AVS › AVS Forum › Display Devices › Display Calibration › i1Pro Dark Readings - I Don't Get It
New Posts  All Forums:Forum Nav:

i1Pro Dark Readings - I Don't Get It

post #1 of 15
3/22/11 at 5:14am
Thread Starter 
I have an X-Rite i1Pro, REV D model.

Last night in a light and temperature controlled room I took a series of grayscale and chromaticity measurements:

1. After dark reading
2. After 1 hour
3. After 2 hours
4. After 3 hours
5. After 2nd dark reading

Other than for the initial measurements I took no further dark readings until after the "after 3 hours" reading. In the attached jpg file I found it unnecessary to display the "after 1 hour" and "after 2 hours" results as they are essential the same as the "after 3 hours" results.

I took a few random readings just to exercise the meter a bit after the initial reading until the "after 2 hours" reading. Because no differences appeared in the readings I thought perhaps meter usage might have an effect. Between the "after 2 hours" and "after 3 hours" reading I had the meter do continuous readings throughout the hour utilizing various patterns. The end result was still no change.

As a final step I took a second dark reading. The resulting readings were the same as the rest - no significant difference.

Is the REV D model different from earlier models? Why am I supposed to be taking dark readings if no difference in accuracy can be detected?

Why the term "dark reading" when it is actually accomplished on a white tile?
LL
Reply
Reply

AVS Top Picks

post #2 of 15
3/22/11 at 7:43am
It is a dark reading, because there should be a total absence of light.

It's against the white tile, because in reflectance mode it also takes a reading with it's internal lamp turned on.

The one thing I don't see in your data is the big Y value. You are starting to see more drift down at the low end , and that is typically were you see the most drift.

If I was using a i1Pro to do grayscale, I'd stop probably every 10-15 minutes. If I was doing CMS, I might not do a dark cal for 30min or more.

Any general recommendation are for people who don't dig into the specifics. If you know how it works, then you can determine what's right for you.
Reply
Reply
post #3 of 15
3/22/11 at 10:56am
Thread Starter 
I see what you mean about the lower end but how significant are the differences in the before and after? Also, for what it's worth, grayscale was for test purposes. When I calibrate I profile a C5 from the i1Pro and calibrate grayscale with the C5 and color with the i1Pro. At this point a question comes to mind: Would I be just as accurate if I did color as well as grayscale with the profiled C5?

On the attached calibration report it can be seen that color dE average remains the same and the largest individual deviation is .3. Blue is .3, Red .2 and the other 4 (5 including white) is only .1 dE.

My goal is to eliminate taking down the i1Pro for dark readings if it can be avoided and with the small amount of error would it make any difference? In fact, the second attachment, "i1ProAfterDarkReadingReduced" pretty much backs me up. The before is just after the intial dark reading and the after is the result of a dark reading I took when finished after three hours. The drift is every bit as large as that with no repeated dark readings.

 

i1Pro Check Plus 3 HourReducedSize.pdf 251.615234375k . file

 

i1ProAfterDarkReadingReduced.pdf 256.3388671875k . file
Reply
Reply
post #4 of 15
3/22/11 at 11:20am
if you're just using the i1 for CMS, I wouldn't worry too much about doing extra dark readings.

Also you don't need to do the dark reading on the white tile, simply affixing the cap to the meter would suffice.
Reply
Reply
post #5 of 15
3/22/11 at 11:41am
Thread Starter 
Quote:
Originally Posted by sotti View Post

if you're just using the i1 for CMS, I wouldn't worry too much about doing extra dark readings.

Also you don't need to do the dark reading on the white tile, simply affixing the cap to the meter would suffice.

That is music to my ears, sotti, and I thank you for the information. It's not that I'm lazy but rather just trying to avoid a PITA.

Steve Shaw of LightIllusion.com gave me the same sort of answer on the LUT thread.

Quote:
Originally Posted by Light Illusion View Post

X-Rite don't actually give a time figure for recalibration of the probes, but do say they can 'vary' due to temp and other changes...

We always say to re-cal before a calibration run, after a good warm-up period, but not during a calibration run. The best is to keep temperature stable.
Reply
Reply
post #6 of 15
3/26/11 at 10:20am
There is a general rule (not sure where it started) that the i1pro has more difficulty with accurate readings below 1 ftL.

This can easily come into play when using the i1pro for CMS work, not just grayscale. Remember that blue is only 7.2% of the 100% white reading. This means that your 100% gray reading should be 13.8 ftL or more if you want to have more confidence with your blue readings with the i1pro. It is very easy for projectors to have difficulty reaching 13.8 ftL once fully calibrated.

So the low light accuracy of the i1pro can easily come into play with CMS, not just grayscale.

Now, having said that, this is what I did to resolve that issue. I got a nice size sample of my screen material and hung it much closer to my projector, roughly 3-6 feet away. (a 12" x 12" sample is plenty big). This raises the light output significantly reflecting off the screen material. Then, use your i1pro to read off the small screen material sample. Just make sure to keep the 100% white reading below 300 cd/m2 (87.5 ftL) to prevent oversaturating the meter. This allows for accurate readings, even with 10% grey readings and blue readings. (both of which can be problimatic with the i1pro).

Since the small screen material is the same as your large screen, the results will be spot on for the large screen as well.

I'll probably post this information in a seperate post to help others that have an i1pro, since it has worked great for me and I'd like to share the information.

Dan
Reply
Reply
post #7 of 15
3/26/11 at 2:09pm
Thread Starter 
A comment: Looking at this attachment, the GS numbers after 3 hours with no dark readings followed by those of a subsequent dark reading are close to the same and seem to indicate that it was the display that had the shift, not the meter. Is that normal/possible? The display is an LCD that had a one hour warm up before I started.
LL
Reply
Reply
post #8 of 15
3/27/11 at 8:35am
Looks very stable to me. Seems as though you may be able to leave the meter up for your calibration run based on your results.

Dan
Reply
Reply
post #9 of 15
4/27/11 at 9:15am
Is it safe to say we can get away doing the same with a DTP-94 meter?
Reply
Reply
post #10 of 15
4/27/11 at 10:31am
Thread Starter 
Quote:
Originally Posted by lodolfan View Post

Is it safe to say we can get away doing the same with a DTP-94 meter?

It may have been my particular meter. It's easy enough for you to test yours.

1. Perform dark reading and take measurements.
2. Wait for some amount of time.
3. Take measurements.
4. Perform dark reading and take measurements.

If 1=3=4 then readings beyond the initial one not required.
If 3=4 but not =1 then the display has shifted.
If 1=4 but not =3 the meter has shifted and requires frequent calibrations.

The 3=4 but not=1 sample for my meter found max difference to be .002 and that was on only two of the 20 separate measurements, the rest being = or +/- .001. It's not like I'm calibrating studio monitors or something so that's close enough for me.
Reply
Reply
post #11 of 15
4/27/11 at 10:35am
Also you'll see bigger changes down at 10% or even 5%

The recommendations from us are based on the worst case scenario. The timers are there as a guideline, we don't require you to redo dark readings.
Reply
Reply
post #12 of 15
4/27/11 at 10:56am
Thread Starter 
Quote:
Originally Posted by sotti View Post

Also you'll see bigger changes down at 10% or even 5%

The recommendations from us are based on the worst case scenario. The timers are there as a guideline, we don't require you to redo dark readings.

I'm sure that is true. Since my first post in this thread Tom Huffman has advised that once a Chroma 5 is profiled to an i1Pro there is no reason to use the i1Pro. Consequently I now do both GS/Gamma and CMS with the C5.

My original test is with one i1Pro meter on one display over one time period. If anyone does a similar test with their meter I'd love to see the results posted here. My meter and test could be a fluke.
Reply
Reply
post #13 of 15
11/10/12 at 7:08am
Quote:
Originally Posted by buzzard767 View Post


I'm sure that is true. Since my first post in this thread Tom Huffman has advised that once a Chroma 5 is profiled to an i1Pro there is no reason to use the i1Pro. Consequently I now do both GS/Gamma and CMS with the C5.


My original test is with one i1Pro meter on one display over one time period. If anyone does a similar test with their meter I'd love to see the results posted here. My meter and test could be a fluke.

Despite profiling the i1Pro to the C5, the i1Pro will still give a better calibration!
After doing a profiled calibration, go back and read your results with the i1Pro. If Tom is correct , the results should be nearly identical. If not, maybe it's time to upgrade to a newer Meter to profile to.
Reply
Reply
post #14 of 15
11/10/12 at 7:28am
Quote:
Originally Posted by buzzard767 View Post


That is music to my ears, sotti, and I thank you for the information. It's not that I'm lazy but rather just trying to avoid a PITA.


Steve Shaw of LightIllusion.com gave me the same sort of answer on the LUT thread.

Buzzard, another thing you can do if you have not, is in calman, set the i1pro side button to do initialization. Then all you need to do is cover the hole and hit the button.. done.
NOTE: the button will not work if you have the initialization screen up..
Reply
Reply
post #15 of 15
11/10/12 at 8:22am
Thread Starter 
Quote:
Originally Posted by p5browne View Post

Despite profiling the i1Pro to the C5, the i1Pro will still give a better calibration!
After doing a profiled calibration, go back and read your results with the i1Pro. If Tom is correct , the results should be nearly identical. If not, maybe it's time to upgrade to a newer Meter to profile to.

No, not always. I'm into real 65x65x65 3D LUTs produced from 17 point (4913 reads) profiles and a lot of these reads are very low luminance beyond which the i1Pro can read with any level of accuracy. I don't need to go back and reread. First I take "before" measurements with the i1Pro, then I profile, then I take "after" reads with the profiled meter. The report is my comparison. One needs nothing more than that. New meter? I'm using a Klein K10-A. New enough? wink.gif

Quote:
Originally Posted by airscapes View Post

Buzzard, another thing you can do if you have not, is in calman, set the i1pro side button to do initialization. Then all you need to do is cover the hole and hit the button.. done.
NOTE: the button will not work if you have the initialization screen up..

I have the X-Rite tripod doodad. Time for a dark reading, slide the switch, take dark reading, reset the slide.

For the time being my tests are valid. I've asked others to try the same thing but nobody has....
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration

AVS Top Picks

AVS › AVS Forum › Display Devices › Display Calibration › i1Pro Dark Readings - I Don't Get It