The only requirement for a meter to adequately measure HDR signals is the ability to read very high light levels. I have tested both the retail and OEM i1 Display Pro meters (Rev. B) for this, and both pass with flying colors. See below.

HDR Test

There had been some suggestion that only an OEM Rev B meter would work with HDR. Apparently, this is not so.
BTW, I didn't have a Rev. A meter on hand to test, but prior experience suggests that it would not read as high as the Rev. B meters.

The statement that Retail i1D3 probes are only 1000 nits rated is directly from X-Rite.
I do not see why they would lie about that.
But who knows?

Steve

I didn't mean to accuse anyone of lying, but the results are pretty clear. The X-Rite specs appear to be overly conservative.

BTW, in looking at the numbers more closely, although luminance continues to measure well over 3,000 nits, the color accuracy takes a severe hit above 2,500 nits, which is still well within what one needs for HDR.

Is this Rev B regardless of date code (there is discussion in the other thread of the date being important for OEM meters).
Out of interest, is Rev A any different?

I had an OEM Rev A meter which I'm sure read much better that my current Retail Rev B at low light (in equivalent burst modes). I since sold the Rev A, but it has got my interest piqued so much I've acquired another bargains new-in-box OEM meter which I think is going to be a rev A due to the timeframe of when it was originally bought.

The clip showed that both Retail and OEM meters were 17B. Unknown if the results would be the same for a pre-17B meter.

The official information from X-Rite is that all i1D3 probes manufactured before Jan 2017 are certified to 1000 nits only.
Post Jan 2017 the OEM version is certified to 2000 nits, and the Retail version is still only certified to 1000 nits.
But, being certified to a specific nits level doesn't mean any probe will stop reading above that - just they are not certified to be accurate.

All i1D3 probes are certified to 0.1 nits min.
But again, will read lower, just not with any certainty.

However, the question regarding min level readings is an interesting one, and one that is really being overlooked in these discussions.
The PQ EOTF really does hold a lot of the main image detail low, with just spectral highlights being above 100 nits.
That actually makes readings above 100 nits a lot less critical than below... as there is little relative detail in highlights.

With many modern displays having very low blacks (SDR blacks are identical to HDR, regardless what some HDR marketing may attempt to suggest...) the ability for a probe to read well in the ultra shadows has become rather critical.

And the question of RevA, or pre Jan 2017, i1D3 probes having better low-light capabilities is an interesting one...

Steve

I should have an old OEM rev A here again soon and will be interested to compare this to my current rev B retail for low light. Maybe my old meter was a fluke, wish I'd held on to it.

The method described in Spectracal's C6HDR meter whitepaper would seem to be the most useful for looking at the HDR capabilities of a given meter in detail. What you really want is an automated jig which moves the meter and a diffuser (not sure if you could just use the built in diffuser?), taking measurements at regular intervals and plotting the x,y and Y - and then comparing them against the expected values at the given distance. You wouldn't even need a reference meter if you assume the meter is already good enough at the 100nit level and you started your measurements around that level.

Someone with such a jig could offer HDR meter characterisation as a service to end users - providing them with the data as to how their individual meter behaves. Maybe I'll get building...

Tom,

Greatly appreciate you performing this test. It appears that, despite X-Rite's statement on the more limited capability of the Retail version of the meter, it matches the OEM's performance in 2017. Still, I think I may purchase on OEM version as a backup for my 2017 Retail unit and do some tests of my own, with regards to low-light.

The low light issue has nothing to do with Retail vs. OEM. This is a software issue regarding the read mode that is used. X-Rite developed a new read mode called All-in-One (AIO) for the Rev B meters that offers good performance for all display types. However, the one downside to it is that it cuts off low light readings at about 0.01 nits. To get readings down to the meter's limit--which is about 0.0025 nits--the software has to use the old LCD or CRT read mode, preferably CRT.

In ChromaPure we accommodate this by offering a separate mode when the user initializes the meter that we call Slow Mode, which uses the CRT mode. You should use this only for taking contrast and black level readings, the only time measurements below 0.01 nits are useful. The rest of the time AIO mode is preferable.

I was not suggesting it was. Merely going to be a curiosity test for me to compare capabilities between different units. I'm well-aware of the improvements of AIO mode. With both Calman (in LLH) and HCFR, the lowest I can get my Retail Rev. B unit to measure a black window is 0.003 cd/m2. Thank you for the additional information.

I've had two rev B retail meters which were both less likely to manage to take a reading at the very darkest levels than an old rev A OEM meter, in the same modes, so I'm not convinced yet that there isn't some difference between the behaviour of either the hardware or firmware of the rev A vs rev B. I saw this difference in both Lightspace and HCFR. I've managed to source a new OEM rev A meter which I'm going to try again against my current rev B.

I noticed this issue on the first rev B I bought, returned it to store and got another one which was exactly the same.

As a projection user I couldn't care less about the HDR aspects unless they have a negative effect on the low light handling.

Ask your software vendors which of the X-Rite read modes they are using. The default mode is AIO, which only works with Rev B and which also causes this problem. The mode I refer to is behind the scenes and nothing the user would necessarily select or even be aware of.

The different CRT/LCD/Burst/AIO modes are to do with how the probe manages 'sync' with the display being profiled.
Using an incorrect mode on any given display will potentially cause read errors.

Steve

HCFR uses its own code for doing meter access (I believe it is using @gwgill 's ArgyllCMS under the hood). This bypasses the X-rite drivers.

Lightspace provides the option to choose meter modes, and as I say I was testing like for like mode. For JVC projectors it seems the "best" mode is Burst; so I was describing rev A OEM vs rev B retail meter both in Burst mode.

Odd, since HCFR doesn't use the AIO mode of the Rev B...

In HCFR, with my i1 Display PRO set to 1 sec. read time, I get the following with my Vizio M55 with White LED correction table:
:
18% Window 18 APL (GDI & MadTPG): .005 cd/m2

50% Window 50 APL: .004 cd/m2

100% Black Window, with TV menu pulled up: .003 cd/m2

What should I be getting?

0.003 sounds about right to me.

Pretty stoked to have picked up this new in box i1d3 OEM (unlocked) for €58 euros delivered from Germany... Going to test it for low light against my 2016 rev B retail. This is a 2014 OEM, so certainly rev A.

Sent from my D5803 using Tapatalk

Some great info in this thread and thanks to Tom for performing the experiment.

I've just checked my recently bought i1D3 retail variant and it has a 17B serial number, so should be future proof for a while at least.

Paul

Started a new thread on the subject of i1d3 low light handling as I don't want to polute this HDR thread.
https://www.avsforum.com/forum/139-di...-handling.html

The question is, is the Rev B good enough to calibrate HDR home theatre projectors ? None go to 1K nits. Is the SpectraCal C6 HDR2000 Colorimeter the same as the Display 3 Pro by ChromaPure ?

Yes and no. They are similar in that they both have extra display profiles provided and are more-or-less based on OEM I1D3s. They are different in that those extra profiles are only available in their respective software packages (CalMAN for the C6 and ChromaPure for the Display 3 Pro).

Thanks Rolls, is the standard retail version of the Rev B retail i1D3 close to the OEM version in terms of real world performance ? or is the OEM version with calibration profiles and ChromaPure/CalMAN much better ?

Which of the 2 apps is better in terms of ease of use and performance ?

You'll get lots of opinions on all your questions. The Retail I1D3 is supported directly or indirectly (one commercial package uses a modified driver for the Retail meters) by all software packages that I know of, as is the plain-Jane OEM version. The Retail is certified - whatever that means - to 1000 nits by X-Rite, and the post-Jan '17 OEMs are certified to 2000 nits. However, it has been reported that some post-Jan '17 Retail meters have also checked good to 2000 nits. X-Rite doesn't guarantee them to be, though, so your guess is as good as mine.
If you don't have or plan to have a spectro to profile your I1D3, one of the enhanced OEMs would be a good choice. You'd want to make sure whichever route you go has a specific meter profile included for your make, model (and now year) of display.

Projector calibration is my main interest and it does not have to be to reference levels because I'll always end up tweaking the image to my preference. How close does the retail version come to the OEM version ?

Most programs see and address them in exactly the same way, unless you have one of the "enhanced" versions. AFAIK, there is no difference in the default display profiles that X-Rite loads into retail and OEM EEPROMs. The OEMs cannot use the X-Rite profiler software to build ICC profiles, as they are locked out by X-Rite.

The latest generation of i1Display PRO (Rev.B) (OEM/Retail) it features a new refresh rate detection and synchronization AIO (All in One) measurement mode which can improve the measurement stability for certain displays and provide lower reading times.

AIO mode is supported only from LightSpace and ChromaPure.

HCFR/DisplayCAL/ArgyllCMS has also custom code for providing similar results with adaptive exposure times/sync modes without using AIO mode. CalMAN is using custom code also (but not adaptive exposure times for i1Display PRO)

i1Display PRO OEM meters manufactured from January 2017 or later are certified from X-Rite up to 2000nits, according to X-Rite the Retail version are certified for 1000nits.

SpectraCAL C6/C6 HDR is a OEM Branded i1Display PRO meter as we know, all i1Display PRO meters (Retail/OEM/OEM Branded) feature the exact same hardware/firmware, just C6 has different internal meter unlocking code (to make it work with CalMAN only officially....but it can work with HCFR/LightSpace/DisplayCAL/ArgyllCMS also unofficially; with exact same drivers; just with different unlocking code inside).

HCFR/Argyll is using their own custom drivers to operate the meter, the same SpectraCAL is doing also, they have written exclusive code to operate C6 (and i1D3 also) from inside CalMAN.

This is one reason why AIO mode of i1Display PRO Rev.B meters is not available from CalMAN/HCFR/Argyll, because they are not using X-Rite SDK to operate the meter, they are using their own custom code modes.

LightSpace has an additional feature called 'Intelligent Integration' where you set a nits value below which the i1Display PRO will perform a second, third, fourth, etc. reading, with increasing integration times, up to the maximum integration time of 6 seconds. The readings will stop as soon as a stable and valid reading is achieved. If no valid and stable reading is achieved a 'zero' value will be returned, enabling LightSpace to intelligently calculate a replacement value when generating a calibration LUT.

SpectraCAL has equipped the C6's meters with additional EDR (Emissive Display Reference) tables for various consumer or professional display types, which are installed to your PC every time you will install CalMAN. They can add new tables in a future software release also.

C6 or i1Display PRO, both are field upgrade-able, so X-Rite can add new EDR tables when they will release new SDK files to calibration software developers. (but X-Rite hasn't released something new from 2012).

Something additionally, X-Rite is storing the default calibration to the meter EEPROM (Generic CMF), all the additional tables are loaded 'temporary' to EEPROM when you select each one from the software you are using which is support these additional tables (EDR files), I mean these selections the i1Display PRO (OEM/Retail) is coming: CCFL-LCD, Wide Gamut CCFL-LCD, White LED, Wide Gamut RGB LED, OLED, Plasma, RG Phosphor, Projectors (DLP or LCD) with Ultra High Pressure (UHP) lamps.

The EDR files are available as files inside to the calibration software folder (for example to LightSpace or CalMAN folders).

ChromaPure is not supporting EDR files, so you will not see any of these tables as selection when you will use a normal OEM or Retail i1Display PRO instrument with ChromaPure, you will see the Generic Default.

ChromaPure supporting extra matrix tables (not EDR) to the ChromaPure's edition of the i1Display PRO called as 'Display 3 PRO' colorimeter.

As you're dealing with projection you probably don't care about 1000 or 2000 nit max as you're unlikely to ever go there.

My old rev A OEM reads a little lower than the rev B retail I have, though neither are particularly accurate down in the really dark levels.

The OEM is a bit less hassle if you use Lightspace as you don't need to replace the DLL every time you update. It is a simple process but nevertheless one I seem to forget to do often.

Has anyone measured what a pre-17B Retail can read?