! Please chime in.

wow, big differences between them.

Yes, it's discouraging to me as a calibrator to find such big differences.
The good news is that color wise the two i1's are nearly identical at the high end. It is only as the light output goes down that the older meter's red output diverges from the newer meter.
The 2 colorimeters are roughly similar to each other, but their difference from the i1 is huge.
The CP III, AKA Trichromat-1 has always read luminance very low compared to all the other meters, and I believe it is also the least directional. I've wondered, could the directionality of a meter have an impact on it's luminance reading with contact readings of RPTVs? The difference in luminance is not nearly as big with direct view sets. In old magazine test reports of RPTVs I used to read comments like "light output was 20 fL, not including screen gain." That was back when colorimeters rather than spectroradiometers were widely used.
This leaves a huge question mark over the practice of always calibrating a display to a certain footlamberts reading (I have never done this, but I have heard of others always calibrating to 20 fL, etc), as the meter used and the display type would have a tremendous impact on the outcome.

Greetings

The readings are not to be trusted below 20 %/ire ... we don't trust them on the CS200 ... or the photo research stuff so they should definitely not be trusted on those units you have.

The only thing you might believe is the fL reading ... but nothing else.

Regards

Greetings

The readings are not to be trusted below 20 ft-l ... we don't trust them on the CS200 ... or the photo research stuff so they should definitely not be trusted on those units you have.

The only thing you might believe is the fL reading ... but nothing else.

Regards

Michael, surely this is a typo... You don't trust anything below 20 footlamberts???

The head-to-head comparisons would be easier if you provided raw xyY data.

Michael, surely this is a typo... You don't trust anything below 20 footlamberts???Yea, surely he meant below 20% stim.

Greetings

We need to trust our eyes more than what the device is telling us. None of this stuff is that great in low light conditions.

It's time to pull out the optical comparator at the low end if you want better accuracy.

20 % might be about as low as anyone might ever want to go. But if your gamma is nice and high (low number ... high brightness) then 20% would be fine for the device.

Used on a display doing a proper 2.4 gamma ... you really have to be careful what the meters tell you down there. Use the wetware between the ears. :)

Regards

I decided to do a quick comparison of the 4 meters I have available.

Chad,

thanks much for the comparison.

In looking at the Calman graphs - assuming I'm reading them correctly - the biggest difference seems to be between the i1Pros vs. the colorimeters. Color gamut and color brightness readings seem fairly similar for all 4 with a bit of deviation but not dramatic. But the grayscale luminance and RGB tracking do seem to diverge a bit. It would definitely lead to different calibration settings for grayscale and if you were targeting a certain Y value at 100%, your settings would also be a good bit different.

cheers,


--tom

Yes ... 20% ... it was a typo .. sorry

I will fix it


regards

The head-to-head comparisons would be easier if you provided raw xyY data.

Yes, I think I may redo the comparison today and save the raw data. The display will also be calibrated better.
I may also save some larger graphs; these are a bit hard to see.

New test results, with raw data.
For this run I put the units standard to metric, so we are no longer reading footlamberts. To calculate fL, multiply Y by .29.
The display was recalibrated since the last run and a black blanket was draped over it.
Not surprisingly, the i1's red response at low light was effected by the dark reading. The i1 Pro results shown were from a second run after retaking the dark reading, since the 1st run was not consistent with what I saw or what I measured while previously calibrating the display.

New test results, with raw data.
For this run I put the units standard to metric, so we are no longer reading footlamberts. To calculate fL, multiply Y by .29.
The display was recalibrated since the last run and a black blanket was draped over it.
Not surprisingly, the i1's red response at low light was effected by the dark reading. The i1 Pro results shown were from a second run after retaking the dark reading, since the 1st run was not consistent with what I saw or what I measured while previously calibrating the display.

Chad, which device was used to recalibrate it? Was it a different device outside of the test group?

cheers,


--tom

Chad, which device was used to recalibrate it? Was it a different device outside of the test group?

cheers,


--tom

The new i1Pro was used to recalibrate the TV.

It also seems interesting that the Spyder II doesn't do that bad compared to the much more expensive colorimeter

much more expensive ?? (nope).

much more expensive ?? (nope).

Yeah, I was doing a search for the Sencore ColorPro III and getting back 4 figure numbers - but perhaps that is because of the particular software bundled with it. How much is just the colorimeter?


--tom

The new i1Pro was used to recalibrate the TV.

Thanks Chad. I guess my next question would be what is your feeling and expert opinion on whether or not the i1Pro is the more accurate of the set. If you were to recalibrate using the Spyder II or Sencore colorimeter do you feel that the results would be "off" to your eyes?

cheers,


--tom

Greetings

The sencore hardware is bundled with the software. That is how they do it. The stuff is aimed at professionals ... not enthusiasts. Not that they won't sell to enthusiasts ... they will sell to anyone.

Being aimed at the professional, you also expect professional pricing. There calibration packages range in the $3500 to $8000 range.

Regards

Greetings

pod based hardware seems to be weak on the LCD/LCOS/DLP technologies. On many LCD displays, there is a trend in their behavior. The trend is toward red. No distinct trends on the other technologies so it is hard to tell where you end up. (Just not going to be accurate ... figure a dE of 5 or more versus spectro devices.)

Regards

pod based hardware seems to be weak on the LCD/LCOS/DLP technologies. On many LCD displays, there is a trend in their behavior. The trend is toward red. No distinct trends on the other technologies so it is hard to tell where you end up. (Just not going to be accurate ... figure a dE of 5 or more versus spectro devices.)

Thanks Michael for this info - when you say "the trend is toward red" for LCDs do you mean that the pod registers too little red or too much red? I'm guessing too much but I've already got one strike against me in this thread on colorimeter pricing :) so wanted to double check.

cheers,


--tom

Greetings

The pod tells you it is 6500K ... let's say ... but it really is 5500K ... and explains why the image is way more red than you thought it should be.

From an enthusiast perspective, if you know about this weakness, you shoot for 7200K to the blue instead ... knowing that the real end result is somewhere in the 6300 to 6700 range.

But you don't give customer graphs that say it is 6500 when the image is too red. It will catch up to you ...

And conversely you don't give them graphs that say 7200K either

regards

I do know that the i1Pro is much more accurate, because of the research I've read and also going by eye.
I have also noticed significant differences on plasmas.

Hey Chad, have you tried profiling any of your pod based meter to your i1Pro?
Would love to see some graphs of say your Sencore Pod meter profiled to the i1Pro and differences betwee the Sencore profiled to the i1Pro vs the i1Pro itself. I need to start tinkering with the profiling feature of Calman Pro and see what kind of results I get.

I do know that the i1Pro is much more accurate, because of the research I've read and also going by eye.
I have also noticed significant differences on plasmas.

Significant differences in that the i1Pro is much more accurate than the others with plasmas?

BTW, We just opened up the CalMAN V3.1 Pro and standard betas to an open beta. We also now have hardware bundles that include the i1Pro, Chroma5 and/or Hubble for the Professional.

CalMAN V3.1 has a very long list of new features and hardware support many aimed at the Professional.
http://www.datapopuli.com/forums/viewtopic.php?f=30&t=1344

I decided to do a quick comparison of the 4 meters I have available.

Meters:
1. Milori Trichromat-1, AKA Sencore ColorPro III, AKA Chroma 4. Approximately 5 years old and never factory recalibrated.
2. Spyder2. About 7 or 8 months old.
3. GretagMacbeth i1 monitor, older version of the X-rite i1Pro. Approximately 3-4 years old and factory recalibrated 2 years ago.
4. X-Rite i1Pro. Around 6 months old.


Chad, feel free to post your CDFs with raw data if you like since anyone can download V3.1 and load CDFs.

Significant differences in that the i1Pro is much more accurate than the others with plasmas?

Yes, that's right. When I first got the Spyder2, after hearing from the manufacturer how wonderful it was for all display types, I used it to calibrate a Pioneer Elite 1130. I was not happy with how it looked... Kinda sickly and green. I hooked up the i1Pro and yes, after calibration with the Spyder2 the grayscale was positive green and negative blue.

Greetings

That's probably not indicative of a weakness in the spyder2 pod ... but rather that you have one that was severely out of tune.

I recently had a similar experience with the spyder 3 pod. A new one out of the box was measuring everything as too green ... the solution ... make it too blue.

At our recent THX class at the Datacolor facility, they took that pod of mine and compared it to the reference instrument and a box of 10 other spyder 3 pods. Turns out I had the worst one out of the lot. It went into the garbage right away and I was given the one that measured closest to the reference device.

$100 pods can have quite a lot of variance.

Regards

Hey Chad, have you tried profiling any of your pod based meter to your i1Pro?
Would love to see some graphs of say your Sencore Pod meter profiled to the i1Pro and differences betwee the Sencore profiled to the i1Pro vs the i1Pro itself. I need to start tinkering with the profiling feature of Calman Pro and see what kind of results I get.
I haven't yet, but I plan to.
I did use the meter training with Colorfacts; I trained the Spyder2 to emulate the i1 on many different display types. It worked well enough to do a calibration with minimal (maybe 2-3% at most) error with most displays. The advantage was I could go an entire calibration without taking the meter down for a dark reading. Occasionally it wouldn't work as well, like with a Toshiba LCD flat panel.
I think the reason is Colorfact's training only used 100% stim gamut results to match the meters. From what I understand, CalMAN uses 30-100% grayscale readings, which I would think should be better. I don't think CalMAN profiles for the gamut, though, but like I said I haven't tried it yet.

Greetings

That's probably not indicative of a weakness in the spyder2 pod ... but rather that you have one that was severely out of tune.

I recently had a similar experience with the spyder 3 pod. A new one out of the box was measuring everything as too green ... the solution ... make it too blue.

At our recent THX class at the Datacolor facility, they took that pod of mine and compared it to the reference instrument and a box of 10 other spyder 3 pods. Turns out I had the worst one out of the lot. It went into the garbage right away and I was given the one that measured closest to the reference device.

$100 pods can have quite a lot of variance.

Regards

Interesting. I'm interested in how the Spyder 3 compares.
What's your overall impression of it?

Greetings

Overall ... good. However ... I still would not use it on LCD or LCOS or DLP as I still don't trust it. :)

regards

I decided to do a quick comparison of the 4 meters I have available.
Display was a Sony 60XBR1 SXRD rear projector.
Room lighting was subdued; curtains drawn and the room was dim overall, but not black.
Test pattern source was consistent throughout, and each meter was contact mounted at the exact center of the screen. All readings were done in about 1/2 hour.
Times taken to run through the gamut and grayscale readings are posted next to the meter name.
Software was CalMAN Pro v3.1 (I'm a big CalMAN fan!)
Low light handler was on.
Units standard is Imperial.

Meters:
1. Milori Trichromat-1, AKA Sencore ColorPro III, AKA Chroma 4. Approximately 5 years old and never factory recalibrated.
2. Spyder2. About 7 or 8 months old.
3. GretagMacbeth i1 monitor, older version of the X-rite i1Pro. Approximately 3-4 years old and factory recalibrated 2 years ago.
4. X-Rite i1Pro. Around 6 months old.

Results are attached in a Word document. Zoom to 150 or 200% to view clearly.
I welcome everyone's comments! Please chime in.

Interesting, I did the same about 30 days ago.

My SXRD had been calibrated 30 days earlier with a PR-670.

The only difference I had was I had a new ColorPro 5 that was sent to a client in town for review in the mix as well.

I was shocked to see the older ColorPro 4 was much closer in terms of color temps than the newer ColorPro 5. As was to be expected, the Spyder 2 was the furtherest out.

http://img262.imageshack.us/img262/4169/metercomparisonta9.jpg (http://imageshack.us)


What is the REV number on your i1's?

My SXRD had been calibrated 30 days earlier with a PR-670. The only difference I had was I had a new ColorPro 5 that was sent to a client in town for review in the mix as well.

Do you have the numbers from the PR-670 calibration? How well does the i1Pro compare to that?

cheers,


--tom

The SXRD had around 1,000 additional hours on it when I ran these 4 meters through it compared to the PR-670 and the picture had visibly changed in that time frame - thus the comparison would be meaningless.

I wish I had done them at the same time.

Hindsight is 20/20.