i1d3 rev a vs rev B low light handling - Page 2 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 19Likes
Reply
 
Thread Tools
post #31 of 70 Old 08-10-2017, 07:16 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by Bloodwound View Post
I get output from 2 separate C6 (which are rev A) at 0.002 cd/m2. I dunno what Spectracal does or does not do to them (including their software), but it seems to be real since I've verified it with a couple of K10s. Of course, they're slow as hell at these leves. Anyways, also at these levels you need to use your eyes anyways to calibrate the right brightness level, so for all intents and purposes I think they are suited for calibrating OLEDs correctly. At least when profiled.

I've not verified the max light output they will read, or at which luminance levels they get unstable, but my guess they're good enough for the foreseeable future. The included Vizio QD profile matched Samsung JS9500 perfectly when profiled with a spectro. Pretty amazing. (But the luminance levels between 2 C6 didn't quite match)
I don't think spectral have done anything to them. I get readings around that low in CRT mode on the Rev A OEM meter that look like they might be plausible, the rev B meter gets no where near. When you say you've verified them do you mean you've verified the luminance reading or the chroma reading or both?

Sent from my D5803 using Tapatalk
bobof is online now  
Sponsored Links
Advertisement
 
post #32 of 70 Old 08-10-2017, 07:34 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,949
Mentioned: 130 Post(s)
Tagged: 0 Thread(s)
Quoted: 4884 Post(s)
Liked: 1848
Quote:
Originally Posted by bobof View Post
What model JVC do you have that gets 0.004 sat IRE6? Or is peak white quite dim?
The 0.004 reading was at 6/255 or about IRE2.5, not IRE6.
bobof likes this.
Dominic Chan is offline  
post #33 of 70 Old 08-10-2017, 07:47 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by Dominic Chan View Post
The 0.004 reading was at 6/255 or about IRE2.5, not IRE6.
Thanks. Reading/replying on my phone usually results in not catching everything!

Sent from my D5803 using Tapatalk
bobof is online now  
Sponsored Links
Advertisement
 
post #34 of 70 Old 08-10-2017, 08:03 AM
Member
 
Bloodwound's Avatar
 
Join Date: Apr 2007
Posts: 199
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 128 Post(s)
Liked: 39
Quote:
Originally Posted by bobof View Post
I don't think spectral have done anything to them. I get readings around that low in CRT mode on the Rev A OEM meter that look like they might be plausible, the rev B meter gets no where near. When you say you've verified them do you mean you've verified the luminance reading or the chroma reading or both?

Sent from my D5803 using Tapatalk
I believe it's both. It's been a year since I've calibrated my own display, so I'm probably going to calibrate it again next week. I'll be sure to do a check on the C6 meters then. ( My kuro is almost OLED black after tweaking the voltages).
ConnecTEDDD likes this.
Bloodwound is offline  
post #35 of 70 Old 08-10-2017, 02:30 PM
Member
 
fatjulio's Avatar
 
Join Date: Dec 2004
Location: New Zealand
Posts: 170
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 14
Quote:
Originally Posted by Dominic Chan View Post
If you have a neutral density filter you can confirm which integration time gives you the best reading.
For example, with a 10x ND filter you can first take readings of the 200,200,200 pattern with and without the filter to characterize it, then read the 20,20,20 pattern with and without the filter. That way you can deduce the correct reading at 2,2,2.
Just had a thought. My JVC is zoomed to max image size. I could zoom to min size a take a reading of that for "reference", it should be a lot brighter. I do have a 2x ND filter, but not sure it'd be enough.
fatjulio is offline  
post #36 of 70 Old 08-10-2017, 02:51 PM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by fatjulio View Post
Just had a thought. My JVC is zoomed to max image size. I could zoom to min size a take a reading of that for "reference", it should be a lot brighter. I do have a 2x ND filter, but not sure it'd be enough.
Aperture can definitely and I imagine zoom can also affect the colour balance coming out of the projector, so it must be best to have the projector set up as it is expected to be used.

Sent from my D5803 using Tapatalk
bobof is online now  
post #37 of 70 Old 08-10-2017, 03:05 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,949
Mentioned: 130 Post(s)
Tagged: 0 Thread(s)
Quoted: 4884 Post(s)
Liked: 1848
Quote:
Originally Posted by bobof View Post
Aperture can definitely and I imagine zoom can also affect the colour balance coming out of the projector
Aperture definitely, but zoom should not affect the colour balance. In any case, that can be verified with a high IRE pattern.
fatjulio and bobof like this.
Dominic Chan is offline  
post #38 of 70 Old 08-16-2017, 01:04 AM
AVS Forum Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 7,890
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 1000 Post(s)
Liked: 839
Here's a video showing the low light ability of the i1D3 Rev B meter, at least with respect to luminance, which at below 0.01 is all that matters. It is just a matter of using the right read mode with adequate integration time.

Low-Light Demo

One issue I did see was that the meter will only read once and does not appear to function properly in continuous mode. I gather this is what some of you have been calling "fake readings". They are not really fake. The reading is accurate. It is just not repeatable for whatever reason.
Rolls-Royce, ivan2011 and bobof like this.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
post #39 of 70 Old 08-16-2017, 01:52 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by TomHuffman View Post
Here's a video showing the low light ability of the i1D3 Rev B meter, at least with respect to luminance, which at below 0.01 is all that matters. It is just a matter of using the right read mode with adequate integration time.

Low-Light Demo

One issue I did see was that the meter will only read once and does not appear to function properly in continuous mode. I gather this is what some of you have been calling "fake readings". They are not really fake. The reading is accurate. It is just not repeatable for whatever reason.
I'm more interested in which is better (Rev A vs Rev B). While I can also coax luminance readings out of the Rev B at 0.003 nits, there is something odd going on with it.

Look at the results I captured using LS. The reads at values 0,0,0 and 1,1,1 are cut short for some reason. This adds fuel to the fire of there being something less believable about them. The rev A unit doesn't do this and doesn't have the "fake" readings issue at the same light level - this behaviour is consistent with two other rev A & B pairs I have seen.

Quote:
05/08/2017 23:35:09, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:12, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:16, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:19, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:23, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:27, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:30, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:34, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:37, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:41, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:44, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:48, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:51, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:55, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:58, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:02, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:05, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:09, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:12, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:16, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3016ms
05/08/2017 23:36:19, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:23, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:26, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:30, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:34, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:37, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:41, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:44, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:48, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:51, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:55, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:58, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:37:02, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:37:11, 2, 2, 2, 0.0053, 0.2633, 0.3358, 8922ms
05/08/2017 23:37:21, 2, 2, 2, 0.0053, 0.2652, 0.3356, 8922ms
05/08/2017 23:37:30, 2, 2, 2, 0.0054, 0.2668, 0.3369, 8859ms
05/08/2017 23:37:39, 2, 2, 2, 0.0053, 0.2665, 0.3362, 8890ms
05/08/2017 23:37:49, 2, 2, 2, 0.0054, 0.2643, 0.3381, 8829ms
05/08/2017 23:37:58, 2, 2, 2, 0.0053, 0.2640, 0.3373, 8859ms
05/08/2017 23:38:07, 2, 2, 2, 0.0057, 0.2654, 0.3487, 8469ms
05/08/2017 23:38:15, 3, 3, 3, 0.0067, 0.2774, 0.3711, 7703ms
05/08/2017 23:38:24, 3, 3, 3, 0.0067, 0.2771, 0.3706, 7719ms
05/08/2017 23:38:32, 3, 3, 3, 0.0067, 0.2771, 0.3706, 7719ms
05/08/2017 23:38:40, 3, 3, 3, 0.0067, 0.2784, 0.3701, 7735ms
05/08/2017 23:38:48, 3, 3, 3, 0.0067, 0.2777, 0.3719, 7672ms
05/08/2017 23:38:57, 3, 3, 3, 0.0067, 0.2783, 0.3704, 7719ms
05/08/2017 23:39:05, 3, 3, 3, 0.0067, 0.2783, 0.3703, 7719ms
bobof is online now  
post #40 of 70 Old 08-16-2017, 02:32 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,398
Mentioned: 207 Post(s)
Tagged: 1 Thread(s)
Quoted: 3258 Post(s)
Liked: 4050
Quote:
Originally Posted by bobof View Post
I'm more interested in which is better (Rev A vs Rev B). While I can also coax luminance readings out of the Rev B at 0.003 nits, there is something odd going on with it.

Look at the results I captured using LS. The reads at values 0,0,0 and 1,1,1 are cut short for some reason. This adds fuel to the fire of there being something less believable about them. The rev A unit doesn't do this and doesn't have the "fake" readings issue at the same light level - this behaviour is consistent with two other rev A & B pairs I have seen.


Hi, if you are using a notebook output to generate the patterns (it will be unknown if its bit-perfect) so better to use your player with my calibration disk and measure the CalMAN Near Black Chapter which has 0% (16.16.16) and 0.5% (17.17.17) grayscale patterns inside. See if the i1d3 can measure the difference between these 2 levels...in case your notebook is displaying the same patch when you use the software pattern generator.

Display that 2 patterns from my disk and use LightSpace with measure & log to check the results.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #41 of 70 Old 08-16-2017, 02:40 AM
Member
 
ivan2011's Avatar
 
Join Date: Jun 2011
Posts: 49
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked: 16
Quote:
Originally Posted by TomHuffman View Post
Here's a video showing the low light ability of the i1D3 Rev B meter, at least with respect to luminance, which at below 0.01 is all that matters. It is just a matter of using the right read mode with adequate integration time.

Low-Light Demo

One issue I did see was that the meter will only read once and does not appear to function properly in continuous mode. I gather this is what some of you have been calling "fake readings". They are not really fake. The reading is accurate. It is just not repeatable for whatever reason.
It is nice to see a direct comparison to a higher class meter compared with the i1d3. I understand that they could differ even if slightly from unit to unit, does the i1d3 offer the same level of luminance accuracy at higher luminance levels?
ivan2011 is offline  
post #42 of 70 Old 08-16-2017, 03:08 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by ConnecTEDDD View Post
Hi, if you are using a notebook output to generate the patterns (it will be unknown if its bit-perfect) so better to use your player with my calibration disk and measure the CalMAN Near Black Chapter which has 0% (16.16.16) and 0.5% (17.17.17) grayscale patterns inside. See if the i1d3 can measure the difference between these 2 levels...in case your notebook is displaying the same patch when you use the software pattern generator.

Display that 2 patterns from my disk and use LightSpace with measure & log to check the results.
I'm using the internal pattern generator in the Lumagen 2143, so I guess it is accurate enough for this... The following were done on the same display within 20 minutes of each other. They are obviously quite different.

i1d3 OEM A 2014 CRT 3s int
Quote:
05/08/2017 23:23:37, 0, 0, 0, 0.0032, 0.3249, 0.2266, 12375ms
05/08/2017 23:23:50, 0, 0, 0, 0.0032, 0.3247, 0.2265, 12329ms
05/08/2017 23:24:03, 0, 0, 0, 0.0032, 0.3248, 0.2266, 12328ms
05/08/2017 23:24:16, 0, 0, 0, 0.0033, 0.3329, 0.2276, 12031ms
05/08/2017 23:24:28, 1, 1, 1, 0.0033, 0.3330, 0.2276, 11906ms
05/08/2017 23:24:41, 1, 1, 1, 0.0033, 0.3330, 0.2276, 11984ms
05/08/2017 23:24:53, 1, 1, 1, 0.0033, 0.3323, 0.2275, 11969ms
05/08/2017 23:25:05, 1, 1, 1, 0.0036, 0.3586, 0.2309, 11110ms
05/08/2017 23:25:16, 2, 2, 2, 0.0038, 0.3720, 0.2327, 10297ms
05/08/2017 23:25:26, 2, 2, 2, 0.0038, 0.3722, 0.2327, 10266ms
05/08/2017 23:25:37, 2, 2, 2, 0.0038, 0.3731, 0.2328, 10234ms
05/08/2017 23:25:48, 2, 2, 2, 0.0038, 0.3730, 0.2328, 10235ms
05/08/2017 23:25:59, 2, 2, 2, 0.0038, 0.3724, 0.2327, 10250ms
05/08/2017 23:26:08, 2, 2, 2, 0.0043, 0.4098, 0.2376, 8922ms
05/08/2017 23:26:20, 3, 3, 3, 0.0059, 0.3078, 0.3319, 11032ms
05/08/2017 23:26:31, 3, 3, 3, 0.0059, 0.3109, 0.3310, 11031ms
05/08/2017 23:26:43, 3, 3, 3, 0.0059, 0.3077, 0.3320, 10985ms
05/08/2017 23:26:54, 3, 3, 3, 0.0059, 0.3091, 0.3318, 11000ms
05/08/2017 23:27:06, 3, 3, 3, 0.0059, 0.3079, 0.3320, 11032ms
i1d3 Retail B 2016 CRT 3s int
Quote:
05/08/2017 23:35:09, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:12, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:16, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:19, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:23, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:27, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:30, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:34, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:37, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:41, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:44, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:48, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:51, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:55, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:35:58, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:02, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:05, 0, 0, 0, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:09, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:12, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:16, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3016ms
05/08/2017 23:36:19, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:23, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:26, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:30, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:34, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:37, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:41, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:44, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:48, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:51, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:55, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:36:58, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:37:02, 1, 1, 1, 0.0025, 0.2171, 0.2243, 3000ms
05/08/2017 23:37:11, 2, 2, 2, 0.0053, 0.2633, 0.3358, 8922ms
05/08/2017 23:37:21, 2, 2, 2, 0.0053, 0.2652, 0.3356, 8922ms
05/08/2017 23:37:30, 2, 2, 2, 0.0054, 0.2668, 0.3369, 8859ms
05/08/2017 23:37:39, 2, 2, 2, 0.0053, 0.2665, 0.3362, 8890ms
05/08/2017 23:37:49, 2, 2, 2, 0.0054, 0.2643, 0.3381, 8829ms
05/08/2017 23:37:58, 2, 2, 2, 0.0053, 0.2640, 0.3373, 8859ms
05/08/2017 23:38:07, 2, 2, 2, 0.0057, 0.2654, 0.3487, 8469ms
05/08/2017 23:38:15, 3, 3, 3, 0.0067, 0.2774, 0.3711, 7703ms
05/08/2017 23:38:24, 3, 3, 3, 0.0067, 0.2771, 0.3706, 7719ms
05/08/2017 23:38:32, 3, 3, 3, 0.0067, 0.2771, 0.3706, 7719ms
05/08/2017 23:38:40, 3, 3, 3, 0.0067, 0.2784, 0.3701, 7735ms
05/08/2017 23:38:48, 3, 3, 3, 0.0067, 0.2777, 0.3719, 7672ms
05/08/2017 23:38:57, 3, 3, 3, 0.0067, 0.2783, 0.3704, 7719ms
05/08/2017 23:39:05, 3, 3, 3, 0.0067, 0.2783, 0.3703, 7719ms
bobof is online now  
post #43 of 70 Old 08-16-2017, 03:17 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,398
Mentioned: 207 Post(s)
Tagged: 1 Thread(s)
Quoted: 3258 Post(s)
Liked: 4050
Quote:
Originally Posted by bobof View Post
I'm using the internal pattern generator in the Lumagen 2143, so I guess it is accurate enough for this...
I have checked with DVDO AVLab TPG Color Checker (which displays the RGB or YCbCr triplet of a selected pixel) the accuracy of Lumagen 2143, the bit-perfect input/output was only the YCbCr 4:2:2, all other options (RGB-Video or 4:4:4) were adding a random +-1 level difference to any subpixel level of a any pixel, when you measure a solid patch for example which supposed all pixels to have the same RGB triplet, seems that some additional internal processing which you can't bypass it active.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5

Last edited by ConnecTEDDD; 08-16-2017 at 03:23 AM.
ConnecTEDDD is online now  
post #44 of 70 Old 08-16-2017, 03:41 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by ConnecTEDDD View Post
I have checked with DVDO AVLab TPG Color Checker (which displays the RGB or YCbCr triplet of a selected pixel) the accuracy of Lumagen 2143, the bit-perfect input/output was only the YCbCr 4:2:2, all other options (RGB-Video or 4:4:4) were adding a random +-1 level difference to any subpixel level of a any pixel, when you measure a solid patch for example which supposed all pixels to have the same RGB triplet, seems that some additional internal processing which you can't bypass it active.
Sounds like some kind of dithering going on internally. Useful info. Makes sense I guess as probably all the processing is being done in one colour space.

In any case my display is set up for YCbCr 4:2:2 so I guess I have the "good" path (internal TPG -> YCbCr 4:2:2)
bobof is online now  
post #45 of 70 Old 08-16-2017, 03:43 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,398
Mentioned: 207 Post(s)
Tagged: 1 Thread(s)
Quoted: 3258 Post(s)
Liked: 4050
Quote:
Originally Posted by bobof View Post
Sounds like some kind of dithering going on internally. Useful info. Makes sense I guess as probably all the processing is being done in one colour space.

In any case my display is set up for YCbCr 4:2:2 so I guess I have the "good" path (internal TPG -> YCbCr 4:2:2)
Exactly, YCbCr 4:2:2 is accurate/bit-perfect setting.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #46 of 70 Old 08-16-2017, 03:51 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,836
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 780 Post(s)
Liked: 1146
The latest build of LS allows for 6 seconds Max Integration time for the i1D3.
It should be easier to see if there is something strange going on when the apparent 'fake readings' happen...

Steve
bobof likes this.

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #47 of 70 Old 08-16-2017, 04:01 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by Light Illusion View Post
The latest build of LS allows for 6 seconds Max Integration time for the i1D3.
It should be easier to see if there is something strange going on when the apparent 'fake readings' happen...

Steve
I will have a play. Thanks Does this also apply to the AIO mode in rev B?
bobof is online now  
post #48 of 70 Old 08-16-2017, 06:57 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,836
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 780 Post(s)
Liked: 1146
Integration Time is totally separate from any other probe setting...

Steve
bobof likes this.

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #49 of 70 Old 08-17-2017, 02:49 AM
AVS Forum Special Member
 
TomHuffman's Avatar
 
Join Date: Jan 2003
Location: Springfield, MO
Posts: 7,890
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 1000 Post(s)
Liked: 839
Quote:
Originally Posted by ivan2011 View Post
It is nice to see a direct comparison to a higher class meter compared with the i1d3. I understand that they could differ even if slightly from unit to unit, does the i1d3 offer the same level of luminance accuracy at higher luminance levels?
That's an interesting question worth investigating.

i1D3 Luminance Accuracy

According to this test, the i1D3 under reported video white by about 6.7 nits or 5.5%. It also reported a black level that was off by 0.0007 nits. Also, there is a non-trivial unit-to-unit variation among i1D3s, so I would only take this as a rough approximation, rather than as a precise comparison.

Tom Huffman
ChromaPure Software/AccuPel Video Signal Generators
ISF/THX Calibrations
Springfield, MO

TomHuffman is offline  
post #50 of 70 Old 08-17-2017, 05:38 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,949
Mentioned: 130 Post(s)
Tagged: 0 Thread(s)
Quoted: 4884 Post(s)
Liked: 1848
Quote:
Originally Posted by TomHuffman View Post
According to this test, the i1D3 under reported video white by about 6.7 nits or 5.5%. It also reported a black level that was off by 0.0007 nits. Also, there is a non-trivial unit-to-unit variation among i1D3s, so I would only take this as a rough approximation, rather than as a precise comparison.
The specification of the luminance accuracy of the i1D3 is +/-4%, so this particular meter seems to be out-of-spec.
Dominic Chan is offline  
post #51 of 70 Old 08-17-2017, 05:48 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,398
Mentioned: 207 Post(s)
Tagged: 1 Thread(s)
Quoted: 3258 Post(s)
Liked: 4050
Quote:
Originally Posted by Dominic Chan View Post
The specification of the luminance accuracy of the i1D3 is +/-4%, so this particular meter seems to be out-of-spec.
Hi Dominic,

The CR-100 has also +-x% luminance accuracy, Colorimetry Research specify +-2% @ 1.37 nits only.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #52 of 70 Old 08-17-2017, 12:17 PM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
If I get time I will build a jig for testing meter reading linearity. So I will probably never build it.

The problem with referencing another meter across different luma levels is unless you have detailed data about that other meter who knows, right? Even if it was calibrated at the factory, it is only calibrated to their standards (which sometimes might be "looser than a wizard's sleeves"...!)

The idea for the jig is like Spectracal use for HDR meter calibration. Get a light source at one end of a long box with tripod mount at other end for meter. Moving diffuser in between the light source and the meter, while performing measurements and logging the distance of the meter to the light source. Automate it if possible.

This jig should be able to show for any meter charts of the linearity of both the luma and chroma readings over a range of luma levels. In this most basic form the readings would be self-relative - so it would just show linearity and not absolute values, but if you had a meter which had known highly accurate calibrated performance at just a single luma level it would then be possible to generate absolute calibrated data referencing this meter.
bobof is online now  
post #53 of 70 Old 08-17-2017, 04:25 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,949
Mentioned: 130 Post(s)
Tagged: 0 Thread(s)
Quoted: 4884 Post(s)
Liked: 1848
Quote:
Originally Posted by bobof View Post
The idea for the jig is like Spectracal use for HDR meter calibration. Get a light source at one end of a long box with tripod mount at other end for meter. Moving diffuser in between the light source and the meter, while performing measurements and logging the distance of the meter to the light source.
Are you using some mathematical model to "predict" the measured lumanance as a function of the distance? I suspect the errors in the mathematical prediction and distance measurements will far exceed the meter linearity error. Spectracal compares the measurements against a reference meter, when using the jig.

Last edited by Dominic Chan; 08-17-2017 at 04:40 PM.
Dominic Chan is offline  
post #54 of 70 Old 08-17-2017, 04:45 PM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by Dominic Chan View Post
Are you using some mathematical model to "predict" the measured lumanance as a function of the distance? I suspect the errors in the mathematical prediction and distance measurements will far exceed the meter linearity error. Spectracal compares the measurements against a reference meter, when using the jig.
No they don't.

Quote:
The distance of the diffuser from the light source was then incrementally decreased to produce higher luminance levels at the meter test position. The required shorter diffuser distance to produce a desired higher luminance could be easily calculated, since the diffuser luminance is proportional to the square of the distance from the light source to the diffuser.
At some luminance level, the luminance measured by each C6 stopped increasing proportionally as the brightness of the diffuser was increased. It was found that this luminance saturation level was about 1700 cd/m2 for most C6 meters.
Edit: I should add, that is the whole point of creating a jig, to carefully control the distance. Using reasonably cheap lead screws, stepper motors and linear bearings it is possible to position something to within a fraction of a mm. I have a 3D printer running in my workshop at the moment doing exactly that .

Last edited by bobof; 08-17-2017 at 05:07 PM.
bobof is online now  
post #55 of 70 Old 08-17-2017, 05:54 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,949
Mentioned: 130 Post(s)
Tagged: 0 Thread(s)
Quoted: 4884 Post(s)
Liked: 1848
Quote:
Originally Posted by bobof View Post
No they don't.
Quote:
The distance of the diffuser from the light source was then incrementally decreased to produce higher luminance levels at the meter test position.
SpectraCal only do that for the "first round", to screen the meters for the ones that have higher saturation levels. That part of the process does not require any precise measurements.

The second round is when they actually certify the meters:
Quote:
Certification of new C6 HDR Colorimeters is now a simple process of comparing a C6 to the CS-2000 reference meter in the HDR certification fixture, to check whether the C6 exceeds a tolerance of ±2.0% for luminance and ±0.002 for x,y chromaticity at a luminance of 1300 cd/m2 or greater.
Ref: http://files.spectracal.com/Document...20C6%20HDR.pdf

Quote:
I should add, that is the whole point of creating a jig, to carefully control the distance. Using reasonably cheap lead screws, stepper motors and linear bearings it is possible to position something to within a fraction of a mm.
That gives you a very precis measure of the travel. For the mathematical model to hold, you need to be able to determine accurately the absolute distance between equivalent point source light and the diffuser. And even then the inverse square mathematical model is still only an approximation.

Last edited by Dominic Chan; 08-17-2017 at 10:25 PM.
Dominic Chan is offline  
post #56 of 70 Old 08-17-2017, 06:15 PM
AVS Forum Special Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 1,106
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 358 Post(s)
Liked: 259
Quote:
Originally Posted by TomHuffman View Post
According to this test, the i1D3 under reported video white by about 6.7 nits or 5.5%. It also reported a black level that was off by 0.0007 nits.
I think that's close to the quantization limit for the i1D3 at its max integration time, so no real surprise.
gwgill is offline  
post #57 of 70 Old 08-18-2017, 04:43 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by Dominic Chan View Post
SpectraCal only do that for the "first round", to screen the meters for the ones that have higher saturation levels. That part of the process does not require any precise measurements.

The second round is when they actually certify the meters:


Ref: http://files.spectracal.com/Document...20C6%20HDR.pdf


That gives you a very precis measure of the travel. For the mathematical model to hold, you need to be able to determine accurately the absolute distance between equivalent point source light and the diffuser. And even then the inverse square mathematical model is still only an approximation.
It looks like they only certify the meters at a single maximum point with the fixture and Minolta.

I take your points on practicalities of plotting the luma linearity using this method re: the absolute vs relative distance vs point source lighting vs mathematical model. I'm sure making a true non-reflective light sealed box isn't trivial either.

I'm sure the diffuser to light source distance could be measured accurately enough and verified using a small range of travel well within the meters known good range. In any case I'm in the middle of building a new DIY 3D printer with as it happens a whole load of stepper motors and linear bearings; I might divert some of the parts into having a play with this if I get a chance.

However, the most interesting aspect of the Spectracal document is their observation that the x,y divergence away from +/- 0.002 actually happens MUCH sooner than the non-linearity of the Y measurement. So if you choose to believe that then you can probably garner a lot of useful information about your meter's performance without even bothering to look at the Y measurement - you simply plot the x,y values against Y while moving the diffuser relative to the lamp with the meter in a fixed position.

At the point that the readings diverge from being within +/-0.002 of the values at a sensible level you know your meter is no longer useful at that luma level.

And as the device is a colorimeter - you probably don't even really care about the offset of the reading against the "true" x,y value as you'd be offsetting against your display anyway with your own spectro.

I wonder how difficult it is to assemble an adjustable diffused light source which is sufficiently homogeneous over the kind of distance required.
bobof is online now  
post #58 of 70 Old 02-12-2018, 05:18 AM
AVS Forum Special Member
 
shoman94's Avatar
 
Join Date: Jun 2006
Location: Maine
Posts: 7,207
Mentioned: 203 Post(s)
Tagged: 1 Thread(s)
Quoted: 5071 Post(s)
Liked: 3581
This is an interesting topic. I'll have to look at my box but I have a retail version I bought in April of 2016. I get consistent black readings even on continuous of 0.0014-0.0016 cd/m². To get these reading though I need to set SYNC to on in CalMAN.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Sony LCD Master Series Calibration Thread
My P75-C1 Calibration Settings (5.0.14.1).
Sony XBR-75Z9F, Vizio P75-C1, Pioneer Kuro 5020, Pioneer VSX-1131, XBox One(X), Wii, NVIDIA SHIELD.v1, FireTV 4K-HDR, HDHR Connects, QNAP 431+
shoman94 is offline  
post #59 of 70 Old 02-12-2018, 06:10 AM - Thread Starter
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,278
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 2646 Post(s)
Liked: 1744
Quote:
Originally Posted by shoman94 View Post
This is an interesting topic. I'll have to look at my box but I have a retail version I bought in April of 2016. I get consistent black readings even on continuous of 0.0014-0.0016 cd/m². To get these reading though I need to set SYNC to on in CalMAN.
I guess they may behave differently with different display tech; my retail i1d3 B retails all struggle to return something around 0.003cd/m2 from my projector (as you say, sync helps). The OEM A returns "something" much easier.

You don't have a DIY meter profile loaded do you? That could skew your reported levels.
bobof is online now  
post #60 of 70 Old 02-12-2018, 06:41 AM
AVS Forum Special Member
 
shoman94's Avatar
 
Join Date: Jun 2006
Location: Maine
Posts: 7,207
Mentioned: 203 Post(s)
Tagged: 1 Thread(s)
Quoted: 5071 Post(s)
Liked: 3581
Quote:
Originally Posted by bobof View Post
I guess they may behave differently with different display tech; my retail i1d3 B retails all struggle to return something around 0.003cd/m2 from my projector (as you say, sync helps). The OEM A returns "something" much easier.

You don't have a DIY meter profile loaded do you? That could skew your reported levels.
I have a ColorMunki Photo Spectro for my profile, but with the profile disabled using the standard BG LED profile on my Vizio P I get the same results. As of right now I don't know what REV mine is, but I thought it was a Rev-B.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Sony LCD Master Series Calibration Thread
My P75-C1 Calibration Settings (5.0.14.1).
Sony XBR-75Z9F, Vizio P75-C1, Pioneer Kuro 5020, Pioneer VSX-1131, XBox One(X), Wii, NVIDIA SHIELD.v1, FireTV 4K-HDR, HDHR Connects, QNAP 431+

Last edited by shoman94; 02-12-2018 at 08:15 AM.
shoman94 is offline  
Sponsored Links
Advertisement
 
Reply Display Calibration

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off