piano samples vs. real piano

[koolkid]

Originally posted by jmitch:

Okay, I sampled each and every last sample. First of all, the real Steinway is amazing. I was blown away right from the getgo. The samples immediately following were rather unimpressive- far below the musical capabilities of the real thing. However, as I got further down, there were a few that sounded quite similair to original Steinway. I was quite amazed at the realism many achieved.

 

My original perspective on this matter has not changed. Digital recreations will continue to come ever closer the real thing, and take more and more from it. Nevertheless, there will always be a place for real, acoustic, traditional instruments.

the new 4front module is almost incomparable,dont you think??

[koolkid]

Originally posted by sign:

Try the following: put your foot on the sustain pedal of a grand piano and press it down.

 

Tap with a piece of wood on the grand's frame and listen carefully, the ringing of 200+ strings. Nice sound isn't it?

 

Try the same with a digital piano, press the pedal and tap somewhere.

 

Play a chord on the grand and press the sustain pedal, most of the strings will add something special to the overall sound, a sampled piano will not do this.

 

And this is only just one of the differences.

why not?

 

the digital sample contains most or all of the same overtones when it was sampled from a real piano. just because it doesnt ring when you tap it doesnt mean the notes do not have the same tonal qualities

[Duddits]

Originally posted by koolkid:

why not?

 

the digital sample contains most or all of the same overtones when it was sampled from a real piano. just because it doesnt ring when you tap it doesnt mean the notes do not have the same tonal qualities

You misunderstand --

 

Yes, that note has the same tonal qualities as the note on the real piano, but as soon as you add another note or pedal on an accoustic piano, you get all sorts of sympathetic vibrations that create the real piano sound: all the other strings ring to different extents, excited by the vibrations propagating from the note that is struck! And every combination of notes (plus the use of pedal), incurs different sympathetic vibrations.

 

Digital samples cannot vibrate sympathetically -- if you play a C, E, and G, for example, 100 other samples don't ring their upper partials in response as they would in a real piano. The sampled piano only gives you back the C, E, and G. But on a real piano, the sound is

 

C, E, and G

 

PLUS

 

All the sympathetic vibrations that ring from all the other strings in response to the mixture of the C + E + G (plus pedal).

 

On a sampled piano with 88 keys, for example, you have 88 general samples, or in multiple levels of loudness, e.g. 88 X 4 samples.

 

But on a real piano you have 88 keys X 100s of strings X MILLIONS of different sympathetic vibrations that respond to different combinations of notes and pedals.

[fantasticsound]

Originally posted by koolkid:

Originally posted by sign:

Try the following: put your foot on the sustain pedal of a grand piano and press it down.

 

Tap with a piece of wood on the grand's frame and listen carefully, the ringing of 200+ strings. Nice sound isn't it?

 

Try the same with a digital piano, press the pedal and tap somewhere.

 

Play a chord on the grand and press the sustain pedal, most of the strings will add something special to the overall sound, a sampled piano will not do this.

 

And this is only just one of the differences.

why not?

 

the digital sample contains most or all of the same overtones when it was sampled from a real piano. just because it doesnt ring when you tap it doesnt mean the notes do not have the same tonal qualities

Yes, as a matter of fact, it does mean you're not getting the whole, timbral picture. The point is, every note on a sampler is a sample of that note, by itself, with no input from other notes played simultaneously.

 

It's as individual as single lines of guitar played through individual distortions rather than chords with all those same notes played on one guitar through a single distortion. The interaction of notes prior to being fed through distortion yield a very different timbre than indiviudally distorted notes.

 

So when you play any two (or more) notes on a sampler, there are interactions that would occur in a real piano which are absent when two, distinct, non-interacting notes are played simultaneously on a sampler.

[Duddits]

Hendmik : Thanks for the great link comparing a real Steinway B to the fake ones. I find the differences to be as (or even more) dramatic in my own work -- I too use a Steinway B as my main "real piano," and use a variety of fake pianos as well, from Roland to Kurzweil.

 

I think the moral of all this is that there is a big difference between a real and a fake piano, unless you work for a company or reseller that sells fake pianos!

 

[jmitch]

I wouldn't go so far as to commit to one moral yet, and say there is a HUGE difference. Plus, I don't think everyone here agrees with that moral to begin with. I think there are definitely some differences between the two, worthy of noting.

 

My moral is that digital pianos are are amazingly real for being fake to begin with. They are always coming closer and are becoming more and more indistinguishable from the real thing. I don't anyone can disagree with this.

[Duddits]

Originally posted by jmitch:

I wouldn't go so far as to commit to one moral yet, and say there is a HUGE difference. Plus, I don't think everyone here agrees with that moral to begin with. I think there are definitely some differences between the two, worthy of noting.

Just being catty But I do think that anyone who has had significant experience with real and fake pianos knows that there is a HUGE difference between them -- H - YOU - GEE -- EE -- HHHHUUUUUUGGGGGGEEEEE.

 

HUUUUUUUUUUUUGGGGGGGGEEEEEE.

 

HEUUUUUUUJJJJJJJ

 

HHHHHHUUUUUUUGGGGGGGEEEEEEE.

 

huge

 

meow.

[jmitch]

Yes, yes catty. But, that difference is becoming smaller and smaller and will continue to become smaller until there are relatively few and minor differences.

[gtrmac]

It seems to me that if people can hear obvious differences between the real piano and the sample when they are recorded as MP3's there must be a significant difference. I haven't had time to check them out myself but that's the consensus isn't it?

 

I'm glad there are good samples available, I don't see myself getting a real grand piano anytime soon! I would always rather record the Yamaha at the music school studio where I work than use the samples though. I just like moving microphones around though, recording samples is boring.

[Duddits]

Originally posted by jmitch:

that difference is becoming smaller and smaller and will continue to become smaller until there are relatively few and minor differences.

Yes, I agree. In about 100 years, I figure that computers will proabably be fast and cheap enough to model the hugely complex sympathetic vibrations that give real pianos their characteristic "blur." And anyone who wishes to invest between several and twenty thousand dollars for a piano-like action to control the samples (assuming something like that is produced for anything but a real piano) will be able to play the samples like a piano.

 

Until then, I guess we'll have to settle for real pianos and fake pianos coexisting peacefully together, and use each based on availability, logistics and musical appropriateness.

[jmitch]

Originally posted by Duddits:

Until then, I guess we'll have to settle for real pianos and fake pianos coexisting peacefully together, and use each based on availability, logistics and musical appropriateness.

Now that, I can agree with.

[Tedly Nightshade]

Some great perspectives here.

 

I think we are doing everyone a disservice by lauding the latest techie breakthroughs too highly- we must demand far, far better, or we will never get it. Make those turkeys work for our money.

 

Acoustic instruments are here to stay, in a very big way. There is fortunately more to music than recording, and the difference between a live banjo and a sampled banjo, or pianos for that matter, will remain huge- the medium is the message.

[jmitch]

<>

 

Did you not just read the last few posts? It was agreed upon, that this difference would not remain the same. In fact, it is common sense that the difference will not remain the same. I'm sorry but anyone who disagrees with this needs to check out the latest on world technology.

[fantasticsound]

Lest we forget the obvious.

 

When the power goes out musicians like to pull out their instruments and relax with music. Now, to be sure, there are electronic keyboards that use batteries, but nothing on the level of a real piano-action keyboard. Batteries die, too.

 

And listening to unamplified music provides that special feeling that can only be felt when near a piano.

 

So anyone who thinks real pianos are going to disappear, IMO, is fooling themselves.

[Tedly Nightshade]

Originally posted by jmitch:

<>

 

Did you not just read the last few posts? It was agreed upon, that this difference would not remain the same. In fact, it is common sense that the difference will not remain the same. I'm sorry but anyone who disagrees with this needs to check out the latest on world technology.

Please note I am not talking about a recorded piano or banjo, but a live one. In person. Preferably, very preferably, unamplified. Truly acoustic.

 

Common sense has little application to predictions of the future- except perhaps that common sense dictates that if world technology barrels along at the current pace, we'll be back to hollow logs before long.

 

I'll tell you what- we'll set up a quality real piano, unamplified, in a nice theatre, and whatever samples and reproduction system (just as crucial) that the world can come up with, tomorrow or in 20 years, and have an audience of people who actually know what the hell a real piano sounds like, if we can find anybody like that in 20 years... and hear what we hear. Common sense tells us nothing at all about it.

 

To the extent that a virtual experience can replace a real experience, it is because we have forgotten what the real experience is really like, and have lost the ability to appreciate the real thing. For people who have never spent much time in the woods, a convincing virtual woods experience will not be a problem. For the Indian scout who sees volumes of history written where most are blind, in that same woods, it will be a hopeless attempt to replace the incredible multiplicity and complexity of genuine experience with a virtual substitute.

 

Acoustic instruments propagate sound in all directions, different tones in different directions, and all that reacts with the acoustic space in a way that is simply impossible with directional speakers, although attempts have been made by pointing speakers everywhere- speaker covered spheres, etc. But of course unless each speaker propagates a different sound in each direction, and the speakers have far less distortion than 99.999% of *quality* speakers, it's a hopeless exercise.

 

I'll put my favorite hand-hammered and hand lathed Zildjian K (Istanbul made) up against any cymbal simulation, anytime, for as many years as I may be alive- currently there is nothing even beginning to be remotely close. Every time I touch the cymbal it will make a different noise. Percussion is a very difficult field for recording at all, let alone sampling- and piano is a percussion instrument.

[theblue1]

Originally posted by Raymar:

Can't do pitch bends with a real piano. When is somebody going to get around to implementing that feature?

 

steve

Don't pitch bends on digital pianos already sound awful enough?

 

[fantasticsound]

Originally posted by jmitch:

<>

 

Did you not just read the last few posts? It was agreed upon, that this difference would not remain the same. In fact, it is common sense that the difference will not remain the same. I'm sorry but anyone who disagrees with this needs to check out the latest on world technology.

As Ted aptly explained (and I alluded to in my last post), it'll be a cold day in hell when music emanating from speakers sounds like unamplified, acoustic instruments.

 

Bluegrass and folk music lovers understand this principle very well, which is why they attend gatherings across the U.S. dedicated to small groups of musicians sitting in a circle, playing music. No stage. No amplification. Just a bunch of musicians "mixing" themselves by listening to one another.

 

My job, working for Mike Snider, was to mix FOH and monitors at Opryland. But no matter what system they gave me to mix on, no matter how good my skills are, there was no way on earth it was going to sound as good as the 4 or them playing to one another, completely unamplified.

 

Put a keyboard... any keyboard through any sound system next to an acoustic piano, unamplified, and then try to tell me the difference between the two isn't the size of the Grand Canyon.

[Tedly Nightshade]

I often wish I played bluegrass to participate in such things- I dream of acoustic sized venues with nice, well, acoustics, for a great variety of music- certainly jazz would apply.

[jmitch]

Yeah, a keyboard! They have done a lot more than just a keyboard! Maybe, in the late 60s when the first keyboards were introduced. People, this is the digital age. This day and age will be known as the digital revolution.

 

The discussion was based on recording a real piano versus using pre-recorded samples for recordings. Hence, the topic title. It wasn't about comparing a real live piano next to speakers! That's like comparing a photo of something to looking at the real thing. Whenever something is re-produced, which is what recording is, it is obviously going to lose some quality. Getting to that point is much further down the road if we ever achieve "perfect recreation".

[jmitch]

See, acoustic sets and such are nice. But, it's impossible to play for a decent sized audience with just the loudness of acoustic instruments alone. You have to be right near the performers to really get good sound. That's why for many musicians today it's just not practical or else they would be doing 24/7 shows a day just to sell the enough tickets to their concert and not even.

 

Even when artists/bands have a so-called "acoustic set", where the guitarist doesn't have his floor of pedals and racks of effects and amplification, everyone is still miced up and acoustic instruments are still amplified. They refer to it as an acoustic set I guess, because all the instruments are genuinly acoustic (not electric) But, they are still all amplified for ample volume.

[Tedly Nightshade]

Originally posted by jmitch:

The discussion was based on recording a real piano versus using pre-recorded samples for recordings. Hence, the topic title. It wasn't about comparing a real live piano next to speakers! That's like comparing a photo of something to looking at the real thing. Whenever something is re-produced, which is what recording is, it is obviously going to lose some quality. Getting to that point is much further down the road if we ever achieve "perfect recreation".

If you are recording a real piano, you are recording a real piano in a room. That real piano with interact with that real room in a way that no speaker will ever react in a real chamber, and no room simulation will be achieving either. Basically you are putting up a real room against a room simulation.

 

Now, capturing the beauty of a real piano in a real room is a very exacting business. It can be done absolutely divinely. That's not real common, but fussing with knobs instead of placing real mics on real pianos in real rooms doesn't get us any closer to the possible results. Putting in the time and effort and learning to know what to listen for and how it translates will yield far better results.

 

The sampled piano is a shortcut to some kind of result instead of the epic journey it can be to bring back a compelling recording of the real goods. Shortcuts have their uses!

[Tedly Nightshade]

Originally posted by jmitch:

[QB]See, acoustic sets and such are nice. But, it's impossible to play for a decent sized audience with just the loudness of acoustic instruments alone. You have to be right near the performers to really get good sound.

 

[/b] What's a decent sized audience? I saw McCoy Tyner play a 9' Steinway for more than a thousand people completely acoustically. Heard it too- damn it sounded good. I was quite a ways from the stage. I was so grateful no reinforcement speakers were used except for announcements!

 

I've seen various other acoustic instruments play to hundreds at a time. World Saxophone Quartet was pretty damn loud. Big bands can be a spine-tingling thrill for quite a crowd. Ever been to the symphony?

 

Personally we play to a couple/few hundred comfortably with acoustic guitar vox and vibrophone. Yeah you need decent acoustics. Without decent acoustics PA systems sound crummy too, just louder, so you can't get away from acoustics.

 

Concert instruments are a beautiful thing. Our collection includes voice, trumpet, acoustic guitar, vibraphone, concert bass drum & tom & snare, all calfskin, and a whole lotta other percussion. It's not a problem to be heard well and sound great. It's not easy to be real loud, but we get our share of noise complaints. Concert instruments can be a revelation.

 

Indeed it's not practical for many musicians, and practicality seems to beat out sound quality every time these days... not a good thing. Practicality + SPL and nerve damage be damned seems to rule the day in a big way. Repeat, this is not to be confused with sound quality.

 

To my mind it is a total pain in the ass to mic everything and put it through a PA, especially for crowds of a couple/few hundred or less, and we get less a lot of times, when you can just use concert instruments of sufficient projection, and project well as a musician as well. Of course, you may actually have to hold the attention of the audience instead of forcing them to listen through sheer SPL...

 

If the idea is that recording real pianos in real rooms with real mics and real skills is not practical for many musicians these days, I would have to agree. Piano samples may well be more practical. Again, not to be confused with sound quality.

 

And yes, samples may sound better than a poorly mic'ed up piano, and will likely be far easier for mediocre sound systems to handle- real pianos are quite volatile dynamically. And yes it's very difficult to make quality acoustic recordings, much more difficult that pulling up the best available samples and declaring victory.

 

Personally I think you owe it to yourself to find someone who can record that Steinway B really well and luxuriate in the increasingly rare experience of the real thing.

[jmitch]

<>

 

Well, yeah a grand piano with the top open can radiate quite well. I was reffering to a band like group with instruments like guitar, vocals and percussion.

 

<

 

I'm in the symphony, and yes they can be loud. I wasn't talking about a whole symphony of hundreds of instruments. Like I said, a small group/band.

 

<>

 

Yes, a few hundred is small for big name bands. They need to play for thousands to make the kind of money they look for. (not that they need it, but)

 

<>

 

Can you play on stage for over a thousand people in a crowd and still be heard?

 

<>

 

Yes, practicality does come before sound quality for live playing mostly.

 

<>

 

Yes, I agree. Again, practicality over quality for big name bands.

 

<

 

And yes, samples may sound better than a poorly mic'ed up piano, and will likely be far easier for mediocre sound systems to handle- real pianos are quite volatile dynamically. And yes it's very difficult to make quality acoustic recordings, much more difficult that pulling up the best available samples and declaring victory.>>

 

Yes.

 

<>

 

Indeed. This was my plan from the beginning! Thankfully, I have some pretty decent mics, now I just need some decent stands...

[Duddits]

Originally posted by jmitch:

<>

 

Indeed. This was my plan from the beginning!

To help understand where you're coming from -- what exactly are you going to record:

 

What is the condition of the instrument you have access to?

What are the mics and recording chain you plan to use?

What is the style of music, and how prominent is the piano part?

What is the background / expertise of the piano player?

What is the recording for?

What sort of production are you involved in?

 

Just curious in order to get a better sense of the context.

[Tedly Nightshade]

Originally posted by jmitch:

 

<>

 

Can you play on stage for over a thousand people in a crowd and still be heard?

 

Depends what you mean- usually the people who really want to see what's going on come up and listen and those who don't need to hear the details, hang back and catch the beauty wafting on the breeze. Usually we don't play on stage though- I don't care to be up on a pedestal above the audience, with blinding lights in my face- much better to be on a level with the crowd and actually play to people, from our perspective. I think the whole stage thing is most often a power trip so people can feel like a rock star and force people to listen to whatever is left of their music after the PA is done with it- in most cases, damn little is left, but the essential goal is to stand on stage being a big deal, and that is accomplished.

 

A single quality microphone picking up the entire ensemble in acoustic blend is our preferred method when reinforcement is desired. The gain before feedback is enormous compared to multiple mics, and there are no phase issues between mics, so things sound a whole lot better than they will for any multi mic act. Of course, sampled sounds don't involve microphones, but they are incapable of blending with non-sampled sounds, like vocals, in the acoustic realm in anything like the same way.

 

Why the continuing focus on "big name bands"?

 

<>

 

Indeed. This was my plan from the beginning! Thankfully, I have some pretty decent mics, now I just need some decent stands...[/b]

 

I hear of an engineer who says he'd rather have good stands on a gig than good mics- you can always rent good mics, but if you can't get them into the exact right place, you're sunk...

 

good luck, and have a ball!

[Tedly Nightshade]

Just for the record I see a real need for high quality fake pianos for such occasions when a nice well maintained, tuned, and presented piano is not available. I could really use such a thing. Until the soundboard issues and sympathetic string issues are really tackled, we won't be there. Here's hoping we can do it.

 

Nut that I am I imagine sampled strings interacting with a real soundboard. I suppose soundboard modeling is more likely. Or, possibly, vast memory will enable us actually to sample the whole repetoire of chords and dynamics, instead of just single notes...

[jmitch]

<

 

good luck, and have a ball!>>

 

Yeah, haha. That was my thinking too. Invest in some good, quality stands that will last and be able to hold quality mics. If you don't have a good stand, you're in trouble.

 

<

 

What is the condition of the instrument you have access to?

What are the mics and recording chain you plan to use?

What is the style of music, and how prominent is the piano part?

What is the background / expertise of the piano player?

What is the recording for?

What sort of production are you involved in?

 

Just curious in order to get a better sense of the context.>>

 

It's a Steinway Grand, I don't know the exact model, I have to go and look at it again. I think it's a Steinway B. It is in peak condition- practically new, and hasn't been moved everywhere. The mics I intend to use are a pair of Audio Technica AT859 Miniature Unidirectional Elecret Condenser Telescoping Microphones. They sound very flat and neutral and true to the real sound. The frequency response is very good. I plan to go into a Tascam US-428 and record at 24-bit 48khz. The part is a main part in the piece, so good sound is critical. The player is very good. He is classically trained and can play very difficult classical pieces. This is completely my production. I am the producer/engineer. But, we are working together on it.

[Tedly Nightshade]

Quoted from myself from another forum:

 

With a piano you have this huge soundboard and all these strings, like 200 of them plus, radiating sound in all directions, but mostly like a figure 8 centered on the soundboard. The front lobe has string sounds, the back lobe (under the piano) doesn't have direct string sounds, only wood. Different, take your pick which you like today. Or use both.

 

The thing is, you have a little mic diaphragm or two to catch this whole cascade of sound with. The trick is finding little mic-sized places where the entire instrument is in phase and focus and the direct to reflective ratio is just right to where beautiful piano things are happening, and putting the mics there. Then having the two work together so everything's in focus and phase and things laid out nice in the stereo field... It's quite a challenge and asks a lot of the room. Unless the room is simply divine where *exactly* the piano is in the room can make all the difference.

 

If not all registers of the piano are to be used, you can favor the relevant registers for the selection. If the extreme bass register is very important, you will have to make sure to get that fundamental- if you don't have to do that, you will have more flexibility in other registers. If the piece has dense chording in the lower midrange, you will have to be careful to represent that range especially clearly, very possibly at the expense of some other registers.

 

When you place the mic, you have to be very conscious of what all parts of the whole vast piano sound you *need* to capture clearly, and make sure you get them. Sometimes, a more diffuse roomy sound is fine as long as certain key parts have sufficient presence and clarity.

 

The role of acoustics in the whole business is overwhelming. You need some reflected sound in there and some phasy things going on there to get it to sound like pianos sound, which you don't often hear on recordings, but you have to get something tidier than what you get in the room most times, because our ears are so happy in a sea of reflections that would be chaos to a microphone.

 

I'd love to have a Steinway B, and a wonderful room to let it do it's thing in. I have the distinct feeling that you will take on class by pure osmosis just contemplating the sound of such a thing. Trying to do justice to a capture of it's sound is an appropriate form of reverence. It's good for the soul, I'm sure.

[MusicWorkz]

Originally posted by koolkid:can you really hear the difference between the steinway and the 4front module in hendmiks page??

 

and if you do,do you think its of any significance ,,soundwise??

 

honestly??[/QB]

The samples on that page are good, and honestly, yes I can tell the difference. Like every one says, one note at a time makes it hard, but there is something that happens to the sound when you start adding notes and sustains that cannot be captured by a sampler no matter how much they velocity switch samples.

 

Like so many have said, the context is what matters and is what dictates what shouls be used. Understand my paradigm-I produce gospel and jazz with a smattering of R&B. Outside of an occassional R&B tune where I need a piano sound vs. a Rhodes, I use live piano almost exclusively. Consider me a purist, so whenever I hear a sampled piano being used when I KNOW an artist has access to a real one, I consider it lazy an uninspired on the producer's part.

[MusicWorkz]

Originally posted by jmitch:

The discussion was based on recording a real piano versus using pre-recorded samples for recordings. Hence, the topic title. It wasn't about comparing a real live piano next to speakers! That's like comparing a photo of something to looking at the real thing. Whenever something is re-produced, which is what recording is, it is obviously going to lose some quality. Getting to that point is much further down the road if we ever achieve "perfect recreation".

Ummm...how are you going to hear the sampled piano without using SOME form of speaker....so you are infact comparing a real piano next to one being played out of speakers. A sample is just a digital recording anyways, so it will never capture the realism of a live one no matter what you do to it.