Site Links

THX Tech Training

Posted Wed Jul 30, 2008, 3:13 AM ET

My blog is a bit later than usual this week, but I've been pretty busy. Tom Norton and I are taking the newly developed video-technician training course offered by THX at the company's headquarters in San Rafael, California, just north of San Francisco in Marin County. Tuesday was the first of three full days of instruction and hands-on lab work, after which some of us went out to dinner and caught Hugh Masekela's set at Yoshi's, a famous jazz spot in Oakland. After a wrong turn by Laurie Fincham, THX's brilliant but directionally challenged chief scientist—thanks for the grand tour of San Francisco, Laurie!—I just got back to my room.

Now, you might wonder why Tom and I need to take such a course—after all, we calibrate and review TVs for a living. In fact, we don't need it; most of the material is very familiar to both of us. But there's always something to learn, even in one's field of expertise. Also, we wanted to see how THX was training those who would then go out and perform calibrations for paying customers.

The curriculum is well-planned, and instructors Gregg Loewen and Michael Chen present the information in a tag-team fashion, which helps keep it fresh even after eight hours. They alternate between PowerPoint slides outlining the basic concepts and procedures for calibrating a video display and hands-on sessions with 13 different flat panels and front projectors located around the room. The pace is fast but relaxed, and questions are encouraged.

One thing I learned is that film has an effective resolution of about 2600x1400, or 3.7 megapixels. This is roughly twice the resolution of 1080p, so film should look much sharper than HD video, right? Wrong. As film is pulled through a projector, it starts and stops 24 times per second and, as a result, it jiggles around quite a bit—a phenomenon called "judder"—which reduces apparent resolution.

Also, when the projector's shutter, or "gate," opens for each frame, the heat of the lamp causes the film to move back and forth, which changes the focus. These factors reduce film's actual effective resolution to something like 1.5 megapixels, which is less than 1080p's 2MP. As a result, Blu-ray and commercial digital presentation can actually look more sharply resolved than film. I've preferred digital cinema over film since I first saw it, and now I have some evidence that it is actually better. Take that, film snobs!

Of course, the amount of detail you see is directly related to your seating distance. The figures I've used here assume a seating distance of 1.5 times the screen height, which is usually about halfway back in a typical commercial cinema.

But back to calibrations. As I've maintained for a long time, the course emphasizes that the single most important thing you can do to optimize any video display's performance is to set the black level correctly. The eye is much more sensitive to small changes in low levels than in high levels.

Next in importance is maximizing the dynamic range (contrast ratio) by setting the white level correctly. The THX spec for peak contrast ratio (full on/full off) is 2000:1, and the ANSI (checkerboard) spec is 200:1, which is better than film can manage. I've seen many digital video displays that meet or exceed these specs, which is another reason I prefer digital video to film.

Speaking of ANSI contrast, I haven't been measuring that in my TV reviews because there is a lot of controversy about the methodology by which a meaningful result can be obtained. Loewen and Chen advocate a straightforward approach—simply measure the light levels in each of the checkerboard squares in a dark environment, average the white and black levels, and divide the white average by the black average. This is exactly how I've done it in the past, so I'm considering adding it to my measurements in the future.

Tuesday was all about setting the basic user controls—brightness, contrast, color saturation, tint, and sharpness. Wednesday, we'll tackle grayscale calibration, and Thursday will finish up with tweaking a set's color-management system (if it has one). Don't tell our instructors, but Tom and I plan to cut class for a while to visit THX's video-certification lab, a new service that complements the company's long-standing audio-certification program.

I'm glad I made this trip to the Bay Area—where I lived for four years in the '80s and love to this day—and I look forward to honing my calibration skills further. Anyone who wants a solid grounding in the art and science of video calibration would be well served by enrolling in this course, which THX offers at various locations around the country, including the CEDIA Expo in September. Check the company's website for more info.

And now, to sleep—another early day awaits.

If you have an audio/video question for me, please send it to scott.wilkinson@sourceinterlink.com.

Permalink  ::  Blog Home  ::  Email This

Previous Post | Next Post >

Reader Comments 

Posted Wed Jul30, 2008, 3:22 PM — By Dave Anderson

Sorry to be a crank, but here goes: I have yet to see or hear the word "methodology" used where the person didn't mean "method". Methodology would mean something like "a study of methods". Adding three syllables to a perfectly good word doesn't make it a better one. Thanks for letting me vent!

Posted Wed Jul30, 2008, 3:24 PM — By The Flap

Any information about TMH certification with Tom's Newest venture? AS well as 10.2 surround? Also contrast is an issue of uniformity as well as black levels so a manufacture can make a display that varies with the ANSI method ( I have measured up to 200cd/m2 difference) and gets a good number with the averaging method, but still looks terrible. A better number for the actual contrast would be brightest black to darkest white, as that is what the eye would see.

Posted Wed Jul30, 2008, 8:59 PM — By Scott Wilkinson

Dave, I'm happy to let you vent, but I used the word in a different context than you mention. According to Wikipedia, the definition of "methodology" is

1. the analysis of the principles of methods, rules, and postulates employed by a discipline;

2. the systematic study of methods that are, can be, or have been applied within a discipline; or

3. a particular procedure or set of procedures.

I used the word as in definition 3, not as in definition 2.

Flap, no word on anything audio; this is strictly a video-calibration course. I agree completely that uniformity and black level affects contrast, which is why peak contrast measurements should be taken in the center of the screen, where most of the viewer's attention will be focused. I don't know what you mean by "actual contrast" being "brightest black to darkest white." Can you elaborate?

Posted Wed Jul30, 2008, 10:34 PM — By Dave Anderson

Scott, I'll let this go after this, but let's please not use Wikipedia as our guide to the use of the English language. Try replacing "methodology" with "method" and see if it doesn't read more elegantly, and with the full meaning of your idea. We may just have to agree to disagree.

Posted Thu Jul31, 2008, 2:35 AM — By Scott Wilkinson

Okay, let's take a look at Webster's Dictionary, which, I hope you'll agree, is a definitive source:

Methodology:

1. a body of methods, rules, and postulates employed by a discipline; a particular procedure or set of procedures

2. the analysis of the principles or procedures of inquiry in a particular field

Of course, I was using the word in the sense of the first definition here, not the second. With respect to measuring ANSI contrast, there are various controversial methods used for that purpose, which comprise a methodology. Also, each of these methods is a "particular procedure or set of procedures," and thus can be rightly called a methodology.

(continued)

Posted Thu Jul31, 2008, 2:38 AM — By Scott Wilkinson

Here's Webster's definition of "method":

1: a procedure or process for attaining an object: as a (1): a systematic procedure, technique, or mode of inquiry employed by or proper to a particular discipline or art (2): a systematic plan followed in presenting material for instruction b (1): a way, technique, or process of or for doing something (2): a body of skills or techniques

Clearly, "a procedure or process for attaining an object" and "a particular procedure or set of procedures" are essentially identical, especially if you accept that a "process" can be a "set of procedures." So in this sense, the two words are synonymous.

I maintain that my usage of the word is correct. I chose to use "methodology" because I like the "sound" of the word, and I think it worked better rhythmically in the sentence. Of course, you are certainly entitled to not like my artistic choices, but you haven't convinced me that this p

Posted Thu Jul31, 2008, 2:40 AM — By Scott Wilkinson

...but you haven't convinced me that this particular choice was technically incorrect.

Posted Thu Jul31, 2008, 11:46 AM — By The Flap

Sure can: If you use the ANSI method (I use methodology BTW!) with 16 points of measurement you want to ratio the black measurement with the highest illuminance. I.e. if you get a range of 18-32 cd/m2 for black levels you should use 32 cd/m2 for your calculation. Conversely if you get for white levels a range of 750-850 cd/m2 you should use the 750 cd/m2 for calculation giving us a better measurement. We tricromats are dependant on uniform illumination (not color shifted or gamuted illumination) for our vision to distinguish between colors and contrast these (see Kries Ives) but a non uniform illuminance can shift our color gamut effectively compressing the colors that we can see. Manufacturers know this and make sure to get good levels in the center to counteract this (I don't buy into center levels as the image is usually so much brighter than surroundings that it washes that factor out). Given the relation ship for uniformity relates directly to color I feel this would be more descriptive.

Posted Thu Jul31, 2008, 12:35 PM — By Javan

Dave Anderson will not receive an invitation to my birthday party. Anyhoo... Scott, I just wanted to say that information on film resolution is one of the most useful and interesting pieces of technological trivia I have learned in quite some time. I was always curious how digital cinema projection could look sharper than film even though the resolution of the former is technically lower. Now I'm even more excited for 4K. Speaking of which--do you know if there are any 4K digital video cameras available for use in the film industry?

Posted Thu Jul31, 2008, 3:58 PM — By Dave Anderson

Javan: Can't say I blame you. Who wants a crank at his party? Now if the store will only take back that present I bought...

Posted Thu Jul31, 2008, 9:04 PM — By Paul Matwiy

While I support whole-heartedly the move from film to D-Cinema, resolution is not the first concern. I'd be curious to find out where the THX folks came up with the 2600 x 1400 pixel resolving power of film. It might be that they are concluding that after taking a camera negative, running it through the inter-positive process, to the printing negative and onto the release print, the effect of all these steps to to reduce the effective resolution of the original negative to the 2600 x 1400. Practical film resolution depends on the grain of the film as well as the shooting and projection lenses. The original camera negative should be able to support between 6000 lines of resolution. Kodak R&D shows capture limits of line widths per picture height of 1800 for sharpened film, 1520 for native film, 770 for sharpened HD, and 764 for uns-harpened HD: Reference: http://www.kodak.com/US/en/corp/researchDevelopment/productFeatures/dCinema.shtml

Posted Thu Jul31, 2008, 9:07 PM — By Paul Matwiy

Also to be considered in D-Cinema is the regular size of the pixels as opposed to the variable grain of film, and the problem of interpixel spacing which can be visible to some at large magnifications. D-Cinema has a brilliant future in quality and consistency, but at this moment resolution isn't it's primary strength when compared to film.

Posted Mon Aug 4, 2008, 4:43 PM — By The Flap

Funny Paul but every time I watch a movie on both screens I ALWAYS LIKE THE QUALITY OF THE DIGITIAL SCREEN BETTER. However the right movie should be watched after the film is a little worn out as well (almost like transistor vs. tubes debate). How can you explain my un techy date and My experience??

Add Comment

Name (Required):

Email (Required, will not be shown to public):

Comment (Required, max chars: 1024):

You have characters left.

Type the characters you see in this picture

  

Save a link to this article and return to it at www.savethis.comSave a link to this article and return to it at www.savethis.com  Email a link to this articleEmail a link to this article  Printer-friendly version of this articlePrinter-friendly version of this article  View a list of the most popular articles on our siteView a list of the most popular articles on our site  

Sponsored Technology Center

Stereophile    ::     Home Theater    ::     Ultimate AV    ::     Home Theater Design    ::     Shutterbug    ::     Home Entertainment Show
Home/News • Print & Web Media Kit • Privacy • Terms of Use • Contact UAV
RSS News & Reviews • RSS Blogs

Copyright SOURCE INTERLINK MEDIA All rights reserved.