Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards The DPBSD Project!

For that you don't need a robot. Even an unskilled beginner pianist can bang on one key fast (using both #2 fingers alternating) while recording to a MIDI file. do that for 15 seconds and vary the rate randomly then look at the file to see the closest notes. If the key is played to fast it simply will not record. So examine the midi file later to find the threshold of recording.

I suspect it will be years, if ever before I can over play my keyboard.

What is the typical fastest repeat rate on a grand piano?

I'd be curious to see the BSD file run through a CP-1 Electric Piano not an AP. Aren't the EP's fully modeled in the CP-1 while the AP's are sampled with modeled effects added on? Not sure what might come of such a test or if the BSD file is only relevant to AP's.


edited for speeling.

Chris,
Yes, the method you describe would be pretty good, but I would still prefer a mechanical device to do the test, to eliminate the human factor. For example, how would we know for sure that the person who did the test was lifting their fingers up far enough for the keyboard to register it? When I do the two-finger test with my keyboard, it is difficult to get a consistent behaviour. I can sometimes get a short burst of rapidly repeated notes, but then it will stop. This may be my fault, but it could also be the keyboard, to some extent.

By your own admission, you are not good enough to test your own keyboard. ;^) (yes, I realise that if your keyboard is behaving as fast as, or even faster, than a good grand piano, then the fact that you can't reach it's limit is mostly academic)

I don't know what the repitition rate of a grand piano is - good question.

Greg.

My point was that it does not matter how fast you bang on the keys or if you even know how fast you are banging on the keys all you have to do is randomly drift the banging speed to both faster and slower than the DP's threshold. Then examine the MIDI file to find the threshold Find the minimum space between two notes that is statistically repeatable and the reciprocal of the space is the repeat rate. I don't think you have to control the keyboard technique.

You say some time the string of notes stop and some times you get a short burst. You'd expect it to stop if you play to fast. Artificial "impossibly fast" short burst are also a failure mode when over driven. Look for the place where a histogram falls off a cliff. I think a good statistical analysis is cheaper than building a robot.

The way these key sensors work is with a timer. They measure the time between two points and assume this is proportional to velocity.

But I think for 90% of us this is moot.

No you don't I thought it was funny ... you are beginning to remind me of Geoffry Rush in Quill's .....

Chris,
I understood what you meant completely. My point is that I am not certain that all of us can reliably strike the keyboard fast enough, and consistently enough, to produce a quality measurement. I am also not confident that the keyboard will cease "repeating" in a reliable fashion. I.e - as the repetition rate is approached, I think the keyboard will start to behave a bit erratically. (I think my keyboard does, but I can't prove it). If we use a precise mechanism to do the repetition, that eliminates the human error. If we had the hardware, we could gradually increase the rate, and we could find the threshold at which the action starts to become erratic, and then keep increasing the rate to see where it stops completely, for example.

For your keyboard, you have admitted that you can't strike it fast enough to reach the keyboard's limit, so you simply cannot measure your keyboard's repetition rate. For you, it is COMPLETELY academic, of course, but others may want to know what your keyboard is capable of - not what YOU are capable of. ;^)

I think it would be a very interesting test. For example, if the RD700GX, which has a high quality action, but without the triple sensors, can repeat faster than a Casio PX-130, which does have three sensors, I'd be very very interested to know that.

Greg.

@dewster, thanks, here's the original file directly recorded with CA-63:

http://www.mediafire.com/file/rhuwxhnz32i/KAWAI_CA_63_ORG.MP3

But I will redo this anyway, once I have the firmware update, and I'll try to also get the sympathetic resonance included somehow since this is a very important part of the beautiful sound of the new CA models.

Your observation seems to be very much like my first impression. On my previous HP-3e I was hearing loops all the time. When I first tried HP-307 at store I was overwhelmed by the difference. One of the things I was trying to figure out was if there's audible looping or not and I couldn't really tell what Im hearing.
The samples or whatnot were really convincing and keyboard years ahead of my former piano so I didn't think long. I refrain from doing any review on the piano because I can't compare it to any other model, not even previous Roland's HP-207.
Im eagerly awaiting your test results

IMHO your way to choosing which DP to buy is not very good. When deciding about a new DP purchase, there are 4 possible sources of information:

(1) Marketing material.
(2) Technical analysis (dewster).
(3) Reviews of other players (Dr Popper, Lawrence).
(4) Playing the damn thing yourself, obviously.

In the end (4) should always be the deciding factor. Deciding on (2) alone would be quite stupid. Deciding on (3) alone is IMHO not very clever, either. If you really want to make a clever choice, you should collect all information you can get, weigh them in some way ((4) should have the highest weight by far), and then decide based on that.

What exact purpose do dewster's tests have? Are they intended to make you decide which DP to buy? *Definitely not*. They are intended to cut through the marketing bullshit and give us the information which should have been part of the manufacturer's "technical specification" from the get go.

So what do dewster's tests really say about the CP1? Do they say that it's a bad instrument and nobody should buy it? Nope! They only say that the basic underlying technical implementation of the CP1 is still in some parts similar to most older DPs and not the "no-compromise" approach some of us have been hoping for. But the instrument can still sound and play great! Actually better than most others (since most others share the same limitations). But what dewster's tests clearly show is that the CP1 could have been even better than it actually is. Which also means that there will probably be future DPs which will sound even better (noticeably better) than the CP1. Or to take it directly from the horses mouth:


Basically the Yamaha DP engineer said that longer samples result in greater authenticity. But the CP1 does not have longer samples. There you have it.

Again: The CP1 can still be great - compared to most current DPs! But it does use short length samples. So, according to the Yamaha DP engineer, there's still room for greater authenticity.

Lucky! I listened to the file, and it sounds amazingly realistic compared to my piano..

One interesting subjective test I would enjoy seeing here on the digital forum as a complement to the very interesting work dewster has been doing for us would be picking a few midi performances, for example from this site: http://www.piano-e-competition.com/ecompetition/midi_2009.asp#T and having them played on the major contenders' standard pre-sets being discussed here on the digital forum and then posted as lossless files for comparison listening by us all :

- Roland SuperNatural Piano
- Kawai Ultra Harmonic Resonance
- Yamaha Spectral Component Modeling
- Yamaha CLP Technology
- Roland VPIANO modeling
- Pianoteq modeling

You could also add a (5) to your list which would be

(5) listening to comparison performances of real music on the instruments.

Admittedly the player experience is very important but for those who would use the digital piano for performance, the actual total sound produced or recorded would also be of prime importance.

Finally, for those of us using a digital as the most authentic (silent) way to prepare to polish and perform pieces on an acoustic piano it is most important that the instrument acts as closely as possible to an ideal acoustic but it is certainly an added benefit if the tone/sound of the piano is also pleasurable to listen to for all those hours.

I believe it would be safe for you to understand, and if you read my posts thoroughly, that I intend to play the damn thing myself before I buy it. I am not committed to buy it if I don't find it suits my purpose.

So, either you aren't very clever, or you seem to think that I am not.

Based on your interpretation of my posts, I'd say the former, is by far, the more likely possibility, and the latter is a mistaken assumption on your part.

Considering the technical analysis is based on dubious quality MP3 recordings, that may or may not be accurate enough, and the possibility of error, either in the material uploaded (we really can't be certain it is from the actual instrument), I'd say, out of the four possible sources of information, #2 is not a source of information in which I would put a lot of credence.

As I said earlier; you can't measure the soul of an instrument with a computer, or, by analysing squiggly lines on a screen that may or may not be displaying accurate data, depending on the quality of the material being investigated. Believe what you want, and so shall I; after all, it is my money.

If given a choice of only two possibilities; #2 and #3, I'd say I'd trust the word of professionals a long way over an amateur doing a technical analysis.

And, in closing this matter between us, let me be clear on another thing; I do not favor Yamaha over another brand. I'm also very interested in instruments by Roland and Nord, and of course, just so you aren't confused again, I intend to play the damned things before I decide to buy.

Thank you for your concern.

Best regards,

Colleen

Excuse me, there is no room for interpretation of what you said. This is what you said:



I think this speaks for itself. If you now state that you've made up your mind and did not order one then okay, that's good, but it's not what you originally stated.



This is by no way acceptable and just plain offending... You should really think about the style of your discussion. As far as I am concerned, I'll no longer get bothered by your posts since I just put you on my ignore list.

Coleen, If you think mass produced consumer products like digital pianos (essentially computers with plastic keys) have "souls" and that their sound produced cannot be measured or viewed by other computers or people, I can see why you might get so excited about the retro reincarnation of the look, sounds and technology of Christmas Past that Yamaha is packaging up with the designed by committee CP series.

I am sure Yamaha marketing will be thrilled with all the low class ad hominem attacks by its blind supporters on public forums. With friends like that...

You wrote "I ordered a CP-1 this morning". That's usually what you do *after* having decided which DP to buy.


Wow, thanks for the personal insult. I never said that you weren't clever. I just said that deciding on a DP without having played it yourself is not a clever thing to do. That's quite a different thing.


I find it amazing how many funny reasons some of you can find for ignoring pretty useful technical information. FWIW, I could find an equally high number of funny reasons why reviews of forum members should be mistrusted. But I won't, because there's a difference between funny reasons and reasonable reasons.


Do you think that describing a scientific technique with childish language somehow makes your point stronger?


Why limiting yourself to only two possibilities, if it's easy enough for anybody to make use of all four? That makes no sense at all.


And I never hinted or even thought you were.

I also wrote to Steve, " I will post my impressions as soon as possible."

I also said I was under no obligation to buy.

It's pretty hard to give someone "impressions" of an instrument without playing it.


About the "clever" thing; you can try and justify your words all you want, but you tossed out the insult first. I was merely returning the favor.

Again, please read a person's post carefully, before you decide to insult or attack them.

Have you read this one carefully?

I hope so.

Regards,

Colleen

Thank you. The anticipation of playing a piano with a retro look, always excites those of us with an imagination and an appreciation of a well designed instrument, regardless if it was allegedly conceived by a committee or what company makes it.

It does remind me of Christmas past and the thrill of opening a nicely wrapped present. What a lovely analogy.

You are far more clever than Madshi.

I'm sure all the keyboard manufactures will be thrilled at the appearance of inaccurate data about their instruments on a public forum by an amateur tester.

Regards,

Colleen

Thanks! That one sounds better near the noise floor. I'll archive it instead of the other and wait on your improved version.