3D LUT Calibration Software Comparison

[Ted Aspiotis]

Steve Shaw performed a comparison between current versions of LightSpace and CalMAN to evaluate the 3D LUT profiling correction performance.

Devices used for this test were the following:

BON Grade 1 Broadcast Monitor
X-Rite i1Display PRO Colorimeter (i1d3 OEM)
DVDO AVLab TPG External Pattern Generator
eeColor 3D LUT Holder Device (65-Point Cube)

The point of this test was to compare the results between software solutions based on maximum required time of measurement of 2 Hours 30 Minutes.

LightSpace CMS took 2H30M for a 21-Point Cube Profile (9.261 Color Points).

CalMAN IR Profiling (time based selection) used to measure for 2H30M and it took about 6.000 measurements.

Later both software solutions generated the 3D LUT correction table for eeColor 3D LUT Box.

For verification, each correction loaded as Active LUT to eeColor 3D LUT Box and measured with LightSpace with 21-Point Cube (9.261 Color Points) measurements to maximize the verification results.

To view the results of this test you can visit that page: http://www.lightillusion.com/calibration_comparison.html

 

[Mike Nagel]

The comparisons that I ran many many times looked much worse for CM, but I run much larger custom patch sets, which isn't an excuse (for CM)... speed, accuracy, feature set and modularity of LS in unparalleled in 3D LUT world...

Ted, have u by any chance run a full 3D LUT test run with your disc on LS and CM and compared the results ?

 

[Ted Aspiotis]

http://www.my-hiend.com/vbb/showthread.php?6671-【訊源】OPPO-BDP-105-BDP-103-儀器實測分析

The comparisons that I ran many many times looked much worse for CM, but I run much larger custom patch sets, which isn't an excuse (for CM)... speed, accuracy, feature set and modularity of LS in unparalleled in 3D LUT world... Ted, have u by any chance run a full 3D LUT test run with your disc on LS and CM and compared the results ?

Hi Mike, for my HT calibrations I'm always using my calibration disk a pattern source for 21-Point Cube Profiling I'm running, so doing this I'm calibrating my whole video chain.

So the correction table is including any errors that a device like a blu-ray player can add during it's the internal colorspace conversions and processing.

This not possible with other software solution where you have to use an external pattern generator to display the required patches, but doing this you are not calibrating the whole video chain, you just ignoring your player output. It's interesting to see a comparison of the output between of two OPPO Blu-ray players and a reference pattern generator, the differences of each player output have measured and it's easily detectable by looking the reference images used for that test.

 

[Mike Nagel]

yeah, always interesting to see which devices distort output... for single device chains with distorted output this is the way to do it.