Keohi HDTV



Powered by Keohi Web Design


PT47WX51 Review (First Look)

<< Back to Panasonic Tips

Tip from Michael TLV:  "On a lazy afternoon, I found some time to visit the local home theater store to check out the new RPTV. It had been set up in one of their home theater rooms so the lighting conditions were well under control. They had the TV set at 50% contrast (31/63), but still the image was hurting both my eyes and that of a friend who accompanied me to watch what I actually do at these stores. Obviously, contrast was still too high at 50%. I turned it down to 30% and the pain subsided. I also turned the sharpness down to 30% or so because it was terribly annoying at 50%.

This was my first chance to play with the new 2001-2002 line up. So for the record, consider this a "work in progress" report on how this set stands up to the rigors of calibration as well as a first look at some of the issues that complicated matters.

So the first thing up was to examine how this set was out of the box with only an AVIA/VE set up in place. After setting it up optimally via the discs, it was time to take some readings to check on colour accuracy in the TV’s warm mode. "Bad news coming, Dr. Jones." Warm hurt our eyes … especially the whites or what passed as white. Too blue. 13,000K on the dark end and +20,000K on the light end. No wonder we were getting a headache. Now just for kicks, I looked at how different the Normal and Cool colour temp settings were. Too blue as well, but surprisingly close to warm and that would seem to track the whole session. Temperature deviation between all three colour temp settings was at most only 1,000K.

The source DVD player used in the calibration was the new Progressive Scan Panny 91 unit. A quick check of the needle pulse pattern showed that the SVM was firmly in place. (A Mr. Yuck sticker). The SMPTE colour pattern didn’t look quite right at all. The colour yellow was missing as it was a sick orange – yellow hue. Some work on the colour decoder was obviously required here.

So it was calibration time … a first round of changes and experiments to see what this set was all about. Entering the service menu, the two colour decoder parameters were right there for easy manipulation. There was actually not much red push, but the green was significantly lacking. Fortunately, the decoder adjustments took care of that. Very nice of Panasonic to include this feature in the service menu. Virtually perfect colour decoding. I can live with that. Just being able to do that was worth its weight in gold. Picture getting better and the colour bar pattern now looked normal.

Next up, the SVM. Readily visible and nothing the end user could do about it. I took two approaches here to see what I could do about the SVM. The first was the software adjustment approach. I had some notes on accessing the design menu of the Panasonic so I went in to see what all was there. There are many, many parameters in the design menu. Many 100’s of things to change. Unfortunately, the information for SVM taken from the 51" 4:3 Panasonic set could not be applied to this design menu. A real bummer. I have access, but no way to figure out what the parameters mean. It will definitely take some exploring to figure out this menu. So software SVM disabling was not possible at this time.

I opened up the front end of the set and looked for the hardware fix. I found the three SVM wires and where they connected to the various circuit boards in the TV. It was a matter of unplugging them and all would be fine. A second problem arose. Now unlike the Toshiba sets, for instance, where disabling SVM is as easy as plugging in a sound card into a computer, the plugs on the Panasonic were wedged in so tightly that it was virtually impossible to pop out the three SVM connectors. It looks like the only way to disable the SVM would be to cut the three wires. (Since this was a floor model, I left it like that. But I know where the wires are for future clients. )

Moving around in the service menu was very straight forward and one feature actually allows you to work with the colour decoder adjustments without the need for the three colour filters.

A quirky aspect about the service menu parameters is that not all the items are available for you to adjust in all signal modes. For instance, on the S-video signal, all service parameters are available for adjustment. Now for 480P, all but five parameters are locked out. You therefore cannot adjust grayscale using the 480P signal from your DVD player. You have to default to either S-video or Component 480i.

For now, I will speak to the S-video signal type and how well it calibrated. "My God, it’s nearly perfect." That is how I describe the grayscale tracking via the S-video input. Literally linear from 20 ire to 100 ire with no more deviation than +/- 30K. Unfortunately, it gets worse from there on in. With the TV calibrated to D6500K via the S-video, things looked really nice … (another big issue to come though) One issue that popped up here was grayscale uniformity from the left of the screen to the right side of the screen. I put up the vertical grayscale step pattern and the grays in the 35 ire to 50 ire range (left of center) looked awfully red tinted. We were quickly approaching the world of "Lens Striping," but that is something that I will discuss in the next installment. A reading at the 1/3 point left of screen center read 5900K compared to 6500K in the center. This explains the red tint. The left edge actually reads 6700K and the right edge was 5700K.

I wish I could have stopped there, but I could not. Because we are feeding the TV with a 480P signal as well, that is the desirable input to calibrate correctly.

So what did the 480P signal look like now that S-video was 6500K? I crossed my fingers hoping that we would need no more changes. We’re not that lucky. 100 ire in 480P measured 6600K … so far so good. But now 30 ire was 8100K … Too blue. The dilemma here is to make 480P track properly without impacting the rest of the TV input signal types significantly. I know that I have to sacrifice the S-video for the sake of 480P. (But before I go too far, I did not have a chance to try 480I component in the service menu, but I believe that the RCUT and BCUT parameters are available there. This may be the solution to one of my problems, but for now, I will describe what I actually did to make 480P track closer.)

Problem … S-video tracks linearly at 6500K. 480P rises sharply on the dark end.

Solution … Bring the S-video grayscale down to 5900K from end to end. The resulting pattern range on the 480P signal was a 5900K reading at 100 ire and a 7000K reading at 20 ire. With both ends of the grayscale slightly compromised, this allowed the entire midrange of the grayscale to be much closer to 6500K. The added benefit of doing this was that it solved the uniformity problem on the screen left of center.

Taking a step back, the images were now looking far more pleasing than where we started … but only on the 480P input. The compromise that had to be made to the S-video made that signal type now look positively dreadful. I am hopeful that I can address this issue in the next session with the TV.

A remaining major issue with the set is something that I dread. A look at the green CRT convergence grid alone showed that the entire left side of the screen was significantly out of focus. And badly too. It is not as noticeable on movie material, but clearly the image should be better.

So what remains for this set:

  • A Mechanical focus
  • Electronic focus
  • Lens Striping
  • Green CRT refocus … shimmy perhaps … re-torque the screws …
  • SVM
  • Screen protector …

We’ve come a long way and the image is shaping up nicely. The sales staff and the various customers obviously like what they are seeing so far. But would I get this TV right now as is … nope. Fully calibrated Toshiba sets are still better. Give me a couple more weeks with the set to finish up the other calibration aspects and ask me again.