|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
69.105.45.111
The OP discussion and the Van Cliburn thread below made me wonder about this topic. There are many recordings where the other instruments in the band/orchestra sound perfectly ok, but the piano sounds like it's buried under blankets and put into a cardboard box. Or the piano sounds distorted on the overtones. Or there is no real depth and weight. I have many jazz recordings on CD in the RvG series and it seems the piano is always worse of than the rest. Many SA-CD's also display similar traits. Same on vinyl. The tsax sounds great, but the piano flat and unnatural. There are exceptions (Evans on APO, Lomheim on Artegra, Lazic on Channel Classics for example), so it can be done. I would be interested to know why this is? I can speculate but I am not a recording professional.
Follow Ups:
Joel, I'd like to share exceprts of some observations I posted during the early days of SACD, when its special ability with reproducing piano first struck all of us:From 2001: "The Horowitz SACD with the Chopin Sonata, Rachaninov Preludes, Liszt Rhapsody etc. is excellent, in my opinion.
There is very little residual hiss from the tape. The transfer from analog tape gloriously retains a great deal of the sonic complexity and grandeur of Horowitz's grand playing style.In piano music, you don't just hear individual notes. You hear the myriad harmonics of the resonating strings interact and modulate against each other, producing a magical sonic cushion of sound. A good pianist does not just press keys that produce notes; he/she becomes a master sculptor moulding rich tapestries of sound by controlling the interplay of the resonating strings using the pedals and holding down the keys.
Unfortunately, recording piano music in PCM 44Khz 16bit does not capture that rich tapestry of overtones. Often, you would only hear the fundamental (basic)notes being produced, and miss out on the profound and transcendant drama the pianist is enacting in the realm of the overtones. Piano sound on CD sounds hard and lifeless. Loud climaxes sound constricted, while soft subtle passges don't quite "whisper" as they should. This is something about CD people don't realise until they hear an SACD like the superb Horowitz re-master.
In summary, if you like romantic piano music and Horowitz's style, this SACD captures the grandeur and sweep of Horowitz playing in his prime. His interpretation has amazing depth and breadth, penetrating insight and transcendant vision."
From 2005 "Vinylphile, I also find that piano benefits particularly when reproduced through SACD.
As I posted many years ago, piano timbre is very complex, resulting from the intermodulating overtones of hundreds of resonating strings upon a huge wooden soundboard, especially when the sustain pedal is depressed. The gradations of force with which the hammers hit the strings is also a very delicate range of sounds that need fine dynamic resolution to reproduce.When CD tries to capture this in PCM, it sounds like a Clavinova or sampled piano; banal, dry, thin, small and simply simpified. With DSD, you hear the glorious complex interactions between the vibrating strings, which is as rich as an entire orchestra! You get that infinite range of expressive gradations with the hammers, showing the subtle mastery of the pianist."
SACD has an even better chance of capturing piano than vinyl, in my experience. Because the sustained pitches produced by pianos are so steady and unwavering, more so than other stringed or wind instruments, the subtlest pitch fluctuation caused by wow and flutter is immediately and painfully perceivable. In this aspect, even the best vinyl rigs are at a disadvantage compared to SACD.Also, the dynamics range of a live piano is unusually wide, with very powerful peaks. With vinyl, this leads to compression and distortion during climaxes. Each onset of a note, even softly played, is very often accompanied by a slight "fuzz". The problem gets worse as we progress to the last track of the vinyl disc.
Finally, during soft piano passages, the slightest vinyl pop or crackle becomes very obvious. You really need silence to appreciate the delicacy of these quiet passages, although a soft continuous background tape hiss is not distracting though.
I heard that the Glen Gould Mono recordings used MANY mics, positioned in and around the piano. No wonder his warblings are so distinct, I suppose. I don't remember the number of mics used, but it may have been 15 or 20!
in my experience is playing them back through speakers so that they sound realistic. To be sure, as many have pointed out, a piano produces an extremely wide range of dynamics and a broad frequency response, perhaps broader than any other instrument except for the organ. Excellent transient response is required to reveal it is a percusion instrument. So there is a real challenge for microphones and their placement and down-stream electronics used in the recording process.However, all of the above is well within modern technology. But the challenge presented to speakers in realistically reproducing piano music is enormous. As with all instruments, the fundamental of a note does not not define the character of the instrument very well; it is the harmonics. Getting those harmonics correct in realtive loudness is crucial if an instrument is to sound real or at least convincingly like the real thing. A glance at the frequency rersponse curves of most speakers shows that they look something like the outline of a mountain range. Where various harmonics fall in that succession of ups and downs can significantly alter the character of instrument's sound as reproduced by a speaker. Pianos, with their incredibly rich and complex harmonic pattern for each string interacting with the rest of the instruement, can challenge the speaker as no other instrument can do so fully.
In my experience, speakers with extremely smooth frequency response do much better in reproducing piano. They all need broad, flat frequency response to handle transients (high frequency response) and the low bass that can give a piano such authority. So, in my opinion, reproducing a piano well can be more of a challenge for the speakers than actually making the recording itself. Robertc88 expresses similar views elsewhere in this thread.
I always include piano recordings when I audition speakers because of what I wrote above. I also include other solo instruments with complex overtone structures, such as the cello. The advantage of the cello is that it easier to record and I have an excellent ides what it should sound like, but it doesn't test the full frequency and dynamic range that the piano does.
Finally, a query: Could people list some of their recordings that they feel reproduce the piano the best?
Though it might not still be available, the Kissin peformance of Pictures at an Exhibition on RCA CD sounds very good to me, and I like the performance, too. There is some other good music on this disc. Though it is a pianoforte, I like the sound on the recent Brautigan SACDs of him playing Beethoven sonatas. Several of the London vinyl recordings of Aschkenazy playing Beethoven sonatas have very good sound, e.g., his performance of the Hammerklavier.
Both are Keith Jarrett recordings (not SACD) and they are excellent.
Apropos your last request (good piano recordings), here's a stunning newcomer:Arkadi Volodos "Plays Liszt" (SACD from Sony Germany, type B000MGB0JW in Amazon). Super processed but possibly the best reproducton of the piano sound I've ever heard. The dynamics (which with this player are awesome) don't feel compromised at all and the tone & stage are superb.
In jazz there are very successful piano recordings by Winter & Winter, like for example the Tethered Moon CD's, all of them (Masabumi Kikuchi with Motian and Peacock). They like an opposite approach and until quite recently only made unprocessed analogue recordings.
I've found that it's quite hard to mix a piano with other instruments in a multi-track scenario.If you surf the pro audio forums, it's a typical "newbie" question ("Help! My piano is drowned in the mix. What should I do?")
Often the advice is - compress or EQ. But this tends to make the piano sound "unnatural" (ever encountered a "bright" piano sound in a recording? :-> )
I think part of the reason is that a piano has a wider "dynamic range" (level of ppp vs fff) than some other instruments.
In a live situation, most pianists would adjust their playing to blend in with the other instruments. The problem in a multi-track arises when the pianist records first and other instruments follow (which is typical).
for Norweigan Wood?
A nightmare.
nt
The dynamic range
One label that came pretty close to perfecting piano sound in their digital CD's was Dorian, unfortunately they are out of business.
If you can ever find a recording of Schumann piano works played by A. Kubalek try it, to me this is benchmark for how piano should be recorded, digital is not to be blamed, it can be done and this CD demonstrates it.
Arguably pianos are the hardest, but I think the one we most always recognize as "recorded" is the human voice.Pianos are this hellish combination of primary percussion (hammer hitting string), with brutal transients and dynamics, and all of this harmonic interplay across all of those octaves. Then there are all of these pieces and parts vibrating their little hearts out.
Analog, whatever your feelings about the attractiveness of its euphonies, simply lacks the huevos. It never gets the leading edge right and 60db of dynamic range is simply inadequate to the task.
Clearly superior to the best analog, 44.1k and 96k PCM digital, DSD - Direct Stream Digital (SACD) offers a flat frequency response from essentially DC to 100,000k, and a usable dynamic range of over 110db (compared to about 60db for the best analog). If you want to know what that means, ask a recording engineer about compression of solo piano recordings.
Using a short path (no more than 50' of microphone cable to preamplifiers, no mixing board, and appropriate microphony (such as the Sennheiser MKH-800) ... feeding a DSD recorder, the results of the actual recording and live microphone feed are indistinguishable.
I have been there, and heard that, in analog and PCM, and many times with DSD. Talk to someone who has actually heard the microphone feed/master recording relationship (as opposed to someone who simply has an opinion, uninformed by empirical fact) and they will tell you the same thing; DSD done correctly is indistinguishable from the microphone feed.
And yet, here we are, nearly a decade later, nattering away about recording distribution formats, I-pods and other such complete nonsense, and perhaps pissing away the best chance we have ever had for being able to record pianos (and everything else) properly.
And don’t even bother to try to bait me in to a “mine is bigger than yours,” flame war. When you have actually had the experience, you will have a reason to speak.
The Good (if frustrated) Doctor
this presumes that the "indistinguishable" DSD copy of the microphone feed is completely indistinguishable from the sound of the actual piano being recorded. I would maintain that most listeners could easily identify nearly all recordings of a piano as such when compared to the sound of the instrument itself. Charles Hansen's post below pointed this out first, but to elaborate, the majority of piano recordings suffer for two reasons: the microphones used simply are inadequate, and they are placed poorly.The best attempts at portraying the sound of the piano — in my experience anyway — are on a few of the Opus 3 recordings, and the Harry James releases for Sheffield (these are all recorded in analog, BTW). IMHO, these succeed because of the minimalist placement techniques used. "Talk of the Town" on the Opus 3 Showcase SACD may, in fact, be the single best recording of a jazz piano.
Of course, on the other end of the spectrum are the mostly terrible sounding Concord Jazz recordings that unabashedly spread the piano across the entire soundfield.
The discussion point was "recordings," not reality. The perception of a thing, and the thing itself are always, irrevocably, different.A recording of an event is not the event itself, but another catagory of experience. The "absolute reference" is a catagorical emphemera. A representation of a thing, cannnot the thing itself. The map is not the territory, nor is the play, or the movie, or the photograph. A painting of something is not a lesser catagory of reality, it is simply different from the original event.
The only useful reference for evaluating a recording is the variation from what is recorded, and what was in the microphone feed.
My illustration with the human voice was just that, an illustration of how hard it is to "fool" human perceptions with recordings of human voices.
You may LIKE certain recordings, or families of recordings, but the simple physical fact is that both analog and lesser digital lack both the frequency response and dynamic range to capture even what the microphones can hear.
Try this ... compare two excellent piano recordings, Malcolm Frager's Chopin and the SACD of Manfredo Fest playing Jobim on a first rate SACD player and system. I truly believe you will hear what I am describing.
In the case of the PCM Chopin recording, you will literally hear them "stuffing" that huge Bosendorf into the recording, like a shoe horn and fat feet into too small shoes. The Fest recording, sparkles with this ease and breath that effects you physically. You can relax, sit back and not clench or grit your teeth, subconciously.
.
I thing overall the bigger difference though is when I switched over to the Monitor Audio GS10s from the B&W 603S2s. The GS10s aren't speakers that are real costly either.The ECM label seems to do a very good job with reproduction of the piano for jazz. And overall on classical music on other labels, its much more pleasing to listen to Chopin and also piano concertos from other composers.
This doesn't address the difficulty in recording the piano but I guess I shouldn't be all that surprised that the speakers and CDP utilized in my case made a whole heck of a difference.
Besides the excellent points raised by the other posters, don't forget that most microphones suck. Even the really good ones are barely acceptable. This is also one reason why it is hard to get a good recording of massed strings.
When our band was playing regularly, - our singer brought along 4 mics to every show. This is just one reason for the many variables in the recording process. I remember the bass sounding odd without even using a mic and going straight from the amp to the mixing board, via a direct box. It was probably more "accurate," - but sounded so "different," - that we always chose to mic the amp's speaker.
The problem of leisure, what to do for pleasure. Ideal love a new purchase, a market of the senses. Dream of the perfect life.
I have always had. I have never minded piano recordings but I have always felt that massed strings are the ultimate torture test for any system. The first thing that sold me on SACD was how good violins sounded and one of the things I was most struck by on Kimber's IsoMike recordings was the improved resolution of the string section.
A piano is probably the most difficult instrument to reproduce in a recording. The sound is extremely complex, and its tonal quality relies on many factors that aren't yet fully understood. There are many aspects to the tone that we haven't yet figured out to measure. If you can't measure it, it's hard to develop a methodology to record it.I have heard VERY few digital recordings that come close to accurately reproducing the timbre of a real piano. My opinion is that the few recordings that do come close got there pretty much by trial-and-error or accident.
....In general the piano is tough because it has one of the widest tonal ranges of any instrument and is highly percussive in nature in addition to its sonority. Nonetheless, most classical recordings get the piano approximately correct. Those that don't do so for several basic reasons...most often, the piano is recorded in a bad sounding acoustic environment. For classical music, it is almost essential that some "room sound" be included, and if that "room sound" is bad, the piano sounds bad. And sometimes (most notably at DG) the engineers use a misguided attempt at multimiking rather than the tried an true "purist" approaches used for decades (XY, ORTF, spaced omnis)....When it comes to jazz and pop, the problems are usually different. The piano is treated like any other instrument, and the engineer wants to record it on a separate "track" or "tracks". In the case of jazz this is usually in conjunction with some overall ensemble pick up wich provides a stereo spread and some "room sound". In the case of pop, the engineer usually wants no "room sound" at all and only wants a direct, non-ambient feed of the piano into the mixing console for further manipulation (eq, compression, reverb, etc) before feeding the modified piano sound into a "mix" with other instruments, the whole than often treated to more manipulation when mastered.
So, with the type of recording, there is plenty of room for bad sound to creep in, starting with mic placement. Most jazz-pop engineers want to put mics in or attached onto the piano to get a "bright, articulate" sound. This by itself can lead to overload if they are not careful, and can easily lead to a disjointed sound with no imaging, with disproportionately bright high end and weak and sometimes muddy bass. Then, of course, to the degree that the resultant sound is subject to all kinds of post-tracking manipulation, the chance for poor human judgement is rife. And then when the final mix is "mastered", increasingly the sound is manipulated still further with even more chance for the piano sound to be botched up. Because most other instruments range is not as wide, it is often easier to overlook the effects of this manipulation on their sound...but not so the piano.
Hope this helps a bit. Frankly, for the person interested in great piano sound, they could spend half a lifetime just perfecting their art, and some classical and jazz engineers have done just that. Would they were the rule, not the exception.
Harry
This post is made possible by the generous support of people like you and our sponsors: