Did THX ever have that "discussion" about the EE on Phantom Menace DVD??? (1 Viewer)

[Michael Reuben]

the string of releases with audio sync problems, atrocious edge enhancement, player lockups, etc.

We must live in different HT worlds. Between TV series and movies, I usually load up several DVDs each day, and I can't remember the last time I saw an audio sync problem (if one wasn't there in the source material) or had a disc lock up in my player.

As for EE, it's much less prevalent than it was during the LD era and the early days of DVD. The fact that it's still used is, I suspect, not a failure of QC but a deliberate compromise by transfer houses in the translation of film to what is still a relatively low-resolution medium. Apply some degree of edge enhancement, and people like the participants in this thread will squawk; leave it out, and others will squawk that the transfer looks "soft" or "out of focus" (an example is in the recent thread regarding Fox's Barton Fink disc).

M.

 

[Charles J P]

Who said anything about Moulin Rouge. You make wild assumptions here just like the rest of the "evidence" you present. I actually have a Panasonic AND a Toshiba DVD player and both play Moulin Rouge with no lip sync problems. What about the framing issue with BTTF? What about the lip sync problem with Die Another Day and License to Kill? On Her Majesty's Secret Service had several widespread technical problems too. Since you brought it up, what about Moulin Rouge? These problems are widespread. What about non-technical problems that are not player dependent. Like I said, TPM looks like SHIT and so does Pulp Fiction. There are halos everywhere. You are just arguing your stance with words, The Secrets... and Bjoern backed up their claims with visual evidence. There is a massive thread about the widespread horrible quality control of a major release on this forum every 3 months I would say on average.

Like I said, it appears that quality control is pretty non-existant, so what are they getting paid to do?

 

[dpippel]

Michael - The "softness" or "out of focus" issue with Barton Fink that some are complaining about has nothing to do with lack of edge enhancement.

All we're asking for in a DVD transfer is a faithful reproduction of the theatrical presentation using the best source materials available, with nothing added or subtracted in the video realm. If you want good examples of what a great transfer looks like check out Adaptation for a recently shot film, or Jaws (DTS) for a "vintage" film. Two of the Hayao Miyazaki films recently released by Disney perfectly illustrate the apparently random nature by which the edge enhancement card is played. Spirited Away features a beautiful, filmlike transfer and is truly a pleasure to watch. Kiki's Delivery Service on the other hand is plagued by some pretty heavy EE. It is a huge distraction and whenever I watch this disc I'm constantly pulled out of the film by the problem. Two films by the same director released on DVD by the same studio on the same date. Why is one great and the other mediocre?

Transfers with edge enhancement do not equal sharp and detailed. In fact EE *masks* detail. Transfers without it do NOT equal soft or "out of focus". The simple fact is that at the resolution of DVD there is absolutely no need to "enhance" edges. Studios can put a stop to this easily and cost effectively, and they should start do so.

You know, before television manufacturers got wise, listened to consumers, and offered defeatable SVM circuits on their sets, people resorted to opening their TVs and CLIPPING WIRES to get rid of edge enhancement. Why are we even arguing here, of all places, that software EE is somehow justified? I don't get it.

 

[Charles J P]

As for EE, it's much less prevalent than it was during the LD era and the early days of DVD. The fact that it's still used is, I suspect, not a failure of QC but a deliberate compromise by transfer houses in the translation of film to what is still a relatively low-resolution medium. Apply some degree of edge enhancement, and people like the participants in this thread will squawk; leave it out, and others will squawk that the transfer looks "soft" or "out of focus" (an example is in the recent thread regarding Fox's Barton Fink disc).

And that makes it OK? even if 90% of the DVDs were perfect (they're not BTW) that still leaves 10% that arent. What are they? acceptable losses? I dont see how this is different from the Pan and Scan debate. The studios should do it RIGHT and let the consumer worry about whether or not they want to buy it. Dont throw some piece of monkey shit out there and say well, our market research (conducted entirely of people with 19" Daewoo TV's and an Apex DVD player connected via an RF demodulator) showed that the picture quality is fine.

 

[dpippel]

Not to change the subject Charles, but I love your tagline.

 

[DaViD Boulet]

The "softness" or "out of focus" issue with Barton Fink that some are complaining about has nothing to do with lack of edge enhancement.

Exactly. EE was applied to laserdisc and VHS because those formats lacked the horizontal resolution to produce a sharp picture...the "ringing" from EE helped "fake" the appearance of a sharp picture on NTSC displays (it *still* looked bad on reference gear).

DVD with 720 horizontal pixels/500 lines had enough resolution to be "sharp" on it's own. Any standard source material...even a 16 mm film print...looks detailed and sharp all on it's own to a Walmart shopper without being "tweaked" with EE.

It's a mind-set of what "video production" mean that's been established for a long time and DVD and large-screen dispays are not helping us to change that mindset to what it ought to be.

 

[Michael Reuben]

And that makes it OK?

I never said that either.

M.

 

[dpippel]

I'm sorry Michael, I took this comment to mean you were implying just that.

 

[DaViD Boulet]

Toy Story, Braveheart, and Titanic have no EE and no consumer has ever complained about the image looking "soft".

If stuidos employ minimal filtering prior to compression and use a generous bit-rate with challenging material, there's no problem. And adding EE to a soft transfer just gives you a soft transfer with ringing.

These arguments should have all been laid to rest by now. EE isn't necessary and is a legacy practice from obsolete NTSC formats that had insuficient resolution (VHS and Laserdisc).

Technology and viewing habits have moved on and video mastering needs to move along with it.

-dave

p.s. for those who think otherwise...yes, laserdisc masters typically had copious amounts of EE in them. The reason it's not as obvious is because the LD format can't deliver those high-frequencies upon playback very well, and so it sort of "evens out" a little...the EE adds ringing to the master which then gets "blurred" by the laserdisc. DVD gives you back what that 720 x 480 Digital (compressed) master looked like...and so it shows off the ringing and haloing from EE unforgivingly.

 

[Michael Reuben]

I'm talking about people's perceptions, Doug. I'm not in favor of edge enhancement, and if you read my comments in the Barton Fink thread, you'll see that I commented very favorably on the transfer. To me it's a cheat to use EE to sharpen up discs, because, as you note, it doesn't add sharpness, only the illusion of sharpness (and only for some people).

My point was that I don't think the use of EE represents a failure of QC, as is being claimed here. I suspect it's a deliberate choice being made by someone in the process, either the telecine colorist or someone to whom the colorist must answer. I don't like the choice, and I don't agree with the choice. But I do think that a critical first step in effective lobbying to persuade technical people to choose differently is to purge the discussion of wildly exaggerated rhetoric and claims such as the ones to which I initially responded.

M.

 

[Michael Reuben]

These arguments should have all been laid to rest by now. EE isn't necessary and is a legacy practice from obsolete NTSC formats that had insuficient resolution (VHS and Laserdisc).

Technology and viewing habits have moved on and video mastering needs to move along with it.

It should be clear by now that I agree with these points.

M.

 

[RobertR]

But I do think that a critical first step in effective lobbying to persuade technical people to choose differently is to purge the discussion of wildly exaggerated rhetoric and claims such as the ones to which I initially responded.

But I think it's also necessary to stop with the "problem? What problem? I don't see a problem! Stop getting your panties in a bunch over what I consider to be no big deal" attitude that some people in this thread have exhibited. Can everyone agree that EE is both bad and unnecessary (especially given that I've NEVER seen an instance of someone complaining that a transfer has too LITTLE EE, and that the most critical viewers DON'T want it)? I think David has expressed quite reasonable approaches to alleviating the problem, and it doesn't help to constantly deny that there IS a problem or claim that it's "impossible" to solve.

 

[Michael Reuben]

it doesn't help to constantly deny that there IS a problem or claim that it's "impossible" to solve

Agreed. Wildly exaggerated claims and rhetoric come in many varieties.

M.

 

[Mark_Wilson]

Kiki's Delivery Service on the other hand is plagued by some pretty heavy EE. It is a huge distraction and whenever I watch this disc I'm constantly pulled out of the film by the problem.

It was so bad on KDS that Kiki has three eyebrows in close ups!

 

[DaViD Boulet]

Michael,

wasn't trying to argue! Just pointing out that EE isn't "demanded" by consumers and there are some notable transfers that do it right and no one complains. Yes...different source material will affect the apparent "sharpness" of a DVD image and perhaps make a DVD producer think he/she *should* "boost" the sharpness dial (soft focus lenses etc. like you mention). However...what usually happens when EE is applied to such images is not a "sharp" picture but rather a soft picture with the addition of some odd halo-ringing artifacts around sharp dark/light transitions. In this case, neither group gets what they wanted. Might as well leave the image looking as true-to-film as it can and only boost HF to the degree that it doesn't produce any ringing/halo artifacting.

In any case...the only point is that EE should never be applied to the *degree* that it creates distarcting artifacts from 1.5 screen-widths viewing distance. And like you say we all agree about that!

 

[Geoff_D]

It might be interesting, if one could get the necessary information, to analyze the technical personnel responsible for transfers that result in complaints of EE. I'll bet there are discernible patterns.

Great idea. That would make for very interesting reading indeed. I wonder if people would start boycotting dvds handled by certain parties if their work was continously proven to have been 'compromised'? But as others say, there really are so many elements to take into account when transferring film to video so would definite patterns of behaviour really exist?

 

[RobertR]

But as others say, there really are so many elements to take into account when transferring film to video so would definite patterns of behaviour really exist?

If we can identify the point in the chain where EE is added (I don't think it's a cumulative, "layered" problem), then we can indeed pinpoint patterns.

 

[Cees Alons]

Can everyone agree that EE is both bad and unnecessary...

No we cannot. The filtering techniques that, unfortunately, may also lead to the image fault that some people simply call EE are both necessary and may be very advantageous to the quality of the image as we will perceive it.
What we're talking about here is simple too much, or too visible edge enhancement, not just the mere presence of EE.

Edge enhancement (a very clever form) is part of the normal processing of images in our eyes, on our retina. It isn't something bad or unnatural. It's necessary in any environment where images may be blurred by physical limitations (like for instance bandwidth limitations or sampling errors), which results in errors in the images that have to be countered - by applying EE.

Personally, I think one or more of several things can have gone wrong if we see the EE: the filtering may not have been applied judiciously different on different scenes, or the image may have been judged on a screen that was too small (as suggested above), or some extra EE may be applied by the display devices in some homes (the fact that some people swear they don't see EE on DVDs where others see it clearly may at least suggest this possibility to be present too).

If it was as simple as some people suggested in this thread (and elsewhere), we wouldn't even have to know where EE was added: simply ask the studios not to do it at all again. Ever. Anywhere.
But, as several others have said, and I gladly repeat it, it isn't that simple at all. It's a valuable and necessary tool.

Rather recently, several forms of bad EE ("unsharp mask" which comes originally from a popular technique in analog photography) are clearly visible for the (even moderate) expert on photos in magazines and newspapers and postcards (small textures, like grass or patterns on people's skins being "wrong", "too detailed"). Apparently even some skilled photographers don't mind - or don't see it.

Perhaps it's also in the eyes of the beholder.

Cees

 

[DaViD Boulet]

No we cannot not. The filtering techniques that, unfortunately, may also lead to the image fault that some people simple call EE are both necessary and may be very advantageous to the quality of the image as we will perceive it.
What we're talking about here is simple too much, or too visible edge enhancement, not just the mere presence of EE.

Yes and no.

You're absolutely right that what we're talking about isn't "EE" as a concept...but more "visible *artifacts* from too much applied EE". With you on that one.

However...it's all part of a larger problem where people who master auido/video don't have a concept of fidelity.

A hi-fidelity signal is one that looks/sounds as close to the original as possible.

Within the confines, limits, and mastering difficulites in our 720 x 480 format, this should be our goal.

Right now some techs see natural film grain...and decide that it's "bad" and DNR a signal and strip out all detial. Then they add "EE" to try to sharpen up the image they've just overfiltered.

I know you wouldn't disagree with those things...but it's important that we don't fall into the trap of saying "sometimes EE is necessary to fix a filtered transfer". Perhaps the problem was that the transfer should NOT have been filtered to such a degree in the first place.

As you say, low-moderate levels of EE aren't necessarily a bad thing when used appropriately. It can even improve our perceived sense of detail/clarity when administered with care and precision. So we agree.

but we need to make sure that technicians get the "less is more" concept when it comes to processing film images.. Doesn't mean those things can't ever help out an image that's objectivly poor in quality...but it shouldn't be considered automatic (you'd be surprised at how some technicians really think this way) that the more you turn the dials the better the image (or sound) becomes. That's the key to success whether it's EE, DNR, changing color or removing film-grain.

-dave

 

[dpippel]

It's very evident from the discussion in this thread that we're talking about edge enhancement that causes noticable haloing or ringing, and the resulting degradation of the image in a DVD transfer. It's very easy to spot and should be easily identified and eliminated by a thorough QC process during tha mastering of a DVD. Yes, it should be simple to eliminate. The question remains as to why it's ever introduced int he first place.

While interesting, discussing photographic reproduction in magazines is something else entirely. Besides, commercial photographers have no control at all over how their images are processed during the magazine printing process. There are also technical limitations, depending on the type of printing being done, that limit the quality of reproduced photography. DVDs are an entirely different beast.