ee/ColorBox, LightSpace, and 3D LUT Calibration - Page 2 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 68Likes
Reply
 
Thread Tools
post #31 of 496 Old 03-05-2013, 06:10 PM
AVS Forum Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 7,025
Mentioned: 7 Post(s)
Tagged: 1 Thread(s)
Quoted: 331 Post(s)
Liked: 335
Steve you are out of line making comments like that in Buzz's thread.

This isn't the place for such a debate. But it is safe to say that you have no idea what we are doing with our math to create our interpolation data, so while you might disagree, you also have no technical footing to make such a statement.
sotti is offline  
Sponsored Links
Advertisement
 
post #32 of 496 Old 03-05-2013, 11:22 PM
Senior Member
 
Jeroen1000's Avatar
 
Join Date: Nov 2008
Posts: 380
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 16
A technical discussion on the other hand is sometimes enlightening for us lesser versed in this matter enthusiasts.

I also find this box interesting and although I suspect this has been answered somewhere already, so I hope you do not mind me asking:

What is the benefit from separating the "characterization" and "calibration"? The way I see the process is that the software takes a measurement and then corrects the display like it does with autocal with, for example, a Lumagen Radiance.

But from what I understand from the discussion, you can first gather all the measurement data and subsequently generate a LUT which will correct the gamut? However, should not it be checked whether the adjustments (the LUT) have (has) had the desired effect? Because autocal reiterates measuring and adjusting. So after every adjustment, it checks whether everything is now "better" until it sees the adjustment has succeeded in lowering the dE below a specified threshold. In the event the dE is still too high it makes another adjustment until it gets the dE below the threshold or cannot get any closer. But "characterization" seems to be a one time event where after it generates the corrections in a LUT.

My gear: Panasonic TH-42PF11EK pro plasma display + TX-P55VT60E -- Iscan Duo video processor -- i1 display 3 colorimeter -- i1 pro 2 spectrometer
Jeroen1000 is offline  
post #33 of 496 Old 03-06-2013, 04:42 AM - Thread Starter
AVS Forum Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 5,907
Mentioned: 16 Post(s)
Tagged: 1 Thread(s)
Quoted: 1660 Post(s)
Liked: 2439
Quote:
Originally Posted by turbe View Post

the eeColor box has it's own format and its size is 65x65x65...

Yes. The 65x65x65 LUTs load flawlessly with the ee software. It seems like my K10 has been running constantly for the last couple months and hopefully the info is in my notes but I'll edit out the following statement if I find it to be incorrect when I do some more testing later today. I've produced both 64x64x64 and 65x65x65 LUTs and the ee software will not accept the 64s.

Quote:
Originally Posted by sotti View Post

The ColorBox does have a slightly different firmware, but the hardware is the same.

I looked for the answer before but never found it. The question is, what does the FW change do?

Quote:
Originally Posted by turbe View Post

they are writing (to disc) the entire LUT file (their support for the eeColor 64^3 ) but it's important what you are characterizing the display at (display pattern / take reading at that point) - the points beyond this characterization size are interpolated..

I like to use 10^3 but you do want the option to go up to 17^3 (or more?) for characterization IMO. heck, I'll have to look, what is the size limit for characterization...

So, if Enthusiast can do any size up to 17^3 for the characterization process, that's a win for Enthusiast Uses since it's $1000-$2100 less than what's available now.

For final accuracy, the size of the display characterization is inversely proportional to the degree of display linearity. If linearity is near perfect a LUT can be produced with nothing more than an 81 point quick profile in LightSpace. If linearity is good, a 10 point DC would do it. Most consumer displays are untamed and require a 17 point DC. Anything beyond 17 points is overkill and any increased accuracy is inconsequential.

Quote:
Originally Posted by sotti View Post


Our goal with 5.1 was to make it smoking fast and highly accurate. That way if you need to target a second colorspace creating additional LUTs will still be faster than taking 4000+ readings.

A few comments: This thread is not some kind of software shootout. It is meant to be informational and concerned directly with the eeColor Processor, the two software packages that incorporate specific tools for it, and procedures to create the best possible accuracy with each one . Hopefully, as additional people own this Box, more and more information will be available and we'll all be happy. Obviously both softwares are producing LUTs for this Box differently but I don't see any reason for a cat fight.

The above statement about "faster than" etc. is, in my opinion, a cut at your competitor. I watched this same type of posting for two years on the ChromaPure thread before the offenders were more or less booed off. Please let's not do that here.

Buzz
THX Certified Level II Video Calibrator



 
buzzard767 is offline  
 
post #34 of 496 Old 03-06-2013, 05:09 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,394
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 815
Indeed, a discussion of technical items is of use to all biggrin.gif

And it is correct to state that the interior of the profile cube is every bit as important as the grey scale or primary ramps data.
We recommend the use of full 17x17x17 profiling for any critical calibration.
There are options for a Quick profile and Hybrid profile modes, which can produce excellent results, but they will never be as 'accurate' as a full 17^ profile.
(But, depending on the display the variation my be next to invisible - but never as accurate).

The use of skin tones as an example is a perfect one, as they are critical memory colours and you can only get them accurate by accurate profiling - no interpolation will ever be accurate is is it - well - interpolated information, not accurate profiling.

And to get such accurate profiling you need to have the profile separated from the final calibration as the time taken otherwise would just be too long.

With the separation of profiling from calibration you use a very fastest and accurate display profiling engine, which then uses an invertible 3D freeform colour space modelling technique to provide unequalled accuracy in the final calibration. The accuracy of the calibration process is a proven, as it is based on a very advanced colour mathematics engine, proven time and again across the world's top creative post-production facilities.

This works without the uncertainty of guess-work based adaptive approaches to measurement and calibration, which require recursive iterative readings, limiting the actual range of colours that can be addressed to a level well below the optimum required for accurate calibration.

Put simply, when measuring each individual point just once you can profile more individual points in the same time frame, giving a far larger data set from which to calculate the final calibration - making the results very accurate.

Using an iterative approach each point must be measured multiple times, with a guesstimated change made in-between each measurement of the same point. As a result you can only actually measure (profile) a limited number of points in the same time scale, and must interpolate the remainder.

From this it is obvious the profiled data set will never be as accurate.

Additionally, with the profile linked to the calibration you must pre-set the target goals before starting the process. With the two separated you have no need to set any initial parameters as you are just 'profiling' the underlying display properties. The later calibration process (LUT generation) will then generate a very accurate LUT for any target specified.

This means multiple calibration targets can be generated from the single profile data.

This means a display can be remote profiled, and the profile data sent to the 'calibrator' for the calibration LUT generation.
A lot of professional facilities use this approach with their clients, enabling the client to view material on their remote display with know calibration, but with no need for the 'calibrator' to have ever visited.

Obviously, a visit is preferable for ultimate calibration as being 'there' to see the display can prevent obvious errors, but it is another example of the difference.

All professional calibration systems - TrueLight, CineSpace and LightSpace - use this approach, with profiling separated from the final calibration.

As for verification of the LUT - that is built-in to the process as you can select the LUT as being 'Active' with the system, and use it to pre-adjust the patch data as it is send to the display, so 'proving' the LUT before it is applied for real. Many users will do a full 17^ profile to generate the calibration LUT, and then verify via a Quick Profile or Hybrid profile for example.

I should add that we did build an initial calibration system based on iterative measurements, but rejected it as a product when compared to the results of using a real colour mathematics engine with profiling and calibration separated.

That's not to say the results were bad or unacceptable, just not as good - and we are only after the best possible results - no compromises.

The real proof is in looking at real images to compare results...

I think that covers your question?

Steve
Jeroen1000 likes this.

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #35 of 496 Old 03-06-2013, 09:59 AM
AVS Forum Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 7,025
Mentioned: 7 Post(s)
Tagged: 1 Thread(s)
Quoted: 331 Post(s)
Liked: 335
Quote:
Originally Posted by buzzard767 View Post

The above statement about "faster than" etc. is, in my opinion, a cut at your competitor. I watched this same type of posting for two years on the ChromaPure thread before the offenders were more or less booed off. Please let's not do that here.

It wasn't meant to be a cut.

Being able to re-target gamuts, is a clear feature advantage for some situations. Our counter point is that we can create two in the same time as one.

I've tried to stay on the topic solely of how CalMAN works and the best practices for using CalMAN successfully. I hope people who read the thread from the start would see I didn't come here to disparage other software products. Only to contribute factual information about how the hardware works and how our software interacts with it.
Quote:
Originally Posted by Light Illusion View Post

That's not to say the results were bad or unacceptable, just not as good - and we are only after the best possible results - no compromises.

If you want to debate this with me, feel free to start a thread with your actual dE error for all 17^3 points after calibration. Your arguments, lack factual data. Without some data to back up your claims, those statements are meaningless.
sotti is offline  
post #36 of 496 Old 03-06-2013, 10:18 AM - Thread Starter
AVS Forum Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 5,907
Mentioned: 16 Post(s)
Tagged: 1 Thread(s)
Quoted: 1660 Post(s)
Liked: 2439
Basic LightSpace procedures added to post number 4.

Buzz
THX Certified Level II Video Calibrator



 
buzzard767 is offline  
post #37 of 496 Old 03-06-2013, 10:21 AM - Thread Starter
AVS Forum Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 5,907
Mentioned: 16 Post(s)
Tagged: 1 Thread(s)
Quoted: 1660 Post(s)
Liked: 2439
Quote:
Originally Posted by sotti View Post

It wasn't meant to be a cut.
I've tried to stay on the topic solely of how CalMAN works and the best practices for using CalMAN successfully. I hope people who read the thread from the start would see I didn't come here to disparage other software products. Only to contribute factual information about how the hardware works and how our software interacts with it.
.

And that is exactly how this thread should operate. Thank you for that, Joel.

Buzz
THX Certified Level II Video Calibrator



 
buzzard767 is offline  
post #38 of 496 Old 03-06-2013, 11:04 AM
AVS Forum Special Member
 
N3W813's Avatar
 
Join Date: Feb 2003
Posts: 1,190
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 45 Post(s)
Liked: 81
Quote:
Originally Posted by buzzard767 View Post

Basic LightSpace procedures added to post number 4.

Can you post before and after charts of the display you characterized in post 4? dE2000 chart for either the saturation sweep or colorchecker charts will be perfect.

My ArgyllCMS/MadVR 3DLUT Creation Workflow
My Sharp Elite Movie THX AV Mode Settings
--Aug 2011 Set, 2.2 gamma [ link ]
--Nov 2012 Set, 2.2 gamma [ link ]
N3W813 is offline  
post #39 of 496 Old 03-06-2013, 01:56 PM - Thread Starter
AVS Forum Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 5,907
Mentioned: 16 Post(s)
Tagged: 1 Thread(s)
Quoted: 1660 Post(s)
Liked: 2439
Quote:
Originally Posted by N3W813 View Post

Can you post before and after charts of the display you characterized in post 4? dE2000 chart for either the saturation sweep or colorchecker charts will be perfect.

I wish I could but those documents are gone, all part of nearly daily LightSpace beta testing on what must be the world's worst candidate for LUTs. The display's on/off MLL is .003 FtL but when you put up an ansi contrast checkerboard pattern it jumps to something like .18. Worst TV I ever could have purchased and did so before I ever even heard of 3D LUTs. Not only that, all readings are relative, not accurate, because it's a Mitsubishi 75" LaserVue which requires a 5nm spectro for profiling, not my 10nm i1Pro which is at SpectraCal, Seattle for recertification anyway. The LS beta testers have been receiving 2 or 3 modified versions per week and we race around trying to verify results of the fixes. I run display characterizations, make LUTs, see what they look like, report, and toss.

Thankfully this insanity should end soon because I'll replace the DLP with the not yet released Samsung F8000 LCD if it comes anywhere near living up to all the CES hype. One way or another, some lucky Craig's list buyer will get one hell of a price on this professionally calibrated, ummm, reference display.

Rant over.

Buzz
THX Certified Level II Video Calibrator



 
buzzard767 is offline  
post #40 of 496 Old 03-06-2013, 11:48 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
Quote:
Originally Posted by sotti View Post

If you want to debate this with me, feel free to start a thread with your actual dE error for all 17^3 points after calibration. Your arguments, lack factual data. Without some data to back up your claims, those statements are meaningless.

THX CineCube, FilmLight Truelight, Light Illusion LightSpace CMS, world defacto software of professional industry that the professionals are creating 3D LUT profilings the last 5 years or more, none of these software are using dE presentation of any type. They have only graphs but no dE report. They are trying to calculate the best possible correction.

To compare the speed of 2 different software, you have to measure the total time it takes to create a profile for a specific number of color points for each software.

For example, run a 4913 color point (17-Point Cube) profiling using 2 different software types and then compare the time it took.

Running a 780 color point profiling and then using interpollation calculate the 4913 color point calibration is not speed, this is guessing method.

Guessing will never be close to the real measurement if you need critical accurancy. Interpollation is not something new...

600px-Interpolation_example_spline.svg.png

Speed means 1 reading per color point, not 2-3 reading per color point.

THX CineCube, FilmLight Truelight, Light Illusion LightSpace CMS are using one read per color point method so these software solutions are faster from any other solution...

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 offline  
post #41 of 496 Old 03-06-2013, 11:59 PM
AVS Forum Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 7,025
Mentioned: 7 Post(s)
Tagged: 1 Thread(s)
Quoted: 331 Post(s)
Liked: 335
Buzz would prefer for us not to discuss the perfomance in this thread.

Feel free to start another if you want to continue this debate.
sotti is offline  
post #42 of 496 Old 03-07-2013, 04:18 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
Quote:
Originally Posted by sotti View Post

Buzz would prefer for us not to discuss the perfomance in this thread.

Feel free to start another if you want to continue this debate.

I have nothing to debate with you, a 10-years old boy who reads my above post will understand what software speed means. wink.gif

When we are talking about speed we have to compare the same number of target color points, it's so simple...

+ I have both CalMAN+LightSpace and i know what each software is capable of for my display.
+ I have over 100 different TV models i can test anytime at my work, with eeColor, LumagenMini, VideoEQ but i don't want to spend my time on testing or debating because these type of topic debates will never end...

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 offline  
post #43 of 496 Old 03-07-2013, 10:14 AM
Senior Member
 
Nudgiator's Avatar
 
Join Date: Mar 2007
Posts: 336
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 34 Post(s)
Liked: 37
Quote:
Originally Posted by ConnecTEDDD View Post

I have nothing to debate with you, a 10-years old boy who reads my above post will understand what software speed means. wink.gif

+1 ! And I agree: it's annoying to discuss about absolute basics of LUT profiling / calibration.
IMHO I don't think, that the guys at THX CineCube, FilmLight Truelight, Light Illusion LightSpace CMS are all "dumb asses" and use "wrong and slow algorithms" for their products, at least, they have the leadership in that part of calibration wink.gif

Source: Samsung UBD-K8500 (UHD), Panasonic BDT-110 (BD), Toshiba HD-XE1 (HD-DVD), Vantage VT-1S (SAT), Kathrein UFS-924 (SAT)
Calibration: LightSpace CMS, SpaceMan, SpaceMatch, CalMAN 5 Prof., Chroma Pure, Lumagen Radiance Pro 4446, Chroma 5, SpectraCal C6, EODIS3, i1 pro 2, Klein K10-A
Target: Panasonic TX-P65VTW60 (Plasma TV)
Sound: Yamaha RX-V2067 (AVR), Teufel Theater LT 3 (Speaker)
Nudgiator is offline  
post #44 of 496 Old 03-07-2013, 10:37 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,537
Mentioned: 63 Post(s)
Tagged: 1 Thread(s)
Quoted: 1732 Post(s)
Liked: 2131
Quote:
Originally Posted by Nudgiator View Post

Quote:
Originally Posted by ConnecTEDDD View Post

I have nothing to debate with you, a 10-years old boy who reads my above post will understand what software speed means. wink.gif

+1 ! And I agree: it's annoying to discuss about absolute basics of LUT profiling / calibration.
IMHO I don't think, that the guys at THX CineCube, FilmLight Truelight, Light Illusion LightSpace CMS are all "dumb asses" and use "wrong and slow algorithms" for their products, at least, they have the leadership in that part of calibration wink.gif

These software solutions are used at professional industry a lot of years already: THX CineCube (since ~2005), FilmLight Truelight (since ~2004), Light Illusion LightSpace CMS (since ~2008) to perfrorm only 17-Point Cube Calibrations.

What to compare about? Industry has set that 17-point cube resolution as a standard, not 16 or 18 or 19 point cube....... that 17 point cube resolution they have selected, so calibrating with this resolution will get you closer to the condition the monitors they are using at the DI/Bl-BluRay Mastering Studios are calibraTED wink.gif

Before one year noone here knew anything about 3D Cube Calibrations, 7-Color Calibrations was enough and reference grade. wink.gif

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 offline  
post #45 of 496 Old 03-07-2013, 01:21 PM
Advanced Member
 
Iron Mike's Avatar
 
Join Date: Oct 2006
Posts: 901
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 71 Post(s)
Liked: 414
Quote:
Originally Posted by sillysally View Post

The only problem with eecolor or colorbox is they don't generate patterns for LUT or any patterns for that matter. The Radiance Mini 3D does, and that sure makes a LUT calibration fast,accurate and easy.
The draw back with the Radiance is it will only do a 125 point cube like what is in Calman. However from what the OP (buzz) says, anything over that for a display like my plasma VT50 to view HQ Blu Ray movies is over kill.

SillySally,

I have the Mini and the color box and I have CM and LS... the only benefits you get with a Mini are the additional VP tools - which I use NONE - and the included pattern generator... but with the Mini being nearly US$1,000 more expensive than a color box, you could buy yourself a better pattern generator than the Mini, not only a more accurate one but one that is full data range 0-255 (Lumagen are too unorganized and have way too many baby issues to realize the actual full potential of their hw)...

Most importantly: the Mini introduces heavy distortion at 2 stages... making it less accurate with the final LUT than the color box which is (almost) perfectly transparent, besides the fact that the Mini is only 5^3 and the color box is 65^3...

If you don't have a LUT holder box yet, get the color box from Buzz, way less trouble...

- M

calibration & profiling solutions: Lightspace, Spaceman, Calman, Argyll, ColorNavigator, basICColor
profiling & calibration workflow tools: Display Calibration Tools
meter: CR-250, Klein K-10 A, i1Pro, i1D3
AVS thread: Lightspace & Custom Color Patch Set & Gamma Calibration on Panasonic 65VT60
Iron Mike is offline  
post #46 of 496 Old 03-07-2013, 01:32 PM
AVS Forum Addicted Member
 
mark haflich's Avatar
 
Join Date: Dec 2000
Location: Venice, Florida, USA
Posts: 21,464
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1418 Post(s)
Liked: 1012
I have 3 EE color boxes and a Radiance. Lumagen i suspect will revise their software to gerate as many points as the calibration industry requires. If you find fault with somethings thaey do, let them know. They tend to fix things.
mark haflich is offline  
post #47 of 496 Old 03-07-2013, 01:39 PM
Advanced Member
 
Iron Mike's Avatar
 
Join Date: Oct 2006
Posts: 901
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 71 Post(s)
Liked: 414
Quote:
Originally Posted by mark haflich View Post

I have 3 EE color boxes and a Radiance. Lumagen i suspect will revise their software to gerate as many points as the calibration industry requires. If you find fault with somethings thaey do, let them know. They tend to fix things.

yeah, they are very nice people, customer support is great... but there are fundamental problems with the box that I - as an end user - don't need to "report" to them...

if they don't discover and fix that in their development, manufacturing process (with a way larger array of tools at their disposal), then that tells you already a lot about a company...

I'm sure they will improve things for future products...

- M

calibration & profiling solutions: Lightspace, Spaceman, Calman, Argyll, ColorNavigator, basICColor
profiling & calibration workflow tools: Display Calibration Tools
meter: CR-250, Klein K-10 A, i1Pro, i1D3
AVS thread: Lightspace & Custom Color Patch Set & Gamma Calibration on Panasonic 65VT60
Iron Mike is offline  
post #48 of 496 Old 03-07-2013, 03:41 PM
AVS Forum Special Member
 
sillysally's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 4,947
Mentioned: 14 Post(s)
Tagged: 0 Thread(s)
Quoted: 1012 Post(s)
Liked: 952
Quote:
Originally Posted by Iron Mike View Post

SillySally,

I have the Mini and the color box and I have CM and LS... the only benefits you get with a Mini are the additional VP tools - which I use NONE - and the included pattern generator... but with the Mini being nearly US$1,000 more expensive than a color box, you could buy yourself a better pattern generator than the Mini, not only a more accurate one but one that is full data range 0-255 (Lumagen are too unorganized and have way too many baby issues to realize the actual full potential of their hw)...

Most importantly: the Mini introduces heavy distortion at 2 stages... making it less accurate with the final LUT than the color box which is (almost) perfectly transparent, besides the fact that the Mini is only 5^3 and the color box is 65^3...

If you don't have a LUT holder box yet, get the color box from Buzz, way less trouble...

- M

Iron Mike,

Yes I have been kicking all this around, and have PM Buzz about the eeColor box.
I really haven't had any problems with my Mini, and wasn't aware of "introduces heavy distortion at 2 stages". What is the distortion you speak of?
With CM 5.1 and the Mini I am getting some very good LUT calibrations ( both on paper and on my display). So I am not in any hurry to make changes, I am just tiring to sift though the hype and reality right now.

Anyway I did want to get your input on all this plus LS verses CM 5.1 for are VT50's.
Thanks for your post.

ss

"Don't worry be happy"
For Sale Lumagen Radiance Pro 4440/2, see link for details
http://www.avsforum.com/forum/210-vi...l#post54911046
sillysally is offline  
post #49 of 496 Old 03-07-2013, 04:08 PM
Senior Member
 
Nudgiator's Avatar
 
Join Date: Mar 2007
Posts: 336
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 34 Post(s)
Liked: 37
Quote:
Originally Posted by sillysally View Post

I really haven't had any problems with my Mini, and wasn't aware of "introduces heavy distortion at 2 stages".

I agree with Mike: the Radiance cannot passthrough the signal 1:1. You can easily check this: use test patterns via a laptop with HDMI-out or TEDs disc and compare the results with a Radiance in the chain. You cannot see differences between TEDs disc and the laptop, but with the Radiance as pattern generator.
Quote:
What is the distortion you speak of?

Especially look at the lower, dark IREs. There you can see some distortion (also in movies with a lot of dark scenes). That's because of the "only" 5^3 cube of the Radiance.

Source: Samsung UBD-K8500 (UHD), Panasonic BDT-110 (BD), Toshiba HD-XE1 (HD-DVD), Vantage VT-1S (SAT), Kathrein UFS-924 (SAT)
Calibration: LightSpace CMS, SpaceMan, SpaceMatch, CalMAN 5 Prof., Chroma Pure, Lumagen Radiance Pro 4446, Chroma 5, SpectraCal C6, EODIS3, i1 pro 2, Klein K10-A
Target: Panasonic TX-P65VTW60 (Plasma TV)
Sound: Yamaha RX-V2067 (AVR), Teufel Theater LT 3 (Speaker)
Nudgiator is offline  
post #50 of 496 Old 03-07-2013, 07:41 PM
Advanced Member
 
Iron Mike's Avatar
 
Join Date: Oct 2006
Posts: 901
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 71 Post(s)
Liked: 414
Quote:
Originally Posted by sillysally View Post

Iron Mike,

Yes I have been kicking all this around, and have PM Buzz about the eeColor box.
I really haven't had any problems with my Mini, and wasn't aware of "introduces heavy distortion at 2 stages". What is the distortion you speak of?
With CM 5.1 and the Mini I am getting some very good LUT calibrations ( both on paper and on my display). So I am not in any hurry to make changes, I am just tiring to sift though the hype and reality right now.

Anyway I did want to get your input on all this plus LS verses CM 5.1 for are VT50's.
Thanks for your post.

ss

first of, me saying the distortion is "heavy" is obviously a subjective statement, but considering how (almost) perfectly the color box passes the signal, I call it "heavy"... others would call it "unnecessary"

The Mini introduces distortion for the first time (a), just by being in the chain. As Nudge explained, simply profile your TV w/o the Mini in the chain and then profile again with the Mini in the chain with an empty CMS slot, you'll see the distortion. The second time (b) the Lumagen introduces distortion is when it actually applies a LUT sitting in a CMS slot to the passing signal.

I'll keep this comparison between LS and CM with regards to the Mini, as it appears CM is still figuring out how to use the color box correctly. It is good to see that the color box is supported in CM E.

I've done heavy testing between LS and CM using the Mini. I'll go a little bit more in depth for other users, I know that you SillySally know most of the information posted below already.

CM

Calman has a very nice interface and a heavy feature set, some are very useful.

No need to mention here that as a paying Calman customer, I was and I am heavily disappointed with the gazillion of bugs in this software. I'm in sw dev for 20+ years, sure my coding standards are ultra high, but man their QA / QC is almost zero to none. With every version they fix 4 bugs and introduce 20 more. As a coder, I'd be so embarrassed about some of these releases, that I wouldn't show up on the AVS until all of these bugs are fixed. Not trying to bore here the many of existing CM customers, they know what is going on, but since this is still a forum for USERS, this is rather information for potential new customers or cross over customers: You will be dealing with frustrating bugs.

I still hope they keep improving the software and finally deliver the stellar product they are not getting tired of marketing here in these user threads. IMO, (still) lots of potential there.

Biggest advantage currently of the software is it's customizable reports. That alone could be a reason to buy the software or Chromapure for that matter. Another plus of CM with the Mini is that it is pretty much a one-click solution, a customer does not need to know much about color science to start a calibration. Results are good (on the reports), but how would you know what you actually have until you visually compare it with results from either CP or LS... (?)

Btw, coming back to the double distortion of the Mini, mentioned as (a) and (b) above: With it's redundant approach of constantly reading / writing (not sure if this has changed recently), CM actually accommodates somewhat for the (b) distortion of the Mini, which LS does not with it's single write method.

Also, it appears CM uses a lot of guesswork not only in the data the probe returns but then also in their LUT creation. Now this does not matter (and could actually be a plus) for users who just want to get decent results quickly, but if you want information of what is actually going on, this software does not consistently return accurate data.

I can easily reproduce this every time I finish a 5^3 3D cube for the Mini, then see the final calibration verification report (all looks perfect). Take a screenshot. Then go to Quick Analysis, run the same verification tests. Results are different EVERY TIME. 2 mins after the 3D cube. This is very concerning and has been reported over and over - this alone makes the sw useless to me, at least in color critical application...


LS

LS is a bare bones interface that has less automated workflows as CM. If you don't know much about color science and display calibration or you don't want to get your hands dirty a little bit and learn about it (at least the basics), then IMO this sw might not be for you. Better get a color box and have an experienced calibrator like Buzz use LS for you.

But LS offers reliable, working tools to get all the data you need. LS reports true, accurate data. You will see what is truly going on with your probe. You will see what is truly going on with your display. Every time. Make a Quick Profile a day later, you will see the same consistent results. Reports are by far not as customizable or in depth as CM, but all necessary ones are there.

It took me quite a while to get a good LUT for the Mini, as at that point in time, LS went through lots of feature improvements, and again it is not a one-click process. If you decide to use Ted's disk, the profiling session requires one click.

The final 5^3 LUT, although it does not compensate for the (b) distortion of the Lumagen, is visually better than the one achieved by CM. Better color accuracy, the picture has more depth. Very easy to compare between LUT's on the Mini, just load both into a CMS slot onto the Mini, load reference video material and switch between LUT's.

Running reports in CM for both of these LUT's showed that CM was on par with (or very slightly better than) LS when it comes to dE, but obviously, CM is reading it's own calibration points in it's reports, while with a LS LUT it does not read actually profiled points. So LS actually competing here shows how good the LUT is.

Also, and this has been mentioned before, dE is just one approach of validating a result... another approach is to LOOK at true reference video material... I personally rather have a slightly less perfect greyscale (as long as there is no visual contamination) with a better overall picture than to see really good dE's but an inferior picture... The LS color engine does a very good job of profiling the ENTIRE cube, including the interior of the cube (very important).

Be aware that if you're trying to use an i1D3, you need an OEM i1d3, the retail version is NOT supported. You might also need a different serial cable to interact with the Mini, although your current cable does work with CM. The profile data is saved on your PC, which you then use to generate a LUT. You can generate any of the many supported color spaces from the profile data. Lots of options there.

The generated LUT might need some polishing, e.g. in the lower IRE's... this can get somewhat complicated for newbies, but LS has now implemented an almost fully automated adjustment for users. You can always go back and tweak more, no need to run a profile again.


Mini vs. Color Box

For new customers, get a color box and avoid trouble and use the rest of the money for a pattern generator or just have a calibrator do it for you - US$1,500 for a 5^3 LUT box is way too much, and ask yourself honestly will you actually need any of the VP features of the Mini ?!

If you're looking for a second LUT holder box, get the color box if you already have the Mini. The color box is much cheaper, has more potential (65^3) and you can use the Mini as the pattern generator.

Edit: the color box referred to in this post is the eeColor box available from Buzz for US$600.00.


- M

calibration & profiling solutions: Lightspace, Spaceman, Calman, Argyll, ColorNavigator, basICColor
profiling & calibration workflow tools: Display Calibration Tools
meter: CR-250, Klein K-10 A, i1Pro, i1D3
AVS thread: Lightspace & Custom Color Patch Set & Gamma Calibration on Panasonic 65VT60
Iron Mike is offline  
post #51 of 496 Old 03-08-2013, 09:25 AM
AVS Forum Special Member
 
derekjsmith's Avatar
 
Join Date: Oct 2003
Location: Snohomish, WA
Posts: 1,884
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 116
Quote:
Originally Posted by Iron Mike View Post

No need to mention here that as a paying Calman customer, I was and I am heavily disappointed with the gazillion of bugs in this software. I'm in sw dev for 20+ years, sure my coding standards are ultra high, but man their QA / QC is almost zero to none. With every version they fix 4 bugs and introduce 20 more. As a coder, I'd be so embarrassed about some of these releases, that I wouldn't show up on the AVS until all of these bugs are fixed. Not trying to bore here the many of existing CM customers, they know what is going on, but since this is still a forum for USERS, this is rather information for potential new customers or cross over customers: You will be dealing with frustrating bugs.

Mike it is obvious you have never managed or participated in a beta program which I understand can be a bit daunting for some and that is alright, just don't sign up for them and wait for the final release. We are one of the few companies that allow or even encourage our end users to actively participate in our betas.

Quote:
CalMAN: Btw, coming back to the double distortion of the Mini, mentioned as (a) and (b) above: With it's redundant approach of constantly reading / writing (not sure if this has changed recently), CM actually accommodates somewhat for the (b) distortion of the Mini, which LS does not with it's single write method.

Also, it appears CM uses a lot of guesswork not only in the data the probe returns but then also in their LUT creation. Now this does not matter (and could actually be a plus) for users who just want to get decent results quickly, but if you want information of what is actually going on, this software does not consistently return accurate data.

I guess what you call our guesswork method then actually works. We call this good science: observe, correct and validate. You can call it whatever you like. But again it does work and in many cases works very well.

Quote:
LS: The final 5^3 LUT, although it does not compensate for the (b) distortion of the Lumagen, is visually better than the one achieved by CM. Better color accuracy, the picture has more depth. Very easy to compare between LUT's on the Mini, just load both into a CMS slot onto the Mini, load reference video material and switch between LUT's.

Running reports in CM for both of these LUT's showed that CM was on par with (or very slightly better than) LS when it comes to dE, but obviously, CM is reading it's own calibration points in it's reports, while with a LS LUT it does not read actually profiled points. So LS actually competing here shows how good the LUT is.

OK I’m really getting tired of this subjective crap. It is either right or it is not and you even admit CalMAN produces a better result but claim to like the LS result better. Which again is your subjective preference and your right to say so.

How about this you calibrate your Mini with CalMAN 5.1 Studio and measure the results with color checker, saturations at 5% steps, saturations at 16 points and white balance/gamma at 21 points. Then do the same with LS and post the results. But you already know what you will find and I suspect is why you don’t.

Anyhow I’ve already said we as a company are getting tired of this treatment of being belittled on AVS for the fact that we came from the consumer side of the industry and then went into Post and Studios. Well all I have to say is the Post houses and Studios don’t treat us this way. In fact are very interested in our work because we listened and are now providing the solutions they have been asking for. Our demo at HPA was blowing people away and as a result we now have a lot of interest going into NAB. But again this has nothing to do with AVS or our consumer products other than they benefit from all this work as well.

Derek

CTO Portrait Displays Inc.
derekjsmith is offline  
post #52 of 496 Old 03-08-2013, 09:50 AM
Senior Member
 
visca blaugrana's Avatar
 
Join Date: Jan 2008
Posts: 346
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Why not respect buzz's wish and prevent this thread from getting out of control. It's seems to become a trend in this forum to junk up more or less every thread. Surely I'm not the only one sick of this! If LS and its followers want to keep arguing over this why not do as digested open a new thread and junk it up but keep it in there.

as an alternative it might be worth to ask admins to exclude manufactures and resellers to be excluded from the forums, but surely that is not in anyones interests.

I think it's clear LS is attacking CM in the attempt to gain new customer's.

it's great with some debate about how the software work, but surely a better tone can be used!!!
visca blaugrana is offline  
post #53 of 496 Old 03-08-2013, 10:29 AM
AVS Forum Special Member
 
derekjsmith's Avatar
 
Join Date: Oct 2003
Location: Snohomish, WA
Posts: 1,884
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 116
Quote:
Originally Posted by visca blaugrana View Post

Why not respect buzz's wish and prevent this thread from getting out of control. It's seems to become a trend in this forum to junk up more or less every thread. Surely I'm not the only one sick of this! If LS and its followers want to keep arguing over this why not do as digested open a new thread and junk it up but keep it in there.

I could not agree more. In fact Joel even asked for this.

But if I feel my companies reputation or employees are being personally attacked I will not let those go unanswered.

Derek

CTO Portrait Displays Inc.
derekjsmith is offline  
post #54 of 496 Old 03-08-2013, 10:51 AM
Advanced Member
 
PE06MCG's Avatar
 
Join Date: Jun 2010
Location: West Yorkshire, UK
Posts: 729
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 51
Quote:
Originally Posted by derekjsmith View Post

I could not agree more. In fact Joel even asked for this.

But if I feel my companies reputation or employees are being personally attacked I will not let those go unanswered.

Hi Derek,

Regarding posting results I'm afraid I do not have a means to calibrate with even 5x5x5 LUT but here is my latest attempt using a DUO and Calman 5 Enthusiast at 80% saturation patterns. 80% Saturation patterns.jpg 233k .jpg file
Attached Thumbnails
Click image for larger version

Name:	80% Saturation patterns.jpg
Views:	160
Size:	233.0 KB
ID:	67747  
PE06MCG is offline  
post #55 of 496 Old 03-08-2013, 12:27 PM
Advanced Member
 
Iron Mike's Avatar
 
Join Date: Oct 2006
Posts: 901
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 71 Post(s)
Liked: 414
Quote:
Originally Posted by derekjsmith View Post

Mike it is obvious you have never managed or participated in a beta program which I understand can be a bit daunting for some and that is alright, just don't sign up for them and wait for the final release. We are one of the few companies that allow or even encourage our end users to actively participate in our betas.

Derek, I only worked with your FINAL RELEASES and they contain a gazillion bugs. You know it, I know it. Let's not argue. I was once told you guys are programmers (not calibrators), I don't know if this is true, but from a programming perspective PLEASE deliver final releases that are working, and have ALL advertised features, in a working state. Features we paid for. Thank you.
Quote:
Originally Posted by derekjsmith View Post

I guess what you call our guesswork method then actually works. We call this good science: observe, correct and validate. You can call it whatever you like. But again it does work and in many cases works very well.
OK I’m really getting tired of this subjective crap. It is either right or it is not and you even admit CalMAN produces a better result but claim to like the LS result better. Which again is your subjective preference and your right to say so.

How about this you calibrate your Mini with CalMAN 5.1 Studio and measure the results with color checker, saturations at 5% steps, saturations at 16 points and white balance/gamma at 21 points. Then do the same with LS and post the results. But you already know what you will find and I suspect is why you don’t.

I never said the CM LUT was a "better" result. Read my post carefully. In the CM report, reading ONLY CM profiling points, CM had in the greyscale a dE that was 0.2 lower in the average. Is this visible ? No. Did your CM report measure LS profiling points ? No. Unless your dE is way off, visual validation is as important. You out of all people should know that.

But here's something you can do for us, your paying customers, I've asked for this in your main CM 5 thread. Please implement a feature in CM where users can provide the points in the gamut that should be validated in the report. I'm talking about complete freedom of choice. Users provide an array of RGB triplets (via XML file or whatever works for you), and CM simply creates a report for these points, very easy since you already programmed the main module. THEN we can talk about validating your LUT vs. competitors.

And Derek, do not think I'm asking about this to use it to bash your LUT results. This feature alone would be another single reason to purchase your software.
Quote:
Originally Posted by derekjsmith View Post

Anyhow I’ve already said we as a company are getting tired of this treatment of being belittled on AVS for the fact that we came from the consumer side of the industry and then went into Post and Studios. Well all I have to say is the Post houses and Studios don’t treat us this way. In fact are very interested in our work because we listened and are now providing the solutions they have been asking for. Our demo at HPA was blowing people away and as a result we now have a lot of interest going into NAB. But again this has nothing to do with AVS or our consumer products other than they benefit from all this work as well.

Man, you're making me laugh, NOBODY is belittling Spectracal. Let me tell you from you a USER's perspective what's going on on the AVS. Spectracal polutes a lot of threads and tries to bully competitors around. That's how it appears to me, as a user. I find it quite entertaining, back in the day with Tom and now you're doing the same thing with Steve. I don't see them hijack the huge CM threads and stir up trouble. That has been MO right from the get go. I've also seen your company post very helpful information in tons of other threads, kudos for that.

Not sure why you're reacting like this. Your company has by far the biggest chunk of the market out of all 3 here (CM | LS | CP). You don't need to be scared that you're wiped off the market, as your product offers some features that other solutions don't (and I was very nice to you in my post and pointed that out). It has a place in the market. Now if you maybe go and fix all the bugs and LET THE PRODUCT SPEAK FOR ITSELF, then you guys would save a ton of time w/o having to post every 5 mins on AVS or your own forum regarding all of these issues.


And just to make this clear for everybody else (most know already): I'm not affiliated with Spectracal or Light Illusion, I was providing my personal experiences as a lot of users had asked me about this. I personally do not care which solution (CM | LS) delivers the final LUT I will be using in my box, b/c I purchased both solutions.

- M

calibration & profiling solutions: Lightspace, Spaceman, Calman, Argyll, ColorNavigator, basICColor
profiling & calibration workflow tools: Display Calibration Tools
meter: CR-250, Klein K-10 A, i1Pro, i1D3
AVS thread: Lightspace & Custom Color Patch Set & Gamma Calibration on Panasonic 65VT60
Iron Mike is offline  
post #56 of 496 Old 03-09-2013, 08:28 AM - Thread Starter
AVS Forum Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 5,907
Mentioned: 16 Post(s)
Tagged: 1 Thread(s)
Quoted: 1660 Post(s)
Liked: 2439
A separate thread will be started for the eeBox and CalMAN -- it's too confusing to be referencing both software packages in one thread, and since there is already LightSpace information here on this one I believe it should stay that way. CalMAN will soon offer eeBox support in its Enthusiast version, something it does not do now, and it deserves its own thread.

Buzz
THX Certified Level II Video Calibrator



 
buzzard767 is offline  
post #57 of 496 Old 03-12-2013, 06:59 PM - Thread Starter
AVS Forum Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 5,907
Mentioned: 16 Post(s)
Tagged: 1 Thread(s)
Quoted: 1660 Post(s)
Liked: 2439
Quote:
Originally Posted by visca blaugrana View Post

Why not respect buzz's wish and prevent this thread from getting out of control.!

Quote:
Originally Posted by buzzard767 View Post

A separate thread will be started for the eeBox and CalMAN -- it's too confusing to be referencing both software packages in one thread, and since there is already LightSpace information here on this one I believe it should stay that way. CalMAN will soon offer eeBox support in its Enthusiast version, something it does not do now, and it deserves its own thread.

Let's keep this thread strictly eeBox/LightSpace.

There is a new eeBox/CalMAN thread here: http://www.avsforum.com/t/1462960/eecolor-processor-calman-3d-luts

Buzz
THX Certified Level II Video Calibrator



 
buzzard767 is offline  
post #58 of 496 Old 03-13-2013, 03:10 PM
AVS Forum Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 7,025
Mentioned: 7 Post(s)
Tagged: 1 Thread(s)
Quoted: 331 Post(s)
Liked: 335
Quote:
Originally Posted by Light Illusion View Post

We use a 65x65x65 formats with LightSpace and the eeColor LUT box, and the results are far from 'unusable' for PC (data) formats.

Actually works very well biggrin.gif

Steve

Hi Steve,

Can you explain how lightspace handles this?


Here is the LUT file I used to create this issue
3DLUT_1.zip 1195k .zip file

If you look at the file, everything should be green and dim all the way to the top, there is no entry in the file that tells 109 to go back to native, yet it happens anyway. You do need to have a generator that can create triplets up to 255. The data between 251 and 255 is splined, so if the pattern generator is off even a little the effect of this is diminished.

Like I mentioned earlier in the thread this isn't really an issue for video, but it is an issue for PC levels.

Also as a footnote, while I used CalMAN to test and gather the data, this is not a specific issue to anything except the EE hardware and therefore is relevant to any thread discussing the EE box.
Attached Files
File Type: zip 3DLUT_1.zip (1.17 MB, 99 views)
sotti is offline  
post #59 of 496 Old 03-14-2013, 09:23 AM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,275
Mentioned: 89 Post(s)
Tagged: 0 Thread(s)
Quoted: 1316 Post(s)
Liked: 1064
Quote:
Originally Posted by sotti View Post


Like I mentioned earlier in the thread this isn't really an issue for video, but it is an issue for PC levels.

Also as a footnote, while I used CalMAN to test and gather the data, this is not a specific issue to anything except the EE hardware and therefore is relevant to any thread discussing the EE box.

Is this really an issue for typical LUTs though? If you have calibrated the native white point close to the LUTted one there won't be the kind of discontinuities you show above.
zoyd is offline  
post #60 of 496 Old 03-14-2013, 10:12 AM
AVS Forum Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 7,025
Mentioned: 7 Post(s)
Tagged: 1 Thread(s)
Quoted: 331 Post(s)
Liked: 335
Quote:
Originally Posted by zoyd View Post

Is this really an issue for typical LUTs though? If you have calibrated the native white point close to the LUTted one there won't be the kind of discontinuities you show above.

For video its really a non-issue. But if you were using one for PC levels and it had a wide gamut it would be a serious issue.

Every single triplet were a value is 255 is completely uncalibrated. For any sort of professional use, that needs to be a known issue.

My only point is that this flaw exists, even though Steve denied it early in the thread.
sotti 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