Standards for Grading Symphony Orchestra Concerts of Symphonies, Concertos, and other Large-scale Compositions

Executive Summary

This article covers HDVD sound (SQ), video quality (PQ), and video content, with emphasis on the disease DVDitis.

The table below shows typical metrics for the 5 best symphony HDVDs published to date (all from NHK). It sets out benchmarks for diagnosing if an HDVD is infected with DVDitis, and if so, how bad the case is:

Title Pace   Supershots   Conductor Shots
Tchaikovsky Symphony 6    15 seconds per clip   39%   24%
Mahler Symphony 1 11 seconds per clip   38%   17%
Brahms Symphony 2 11 seconds per clip   39%   13%
Bruckner Symphony 9 13.7 seconds per clip   30%   20%
Mito Chamber Orchestra 11 seconds per clip   40+%   18%

Stated briefly, a good HDVD should have a slow pace with more than 10 seconds per video clip on average (longer the better). 20 to 40% (higher is better) of the clips should be large-scale "supershots" (whole-orchestra, part-orchestra, multiple-section, and large-section shots). Conductor shots should be less than 20% (way less really) of the clips in the video. It's easier to avoid DVDitis in music like a concerto when a large number of clips of a star soloist are counted as supershots. To diagnose DVDitis confidently, one must keep detailed records while stepping through each video clip in the performance being judged (a bunch of work). This work is made easier by using a Wonk Worksheet and Wonk Worksheet Instructions.

Sound

HDVD recordings (i.e., high-definition video discs that qualify to be reported on this website) of symphony orchestras should have at least 5.0 surround sound. This rule excludes legacy recordings in stereo (that often have weak PQ also). There may be exceptions for extraordinary titles such as a documentary where stereo (or even monaural) sound does the job and more elaborate sound would not be possible or particularly enjoyable. At the moment, the only exception we have made for a recording of a symphony orchestra without surround sound is the C Major disc of historically important "symphony movies" Karajan made with Clouzot in 1966 (taken from black and white 35 mm film with monaural and stereo sound).

Sound should be recorded with 96kHz/24bit technology supported from A to Z. At the moment, only a few discs of symphony orchestra performances issued by NHK were recorded this carefully. We can hear the difference between those sound recordings with 96kHz/24bit sampling and the rest. Many of our recent titles were probably recorded at 48kHz/16 bit sampling rates (the DVD standard) and some sport the claim of 48kHz/24 bit, which is heading in the right direction. The rest of the titles we encounter probably started with 44.1kHz/16 bit sampling (the CD standard). It goes without saying that starting off right does little good unless the recording is carefully handled throughout the entire manufacturing process.

How does one find out about sound recording specs? Well, it's hard to believe that any publisher that could claim 96kHz/24 bit sampling would fail to mention that on the packaging for the disc. This spec might also be included in the metadata on the disc and some players can display that with their "display info" function. For example, on my Oppo BDP-93 player most if not all my Blu-ray discs show the sampling rate. A few also show the bit rate (see Bruckner Symphony No. 9 with 96/24 specs and Bruckner Symphony No. 4 with 48/24 specs). 

When the spec is less than 96kHz/24 bit or no spec is given we have to use special software (to get metadata from the disc) or trust our own observations whether the sound is good enough to allow the disc to be covered. The objective is to exclude stereo sound presented falsely as surround or poorly recorded sound that has been spruced up electronically to be represented as high-fi. So what should the reviewer consider?

First, read the package information for clues. The older the recording, the greater the probability that the sound is obsolete. Since we have video, we have a chance to see the placement of microphones. If the performance is in a venue well-known to the orchestra performing and you see a lot of mics, the audio folks are probably getting a lot of material to work with. But if it's an unusual venue and you don't see many mics, maybe the odds for a great recording are reduced. If you know about people in the recording industry, you may get clues about quality from reading the audio credits in the keep-case booklet. 

Next, consider the accuracy and clarity of the sound of the individual instruments. Do they sound real? How good do the small sections sound (two oboes, three trombones, four French horns). How about those larger string sections? And how does it sound when sections are playing together in various combinations? When the HDVD video is done right (alas, rarely these days) you should also see as well as hear the sections playing together. This is a help in judging how accurate the recording is. The goal is to hear the timbre of all the individual instruments at the same time you hear the colors created by the composer with his combinations of forces.

If the sound if muffled, stifled, or cramped, something is wrong. Good recording sounds clean, free, and brilliant. If there is a picture of the harp playing, but you don't hear it well, this is a red flag because the distinctive sound of the harp cuts through almost anything. If you see all the strings playing pizzicato, do they sound percussive or do you hear a string of beautiful notes? If the latter, you are listening to a good orchestra and a good recording. Can you clearly hear the concertmaster playing solos? Can you hear the notes of the second violins supporting the notes of the first violins? (Well, don't expect that often.)

Sound output normally should be via LPCM, dts-HD Master Audio, or Dolby True-HD, i.e., one of the lossless output flavors. Its seems that dts-HD Master Audio is being used most of the time now. But there are HDVD titles on the market with lesser output specs, and this could indicate a noticeable loss of quality. 

The final call as to whether the sound is "high-fidelity" is pretty subjective. Many Blu-ray users think the sound they are getting is much better than what they remember from DVDs. This suggests that Blu-ray technology is finally allowing us to take full advantage of DVD-level recording standards. These happy viewers are thrilled about their Blu-ray sound quality even if it is not at the audiophile level. (In fact, many of these music lovers think that the improvement in sound quality from Blu-ray technology is more important to them than the improvement in video quality.)

Does this mean that Blu-ray sound on our fine-arts titles is as good as it should be? Or is it just an indication how rotten the sound was on most DVDs? We at HDVDArts tend to think that the sound we are getting from recently-made recordings is mostly very good. But we also support the efforts of publishers like NHK and AIX Records to give us the very best sound that can be provided today. If there is any entertainment segment that deserves the best sound, it would be classical music HDVDs.

Picture Quality

The starting point for picture quality is "let there be light." To show a symphony orchestra well, you have to have plenty of light. Good lighting is usually present in standard symphony venues. But in outdoor or other unusual situations, the lighting may be bad. If the stage itself is heavily colored, say, with gold and rosewood (like Severance Hall in Cleveland or Davies Hall in San Francisco) the camera crews may already be handicapped. On the other hand, there can be too much light that overwhelms the cameras and makes everything look bleached.

Resolution should be sharp and clear. There should be a minimum of edge artifacts (jaggies), pixel blocking, Moiré patterns, or similar issues. Resolution seems to be closely related to proper color balance. Flesh tones should look real. The picture should not be "rosy" with pink or yellow flesh tones or faces that look pasty or plastic.

Focus should always be perfect. Further, with good light, proper cameras, and wise planning, there should usually be enough depth of focus available to frame every shot with sharp focus throughout the whole picture. We expect field-of-focus tricks at a horror movie, but not in a symphony HDVD where the whole point is to "be there." One exception to this seems to be close-up telephoto shots of soloists in front of an orchestra. In that situation the soloist is, of course, in sharp focus, but members of the orchestra just a few feet away usually will be out-of-focus. We don't know if this caused by equipment limitations or if it is done to highlight the soloist. In any event, we are used to this now.

There should be a minimum of motion artifacts. It's great if the conductor doesn't use a baton. If he does, there may be some motion smear. Timpani mallets also smear. The cameraman can try to reduce this by avoiding a dark background with the mallets. When the conductor comes on stage, the TV director should try for head-on shot that reduces or eliminates panning motion. Cameramen should also avoid panning across the orchestra in any way that will introduce jerkiness or smear.

Generally, we seem to be getting truly impressive and beautiful PQ these days in our recently-made videos of symphony orchestras.

Picture Content and the Disease of "DVDitis"

Now we come to the controversial part of this special article: picture content. Good picture content of a symphony performance in an HDVD will be radically different from good picture content for a DVD! Generally, management and technicians in the industry have yet to recognize this fact. Consequently, most of the HDVDs of symphony concerts that have come on the market from 2007 to 2016 were obsolete when published and are shockingly weak. We sometimes call this weakness the disease of DVDitis. There is some hope for improvement, however, because one publisher is taking the lead in making good symphony HDVDs.

Having made accusations about almost an entire industry, we fulfill our duty of explanation in the rest of this special article.

DVD has low resolution. If you make a DVD shot of an entire symphony orchestra, the camera has to be quite far from the stage. Then the viewer can't see the players in the orchestra well enough for the shot to be pleasant. It's all too much of a fuzzy blur. Thus the low resolution of DVD forces the TV director for DVD to shoot mostly (or entirely) close-up shots of the conductor and individual musicians or small groups of musicians. Then the viewer can at least understand what he's seeing. But the string of close-ups gets boring fast. So the director tries to spice up the shoot with rapid cuts, all manner of unusual angles, and extreme close-ups of musicians and instruments. The result of this is a kind of "road runner race" (named in honor of the famous Road Runner character in movie and TV comic movies) with hundreds of short clips (612+ in one HDVD recording of the Beethoven Symphony No. 3) incessantly leading the viewer around tiny parts of the orchestra until the viewer finally gets relief at the end of the piece. Because of this, DVDs of symphony recording are inherently bad and unsatisfying. This approach to recording a symphony has little in common with the normal experience of going to a symphony. In the symphony hall, the viewer sees everything in one clip and he is free to enjoy whatever portions of the orchestra he wishes to focus on. He's not being lead around by the nose.

Fortunately,  HDVD opens up huge and better possibilities for displaying symphonies on video. With the higher resolution of HDVD,  the TV director can shoot the whole orchestra in a way that is pleasant to watch. If the TV director shoots the whole orchestra in a picture framed to take up the entire screen, the viewer can recognize the individual players quite well. (Yes, 4K resolution would be better, but 1080 gets us over the threshold we need for a successful shot of the whole orchestra.)

Now the TV director has the ability to follow "Huang's Law" for the good HDVD shoot of a symphony: "Use the flexible power of the the high-def camera to get a pleasant (not hyperactive) mixture of shots of the whole orchestra, groups of sections, large sections, small sections, small groups, and solo players—depending on what forces the composer commits at various places in the score." (Huang's Law is a precept of wonk William Alexander Huang.)

When the TV director follows Huang's law, the director no longer abuses the viewer with an exhausting road runner race. Now the viewer gets an experience he can relate to because it's similar to going to the music hall to see an orchestra play. The viewer also gets some additional help from the TV director, who (knowing the score almost as well as the conductor) will gently guide the viewer into seeing what the composer does with the orchestra.

It follows then that a DVD shoot cannot be used for an HDVD:  picture content for an HDVD must be planned for HDVD only. A corollary of this is a good HDVD cannot be displayed in DVD because much of the video will be too fuzzy for the DVD viewer to tolerate. Let's twist the prism slightly and say it again this way: You can't shoot a video of a symphony that you can publish as both a DVD and an HDVD. That is, of course, exactly what the industry is doing, but it will not work. If you try it, you will cheat one of your customers. And since the industry continues to shoot DVDs of symphony events and to publish them also as HDVDs, the customer getting cheated is the Blu-ray buyer.

If both a DVD and an HDVD is to be published for the same symphonic event, the HDVD should be shot separately at another performance from the DVD. Or perhaps a clever director with enough gear could use one set of cameras to shoot simultaneously for HDVD and DVD. But we think little like this is happening. We think all the symphony recordings being published at the same time in DVD and HDVD are shot as DVDs. The Blu-ray is just an extra source of profit for the same material. The publisher can truthfully say that the Blu-ray is "better" than the DVD. But this isn't saying much since DVDs of symphonies are bad to start with. And we are not interested is a better DVD. We want and deserve it all,  including video content that takes maximum advantage of what HDVD can do!

We have explained  the problem the industry has with Blu-ray symphonic discs now being published. The next step is to discuss the characteristics of good HDVD video content for a symphonic recording:

  • The TV director should follow the composer's score as his script. Of course, the TV director has to know where the instruments will be located in the orchestra for the event to be recorded. Then the TV director can plan, bar by bar, how to provide the most information possible in each video segment.

  • When more than a few instruments are playing, the emphasis should be on large-scale shots of sections and groups of sections or the whole orchestra. With DVD, you can't do this. With HDVD, you can; and the difference makes for a huge increase in the value of the video to the viewer.  

  • Once a shot is framed, emphasis should be on holding the camera still long enough to give the viewer time to savor the shot. This is the opposite of the hyper-active succession of close-ups that is typical for DVDs. Also, the TV director should not interfere with viewer enjoyment via excessive zooming in and out, panning within a section, change of focus tricks, and the like. 

  • As early in the program as possible, the TV director should try to show the whole orchestra and allow the viewer to see where each section is. 

    • The TV director should show each string section separately. Then the viewer can usually locate the wood winds, the tuba, the trombones, trumpets, harps, and percussion with little difficulty.

    • The TV directors must plan to show all the first violins as a group, etc. We don't want to see just part of a section if we can see it all. Showing three violas out of a section of 8 violas is a DVD trick: this is error in HDVD unless there is a good reason to show just 3 players.

    • But where are those horns? The TV director should always show the whole horn section at the earliest opportunity so the viewer can see how they are organized. Horn players divide up the score in special ways. The camera should allow the viewer to see how the horns team up. Then when only two adjacent horns play, bore in on the duet. When the whole section plays, pull back and show that fact.

    • Sometimes the cellos are spread around the orchestra in odd ways. Then it takes good planning to frame shots that will get all the cellos.

  • After the TV director plans  to show each entire section, he must next plan to show multiple sections playing together. This involves "part orchestra" shots that show for example, the cellos, violas, and basses playing together. This obviously can get tricky depending on how the conductor has located his sections.

  • Finally, the TV director must be ready to pull back and show the whole band playing in tutti. This can  look and sound wondrously impressive in HDVD whereas you can't show this well in DVD.

  • The conductor is an important part of any symphony concert, and the viewer wants to see how the conductor operates. In many symphony DVDs, the conductor is shown innumerable time as the "hub" of the performance while soloists or small groups show up as "spokes." HDVD makes "hub and spoke" obsolete. The good TV director for HDVD will use a more balanced approach that shows the conductor a reasonable number of times, especially in situations where the conductor manages dramatic changes in the music. Some conductors are inherently interesting to watch. Others get great results even though they are boring to see. Ultimately, the average viewer doesn't glean too much from seeing a conductor, so shots of the conductor should not be overdone. Of course, some HDVD viewers might be young aspiring conductors.  The TV director can serve them by providing an alternate angle that shows the conductor only. Just set one camera to capture the conductor throughout the whole program (to date this has been done maybe once or twice).

  • If we don't want to see the conductor all that much, we for sure don't want to see the backs of the musicians! Shots of musicians' backs are anathema! Why would anybody make a photograph of the backs of 25 to 75 musicians, many of whom are world-famous? Well, the answer is that most of the "back shots" we are seeing in HDVDs are really just conductor shots from a distance. All you see is the conductor, and he's a tiny figure seen across a large room over the backs of scores of other people. How dumb can a shot get? Well, chucking out these low-value shots of conductors will also get rid of a lot of shots of backs of musicians.

  • Most of us want primarily to see the musicians playing their instruments. We don't want to see just their faces. We don't want to see just a violin or a tuba. We want to see men and women doing what they have spent their whole lives mastering—playing their instruments. Musicians tend to look ordinary. But when they play their instruments we can see them for what they are: angels from God singing to us.

    •  So these angels are the stars of every scene, as soloist, in duets and trios, in sections, groups of sections, and in tutti.

    • We do not want to see "instrument only" shots. (These are usually just desperate attempts by a DVD director to come up with something different for the road runner race.)

  • The TV director must get the right elevations for his cameras to be able to frame large and small parts of the orchestra. Tall cameras and balcony positions are critical, especially if the concert hall does not provide generous risers to elevate each rank of musicians over the rank in front.

  • The cameramen should be given time to frame shots artistically and get the focus and field of focus right. Cameramen have to be trained to avoid all-too-common errors like those out-of-focus bassoon snouts sticking up in the middle of the picture.

  • One camera should always be able to show the concertmaster from his front (from a vantage point in the middle or rear of the orchestra). Avoid framing a view of his head from 5:00 o'clock even if you can see the violin. When cameramen does this close-up, the viewer will often not know who is playing unless the viewer is very familiar with the concertmaster.

  • Although many of the rules suggested above are designed to slow things down in an HDVD shooting plan, the TV director must also avoid being too static. For example, it would not do to just rotate through, say, a group of nice whole-orchestra and 80% part-orchestra shots for an entire symphony piece (even if that might be better than some of the hyper DVDs we see). Such a simple plan would run the risk of seeming clinical or sterile. Once the TV director gets the viewer oriented by showing whole-orchestra and part-orchestra shots, Huang's Law allows plenty of room for the TV director to be creative.

  • All these rules are made to be broken. If the bass fiddles are playing a creepy sound, then maybe it's neat to show a huge close-up of a bow slowly dragging across a string. If the composer has the melody flying like crazy around the orchestra, maybe you will need a few hyperactive short shots of solo players. Sometimes a shot from the back of the orchestra can be nice because it can show the magnificence of the concert hall or give the viewer a perspective that he will never get from his $200 seat.

Recent Developments

In 2015-2017  the industry started publishing fewer symphony titles. Perhaps sales have fallen short. Perhaps the publishers are holding up wondering what to do about 4K video. Still, the Berliner Philharmoniker has continued to publish some significant titles in Blu-ray video only. If the Berliner Philharmoniker has quit publishing DVDs, then have they started to make Blu-rays that take advantage of HD TV? No. Their recent titles continue to suffer from DVDitis. Why. Maybe when the titles were shot they planned DVDs and then later changed course. Or perhaps it's just a case of old habits dying slowly.

Science Supporting Objections to DVDitis

It was only in the 1950s that science established that humans have quite limited working memory capacity. In computer terms, our brains have a lot of long-term storage but just a tiny bit of RAM. Cognitive load is the term for the mental effort used by short-term memory to do something.

  • Intrinsic cognitive load is the difficulty of the task at hand. Listening to a Beethoven symphony puts a  heavier intrinsic cognitive load on the brain than listening to Für Elise

  • Extraneous cognitive load is the manner in which information is presented to the listener/viewer. In musical terms, the extraneous cognitive load from watching a DVD of a symphony is high because the video is presented (1) with no initial orientation, (2) at a fast pace, (3) in small chunks of material focusing on tiny fragments of the orchestra, and with (4) a multitude of low-value views of the conductor and instruments only. The listener/viewer expends much of his mental energy just figuring out what is happening. But with a well-done HDVD, the extraneous cognitive load is small because (1) there will be a good initial orientation, (2) the pace will be slow enough for the listener/viewer to relax, (3) each chunk will provide more information for the listener/viewer to savor, and (4) there will be fewer low-value shots to bog the process down.

  • Germane cognitive load is the manner in which the brain's working memory adds something from current experience to one's permanent body of knowledge.  At the end of a DVD, the listener/viewer may well be exhausted and irritated. He may not remember much about the music itself. At the end of an HDVD, the listener/viewer may feel exhilarated by the experience and have enough cognitive energy left over to store away a new and even precious memory.

Science has even established a relatively simple way to measure cognitive load in some circumstances. When load is high, the pupils dilate. When load is small, the pupils contract. Has anyone measured cognitive load in listener/viewers of DVDs and HDVDs of the same music? We don't think so. It would be grand for someone to try this. If it could be established this way that DVDs impose a greater cognitive load on the listener/viewer, that should be the end of DVDitis in HDVD recordings.

Summary

Because sound quality is so important in the recording of a symphony, we normally will give an A+ grade only if the symphony music (1) was recorded with 96kHz/24 bit sound sampling, (2) the recording is produced at the audiophile level throughout and, (3) results in lossless output to the speakers.  Further, we normally will give an A+ only the HDVD has video content that takes advantage of the strengths of HDVD. We will give significant grade reductions to discs that have DVD video content republished in Blu-ray form as an additional profit center. There will doubtless be exceptions. For example, some symphony recordings are made under circumstances where ideal conditions are not present and the grade can be A+ if the producer does the best he can. An example of this would be the Britten War Requiem recording made at a special, one-time performance in the Coventry Cathedral---this was as much the recording of a news event as of a concert.

Right now we know of several symphony HDVDs, all from NHK, that meet the standards detailed above for SQ, PQ, and video content. All of these titles get the grade of A+. NHK is the only publisher of symphonic music that has seriously tried to take full advantage of the technology that we have in our high-definition home theaters. NHK has set the standard, and everybody else in Blu-ray symphonic music is mostly bringing out obsolete material. Eventually, the demand for HD classical videos will clearly exceed the demand for DVDs and the industry will turn its attention to making the top-notch HDVDs we deserve.

May 22, 2016.