LCoS Shoot-Out Results

mikemuch writes "DisplayMate founder Ray Soniera has revealed the results of his LCoS HDTV Shoot-Out. He puts five HDTV's through a slew of test pattern measurements, and then lets 34 real people, including home-theater lay people and experts, conduct jury tests and make comments. There was one case where the experts gave low marks to a display that the lay people loved. From the article: 'We spent some time trying to understand why the consumer panelists rated the JVC Consumer unit so highly. It had the lowest objective on-screen resolution of all of the units, because of internal signal processing, but a number of consumer panelists commented on how sharp it looked. The copious artifacts and significant edge enhancement produced so much artificial texture in the image that some panelists interpreted it as superior sharpness. All of the Video Experts recognized this effect and gave the unit the lowest score.'"

Comments

Expert textpert choking smokers

[ExE122]

I actually think this result is just a matter of a having a trained eye... just as a real musician would probably cringe at the sound of most pop songs on the radio, despite the fact that a large number of people actually enjoyed that kind of "music". [flamebait warning]

But seriously, I wouldn't expect a "lay person" to be able to understand the technology involved in these units and to be able to make any intelligent\educated distinctions about their quality. IMHO, there's a reason we call them experts and they are the only ones we should really be paying attention to.

Re:Expert textpert choking smokers

[Red Flayer]

"But seriously, I wouldn't expect a "lay person" to be able to understand the technology involved in these units and to be able to make any intelligent\educated distinctions about their quality"

You've got to define quality here -- it depends on your goal and what metrics you assign to measure achievement.

Is your goal to maximize appreciation of the picture quality in your target market? If so, what's your target market -- video experts or typical consumer? What's the crossover between the two markets?

If my customers are more satisfied with my product than the 'experts' say they should be, then good for me. The problem here is not that experts and customers disagree -- the problem is that they are using different metrics. And to the people actually buying my product, it's their metrics that really matter.

Re:Expert textpert choking smokers

[PFI_Optix]

There's something to be said for a "non-expert" opinion on the matter.

To go back to your analogy of musicians: There is some "music" which is absolutely adored by the experts that sounds like utter crap to the lay person. Why? Because what the expert hears is technical acheivement, innovation, something hard to play that's never been done before. What they lay person hears is an annoying cacophony of seemingly random blarings from an orchestra. I'm thinking of a specific orchestral piece I heard on NPR a few months back. The composer's name eludes me, but his work made a lasting impression...it was impressive to me as a musician that he could write it, but (at best) annoying to listen to.

A monitor can have all the technical features and perfect picture in the world to impress the experts, but if another "inferior" TV somehow fools the average buyer into thinking they're looking at a better picture, which one do you think they'll buy? Last I checked, buyers far outnumber experts.

This article raised an excellent point about the *difference* between what technical experts and average consumers see when they look at a TV. In the end, two things will influence a buyer more than anything: their wallet and their eyes.

No surprise

[Tx]

I know people who watch 4:3 content stretched out to 16:9, and are apparently immune to the completely distorted aspect ratio, they just think whatever they're watching should fill the screen regardless. If a consumer panel contains people like that, I don't wanna know what they think.

Brighter == Better

[engagebot]

To the average Best Buy shopper, the brightest screen in the lineup wins. Doesn't matter if the red tones are blown out, doesn't matter about artifacting.

Just turn the brightness control down a few notches on a particular TV in the lineup, and watch the Best Buy sales numbers change.

Same thing with audio equipment. Room-shaking bass and razorblade sharp piercing highs sell gear. Doesn't matter if its a balanced sound, or if there's any separation between the elements in the mix. More bass? check. Killer sharp highs? check. Go to the checkout counter.

Re:Brighter == Better

[Kadin2048]

I think in retail it's a combination of brightness and contrast/color saturation. If you look at the TVs people are drooling over at Best Buy, they're often the ones that have the contrast and saturation jacked up ridiculously high, also. Sometimes to the point where flesh tones start to look really distorted, everyone looks like they're wearing a lot of blush on their cheeks and stuff. It's pretty bad.

But this same thing happens with photos. A few years ago there was a sort of "contrast war" between the makers of different high end digital minilab equipment (principally Agfa and Fuji). In order to create pictures that "look best," they each would come out with new software for the minilab system that would pre-process the digital image coming from the film scan before it went to the printer. Generally the "automatic" options (on either brand) would compress the dynamic range horribly, then proceed to drive the saturation up to almost unbelievable levels. But customers loved it because it made their vacation photos look like postcards, so what the hell. Nobody really cares about 'accuracy' in the real world -- or rather, not accuracy to the physical world or to the film, they want a product that's accurate to their memory of something, which often is nearly unrelated to reality. Give them that, and you'll get rich.

Same thing with the "bass boosters" or "sound enhancers" on low end stereos. It mucks the music up, but people think it's better that way.

The television thing is the same. People don't really want to see what the actual football field looks like, they want to see what they think the football field looks like, and that means the grass ought to be bright, hunter green, the white uniforms should be almost shiny, and the yellow lines should be just about ready to pop off the screen, walk across the room, and rip your eyeballs out. Being true to the video signal that's coming into them isn't a factor.

This is why if you want accuracy, you generally have to pay for it or expend some effort. With a photo, you have to tell the lab operator to run it though without corrections. With audio, you have to get "nearfield monitors" instead of regular consumer stereo speakers, and with televisions, it's why there are video monitors that are actually made to display what they're being fed, instead of an idealized version.

It's all about giving people what they think they want.

small wonder...

[pulse2600]

One possible explanation for the consumer ratings is that JVC is simply giving consumers exactly what they think they want.

This statement hits the nail on the head...JVC knew what they were doing when they made a technically crappy screen, just like Microsoft cares more about how much users like clippy the office assistant than they do about a buffer overflow. They know what they need to do to sell their product, most other things are irrelevant. Why should JVC give a flying rat's if 100,000 geeks see artifacts when 1,000,000 non-geeks see "sharpness and texture"? They'll probably make more off the geeks by selling them some model they deem "higher-end" than the consumer version for 20% extra, because the geeks will percieve it as being so much better than the "inferior consumer" model. Someone at JVC really knows how to play the consumer perception card real well, and I bet this particular example comes at a manufacturing cost savings as well.

Re:small wonder...

[DiscoOnTheSide]

Having worked in a Sears in Toms River, NJ in the same department... it's not so much that they're doping the signal... it's that retards wire the displays. No TV shop is going to be able to show you the full quality of a TV using an HD signal. Sears had their feed from DirecTV... the non HDTVs were put to the plain ol' Discovery Channel. Now... the splitters they use for the HD screens... are pretty efficient... but there is still a quality loss at each split. The non-HDTVs were hooked up via run of the mill coax going through (as far as I could count) 80 RF splitters... these splits are NOT as efficient. Also, for some reason, some TVs took the split signal fine, others wouldn't touch it and a lot of the LCD TVs (~20 inches) put out the most god awful picture ever. The DirecTV sat box pushing the HD signal was set to 480p... nicer than regular TV but nowhere near the level that the TVs could produce.

I rewired a bit of the store (my manager didn't give a shit cause the better the TVs looked, the more likely we were to sell). All of the top-shelf TVs (particularly the Sony XBR LCoS line) were hooked up to Samsung or Sony upconversion DVD players via HDMI. Pretty much I could say "This is exactly how DVDs will look on your TV, and full HD service even better." And customers ate it up. Eventually I swapped DirecTV boxes out of the break room and into the display and low and behold 1080i went to all of the HDTVs.

The difference was immediately noticable and sales surged. I was then fired for not selling enough warranties, my 9.5% not up to their 10% "desired goal", regardless of the big increase of sales I brought in... If the way that store is run is any indication of how Sears as a whole opperate... I give it a decade until they're all K-Marts. They'd shut off the AC on 100 degree days at 8PM (closing time is 10) to save money. Older folks were about ready to have heat strokes, and as was I, surrounded by CRT and Plasma screens all day...

Re:LCoS?

[Mr Z]

Liquid Crystal on Silicon. It's a reflective (as opposedt to transmissive) LCD technology. You basically get all these liquid crystal mirrors to play with, where the rest of the logic on the silicon switches the mirrors rapidly between "reflect" and "absorb" thousands of times a second. (Similar to how DLP works, but instead of actual mirrors rocking back and forth, it's just LCD switching on and off, playing with light polarization.)

As a (relative) industry insider . . .

[mmell]

(My dad was a TV repairman - I grew up looking at television damned near 24x7)

Most consumers don't want a realistic looking picture, they want the picture they've seen all of their lives. Even with televisions; many of my wife's family and friends upon hearing about my background, asked me to look at their televisions. Most needed minor convergance/pincushion adjustment, all needed brightness/contrast/color/tint adjustment. I made them all look (IMHO) pretty good.

Virtually every set I touched was changed within a week. The single control that was most nudged: color (think saturation). Everybody is used to the cartoon-level, LSD-induced superbright colors of a children's room. Real skin doesn't look like that!. I could even hold my bare arm up next to a character on TV, show my relatives and friends that this is what the picture should look like (gee, flesh looks like flesh. Grass looks like grass), and within ten minutes they'd be cranking up the color.

I gave up. Nowadays, I tell people "I don't do Windows, and that includes televisions". Yes, I get some wierd looks for it, but I also get bothered a lot less.

Buy the television which matches your pocketbook and your expectation of picture quality. Most of you will never miss the extra quality that a 200-300% increase in price will bring; worse, you'll probably adjust the extra quality right out of the set in a quest to get the lurid color balance you want. By the way, on a new set you should have a pretty good picture if both brightness and contrast are set to mid-range. Cranking both of them to max may look like what you want, but you're just cutting the lifespan of your picture tube in half (applies to CRT's only - I have no idea what the effect is on LCD/Plasma displays).