Originally Posted by Matrix7
The blacks can be fixed on the LGs to get them down to 0.0035-0.005 fL as well. (The 5300/5500 are 0.004-0.005 fL.) Blacks on a LCD (IPS/VA) are obtained through turning off or dimming the back-light. Even my phone's LCD has better blacks with automatic brightness enabled, but this also makes the whites look mid-gray and the blacks are pretty bad off-angle. My Vita (Samsung AMOLED) is the most impressive display I have (For contrast ratio/black levels). The only minor note is that the colors (towards blue), contrast ratio and black level shift a bit with the viewing angle. The LG OLED TV reviews also note this happening.
So can they on the 5300. Mine is currently measuring 0.002fL with Cinema Smooth on. Let's agree that in general the Samsungs have better black levels. Voltage modifications can lower black levels but a very small fraction of people are going to do that. Even if you do, an apples to apples comparison will still have Samsung with the advantage.
Originally Posted by Matrix7
Fierge_gt and orion2001: If you look at the numbers from the CNET calibration results, the LG Plasma score better for the color error metrics than the Samsung and Panasonic Plasmas. Check their Samsung F5500 and Panasonic S60 reviews. (The F5500 scored poorly for Red Error and the reviewer noted crushed blacks.) The LG scored good in all of these categories which even the F8500 and ZT60 did not manage.
LG haven't improved black levels since 2010. The PDP60R1 to PDP60R5 (The 2014 models all use the PDP60R6 panel) have black luminance around 0.011 fL to 0.013 fL while the Samsung and Panasonics are in the 0.002 fL to 0.005 fL range.
There are a number of issues with both the calibration CNET did and your comparison. Firstly, I wouldn't compare the F5500 result since the CNET forum post detailing his calibration explicitly states that the TV had no CMS control so that hampered the calibration. So obviously the color results would not look good. The F5300 however does in fact have a very robust and capable CMS system along with 10pt whitebalance, with all controls working independently of each other and as one would expect.
The other big issue with CNET and many other calibrations is that people calibrate the 100% saturation targets and call it a day. This in reality is a pretty terrible idea since hardly any colors in most videos are being displayed at 100% saturation. What you care about, is color accuracy in the 25-75% saturation range. Even though CALMAN allows you to do saturation sweeps, hardly any of these commercial sites actually do them. Let's look at my calibration result with the PN60F5300 again:
The colors line up almost perfectly in the 25-75% range. My measured dE values are typically <0.5 for most colors. I intentionally traded off accuracy at the 100% targets by a small amount so as to better hit the 25-75% targets. Cnet's calibrations don't provide you with any meaningful information on how well the set actually tracks colors at meaningful saturation levels.
Thirdly, after looking at David's calibration results and notes on the PH6700, the PNF5500 and the PNF8500, I'm quite convinced that he either does a shoddy/rushed jobs with his calibrations, or he doesn't have a great grasp of the process in general. Points in support of this are: 1) His post calibration Color Luminance chart for the LG set looks worse than prior to calibration despite having a CMS to play with, 2) He somehow managed to have issues getting a uniform <2 dE gamma tracking on the F8500 with 2 and 10pt white balance controls available. Not sure how he managed this. 3) His overall Gamma curves for all 3 sets look absolutely atrocious for sets that have 10pt white balance controls. While he gets the average value to line up to a reasonable number, the deviation for different gray levels is MASSIVE! It clearly illustrates that he places little to no effort in matching luminance targets for a target gamma and simply gets dE numbers down by targeting neutral shades of gray for each gray level. Whether this is in the interest of time or ignorance on his part is open to interpretation.
On the other hand, you can see just how accurately I could tweak the gray scale to match BT.1886 gamma on my PN60F5300 below. Most dE < 0.5, All <1.0:
So I really wouldn't use his measurements on a clearly non-optimal calibration on all 3 sets to then conclusively conclude that one is better than the other with color accuracy. The truth is that if any TV has a decent CMS and no glaring issues in its color reproduction, a capable calibrator should be able to get the colors quite accurate.
Finally, something that I think is important to this discussion... let's look at how atrocious the out of box white balance was on the CNET LG set:
Compare with how much better it is in Movie mode on both Samsungs:
So, if we base conclusions solely off CNET's results and a single datapoint, and if you don't have a meter or plan on buying one, you are far better off getting a Samsung. At least from the 3-4 folks with meters who have measured the PN60f5300 sets here on the forums, all of their measurements show that the out of box Movie mode measurements are quite good. For someone with the LG set, you'd either hope that all sets have the same drastic inaccuracy and that simply adjusting using Cnet settings will get you an accurate picture, or you will need to invest in a meter.
In either case, I really would not use a blurb from a CNET review to justify the position that the Samsung sets are inferior to the LGs when it comes to color accuracy.