Ask S&V

Pete Starsmann

You're right—there are different ways to measure amplifier power, making this is a very confusing issue. Manufacturers often measure the power output of their amps in whatever way will yield the highest number, and there's no way to know which method they used unless they specify it.

Mark Peterson, who performs the audio measurements on review products for Home Theater, has this to say about it: "We measure and publish RMS continuous (100% duty cycle) output with a specified number of channels driven, in the spirit of the original FTC mandate. This is a completely synthetic test, in that it doesn't represent the demands of real-world program material, but it does generally give insight into the capabilities of the oh-so-important (and relatively expensive) power-supply section.

"Beyond reading the fine print and assuming that the manufacturer's marketing department wasn't in charge of the spec sheet, there often isn't enough commonality or specificity in the conditions of the test to meaningfully compare output numbers from different sources. Keep reading Home Theater's amplifier and AVR reviews, and you'll start to see a pattern of manufacturers whose claimed numbers tend to track with what we measure more closely than others."

Time to Upgrade
I currently have a Marantz SR7000 A/V receiver and a Panasonic PT-51DX80A rear-projection TV. I would like to update the TV to an LCD flat panel with LED backlighting, such as the Toshiba 46SV670U. I'm told that I can operate it without HDMI (since the receiver lacks that interface) using S-video instead. I would feed the receiver with an optical link from a Blu-ray player. Will this work satisfactorily, or should I upgrade the receiver?

Gary Lapman

You can use the Toshiba—a very good TV, as Home Theater's review reveals—or any other modern TV without HDMI, but S-video would not be satisfactory for me, especially from a Blu-ray player, because S-video cannot convey high-definition signals. Unfortunately, the SR7000 had no component-video inputs or outputs, so you're not going to see high-definition through it no matter what you do. I would definitely upgrade the receiver to a modern model with HDMI. In addition to seeing high-def on the TV, you'll also hear lossless Dolby TrueHD and DTS-HD audio with no compromise; if you send audio to the receiver via optical, you'll only hear the older Dolby Digital and DTS audio, which loses some audio information in the compression.

Color My Space
As I understand it, YCbCr is a way of encoding RGB information, and it's the way that DVD and Blu-ray discs store video information. Also, by connecting an HDMI cable between a DVD player and an HDTV, the cable carries RGB information to the television. Doesn't this imply that the DVD player is performing the conversion to RGB? So, if you connect component cables, would the YCbCr information be sent to the television, which would then perform the conversion? My main question is this: where is the optimal place for the conversion to take place? In a player? In an A/V receiver? Or in the HDTV? The TV is likely the most expensive component (and perhaps the most capable) of the group, which might lead me to believe the conversion should be done by the TV. But I've been led to believe we should use HDMI cables in lieu of component cables wherever possible. So what is the best approach?

Alan Goldstein

Great question! You are correct that video information is stored on DVD and Blu-ray in the YCbCr color space, which is a form of data compression applied to RGB. And connecting a player to the TV with component cables does, indeed, convey YCbCr to the TV, requiring the TV to convert that signal to RGB. However, it is not true that HDMI can only carry RGB—in fact, it can carry either RGB or YCbCr. Some players provide a control that lets you select one or the other color space. If there is no such control, the onscreen display of the player or display might indicate which type of signal is being sent.

As for where is the best place to perform this conversion, that depends on the quality of the video processor in each device. Recently, I was talking with Stacey Spears and Don Munsil, the creators of the HD Benchmark setup and test Blu-ray, about this very subject. They pointed out that test patterns let you clearly determine which device in a video-signal chain does the best processing. For example, if you have a player that lets you set the output to RGB or YCbCr, HD Benchmark has several patterns to look at with both settings—if sending RGB looks better on these patterns, the player should do the decoding, but if YCbCr looks better, the display should do it. You can also apply this approach if your AVR or video processor provides similar color-space controls.

Which patterns am I talking about here, and what should you look for? Here's what Spears has to say about this:

"The two primary patterns are chroma alignment and chroma multiburst. Look for delay in the horizontal direction on the alignment pattern. On the multiburst pattern, look at the top left burst; the gradient should be smooth, not stepped. The disc includes sample images that show what happens on both patterns. You also want to look at the last burst on the second row of the multiburst pattern. For example, when you send 4:2:2 into the DVDO VP50/50Pro, it will look nice and bright. When you send 4:4:4 into the DVDO, it will be faint.

"The clipping pattern is also important. It was created to identify a bug in the Silicon Image HDMI transmitter/receiver. If a device uses this chip to color-convert from 4:2:2 to 4:4:4/RGB, it will clip in YCbCr first. If only white is clipped, then it is the bug. The reason it works is because the RGB boxes all fall within 16-235 in YCbCr and outside when converted to RGB. The white box extends above 235 in YCbCr. If white is okay but RGB are clipped, it's probably a color-decoding problem or color is set too high. If green is clipped and everything else is okay, it is most likely 601 color conversion instead of 709."

If you have a home-theater question, please send it to scott.wilkinson@sorc.com.