Datacolor SpyderTV Pro [Archive] - Audio & Video Forums


View Full Version : Datacolor SpyderTV Pro

11-20-2006, 01:37 PM
Even though the thread about the SpyderTV project at AudioReview was on the Home Theater board, I figured that the Video section was the best place to post this report. Datacolor sent me the Pro version late last summer, and I used it for the first time last weekend. To my mind, it is a terrific instrument, more comprehensive than the popular consumer version. Geared toward technicians, the system's software permits dealers and installers to navigate the various tests in whatever order they wish and then to customize their reports by client. To my mind, its main advantage, however, is its ability to adjust grayscale via red, green, and blue gain/bias settings when they are accessible outside arcane service codes. More on this issue later.

The display that I calibrated with the SpyderTV Pro was a Sony KDS-55A2000 SXRD (LCoS) RPTV, which impressed me from the moment that I first set eyes on it in a Sony store, though I could tell that its grayscale was inaccurate. Before I began my own calibration, I set up the Sony with parameters from CNET, which purported to derive from a thorough examination. The A2000 series has an Advanced Video Menu, which is available only within the Custom Picture mode. Among other things, it contains gain/bias settings to alter the grayscale. Without proper test instruments, no one should tinker with these settings. Altering the white/gray/black balance can destroy a display's color fidelity, as well as its brightness and contrast. Enter the Spyder Pro.

Like its baby brother, the Pro consists of a colorimeter that adheres to a display's screen (the spider) and connects to a PC--preferably a laptop--running proprietary software via USB. A DVD player is necessary to run detailed instructions, show the various test screens, and, at the end, assess the results. After the user supplies basic information about the display's characteristics for the software, the colorimeter begins taking the measure of the display through multiple snapshots of the test screens, while the calibrator follows instructions about how to proceed on the computer. The system analyzes contrast, brightness, and color temperature first and adjusts color and tint last. It has no provision for sharpness, but sharpness sliders on modern sets are totally misplaced; the compensation for crude filtering that once justified them has not been necessary for a long time. At each step along the way, the calibrator can refer to a graphic illustration of how the new settings fit into the display's relative scheme of things. Before finishing for good, the Pro briefly double-checks its settings and offers a four-page printout, or PDF file, of its results vis a vis its preliminary readings. A few well-chosen still pictures on the DVD are available for before and after comparisons of black level, grayscale, and color.

Grayscale, the system's centerpiece, comes under study after contrast and brightness. The software shows a cylindrical graph of the relationship between the various intensities of red, green, and blue for both gain and bias. The object is get each color as close to the 100% baseline for intensity as possible. If a particular color cylinder's top edge is too far over or under this reference point, the computer program advises how to correct it. Displays will behave differently when confronted with this information. Some of them will be able to achieve good balance, within certain tolerances, and others will resist until the calibrator is, literally, blue in the face. In this case, the Sony was a real trooper, not quite capable of meeting the standard perfectly for each color's gain and bias but coming quite close. The pop-up target illustration showed that its white balance was just outside the bullseye of D65. It possibly could be brought even closer with a little work in the service menu, but I don't think that the effort would be worthwhile in this case. It might be with other displays. All told, the calibration lasted between 45 minutes and an hour. Conceivably, someone might have to run it twice. It's possible to make a mistake when entering data and elicit a wildly inaccurate reading.

The initial calibration courtesy of CNET looked very good to me. The Spyder Pro's calibration, however, seemed to take it a notch higher. When getting familiar with CNET's version, I had a nagging sense that it left a little too much blue in the mix, despite the CNET reviewer's efforts to eliminate it by adjusting gain and bias. My contrast (known as "picture" on the Sony), brightness, color, and tint ("hue") parameters approximated, if not matched, CNET's, but the Pro's gain and bias levels deviated a little, removing the last vestige of a blue tinge. I could find no trace of it in black and white material (I hate a blue grayscale; if perfectly neutral were impossible, I would trade it for a hint of red).

The question remains, Should anyone but dealers and installers buy this product? With a price of at least half a grand, the Pro isn't cheap, but compared to the cost of dedicated ISF test equipment, it's a true bargain. However, only enthusiasts with flat panels or RPTVs that have accessible grayscale controls would fare better with the Pro than with the less costly Spyder, which omits grayscale adjustment. Those with front projectors have no choice, since the original Spyder doesn't work with them, but front projectors often have gain and bias controls. The upgraded software package is certainly a plus, but for the single user who doesn't have an extensive clientele--paying or otherwise--it's not likely to make much difference. Within these practical constraints, I recommend the SpyderTV Pro wholeheartedly. Colorimeter and software come in a small, lightweight carrying case. User must supply DVD player and PC. [Eric, if you're out there, you're welcome to lift this post for a review.]

11-20-2006, 04:25 PM
Nice review, Ed. When I calibrated with the normal Spyder, I was surprised grey scale wasn't addressed. It sounds like they saved that for the Pro. To be honest, I haven't screwed around enough in my menu system to know where greyscale adjustments are, but I have a feeling they are in the service menus, a place I have no desire to visit.

One thing that has always posed as a bit of a quandry for me: The Spyder calibrates via the picture provided by the dvd player, but what if the dvd player isn't dialed in? The 2910 I have has a mind boggling array of adjustments, all of which are beyond my comprehension. I just left it on the factory settings (except I changed the IRE setting based on info from avsforums) and hope for the best. What if the dvd player isn't calibrated correctly? Doesn't that affect the Spyder?

11-20-2006, 04:49 PM
Hi Speedy,

You're perfectly right. Every calibration is specific to a chain of components--in this case, a DVD player and a TV--unless the calibration equipment is professional standard. However, even if it is, when you revert to your own DVD player, you may have to toss it out the window. That's why it's better in some ways to use your own equipment. If you're lucky, you can turn off any elective enhancements on the DVD player, making sure that if it has a switch for video black or PC black, you pick video. Some of the earlier digital outputs defaulted erroneously to PC levels, which crushed the top and bottom ends of the grayscale.

But all is not lost. DVD players aren't necessarily the scourge of the earth. You may well find that calibration results obtained via DVD will hold up when applied to cable or satellite feeds. I would guess that your 2910 doesn't pervert what it shows. I remember the quandary that you had about the IRE choice of 0 or 7.5. Relative voltage shouldn't affect digital outputs, only analog ones. In any event, you could find out which one to use by running the pluge test on DVE. If you can see blacker than black, you're golden.

I don't know where gain and bias are on the JVCs. But wherever they are, you'd be wise not to touch them without already having a good idea of how they should look. AVS might have recommendations. One thing you can do in the meantime is informally look at black and white test screens to see whether any color seems to dominate. Red in blacks and blue in whites are the usual suspects.


Edit: My name one time is more than enough, and grammar counts.