sustain pedal - why that short? (Page 1 of 2)

You can also increase Direct sound duration, and play with impendance, too.

This is very well noticed. When using Pianoteq in a mix the brevity of the sustain can be very useful, but in terms of academic accuracy you are right; the lowest octaves especially can sustain for a great length of time in a real piano with the sustain predal pressed in.

I would imagine this is something Modarrt will address in the name of realism.

I agree that the sustain is longer on a real piano. In fact, in an earlier version, we had it longer but some users complained about the sound loosing clarity when playing with the sustain pedal down, so we came to the current setting. As most of the digital pianos have no effective sustain, people get used playing with the sustain pedal indefinitely down without any problem, which would of course sound horrible on a real piano.

Increasing the impedance makes the sound longer also at high frequencies, which might be what you noticed. So at the same time you increase it, I suggest that you decrease the soundboard cutoff and/or increase the Q factor. That way, the high frequencies will be less prominent and the low frequencies will be longer.

Another solution which can work consists in increasing the global resonance. That will not change the decay rate, but the harp resonance will have more presence, like in an acoustic piano.

How about trying what Julien said there first? There is more than one parameter that governs sound sustain, and you should tweak them all to get closer to the sound you want. Searching for instant gratification is not really what Pianoteq is about, it is about getting in there, in the sound making, getting hands dirty.

Shortening the sustain of the model because some users don´t know how to use the pedal properly sounds stupid to me. Being a professional piano teacher I would advise those people to learn to play the piano properly. If you ask me, I prefer a Pianoteq with the realistic sustain length of the best quality biggest size concert grand piano. Why should we be happy with less ?

Similar to my previous postings in the "silent sound̈́" thread, I´m surprised, that the Pianoteq people do not choose for maximum realism when designing their instrument models ?

--

Martin

Would the "size" parameter (string length) be relevant to the sound duration too? (I know it's mainly concerned with linearity of harmonics)

I speak only for myself - the sound from my fairly high end Roland isn't as good as Pianoteq.  And the Roland cost a lot more.

http://www.rolandus.com/products/produc...jectId=434

There are many, many good reasons - you will find lots of enthusiastic comments here and posts about what users like with Pianoteq. There are also some skeptical posts and questions and feature requests, and this is perfectly normal as everybody has their own perception and point of view. There is always some detail that doesn't seem quite right to someone. If there are more users disappointed that the decay duration of Pianoteq's notes is about 20s instead of 36s and send feature requests because of this, I am perfectly sure the developers will not hesitate to adjust it.
Maybe users should have the choice whether they want it more realistic or more convenient.
To me it seems quite funny that adults are sitting besides a piano and count the seconds until the sound vanishes (which it never does on real pianos, since their oscillations decrase asymptotically...)
Sorry, this was off the mark.

If there are CPU-related reasons for limiting sustain, I can accept that, but a very capable i7 Quadcore machine can be built for the price of a midrange laptop these days.  Maybe the full capabilities of the engine could be restored with caveats in the popup help warning about CPU usage.
 
I would have thought that the best option would be to model a piano to be as close as reality as possible, and then if the sustain is deemed too long by some users, they can modify the other parameters to make it how they want, i.e. *they* deal with the interaction of things like impedance, Q and cut-off if they want a less-than-real piano, *not* the users who want a genuine grand piano sustain and sound.

Best//Neil

Seems you aren't so we prepared a variant called C3ls which is based on the C3 but has longer notes, particularly in the bass, and a longer harp resonance. The ptq file C3ls.ptq can be downloaded from the user area. Once you have downloaded it, drop it on the Pianoteq interface and the file will find its way to the appropriate folder. We hope that you will enjoy it!

Very nice!!!  I for one like this instrument, the difference to the regular C3 is subtle, but definitely noticeable.  I like the increased richness of the sound.  And it makes good pedaling a bit more critical!

Just wondering: what else did you change to the model besides longer bass notes and harp resonance.  I assume you did more than change a few settings, since you made a whole new instrument out of it.

Cheers for your responsiveness!

Modartt's "agile development" and customer service triumphs again!

Hi guys (and gals),

I very seldomly speak up in these forums unless I feel it is necessary and this one does indeed deserve full attention.

As a few of you know, I have been a concert pianist since 8 1/2 years of age.  I have played pianos from all over the world in just about every type of climate, social surrounding, bar, saloon, concert hall (big and small), outdoor stadium and school/University.  This is not for bragging rights, this is because this sustain issue is a big deal.  Very big deal and my experience allows me to speak up to this.

From my experience (I own a Petrof Grand, a Yamaha upright, A Kohler upright and used to own a Steinway B) Pianoteq was "AT FIRST" a revolation to me.  Going from sampler based engines to Pianoteq finally got me the "Playability" from a velocity control, dynamic range that I needed to truly utilize my recording studio and make beautiful recordings.  Why don't I just use my grands etc?  Because it is pure hell to align microphones, mix pre-amps, reduce surrounding noises, on and on and on (retakes from screwing up one note or the phone ringing).

The problem was, it always sounded "dull" and a bit lifeless to me.  There was something missing from it which has drastically gotten better and the more I tweak it to match my Petrof, the more beautiful it gets!  However, there serious problems I started getting into, and the main culprit was the sustain.

1.  I had literally noticed I was getting into bad habits of pedaling because after hours of using Pianoteq, I would go to my grand and start getting a muddy sound.  I realized I was "Over Doing" the pedal due to spending to much time on Pianoteq recordings.

2.  On top of this I was hitting notes with more force then I would have done on my grand.  Why?  My body was literally adjusting to the fact that the harder I hit it, the longer it would sustain.  I would go back to my Grand and I am not exaggerating when I tell you I had to practice for 3 days to fix a "ppp" passage from Chopin because I had literally changed my finger position to get more stroke power out of Pianoteq.

To hear you say you picked to go with the mass majority to make up for their lack of piano skills was a bad decision.  They need to be held accountable for their lack of ability.  Your first choice should be and always be to the highest level of musicality you can achieve (thus a complete replica of a true accoustical response).  Give us what we expect which is the most accurate responding piano engine available to computers.  Let us (actually them) be the ones if they want to "tame" it down to do so.

Or just make more FXP's standard with Pianoteq because your sustain on some of these is more on the sustain level of an upright, not anywhere near a true concert performance level grand.  Sure we can tweak it to no end but, I've spent a few months doing that when in reality, I should have had proper sustain from the beginning.

Although it may sound a bit negative, this is not a bash by any means to Pianoteq or its users/musicans.  This is my favorite software application in the world and I know you will continue to improve it.  I love it, I was just very disappointed to hear you intentionally tamed it down to appease people.  Those people either don't have, never owned/played a real grand piano or own an upright and are to used to the quick decay of sound to appreciate its full potential/need.

Perhaps next time just add another FXP and call it "once was a steinway grand, now a steinway upright" or UR1 .

Maestro2be

Another C3 with more sustain?

Modartt just keeps on amazing me.

Glenn

Two Comments on the new CSls:

1.  It seem to use less CPU power - Options/Performance - the orange curve doesn't go nearly as high with this version, or am I dreaming?

2.  Feline1:  I have a Roland DP that generates values from zero to 127 for the sustain pedal (Pedal Control), but I can't answer your question.  I make use of it because I have to - any midi recorded on my DP has a lot of Pedal Control values.

Glenn

I really love the sense of humor on this board sometimes lol.  I tried many times "metaphorically" but the results were less then desireable lol.

I am away from home at the moment but I will let you know the exact model tonight.  But I can tell you I have owned it for 14 years.  It is a Technics Digital Piano.  It has adjustable key weight settings etc.  It's great.  Probably not as good as the best ones around today but great for my use.  Also my biggest limitations until Pianoteq came into my life were that all sampled pianos were sub par and offered crap playback ability.  Some were good, but nowhere near the real thing.  Pianoteq changed that.  So there was no need for a new keyboard.

About my finger issue.  Do this experiment.  Hold out your hands and extend your fingers so they are all as straight as you can get them.  Hold them there and notice that your hand is resisting you a bit.  It is wanting to pull your fingers into a curved position.  Now with no force at all, just let them go and let that tension relax.  Your fingers should snap into a natural position.  Almost like a claw with your fingers slightly curved.

The point is that when your fingers are "flat" as in the first step it is easier to play "ppp" because it reduces impact and the amount of force you are directly applying in a downward motion.  From the natural "claw" position your fingers will tend to naturally drive a little more forceful and less elegantly for super soft playing.  You can still do it, but it takes extra practice.  The whole point is, because of the sustain issue I was playing with a natural finger position to make up for the loss.  Hence when I went back to my grand and played with the natural fingering (claw) I was playing "p" and not "ppp".

That's todays piano lesson .

Maestro2be

C3ls haven a problem.  Vibrating metalic exaggerated resonance.

Play C1 (C0 in some countries) strong with sustein pedal down, and will heard a quite anoying metalic vibrating oscilation.
Too much metalic, like a coin vibvrating in a glass table.

I dream that pianoteq in entire tone could sound more warm, wood, and less metalic...

Jope- great post, i live in your world.  my mom would turn on the egg timer for 30 minutes and i would dutifully play my lesson.  once that timer rang signalling freedom, it was all rock and roll.  which eventually turned to jazz.  but i would say this,  even if you aren't the classical virtuoso you refer to, but want to record something that sounds authentic, then the realism of the software piano becomes significantly important.  Pianoteq is clearly a great choice for live playing, what i am hoping for is that i can eventually get the depth, brilliance and complexity of overtones and resonance that i hear on well executed acoustic recordings into my own DAW and dithered down CD's.  that may well be a holy grail that is unattainable.  Does anybody think it can be done with this?.....

Just my 2 cents on the subject (and I've been working on tunings lately, how ironic -- thanks, Kyle Gann!) :

I grew up playing an upright Kawai piano that has simply _magnificent_ sustain across the board (even the higher notes ring longer/louder), and I had the privilege of practicing on several Baldwin/Steinway grands at my various schools.  (Personally, I enjoyed the upright the most.)  I would lift (or even remove) the lids (and even remove the lower panel on the upright) just to enjoy the sustain as much as possible on these pianos.  (And yes, I would also pull a John Cage and did all sorts of things to the strings, though more with Morton Feldman sound-world in mind, unbeknownst at the time...  ;^)  So you _could_ say that I'm a "sustain connoisseur."  (I certainly do!)  Just imagine:  opening up a piano on a cold winter's night (or gentle rainy night), lighting a few candles (or not), and playing away, letting the sounds (and their "phantoms") soak the atmosphere!  (This is why I'm into tuning, too -- sometimes I like "pure" intervals to set the "calm temple" mood, and sometimes I just like to let the wolves run in a pack, watching the partials eat each other up in a cloud of fury.)

I noticed a significant decrease in sustain between versions 2 and 3.  (In fact, version 2's sustain was what brought a smile to my face in the first place!)  However, the improvement in overall _tone_ was so significant that I didn't mind losing a little sustain in the process.  (After all, I almost always end up adding more reverb, internal or external, to lengthen the sound regardless.)  And I could always raise the impedance and related parameters (as well as global resonance) to levels that satisfied me.  (I did this with "Worn-out" rather quickly.)

Personally, I found the "less sustain" to be appealing simply for the reason of making the tone improvement clearer.  I'm one of those pianists Heinrich Neuhaus warned you about with the "velvet paws," who is guilty of "overrating tone."  Guilty as charged!

I also believe that the "sympathetic resonance" feature makes Pianoteq _very_ playable, comparable to a real piano.  I just _adore_ how the notes interact with each other, especially with repeated strokes at varying velocities.  I can _imagine_ playing a real piano with Pianoteq;  I _wish_ I were playing a real piano when I play Ivory or the like.  (Yup, let me check -- confirmed:  sounds as dead as a brick wall, since the notes don't "interact" with each other -- though it does make for nice sustain with the aforementioned reverbs, which definitely liven up the sound a bit!)

Lo and behold, thanks to user feedback, we now have a _new_ C3 setting with longer sustain.  Boy, was that fast!!!  Bravo!  I love it.  I'm going back to the vibraphone now.  (Apparently, I love that even more.  %^)

Thanks philippe.  You are The Flash Digital Piano tuner.     ;-)

In earlier versions of PianoTeq, I think people were partly having trouble because of the natural "envelope" of a piano's amplitude. In early versions, the decay was a little smooth, with no sudden decay after the initial strike. The change in the amplitude after the hammer strike seems to be an important part of getting a natural sound: there is a sudden decay, and then a longer decay. The length of the initial decay does a lot towards creating the individual sound of a piano.

William Coakley, for one, speaks of making a piano sample sing more by extending the initial decay a bit by using an amp envelope with two separate decay stages, and making the initial decay a little longer than it might be according to a microphone. (Even though the two step decay occurs naturally, using an envelope can let the initial stage have a little more volume before the longer, less steep, drop.)

From what I can hear and gather, an entirely different math problem comes up in PT when these decays come into play: although the overall sound is reduced at each decay stage, the partial structure changes, too. Not all of the partials are reduced equally, and there is variation at each note.  I would imagine that the Direct duration parameter can be a problem too--since it extends the time before the detuning of the unisons, the decay stages can reduce the amplitude of the notes at the moment when the detuning occurs, and thus reduce the prominence of the detuning. (The extent to which the detuning is noticable.)

Which makes me want to be able to control the two decay stages. Maybe it could be done with the hammer hardness settings--is it the hammer hardness (along with velocity or force, of course) that determines the length of the first decay stage?

Hm...In that other thread, I was more concerned with the way that hammer placement affected the partial structure. Surprising to hear that it affects the length of the initial decay, too.

The rocking chair is a  nice metaphor. In that case, you must imagine that there are several rocking chairs loosely tightened together. As you say, as soon as you let go of it, it will begin rocking at it's natural frequency, but the meaning of “natural frequency” has to be adapted to this new situation: the fact that they are tightened (and the way they are) makes it a bit tricky and quite interesting! Among the things that can happen, they could rock in phase at the beginning (direct sound), and in opposite phase at the end (remanent sound). The problem is that I have only one rocking chair at home, so I cannot make the experience and this is a pure conjecture

the electric guitarists (and bass guitarists) around here will recognise another side of this coin:

if you use an electronic tuner pedal to try and tune your guitar,
you will nearly always get that annoying thing where, initially for a bit of second, when you first pluck the string, the tuner shows it at a slightly different pitch (sharper) before the note sustains and settles down.
As I understand it, this is because when you first pluck the string and make it vibrate, the effective string length is different because you are kind of briefly making a 'node' at the plucking point.
This confuses bad guitarists who can't use their ears to tell if they are in tune