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

I agree that, given hammer physics, key action to sound event timing is critically important. But there is no way to really tell anything without video or some external timekeeper. MIDI alone is probably not enough.

Not enough indeed, but could be telling. About possible quality issues (consistency) and about whether the claimed overall delay simulation was implemented in sound synthesis module or in keyboard mechanics/output. Important informations both when using the keyboard as a MIDI controller or when using an instrument with MIDI in as a sound generator.

I agree you can't judge pianos using numbers, but I think you can judge TECHNICAL aspect of sample sets. Of course it would be general and include some simplifications, but for some people this thread may be too complicated to look; their alternative is to not look in DPBSD at all. General note (or a few of them) may also be nice intro to further analysis.

Anyway, even if you want to avoid such thing, it may be nice idea to consequently add short verbal summary to every test.

Thanks for your comments, I do appreciate any feedback!

I think I'm pretty much including short verbal summaries, as well as owner's comments, and further clarification in the form of both pictures and picture captions. If people are afraid of or turned off by specs they probably don't visit this thread much anyway. I could certainly do more to collect things in one spot, but there is the clickable list in the first post, as well as the text file of compiled reviews at the share point that people can consult and pretty much do whatever they want with. Beyond that I'm kind of out of time lately and currently have something of a backlog going just about every project I'm involved with.

dewster, to be honest, I think the current format is fine.

While I can understand the desire from others for you to include a 'marks out of ten' scoring system, I believe this may undermine the objective nature of the tests.

After all, the intention of the DPBSD is to provide technical analysis of sound, rather than a subjective review.

Cheers,
James
x

That "the intention of the DPBSD is to provide technical analysis of sound, rather than a subjective review" is precisely what limits its usefulness. There's a wealth of data ... and very little useful information.

Though I would not expect to choose a piano based on the information here, it **could** help people narrow their field of search ... if only it were more accessible.

I really think it would help to have a summarized view. Each piano could be judged on a short list of criteria, and each would get a score on each criterion. The scores would be based on the collected DPBSD data.

This summary would not introduce ambiguity or subjectivity. Instead it would drive usability.

I did some "light weight" measurements of MIDI timing and hammer delay simulation just as proposed. I did just want to see, how big issue Time Delay in reality was and whether it would be worth to be explored further for all of tested instruments (e.g. by extending the DPBSD test). Here is a short summary of my results (without documenting them in detail).

Methods:
I took the DPBSD MIDI 2.0 file and the corresponding audio rendering output from it with following instruments:

• Casio PX-350 (as found in the DPBSD Public Library)
• Galaxy Vintage-D 1.2 (Produced by own Instrument)
• Pianoteq 4.20 (Produced by own Instrument)
• Roland rd-700nx (as found in the DPBSD Public Library)

I used Reaper, arranged the MIDI Track and graphical audio track Windows beneath of it.
I made an optical comparison of MIDI events and the delay times to the Attack they produced.
I estimate a sensitivity (precision) of my method of <= 3ms.
I used the DPBSD2.0 sequences of MIDI events
a.) of the same C4 note with increasing velocities from 1 to 127,
b.) 88 keys Note On with the same velocity
I looked at time offests between the MIDI event and corresponding audio attack.

Results:
All tested MIDI processors produced sound with no detectable delay (i.e. beyond the precision threshould of the simple methods above).

Conclusions:
1. I couldn't find any specific hammer delay simulation with these MIDI to Sound processors.
Especially,

• no decreasing delays for the increasing key sequence could be detected
• no decreasing delays with increasing velocity on the same key could be detected
• no incosistent (variable) delay behaviour could be detected

2. Even Casio's PX350 MIDI Synthesis Module didn't included any hammer delay simulation,
3. consequently, if Hammer Delay simulation is implemented, it should be originated completely from the Keyboard Action Output (as Casio actually claimed).
4. If Hammer Delay simulation is implemented with a DP's Keyboard Module, this feature can safely be used in connection with SW-Instruments (provided they behave like the tested ones).
5. Time Delay and its consistency in MIDI Synthesis is probably not a general quality issue for most DP-s I guess.

As a general conclusion I wouldn't regard it now very important to include MIDI Time Delay Check in the DPBSD project, but these results could perhaps be corroborated by some more precise tools and measurements.

It's an open project and everyone has the same access to the base data as I do. If you want to take some aspect of it in a different direction nothing is stopping you.

Thx, James, I am pleased to learn it now! This information was based on following statements by dewster in his very recent DPBSD review for (the elderly) ES6:
It could be perhaps helpful, James, if You could clarify since when Kawai models should definitely not expose this issue anymore. (Since CAx3 and CNx3 or just since CAx5 and CNx4?)

I would ask dewster then to edit his general remark which in this form could be a bit misleading - his excellent DPBSD page is a most visited central source of information about DPs.

I believe since the CNx3 and CAx3 were introduced.

Cheers,
James
x

1,000,000 views and counting. Wow!! Congratulations Dewster and thank you.

Thanks Steve!

Sincere thanks to all who have contributed DPBSD MP3s, and to all who find this thread useful.



Sugary, diabetes inducing DP cake all around! Cheap Champagne on the house!



Happy Holidays Everyone!

Hello people!
I'm new to this fantastic forum, and I have a question about the CP50.
I'm using the new firmware (1.2) and I noticed something strange.
When I'm playing the CF piano, I noticed that there is a big difference between all the keys under G6 and above this note.
The note decay is very short for anything under G6, but at this point, there is some kind of reverb added to the sound.
It's clearly audible when the reverb function is set to off.
It only happens with the acoustic piano sound, but I find it very annoying.
Do you think it's Yamaha's answer to the critics about the very short decay in the high notes?
Now it's even worse, because it sounds very artificial.

Any comments would be welcome. Thanks in advance!

From the DPBSD test notes (see the OP):

10. The highest keys on a real piano have no damper mechanism. The decay time for these notes is so short that a damper probably wouldn't affect playing much, and undamped strings are sympathetic resonant elements that can add richness to the sound of other played notes. The transition point between dampered and undampered is somewhere between D6 and A6. So obviously notes played above this point should not damp at key up (and obviously the damper pedal should have no influence over these notes either). Almost all digital pianos mimic this behavior by not damping the note at key-up (but probably very few model the sympathetic resonance element).

Great, Dewster! Many thanks for this. I played on a friend's Yamaha p115 or something similar last night and noticed the same behaviour. I don't have much experience with acoustic pianos so I was a bit scared when I heard this.
You reassure me!
Have a nice day and happy new year to all!

Nobody has uploaded the Kawai Ca95 or Ca65 yet??? I'm dying to see the review!

Very sorry, due to my digital Theremin research (hung up on simulating transformer based LC tanks) there's something of a backlog going with the DPBSD project. The CA95 is in the pipe (along with the ES7). Hopefully this month.

Hey dewster, any news? 8)

Hi,

I have a Casio PX850, and I have prepared an upload for it using the 2.0 midi file.
http://www.mediafire.com/download.php?ks5fu1fow71nqo2

I understood that it was still missing from the collection from the opening post, and indeed I looked for it during my purchasing decision.
I am not very experienced in this stuff. I believe I prepared the file right, but let me know if modifications are required.
Procedure followed:
midi playing from PC to piano over USB midi link
Piano recording sound directly to WAV file.
Wav file converted to MP3 192 kBit using iTunes (VBR/CBR unknown).

Best,

Martijn

Hi Martijn,

Got it! Levels are a bit low, but the noise floor looks good and it's in stereo.

I'll send you a PM soon.