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

This has not gone unnoticed by The DPBSD Project! upper management team council and governing body federation alliance.

This must surely mean that the level of BS out there is not diminishing. Talking of which, I wonder if the analytical techniques employed could also be applied to politicians - I'm just afraid that the catastrophic signal-to-noise ratio would mean that no meaningful information could be gathered.

Possibly, but where would you plug in the MIDI cable?


Rest assured that the seasoned professional engineers here at The DPBSD Project! are able extract meaningful information from almost any corrupted scenario (our main clients are DPs, after all).

Don't tempt me!

Yamaha P-155 Re-Re-Review



With 226 MP3 and 146 PIX downloads (and counting) the Yamaha P-155 has been the most popular of all my testing (previous reviews here and here). And no wonder, this DP has a lot going for it in terms of brand, keys, sound, features, and price. So it's great that anotherscott retested it and provided us with a v1.9 file to analyze.

MP3: http://www.mediafire.com/?0epjvm3urobp8l6
PIX: http://www.mediafire.com/?cr78xfuijvcdwok

Here are the updates from the previous testing:

PROS:
- Passes the silent replay test.
- Passes the quick partial damping test.
CONS:
- Fails the key sympathetic resonance test.
- Sympathetic resonance sounds echoy and reverby, particularly during parital pedaling.
- Looping is somewhat audible, loops sound bland rather than cyclic.

I think the echoy sound of the sympathetic resonance during the half pedaling test is probably due to interaction of the single note loop period with one of the delay periods in the resonance effect. Also, I can hear the loops of the multiple notes interacting during the pedal-up portion of the pedal sympathetic resonance test, but this isn't nearly as obvious during the pedal-down portion of the test, so the effect is able to smooth this out to some degree.


----------------
- Yamaha P-155 -
----------------
FILE & SETUP:
- dpbsd_v1.9_yamaha_p155.mp3
- Setup: Mbox interface @ 24-bit, Mac with ProTools LE, MP3 via Quicktime Pro.
- Recorded by "anotherscott".
PROS:
- Passes the pedal sympathetic resonance test.
- Passes the silent replay test.
- Passes the quick partial damping test.
- Passes the half pedaling test.
- Blended velocity layers (Yamaha says 4 layers) with visible layer switch @ vel=90, 114.
- Key-up mute sound?
CONS:
- Fails the key sympathetic resonance test.
- Sympathetic resonance sounds echoy and reverby, particularly during parital pedaling.
- Looping is somewhat audible, loops sound bland rather than cyclic.
- Attack sample lengths are: 2.4,2.0,1.9,1.7,1.6,1.2,0.8,? seconds.
- Loop sample lengths are: 0.75,0.75,0.50,0.46,0.43,?,?,? seconds.
- Visibly and audibly stretched over entire range.
- Stretch distances: 3,3,4(x3),2,4,3(x3),2,3(x3),2(x4),3,1,3,3,4,2,2,3,5,4,3,3 = 30 groups.
- Slightly audible layer switch @ vel=114, not much timbre variation over lower 1/2 range.
- No pedal pedal up/down samples or sound effects.
OTHER:
- Dynamic range 66dB (vel=1:127).
- MP3 levels: peak @ -3.4dB, noise floor @ -81dB.
- Date reviewed: 2010-02-08; updated: 2010-03-17, 2010-04-16, 2011-11-19

Hi dewster, your retest fits very well with my experiences of the P155. It's a solid board for the price with some (slight) deficiencies in the sound, noticeable if you play with good headphones. In terms of playing, the most annoying feature for me is the extreme dynamic range, quite unnatural. In order to get into the realistic subrange you have to play very well-controlled.

Nowadays I use my P155 mostly for controlling a Nord Electro 3HP - the sounds of which I love in playing despite their own little quirks. What about a test of the most recent piano-sounds (especially the XL versions which I unfortunately can't load into my Nord)? I don't think the Bosie XL or the Bright Grand XL sounds have already been tested here.

Interesting, I've often wondered if / how the large dynamic range that Yamaha tends to implement might influence one's playing.


At ~200 MB these are still an order of magnitude smaller than what I would consider to be the very low end of serious, professional, solo recordable sample set territory. Being roughly twice as large as the the next step down I'd hope there would be less stretching (Nord seems to love stretching). I'd certainly be happy to test a sample should someone provide one.

Some people interpret the blurb on Clavia's website - "fully mapped accross the keyboard" - as saying that the XL samples are not stretched at all...

If that's the case then probably all you get is no stretching (since the sample sets below this are ~1/3 the size and have ~3x stretching) - no room left for other improvements.

Nord is so close it makes these baby steps extra annoying. I wish they'd take the next logical leap to no looping. If they want to quit soldering Flash to their PWBs, multi-channel USB 3.0 Flash drives are currently < $1/GB.

AFAIK, the only improvement Nord (or anyone) has every claimed for XL is the lack of stretching.


Even if you're correct in your implication that that technology would work within the existing Nord design architecture, it would still require all new hardware. Rather than call XL a "baby step," I think it has to be looked at through the prism of what can be effectively implemented within the constraints of their existing hardware. Remember that much of Nord's rep rests on their providing years worth of updates and support for existing models... it's one of the main reasons some people choose Nords over other brands.

Kawai ES6 Review



Thanks to Piano World forum member "bajabill" we now have a DPBSD sample of the Kawai ES6!

PIX: http://www.mediafire.com/?pqu8jboqq6yr3qw
MP3: http://www.mediafire.com/?mh2qbuv50l2bin7
MP3: http://www.mediafire.com/?80hd6v84u7dxv9d (resonance testing)

bajabill has this to say about the ES6:


The ES6 employs Kawai's "Harmonic Imaging" technology (not the newer PHI or the latest UPHI) so I initially thought it would be most like the MP5 sample I have. I spent some time looking at phase signatures and listening to both, and they sounded similar, but there were no obvious phase matches. However, upon comparing it to the CN33 (which employs PHI) I saw a good phase match with the note attack samples, so these DPs almost certainly share the same base sample set. I previously determined that the CN33 was also a attack phase match with the Kawai CA63 (which employs UPHI).

Neither pedal nor key sympathetic resonance are present during MIDI playback (this seems to be typical of Kawai DPs) so to evaluate them bajabill manually played the tests and recorded them to the second MP3 file listed above. I definitely hear the key resonance but the pedal resonance is quite subtle, which is kind of odd because normally it's the other way around, though the strengths of these effects are individually adjustable in the menus (the MP3 was recorded using the default settings). Since this issue arises rather often I've added manual sympathetic resonance testing instructions to the dpbsd readme file.

There is almost no stretching to speak of which is nice, but the decay times are rather quick, and the short loop decay samples sound fairly bland. The velocity layers are smoothly blended, and so are neither visible in the phase views nor are they audibly switched. It passes the brief pedal partial damping and the half pedaling tests, but fails the silent replay test. And there are no pedal up/down or key up sound effects that I can hear.

Review below, MP3s and more analysis pics at the share point.



Spectral phase view of the stretch test, mid notes, CN33 above, ES6 below. A good phase signature match of the attack samples indicates the sample sets were derived from the same recording session. Very little stretching, none audible.


Waveform view of the looping test with vertical zoom applied. Decay times are uniformly somewhat short.


Spectral pan view of the layer test. No visible or audible layer switching, smooth timbre variation.


-------------
- Kawai ES6 -
-------------
FILE & SETUP:
- dpbsd_v2.0_kawai_es6.mp3
- kawai_es6_resonance.mp3
- MIDI (USB) sequenced with Anvil Studiom, audio capture / MP3 conversion with Audacity.
- Recorded by "bajabill".
PROS:
- Passes the pedal sympathetic resonance test - the effect is visible and somewhat audible.
- Passes the key sympathetic resonance test - notes above and below the stimulus resonate.
- Passes the quick partial damping test.
- Passes the late pedal partial damping test up to the test limit of 0.5 seconds.
- Passes the half pedaling test.
- Except for a stretch group of 2 and a group of 4, all notes are sampled.
- This is most likely a smoothly blended multi-velocity sample set.
CONS:
- Fails the silent replay test @ pedal up (note plays vel=1).
- Decay times over the full range are somewhat short (~1/2 Pianoteq).
- Attack sample lengths are (C1:C8): ?,1.4,1.2,1.8,1.1,0.9,0.9,? seconds.
- Loop sample lengths are (C1:C8) 1.3,0.7,0.65,?,0.65,0.65,?,? seconds.
- Obviously looped, lower and mid notes could benefit from longer decay loops.
- Stretch distances: 1(x72),2,4,1(x10) = 84 groups.
- No pedal up/down or key up sounds.
- No key or pedal sympathetic resonance via MIDI playback.
OTHER:
- Spectral signatures same as CN33 & CA63 (same base sample set).
- Notes played @ vel=1 produce a sound.
- Dampered | undampered transition: F#6 | G6.
- Dynamic range 42dB (vel=1:127).
- MP3 levels: peak @ -1.3dB, noise floor @ -68dB.
- Date reviewed: 2012-09-05.

Interesting analysis, as always.

It's perhaps worth pointing out that the ES6 was launched in mid-2008, and was indeed the first Kawai DP to utilise full, 88-key note piano sampling. As dewster correctly notes, the same base samples are used in the more advanced/detailed PHI and UPHI sound standards found in the CN and CA models respectively.

Cheers,
James
x

What is the coding that dictates the decay? Is it a limited database size or a programmed volume/time relationship? As a beginner, more time to find the next chord before the current one dies out would be nice.

Thanks for confirming that James!


In a looped instrument, decay time is pretty much just a programmed volume/time thing with maybe some filtering over time as well. It's not hard to imagine that decay times are often intentionally made unnaturally fast in an attempt to somewhat conceal the unnaturally repetitive (too short) or drab (way too short and overprocessed) sound of looping. It's a double whammy.


In more synth type instruments you can sometimes adjust the decay time, but in most DPs you can't. Real pianos ring on forever, DPs not so much, and the ES6 has a shorter decay than many others. Holding the pedal down most of the time might help some, but that will likely make you develop bad pedaling habits. You might try more damper and string resonance, and maybe more reverb.

Hi, all.


Recently I came accross a review about a DP: SoundMagic Blue Grand3D on Virtual Piano Domain (the link is here).

Very interesting that on VPD, they rate it as 5 stars (at the moment of writing, the only one with 5 stars and the highest score). From the look of it, looks like that this piano is quite worthwhile for its sound and features (somebody also post a video using it on YouTube here)

However, I remember seeing on this thread that somebody reviewing one of SoundMagic product (their Imperial Grand), and the result is a recommendation to pass the product.

I'm not sure if these two are using the same or different DP Engine; however, I'm interested to see how Blue Grand3D is reviewed by the DPBSD Project. I've tried to find some trial/demo version of the software, but my search so far returns none. Anybody has this product installed on their machine and kindly help to submit the required audio files for the review?

The result will be interesting to see.


Thanks

Thanks for pointing that out! Quite a gushing review.

From that review it still seems to have a layer jump at velocity 66. They use a more sophisticated measurement of dynamic range than I do which is interesting. But they don't mention looping?

Casio Privia PX-350



A first for the DPBSD project - a corporate submission!

Mike Martin (General Manager of Marketing, Electronic Musical Instruments at Casio) kindly supplied a DPBSD MP3 of their new PX-350 at my request. The MP3 (and pix) are of the default piano voice "Grand Piano Concert" and the MIDI file was rendered directly to WAV on the PX-350 - a very handy feature to have. Those interested can also listen to the compressed layer test:

PIX: http://www.mediafire.com/?vhnww5a41ife41b
MP3: http://www.mediafire.com/?dpi8s15uw8aacb3
MP3: http://www.mediafire.com/?ddltnf1r6f866vs (compressed layer test)

Readers may want to refer to my previous PX-330 review for comparison purposes. The PX-350 is Casio's next generation, with new textured keys, technically better sound, high-resolution MIDI output, USB recording, etc. Read more about the new Casio DPs over on this PW thread started by Kbeaumont.

So what exactly has changed in the piano sound department? Let's take a look:

PX-330:
- Attack sample lengths are (C1:C8): 1.8,1.5,1.4,1.4,1.4,1.3,1.0,1.0,? seconds.
- Loop sample lengths are (C1:C8): 1.6,1.1,1.0,0.8,1.0,1.0,0.9,?,? seconds
- Stretch distances: 3,2,3(x16),4,2,3(x5),4,5,5 = 28 groups.
- Dynamic range ~49dB (vel=1:127).

PX-350:
- Attack sample lengths are (C1:C8): 4.3,3.7,3.4,3.4,3.5,3.4,1.0,? seconds.
- Loop sample lengths are (C1:C8): 1.2,1.3,1.1,1.1,1.1,1.1,0.7,? seconds
- Stretch distances: 3,2,3,2,3,2,2,3,2,3,2(x5),3(x5),2(x3),3,2,2,3,2,3(x4),4,4 = 34 groups.
- Dynamic range ~57dB (vel=1:127).

The attack sample lengths for the low and mid notes are quite a bit longer and really benefit from this, though the loop sample lengths remain roughly the same. The crossfade from attack to loop is well done, and the loops sound "wobbly" rather than static, but the loops could benefit from increased length because the repeat interval is somewhat audible. The PX-350 is less stretched than the PX-330, but the stretch group transitions are still audible over most of the range.

Pedal sympathetic resonance for the PX-350 is rather subtle, but audible with the single note E6 test. I have to strain to see or hear any difference with the chord based test. Casio seems to have increased the dynamic range, and there may be more timbre variation with velocity than previously. There is a pedal down "loom of strings" sound effect, but it is at quite a low level in the test file (-60dB from peak). This may have to do with playing it via MIDI rather than with the keys & pedals.

Again, much thanks to Mike Martin & Casio for the timely sample of this new DP, and for the support of this project!

Some analysis pix and text review:


Figure 1. Spectral phase view of the single note E6 pedal sympathetic resonance test, pedal down @ left, pedal up @ right. Sympathetic resonance is both visibly and audibly subtle.


Figure 2. Waveform view of the entire looping test, vertical zoom applied to see the noise floor. Note decay is nice and long.



Figures 3a & 3b. Spectral views of the note C4, normalized to -1dB to increase clarity, cursor at attack / loop transition, same horizontal time scales. TOP: Spectral phase view of PX-330; BOTTOM: Spectral pan view of PX-350. Attack samples are significantly longer in the PX-350, decay samples remain roughly the same.



Figures 4a & 4b. Spectral phase view of the stretch test, mid notes, normalized to -1dB to increase clarity. TOP: PX-330; BOTTOM: PX-350. With 34 stretch groups, the PX-350 is somewhat less stretched than the PX-330, which has 28 stretch groups. Transitions are audible over the low and mid note ranges. Note phase "fingerprint" match between many of the notes, which indicates that both DP sample sets were derived from the same AP sampling session.


Figure 5. Spectral phase view of the layer test, highly compressed. One visible layer switch, the transition of which is not audible to me (Casio claims 4 layers). You can listen to this as a separate MP3 file (listed above).


-----------------------
- Casio Privia PX-350 -
-----------------------
FILE & SETUP:
- dpbsd_v2.0_casio_px350.mp3
- casio_px350_layers_comp.mp3
- Sequenced and recorded using built-in facilities, WAV=>MP3 w/ Adobe Audition.
- This is the first patch: "Grand Piano Concert".
- Recorded by Mike Martin.
PROS:
- Passes the pedal sympathetic resonance test, though the effect is subtle.
- Passes the late pedal partial damping test up to the test limit of 0.5 seconds.
- Passes the half pedaling test.
- Note decays are nice and long.
- This is a smoothly blended multi-velocity layer sample set (Casio reports 4 layers, one step visible @ v=80).
- Timbre change with velocity is nicely spread out with no audible timbre steps.
- Pedal down sound effects (but at fairly low volume via MIDI).
CONS:
- Fails the key sympathetic resonance test.
- Fails the silent replay test - note damps @ pedal up.
- Partially damped notes don't sound "buzzy".
- Obviously looped, the loops sound "wobbly" rather than static.
- Attack sample lengths are (C1:C8): 4.3,3.7,3.4,3.4,3.5,3.4,1.0,? seconds.
- Loop sample lengths are (C1:C8): 1.2,1.3,1.1,1.1,1.1,1.1,0.7,? seconds
- Stretching is visible over the entire range, audible over the low and mid notes.
- Stretch distances: 3,2,3,2,3,2,2,3,2,3,2(x5),3(x5),2(x3),3,2,2,3,2,3(x4),4,4 = 34 groups.
- No obvious key up sound effects (via MIDI).
OTHER:
- Phase signature match to previous generation (from same AP sample session).
- Notes played @ vel=1 produce no sound.
- Dynamic range ~57dB (vel=1:127).
- MP3 levels: peak @ -7.6dB, noise floor @ -84dB.
- Dampered | undampered transition: D#6 | E6.
- Date reviewed: 2012-09-27

Great! Casio has just gone right up in my estimation - and it was already high regarding reliability, value for money and innovation.

That really is quite a big jump up in specs! Kudos to Casio!

So it's basically the same piano as before, but with more memory for longer attack phase ? Meaning if you like the sound of the previous Casio grand sample , you get a souped up version of the same sample set with the new series. If you didnt like the basic sound character of the Casio Piano's before, you should look somewhere else, since in that respect there is nothing new under the sun (right) ? Or are there also brand new sample sets being used ?