AVS Forum banner

CalMan 5 Release Notes and Discussion

719K views 9K replies 455 participants last post by  Vishwa Somayaji 
#1 ·
#4,279 ·
One last thing about profiling. I highly recommend, as we have spoken about, to use Ted's profile verifier workflow to validate the profile you take. Again, for many reasons, sometimes the profile you take can be off and the only way to know this using Calman for example, is to use his verification work flow.
 
  • Like
Reactions: ConnecTEDDD
#4,286 · (Edited)
#mrtickleuk I should have reposted. Since RC2 came out, I haven't had to set gamma at +3. But you will still want to make sure that the gamma is lower (brighter) than the target on the two-point gamma check. This way the autocal can bring gamma in line by lowering luminance rather than by trying to boost it. Also our KS series TVs work best with brightness at the default of 45 for both SDR and HDR. Use the Samsung 2016 Quantum Dot profile for your C-6. The advice given in the CalMAN document you linked to is for non-C-6 colorimeters. I've also found that you may need to set meter sync to "on" or "off" if you see weird instability in readings (red or blue shooting up or down wildly) at around 40% gray. PM me if you have questions.
 
#4,289 ·


For the users which perform HDR10 calibration using the latest released CalMAN 5.8.2.78, there is a bug to the HDR10 workflow default settings.

If you open the Settings panel, you will see that it's selected by default the REC.2020 SDR but with ST.2084 HDR (PQ) Gamma formula, so it's setting as target the 100 nits and not the 10000nits.

Earlier versions of that workflow didn't had that bug.
 
#4,297 ·
5.8.2a RC2 Build 85

https://calman.spectracal.com/downloads.html

CalMAN 2017 R2
Version: 5.8.2a Build: 85
Official Release: December 18, 2017
New Features
- CalMAN 2017 R2 (5.8.2) requires CalMAN or All Access (Maintenance) purchase as of September 2016 or later.
- CalMAN no longer deactivates itself when upgrading. Licenses at this point should remain stable. If you are running CalMAN 5.7.2 or earlier, you will need to manually deactivate CalMAN before upgrading.
- Updated our method for calculating 3D LUTs. The new 3D LUT calculation will generally give a slight accuracy improvement over the previous method.
- Added SDR AutoCal support for Samsung 2016 models which have the xlink plug. Note - We cannot get the state of the DDC controls upon connection. Performing a full DDC reset in CalMAN will sync the DDC controls with the display.
Ultimate, Professional, Video Pro, Studio, ISF, Home Enthusiast, Control
- Added HDR Analysis Workflow
Ultimate, Studio, Professional, Video Pro, ISF, Home Enthusiast
- Added support for the HDFury SDK. This enables the HDFury Integral support without the need to copy and paste data from their software. It is listed under HDFury for the manufacturer. The old method is still there under AVTOPcontroller. The driver for the HDFury Integral needs to be installed for this to work properly. It can be downloaded from http://www.hdfury.com.
​Ultimate, Professional, Video Pro, ISF, Studio, ColorMatch, Home Enthusiast
Changes
​Ultimate, Professional, Video Pro, ISF, Studio, ColorMatch, Home Enthusiast
- Added options in CalMAN to select VideoForge PRO 4K resolutions and the Multiburst pattern
- Minor update to the HDR10 Calibration workflow
Resolved Issues
- Resolved issue connecting to the SpectraCal - Colorbox, eeColor LUT box.
- Resolved issue that would cause CalMAN to not properly display the C6 HDR2000 label on the meter tab with some installers.
- Resolved an issue with the 2017 Samsung display where CalMAN was unable to make changes to CalDay or CalNight display modes.
- Resolved an issue with the Levels Editor in regions where the comma is used for decimal.
​- Resolved an issue with the Save button on the Color Cube (3D LUT) workflow.


One small but important change
- Resolved an issue which would cause CalMAN to reject a license which was generated in 2018.
 
  • Like
Reactions: mrtickleuk
#4,307 ·
Warning !

I like to rename my laptop to where ever it's being used : ie basement, my bedroom, my wife's loft, etc.
This kills the license ! Have requested support for my Calman, to get it going again.- yipes!
 
  • Like
Reactions: galonzo
#4,310 · (Edited)
I like to rename my laptop to where ever it's being used : ie basement, my bedroom, my wife's loft, etc.
This kills the license ! Have requested support for my Calman, to get it going again.- yipes!
Yow! You're telling CalMAN it's installed on a different computer every time you do that. No wonder you ran out of activations and had to have the license reset...:)
 
#4,308 · (Edited)
5000ES HDR calibration (tone mapping)

Happy New 2018 year to all forumers :)

After successfully done SDR/Rec709, I'm currently trying to calibrate my Sony VPL-VW5000ES projector in HDR10/Rec2020 (without Lumagen Pro).

I can't do correct following of the EOTF curve even from the first screen (precalibration) of the Calman's HDR10 workflow: see attached screen capture.

As per this french calibrator report : http://performances-home-cinema.fr/2016/12/30/test-sony-vpl-vw550es-vw520es-calibration-hdr-rec2020/ Chapter "CALIBRATION HDR /  REC2020 pour la lecture des Blu-ray 4K HDR-REC2020" - Use Google translator to translate it in English ;) :

I can do a "custom" EOTF curve to remap the all ST2084 tones from 1000nits to the 100-150nits peak brightness of the projector.
It's just a matter of including a multiplicator coefficient to the ST2084 curve within the calibration software to rescale the EOTF curve perfectly. As an exemple, in the above link, the VPL-VW550ES must use "x4" coefficient while the VPL-VW520ES use "x6" coefficient within LightSpace calibration software.


As the author said :
With the Sony VPL-VW550ES and VPL-VW520ES there is therefore use of a multiplying coefficient to scale ST2084. For example, for the VPL-VW550ES with its 100 nits, in my room, it is a coefficient of 4. This means that the reference curve corresponds in every point to ST2084 cut from 4 × 100 = 400 nits. It requires an analysis software capable of applying this scaling to measure it.

In HDR, Sony applies ST2084 curve cutted to 400 nits (for my room) on signal processing.
Note: the HDR contrast setting makes it possible to adapt this coefficient according to the environment.


So far, I did not found something similar (multiplier) within Calman 5.8.2,isn't it ?
--> So I can't use my brand new Calman calibration software to do my projector's HDR10 calibration :( :(

Could someone in Spectracal can tell me when we will have this (quite simple) feature (multiplicator coefficient) to let us (projectors owners) having correct tone mapping to proceed with HDR10 calibration workflow in Calman ???

Thank you.

John
 

Attachments

#4,311 ·
Could someone in Spectracal can tell me when we will have this (quite simple) feature (multiplicator coefficient) to let us (projectors owners) having correct tone mapping to proceed with HDR10 calibration workflow in Calman ???

Thank you.

John
I've been asking Spectracal to implement this for about two years, without any success.

In the meantime, simply use the screen offset. Set Y to the multiplier you want to target, leave x and y to zero, and it will scale the graph, allowing you to verify/calibrate you wish.

it works fine, but it doesn't calculate the roll-off for you, you have to do this manually.
 
#4,309 · (Edited)
#JohnHWman check out the LightSpace thread in this forum. You can download a free working version of LightSpace (LightSpace DPS) from Light Illusion to see if that feature will be useful to you without a Radiance Pro like the author has.
Also see this link: https://www.lightillusion.com/hdr_calibration.html
It is a LightSpace guide for calibration of HDR displays including projectors. You might find the paragraph entitled "HDR UHDTV CALIBRATION VIA 3D LUT BOX" particularly interesting. It states that the multiplier function is for use with a 3D LUT box such as the Lumagen used by the author of the article you quoted.
 
#4,314 ·
OK Manni, I've got it...

The 'Y' offset value is just to adapt the mesured EOTF values to fit inside EOTF graph...

So with Calman, there is no automated way to calibrate a 100-150nits peak white projector within the provided workflows in Calman : we have to do it manually by manual graphic drawing...

That's not what I'm expecting from Calman HDR10 workflow :(

As I said, LightSpace is able to do it automatically with a single multiplier value entered manually inside their calibration software --> why Calman can't do it the same way ??
 
#4,321 ·
OK Manni, I've got it...



The 'Y' offset value is just to adapt the mesured EOTF values to fit inside EOTF graph...



So with Calman, there is no automated way to calibrate a 100-150nits peak white projector within the provided workflows in Calman : we have to do it manually by manual graphic drawing...



That's not what I'm expecting from Calman HDR10 workflow :(



As I said, LightSpace is able to do it automatically with a single multiplier value entered manually inside their calibration software --> why Calman can't do it the same way ??


Based on your graphs I don’t know if the projectors actually going into HDR mode. Your graphs look like it’s following a gamma curve.

The HDR multiplier is not a standard, I have been pushing people in the standards community to adopt a recommendation for a HDR home projector EOTF.

We don’t invent calibration standards, we just implement ones that have been standardized.
 
#4,317 ·
Anyone know if the monoprice hdfury integral is supported? I'm assuming they are exactly the same. The monoprice version is much cheaper.

Sent from my LG-V30 using Tapatalk
 
#4,323 ·
Tyler,

Thanks again for your help with the Dolby Vision calibration. I wanted to follow up with another question, in LG's 2017 OLED Calibration Notes, they say:


20 HDR Calibration adjustment points have been provided for 540nit content, 1,000nit content, and 4,000nit content, shown in the table below. A test pattern generator should be set to output a grey patch at the appropriated Code Value for each adjustment point. If using CalMan, launch Levels Editor, and create a new set of levels using the percentages in the table below, then select this range for the greyscale adjustment in the settings tab of the HDR workflow.

[Note: I have attached the table below]​

When I open the Levels Editor, it just gives you a single text box to enter values. How do I enter the numbers, and is it worthwhile to do this?
 

Attachments

#4,331 ·
20 HDR Calibration adjustment points have been provided for 540nit content, 1,000nit content, and 4,000nit content, shown in the table below. A test pattern generator should be set to output a grey patch at the appropriated Code Value for each adjustment point. If using CalMan, launch Levels Editor, and create a new set of levels using the percentages in the table below, then select this range for the greyscale adjustment in the settings tab of the HDR workflow.
This LG document is really confusing, first of all you can't force any pattern generator to display specific 10-bit values for HDR calibration, since software calculations and generation work in 8-bit. so sending 10bit output is just a 8bit -> 10bit value.

LG Notes have incorrect percentages, do 10-bit to 8 bit values conversion and use these values, CalMAN's Level Editor can accept 8-bit values while LG has posted 10-bit...when you convert them to 8 bit (10bit / 4 = 8 bit) you will get numbers with decimal, you can't enter to CM Level Editor decimal, so you have to round to closest 8-bit value.

I have checked all digital levels compared to Dolby's Golden Reference values and the percentages don't match the digital values (some are correct, only a few). LG's percentages are 0.1% off compared to the posted digital level values they have posted.

540nit values LG has posted are really useless information, no movie has 540 nit mastering display peak metadata, currently movies have 1000/1100/4000nits.

To save some time making the conversions, you can enter these values to CM Level Editor:

21-Point Grayscale (LG 2017 HDR10 540 nits)

16, 67, 83, 99, 106, 113, 116, 119, 123, 126, 129, 133, 136, 139, 143, 146, 149, 152, 156, 159, 235

21-Point Grayscale (LG 2017 HDR10 1000 nits)

16, 75, 90, 104, 111, 119, 123, 126, 129, 132, 136, 139, 142, 145, 148, 151, 155, 158, 161, 164, 235

21-Point Grayscale (LG 2017 HDR10 4000 nits)

16, 80, 96, 111, 119, 126, 129, 133, 136, 139, 143, 146, 150, 153, 157, 160, 164, 167, 170, 176, 235
 
#4,335 ·
^^^
HDR is fine but DV is a totally different story. You can't just look at the DE. For DV the whole point of the calibration is to update the config file with your specific set's capabilities. The only way to do this is to calibrate the native panel in "bypass" mode then generate a new config file and load it. :)
 
#4,336 ·
Ok, thanks for that information. I have much to learn here and appreciate everything you and the rest of the guys who know what you're talking about share with us.

I used to calibrate cinemas in a former life and moved into home theaters for a little while when DVD first launched, but a whole lot has changed since those days.
 
  • Like
Reactions: jrref
#4,338 ·
To bad all these engineers, at the various manufacturers, can't all get together and come up with some standard that they'll ALL use to simplify things so that the average videophile calibrator doesn't end up with a spinning head !. Is this a reasonable request, or does every company want to do their own thing?
 
  • Like
Reactions: shoman94
#4,339 ·
All good!

Licences re-activated and Support adds the following onto my previous warning:

Addendum:

Support says to add this to the warning - DE-ACTIVATE your licence BEFORE making ANY changes, including computer name change !
 
#4,340 · (Edited)
@WiFi-Spy
With CalMAN, SpectraCAL is allowing DV patch generation only using the internal software generation (for VIZIO) and for LG (with adding HD Fury to send specific manufacturer vector info frame). This requires those have a bit-perfect HDMI Output, so they need an Intel Notebook and many settings to check etc...

Because the DV signal requires only a bitperfect 1080p RGB signal, why doesn't SpectraCAL give DVDO AVLab users the capability to use it for that job, since it's bit-perfect for RGB @ 1080p?

Why do DVDO AVLab users have to buy an Intel notebook to use for DV patch generation, while we already have a bit-perfect external pattern generator? It will not require much code work, and it will work for all users, since DV require only 1080p signal to work. So with DVDO AVLab and HDFury, we will be able to generate DV patterns.

Thank you.
 
#4,341 ·
@WiFi-Spy
With CalMAN, SpectraCAL is allowing DV patch generation only using the internal software generation (for VIZIO) and for LG (with adding HD Fury to send specific manufacturer vector info frame). This requires to have a bit-perfect HDMI Output, so need an Intel Notebook and many settings to check etc...

Because the DV signal require only a bitperfect 1080p RGB signal, why they don't give the capability for DVDO AVLab users to use their DVDO for that job, since it's bitperfect for RGB @ 1080p, it's really a shame, so I'm sending you msg and I will inform other DVDO users I will see, to request from SpectraCAL to add DVDO also.

Why DVDO TPG users to have to buy an Intel notebook to use for DV patch generation? While you have already bit-perfect external pattern generator. It will not require much code work, and it will work for all users, since DV require only 1080p signal to work. So with DVDO and HDFury to be able to generate DV patterns.
+1 Totally agree!

Since DVDO is already bit perfect with RGB 1080p, why to be not able to use it for DV patch generation, this waste the time from user to make it work with Intel notebooks where it can work one day or not work after an driver/sw update etc.

I remember before years discussions about not using notebook HDMI output for patch generation becasue they were not reliable and use reference pattern generators but now we have to do the opposite. There no need for consumer who has already a 1080p reference external patch generator to spend new money for a notebook with Intel GPU or other external pattern generator which cost more from his TV price.
 
#4,348 ·
It looks like the open Beta for Calman 2018 was released today.

As soon as Robert at Value Electronics gets a 2018 LG in the store, which i believe will happen shortly, we will try the Autocal and report back with how it works. I have high expectations that it will work well and it's going to be really interesting to see the results now that we have a product to manipulate the LUT in the TV! :)
 
#4,349 · (Edited)
And if it doesn't work, I expect you to tell them how to fix it!

http://spectracal.com/forum/viewtopic.php?f=139&t=6609

CalMAN 2018 (5.9.0) requires CalMAN or All Access (Maintenance) purchase as of February 2017 or later.

Resolved an issue where the installer's All Access check was off by one month.

Added support for LG 2018 OLED and LCD displays. Support includes 1D and 3D LUTs in both SDR and HDR.
Two new workflows have been included in order to guide users through the new calibration process.
Ultimate, Studio, Video Pro, ISF, Enthusiast

Note: The new LG workflows are a work in progress.

Added Matrix LUT option. An exciting new 3D LUT measurement option for gamut targeting that can generate in under 1 minute.

Added the ability to disable CalMAN sound effects. This can be accessed under Settings\Application Preferences\Disable application sound effects.

Added HDR10 functionality in the Unified Pattern Generator Control Interface (UPGCI)

Updated Konica Minolta CA-410 support to explicitly set measurement mode on connect.

Resolved compatibility issues with CalMAN editors/utilities in language regions where commas are used
Instead of periods.

Resolved an issue with the Jeti Spectraval 1501 causing CalMAN to report invalid readings.
 
  • Like
Reactions: mrtickleuk
#4,350 ·
Stupid question, but should I uninstall Calman 2017 first before installing this beta, or is it better to install it over the top (will it know to "upgrade")? I don't want to lose or waste any of my licences. TIA.
 
#4,351 ·
Over-the-top worked fine for me ...
 
#4,354 ·
Tyler,

Any chance of Calman supporting DVDO AVlab ATP w/HDfury for DV? This there a any explanation on the reason why its NOT supported?

Thanks,
 
  • Like
Reactions: ConnecTEDDD
#4,356 · (Edited)
Perhaps you should be asking DVDO directly. They have said they will create new firmware if enough users request it. It's how we got custom user-set patterns as the initial release firmware didn't have them. Keep in mind that not all current hardware is DV-capable, otherwise more computer video chipsets than just Intel and NVidia would be usable. These DVDOs are now 2-3 years old and might not have the ability to incorporate DV metadata into their signal.
 
#4,364 ·
USB Input manual calibration learning curve

Calibrating the USB Input Info
USB and USB HDR Input calibrated. A real learning curve !

# 1 MPEG 4 Patterns must be used for SDR.
# 2 For those using a USB for their patterns for calibrating, you have to check whether you're actually calibrating the HDMI Input, or USB.
# 3 It appears static patterns do the HDMI Input, versus moving/timed patterns do the USB. A careful check must be done beforehand that you don't calibrate the wrong Input.
# 4 When doing Manual calibrations, always do a 100 and 0% scan before calibrating the grayscale
# 5 When doing Manual calibrations, always do a 100 and 75% scan before calibrating the CMS
# 6 When doing Manual calibrations, always do a 0 and 100% scan before calibrating the HDR CMS.
 
#4,367 ·
Here is a quote from Tyler to me so we can end this debacle. It's as I suspected.

The DV metadata is encoded in to the pattern itself. Only 3 things can generate DV calibration patches, CalMAN’s Pattern Window, VideoForge Pro, and Murideo Six-G. Any other pattern generator manufacture would have to license the technology from Dolby.
 
#4,368 · (Edited)
We aren't asking for direct DV patch generation. HDfury is doing the DV signalling. The TPG is just sending a RGB triplet. We are asking for the option to send a translated RGB output signal. Since the TPG isn't signalling DV, it does not require a licence.
 
#4,378 · (Edited)
More worms in the can! I'm sure this is old news to some, but Dolby has released a new low-latency DV profile at Sony's request that apparently changes the RGB wrapper to YUV. Sony owners got a DV update in firmware, but the new profile offloads the heavy lifting in DV from the display to the player, and none currently except for the Apple TV 4K are capable. Not even the Oppo, which won't receive an update until April. So this discussion may be academic at best. See here:
http://www.avsforum.com/forum/37-vi...ury-vertex-owners-thread-74.html#post55749376
 
#4,380 ·
More worms in the can. I'm sure this is old news to some, but Dolby has released a new low-latency DV profile at Sony's request that apparently changes the RGB wrapper to YUV. So this discussion may be academic at best. See here:
http://www.avsforum.com/forum/37-vi...ury-vertex-owners-thread-74.html#post55749376
I saw that in the Sony threads, I find it odd that it can stream just fine. I wouldn't think that Netflix or Amazon does any processing before streaming it.
 
#4,422 ·
Apologies for the slightly off-topic post (although Calman is directly mentioned).

Shout out to @WiFi-Spy! :D This has just popped up on my youtube subscriptions feed.

 
#4,423 ·
i1Pros

Has anyone noticed the i1Pro, and to a lesser extent the i1Pro2, scan heads are angled?

The angle is made worse when it's strapped into the L bracket that's used for mounting.

I had to use about a 3/8" spacer to have the meter mount into the L bracket, so it's scanning straight forward.
I going to experiment tonight as to whether scan results are the same with, and without the spacer.
 
#4,424 · (Edited)
Has anyone noticed the i1Pro, and to a lesser extent the i1Pro2, scan heads are angled?

The angle is made worse when it's strapped into the L bracket that's used for mounting.

I had to use about a 3/8" spacer to have the meter mount into the L bracket, so it's scanning straight forward.
I going to experiment tonight as to whether scan results are the same with, and without the spacer.
And it probably won't make a bit of difference at any reasonable distance from the display. Remember, these are primarily contact meters for reading surface colors on paper and other solid materials, with measurement of displays as a secondary ability. I'm not aware of any complaints that the I1 Pro series meters are not looking where they are aimed from either the DTP or display calibration communities...
 
Top