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

Sorry, I've got a bit of a backlog going.

Yamaha CLP-440 Review



Thanks to Piano World forum member "Alexank" we now have a DPBSD sample of the new Yamaha CLP-440!

MP3: http://www.mediafire.com/?i8omdaej9s2gutj
PIX: http://www.mediafire.com/?74uvlvxl914cpsz

Alexank has this to say about the CLP-440:



On their website, Yamaha says "The new CLP Series uses the RGE (Real Grand Expression) Sound Engine, the next-generation piano sound source. It improves upon the Pure CF Sound Engine... and then they go on to talk about their new "Smooth Release function" which varies note damping time with key-off velocity. There aren't any note-off events in the DPBSD mainly due to the limitations of the MIDI editor I use (Sonar and most other editors I've tried use note-on velocity=0 instead) so I'm not currently testing for this, but I assume it works. Using key off velocity to modulate note decay time is probably fine to a first order, but to do it 100% correctly one would use the key position, information impossible to obtain with the on-off sensors in the GH3 (and, to be fair, virtually all DP actions made by other manufacturers).

From what I do test for I will say that the CLP-440 main voice is almost exactly the same as the P-155: same spectral signatures, same attack and decay sample times, same stretch groups, same dynamic range, etc. The only difference I can see is in the the spectral pan view of the layer test. I'm not sure what this means, perhaps a difference in post-processing or in the layers themselves. FWIW the layer test itself has changed some from v1.3 to v2.0 (note-off is a bit earlier in v2.0 so as to separate and therefore better differentiate between note-off and note-on).

I'm a bit confused regarding the key action on this new series. Yamaha says they have the GH3 action, but they also say the "Keys have a long distance to the fulcrum" which makes it seem like this is a new feature? I'm all for longer fulcrums, short fulcrums being one of the main glaring deficiencies in DP actions.

A few analysis pix (see the ZIP file for more):


Figures 1A & 1B. Spectral pan view of the note C3, CLP-440 @ top, P-155 @ bottom (please note difference in time scales). Notes look and sound pretty much identical, with same attack and loop sample lengths. I believe the P-155 sample was compressed (by me, I can't recall) to bring out detail, whereas the CLP-440 sample was merely normalized (by me).



Figures 2A & 2B. Spectral pan view of the stretch test, mid notes, CLP-440 @ top, P-155 @ bottom. Notes look and sound identical, with same stretch groups throughout.



Figures 3A & 3B. Spectral phase view of the layer test compressed 20:1 to bring out detail, CLP-440 @ top, P-155 @ bottom. At least three blended velocity layers are visible for the CLP-440, P155 is inconclusive due to the jumblyness (Yamaha reports 4 layers).

------------------
- Yamaha CLP-440 -
------------------
FILE & SETUP:
- dpbsd_v2.0_yamaha_clp440_piano1.mp3
- This is the first patch: "Piano 1".
- Setup: MIDI on flash drive in CLP-440 => headphone out => Macbook Pro / Audacity V1.2.5 record & convert.
- Recorded by "Alexank".
PROS:
- Passes the pedal sympathetic resonance, the effect is not very audible.
- Somewhat passes the key sympathetic resonance test, the effect is quite faint.
- Passes the silent replay test.
- Passes the quick partial damping test.
- Passes the late pedal partial damping test, note decay is caught even after 0.5 seconds.
- Passes the half pedaling test.
- Note decay times are fairly long.
- This is a smoothly blended multi-velocity sample set with at least three velocity layers.
- Timbre variation is fairly smooth with increasing velocity.
CONS:
- Obviously looped, both visibly and audibly over the lows and mids.
- Loop sample lengths are fairly short and static sounding.
- Attack sample lengths are (C1:C8): 2.0,2.0,1.8,1.7,1.6,1.2,0.8,? seconds.
- Loop sample lengths are (C1:C8): 0.8,0.8,0.5,0.5,0.5,?,?,? seconds.
- Fairly stretched, visible over the entire range, audible over lows & mids.
- 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.
- Visual velocity layer switch @ vel=90,114.
- Somewhat audible velocity layer switch @ vel=114.
- No obvious pedal up/down or key up "clunk" samples.
OTHER:
- Tests almost identically to the P-155.
- Dynamic range 66dB (vel=1:127).
- Notes played @ vel=1 produce no sound.
- Dampered | undampered transition: F#6 | G6
- MP3 levels: peak @ -0.1dB, noise floor @ -78dB.
- Date reviewed: 2011-07-21

I just did a quick recording of our Young Chang with a Zoom H2 recorder. Played middle C as softly as I could (-42 dB) and as loudly as I dared (-12 dB) given the wooden construction and early hour here. Playing it that lightly was almost impossible, had to play the key maybe 10 times in order to get the hammer to hit once.

~ 30 dB dynamic range was less than I expected.

BTW, I'm not sure I'd recommend the Zoom H2. The left front microphone element in ours is about -2 dB mismatched with the right, it eats batteries, and my 8 GB SDHC card (Patriot, class 6) often distressingly goes missing at power-up (though the 1 GB SD card that came with it works fine). It takes somewhat longer to boot with the larger card (underpowered processor?). Postage stamp sized LCD (I knew that going in) that's difficult to read with these old eyes and without the backlight on (which helps eat the batteries even faster). Rather disappointing after reading all the glowing reviews on-line.

Hi dewster, I am not too surprised at this result. Playing my present instruments (an old but good Pfeiffer upright, a Yamaha P155 and a Nord Electro 3 with Boesendorfer Large) side by side I find the P155's dynamic range quite exaggerated in comparison with the acoustic piano. However, the Nord's dynamic range seems (intuitively) a little less than what I get from the piano, but that's comparing apples and oranges since I listen to the Nord with headphones. - But then I'd think a grand piano has more dynamic range than an upright - so a little surprise at your 30dB remains.

Your observations make sense to me. The P-155 (like most Yamaha DPs) has an unrealistically large dynamic range of 62 dB. The two Nord voices I've tested are around 33 dB. Not sure why the Nord strikes me as having less dynamic range than our Young Chang, perhaps it's just me? Or perhaps it's a lack of sufficiently low velocity layers? I wish there was some way to quantify this.

I just did this again and got -50 dB and -9 dB (I'm reluctant to hit the note much harder than that for fear of damaging the mechanism or knocking the strings of this note out of tune), so:

~ 41 dB dynamic range - which is at the player position with the front section of the top flipped open so there is a large hole facing the player through which the sound can escape, but the top itself is closed.

It took me literally dozens of tries to get that -50 dB sound sample, and I can't help but imagine that this is a universal problem when sampling a piano. The variation from key to key when played at very soft velocities must make this a very difficult layer to capture with any consistency. Faced with this do engineers generally punt, i.e. only sample down to where the keys all respond similarly? If that's the case, then they are perhaps truncating much of the softer response from the sample set.

I also tried this with our RD-700NX through the speakers I just built and got -50 dB and -6 dB = 44 dB. This may be a good way to calibrate the DP & amp volume settings in order to make it sound the most realistic.

This is a bogus post to see if I can trick James into checking this thread, looking in vain for a Korg Kronos review.

Duped.

Ah, that sounds more like it. And, yes, to play pppianissimo is more difficult than hammering the keys...

Korg Kronos?


8 samples per key... Prove them wrong man!


Dewster has been dewsted...

Soon and very soon. Give us strength till we reach the other side.

You saw the preliminary results, no?

I checked into this with someone who use to work for GEM.

Piano 1 was indeed the Fazioli. (Maybe I should have sent PIano 2 as well, the Steinway... I might be able to get you that in the future, if you'd like to see it.) The rep explained your noted inconsistency this way:

"The pRP700/800 and the RP-X share the same reference sample data and physical modeling. Where the difference lies is in the post-EQ. The pRP800 has internal speakers and the EQ is set to enhance that speaker system. There are two EQ setting that are user switchable for use with the internal speakers or for external amplification. The pRP700 also has an equalizer at the end of the audio chain. Throughout production changes were being made to the EQ settings of the pRP700/800 as per customer input. The RP-X has a five band EQ as well which is user adjustable. My guess is that the units tested had varying EQ settings and effects settings which would correlate to discrepancies of the spectral analysis."


I'm not convinced the Yamaha p95 is single layer, as I hear a definite timbral shift at one point, much more abrupt than what would expect simply a velocity-based filter algorithm, which would tend to create a more gradual shift. I'll try to get you a recording of that at some point. But more to this particular issue, here is how GEM's approach qualifies as modeling, according to their manual:

"Unlike the velocity-switching methods used in other electronic pianos, Generalmusic’s unique FADE technology utilizes only one specially configured sound source per note." So yes, it is single layer. they continue:

"At the heart of the FADE engine is a extensive database which can be used to lookup the precise harmonic content of any note played at any velocity level. Whenever a note is played, the FADE engine analyzes the velocity of the key-strike and constructs, in real-time, a model of the necessary harmonic content for that particular note played at that velocity. The note’s sound source is processed by the FADE engine with appropriate harmonic content being added or subtracted accordingly." So it doesn't seem to be using a filter at all. The alterations to the harmonic content appears to be entirely based on a look-up table... that's the modeling, and that's different from using a filter, even if both are based on a single layer. I mean, for dynamic timbral changes to a note based on velocity, I think that using a look-up table to generate the harmonic content as opposed to just opening or closing a filter is pretty much the definition of a modeled approach, no?

Other than that, as you point out, they use modeling for key-down string resonance, pedal-down string resonance, and probably the key release string dampening.

Thanks! I've only had limited opportunity with identifying sample sets that have had EQ applied. I certainly haven't had enough experience to definitively say two sets aren't the same when there is no spectral match, but when there is a strong match I feel pretty comfortable saying they are the same.


Is this timbre shift audible to you in the DPBSD MP3 layer test for the P95? If not, perhaps this shift is more obvious on a note other than middle C? I could easily provide a MIDI file with the layer test targeting one or more different notes. Let me know, I would certainly be interested in analyzing this issue.


I think we're arguing semantics more than anything else here. Their process probably doesn't create harmonic content, otherwise it wouldn't require a sample as input. From their description it alters the harmonic content of a sample fed through it - which to me is a filter because that's what filters do. Filters can be simple or complex, and can certainly be controlled via computed look-up tables.

So when I say "filter" I don't necessarily mean a simple low pass 12/24 dB roll off filter like in a Moog. It could be that simple in the P95, but probably not in the GEM.

In the GEM I imagine the single piano sample is routed through a multi-band (FFT based with linear binning?) filter. A series of timed look-up tables set the gains for these frequency bands, with linear interpolation between successive table entries (per band). That's a guess - it could be something even simpler but probably not more complex. Depending on the width of the bands, a sound could be radically altered this way.

Since it is excited by a piano note rather than white noise (i.e. we aren't trying to make a piano note out of whole cloth, just alter it some) the bands can be quite a bit wider and fewer - and so there is strong inherent data compression to this approach. Which made a lot of sense when ROM was small and computers were slow and both were expensive. These days not so much, though I do see the merits of modeling for performance inputs that affect the piano model itself (e.g. pedal and key sympathetic resonance).

Korg Kronos Double Voice Review



Thanks to Piano World forum member "anotherscott" we have two DPBSD samples of the Korg Kronos - thanks! These samples are of the "German D" and "Japanese C" stock SGX-1 Premium Piano voices in the Kronos. In the Kronos literature Korg claims "Each uses superb, unlooped stereo samples, sampled at eight velocity levels for each and every key." - which certainly sounds like what many of us have been waiting for!

Though perhaps in a smaller, lighter, more piano-centric package. I see Korg neglected to include support for a music rest and decided to put the joystick in the keyboard area (both of which are Roland's wont). This seems to be the norm for a "workstation" class product, but I see these design decisions as rather unfortunate. Also, I'm not a huge fan of slide pots, or pots in general as digital input, because the physical position generally has no meaning once the patch is changed, and how position resynching is accomplished is invariably cumbersome - give me digital rotary encoders and dedicated LED displays instead.

MP3: http://www.mediafire.com/?pr5g4i0ymlcrt63 - German D
MP3: http://www.mediafire.com/?2q23t6lmjq9120c - Japanese C
PIX: http://www.mediafire.com/?e166zs7g0i2zng6 - analysis pix of both

anotherscott has this to say about the Kronos:

Some pix:

Figure 1. Spectral phase view of the pedal sympathetic resonance test, stimulus removed, normalized to -1dB to increase clarity, pedal down @ left, pedal up @ right, Japanese C. Because the effect is sampled, the sound is very realistic and pleasant.


Figure 2. Waveform view of the entire looping test, vertical zoom applied to see the noise floor, German D. Note decay is generally nice and long, though C1 on the German D poops out kind of early. The key-off "klunk" sound effect can be clearly seen here too (as well as heard, the default setting is much too loud).


Figure 3. Spectral pan view of the note C2, normalized to -1dB to increase clarity, German D. Fully sampled, no looping. Me likey.


Figure 4. Spectral pan view of the stretch test, mid notes, normalized to -1dB to increase clarity, German D. Here you can see what appears to be one stretch group of two on the left, and I believe there is another located in the higher end (not shown here). They don' jump out at me audibly, I don't think I would notice either without these visual cues. I don't see / hear any stretch groups with the Japanese C.


Figure 5. Spectral pan view of the layer test, compressed 20:1 to bring out detail, Japanese C. Visual evidence for at least seven layers (Korg says 8) with somewhat audible layer transitions. The most audible layer transition is the loudest one on the right, which sounds kind of tacked-on to me. The lowest velocity layer sounds somehow blended to me.


Figure 6. Spectral frequency view of the layer test, Japanese C. Some of the velocity layer groups are clearer here.

--------------------------
- Korg Kronos Japanese C -
--------------------------
FILE & SETUP:
- dpbsd_v1.9_korg_kronos_japanese_c.mp3
- Sequenced and recorded on Mac with ProTools LE, Mbox interface @ 24-bit, MP3 via Quicktime Pro.
- Sequenced and recorded by "anotherscott".
PROS:
- Passes the pedal sympathetic resonance test, the resonance is pleasant sounding.
- Passes the quick partial damping test, but perhaps overdamps.
- Passes the half pedaling test, but it is a subtle effect.
- key up/down "clunk" sound effects.
- Not looped, note decays sound very natural and long.
- Unstretched.
- Visual evidence for least seven somewhat blended velocity layers (Korg reports 8 layers).
CONS:
- Fails the key sympathetic resonance test.
- Fails the silent replay test at pedal-up.
- Key up/down "klunk" sounds are too loud (though they are likely adjustable).
- Audible velocity layer switch @ vel=46,60,74,88,102,116.
OTHER:
- Notes played @ vel=1 produce sound.
- Dynamic range 46dB (vel=1:127).
- MP3 levels: peak @ -1.5dB, noise floor @ -84dB.
- Right side needs approx. +3dB to correct stereo imbalance (recording issue, not problem with Kronos).
- Date reviewed: 2011-07-25.

------------------------
- Korg Kronos German D -
------------------------
FILE & SETUP:
- dpbsd_v1.9_korg_kronos_german_d.mp3
- Sequenced and recorded on Mac with ProTools LE, Mbox interface @ 24-bit, MP3 via Quicktime Pro.
- Sequenced and recorded by "anotherscott".
PROS:
- Passes the pedal sympathetic resonance test, the resonance is very pleasant sounding.
- Passes the quick partial damping test, but perhaps overdamps.
- Passes the half pedaling test, but it is a subtle effect.
- key up/down "clunk" sound effects.
- Not looped, note decays sound very natural and long.
- Unstretched (I see what may be two stretch groups of 2).
- Visual evidence for least six somewhat blended velocity layers (Korg reports 8 layers).
CONS:
- Fails the key sympathetic resonance test.
- Fails the silent replay test at pedal-up.
- Key up/down "klunk" sounds are too loud (though they are likely adjustable).
- A fair amount of timbre variation between the low notes, some sound muffled.
- Lowest note decay time is a bit short.
- Audible velocity layer switch @ vel=46,74,88,102,116.
OTHER:
- Notes played @ vel=1 produce sound.
- Dynamic range 42dB (vel=1:127).
- MP3 levels: peak @ -1dB, noise floor @ -87dB.
- Right side needs approx. +4.5dB to correct stereo imbalance (recording issue, not problem with Kronos).
- Date reviewed: 2011-07-25.

Looks like a keeper!

Thanks for the analysis, and to anotherscott (Kronos owner?) for the submission.

Cheers,
James
x

Yamaha CP50 Review



Thanks to Piano World forum member "jef_citron" we now have a DPBSD sample of the Yamaha CP50!

MP3: http://www.mediafire.com/?u83vpo1qu9abk84
PIX: http://www.mediafire.com/?9o8g1rmzj3dlpbd

jef_citron has this to say about the CP50:


Sympathetic resonance sounds OK in the first and second pedal resonance tests, but sounds really echoy in the half pedaling test. It could very well be the case that the short note loop time (0.5 seconds for the half-pedaling test note C4) is interacting with one or more of the delay times that comprises the pedal resonance effect.

The C5 note is kind of strange: the loop itself sounds somewhat thinner than the end of the attack, which makes the attack / loop crossfade fairly audible from a timbre standpoint.

I only see two velocity layer switches, and the lowest one is in a different location than on the CP1 and CP5. The CP1 and CP5 differ from each other in this respect as well. I probably don't have enough information to say difinitively, but it seems like Yamaha may be yanking layers with decreasing retail price of the models in this line, something I wasn't really expecting to find any evidence of:


Another explaination could be that they are just doing a better job of blending the layers on the less expensive DPs, but that strikes me as rather counter-intuitive.

A few (comparative) analysis pix (see the ZIP file for more):



Figures 1A, 1B, 1C. Spectral pan view of the stretch test, mid notes, CP1 (top), CP5 (middle), CP50 (bottom). Amplitude normalized to -1dB to bring out detail. No significant visible or audible differences, stretch groups are the same in all three models. (The CP50 image is offset 4 notes due to the way I captured the image, the sample set itself is not offset compared to the CP1/5).




Figures 2A, 2B, 2C. Spectral pan view of the layer test, CP1 (top), CP5 (middle), CP50 (bottom). Amplitude compressed 20:1 to bring out detail. There are fewer visible layer switch points across this line of DPs (for the note C4) as the retail price drops, which implies fewer layers. It looks to me like they are using one of the higher velocity layers in the CP1/5 to cover a larger range in the CP50.

---------------
- Yamaha CP50 -
---------------
FILE & SETUP:
- dpbsd_v2.0_yamaha_cp50.mp3
- Default patch 1 "CF Grand".
- Cubase AI => USB (MIDI) => WAV recorded on CP50 w/ thumbdrive, WAV=>MP3 via Cdex.
- Recorded by "jef_citron".
PROS:
- Passes the pedal sympathetic resonance test.
- Passes the quick partial damping test.
- Passes the late pedal partial damping test up to the test limit of 0.5 seconds.
- Responds to half pedaling.
- Has key up note damp samples or effect.
- Lowest notes have long decay.
- This is a blended 3 or 4 layer sample set.
- Visible layer switch @ vel=70,122.
CONS:
- Pedal sympathetic resonance sounds echoy.
- Fails the key sympathetic resonance test.
- Fails the silent replay test @ pedal up (note damps).
- Mid and high notes have somewhat short decay.
- Obviously looped, both visibly and audibly.
- Attack and loop sample lengths are fairly short.
- Attack sample lengths are (C1:C8): 2(?),2.1,2.1,1.8,1.7,1.3,0.8,? seconds.
- Loop sample lengths are (C1:C8): 0.8,0.8,0.6,0.5,0.4,?,?,? seconds.
- Audibly stretched over the low end.
- Stretch distances: 3,3,4,4,4,2,4,3,1(x26),2,3,2,1,1,1,3,2,2,3,4,2,3,3,3 = 49 groups.
- Audible layer switch at vel=122.
- No obvious pedal up/down "loom of strings" or key up "clunk" sounds.
OTHER:
- Notes played @ vel=1 produce a sound.
- Dynamic range 41.8dB (vel=1:127).
- Dampered | undampered transition: F#6 | G6
- MP3 levels: peak @ -12 dB, noise floor below -80 dB.
- Date reviewed: 2011-08-07

hey...im a new people here...i juz wan to know tht how far the yamaha's DGX-640 can reach for abrsm's exam?? can it reach grade 8?? thanks!!!!

http://usa.yamaha.com/products/musi...ries/dgx-640_color_variation/?mode=model

anthonyhor, your question is out of topic in this thread, you should open a new one.

IMHO, no DP I've tried will suit classical playing to reach ABRSM grade 8 well, but they are a very useful practice tool if you complement them with an acoustic one.

I've tried most brands (Yamaha, Kawai, Roland, Nord) but the Avantgrand.

Fatar Numa Piano Review



Thanks to Piano World forum member "voxpops" we now have a DPBSD sample of the Fatar Numa Piano!

MP3: http://www.mediafire.com/?ksxbn84es8ooqo5
PIX: http://www.mediafire.com/?692f5jcjkx9wwit

voxpops has this to say about the Numa Piano:


From a sound technology standpoint, the strong point of this DP is the looping, which (from my sub sample of 8 notes) seems to be pretty well done (as these things go). The attack samples are fairly long, the crossfades seem well blended, and most of the loops themselves are subtle sounding. I can hear looping in the midrange, but it isn't very prominent.

It has many weak points though. There is no sympathetic resonance of any kind, and most key / pedal interaction is not implemented. There are 5 velocity layers, but they are unblended, and the highest layer transition is rather jarring. With 36 stretch groups the sample set is fairly stretched, and some of the groups have left / right pan issues: several higher note stretch groups sound somewhat panned left, and one of the lower stretch groups sounds like it is panned right. It's probably just me, but issues with very basic things like L/R pan in the sample set leave me with a slight queasy feeling towards the entire implementation.

Many may not care for the white motif, but I really appreciate the location of the pitch and mod wheels (not in the keybed area), the inclusion of a music rest, the overall compactness, and the relative light weight of the physical design.

Some pix:

Figure 1. Spectral pan view of the pedal sympathetic resonance test, stimulus removed, normalized to -1dB to increase clarity, pedal down @ left, pedal up @ right. No pedal or key sympathetic resonance.


Figure 2. Waveform view of the entire looping test, vertical zoom applied to see the noise floor. Note decay is generally long, though the lowest notes could probably use more decay time.


Figure 3. Spectral pan view of the note C3, normalized to -1dB to increase clarity, cursor at end of crossfade. Fairly long attack and loop samples, with smooth crossfade. Looping sounds pretty well done.


Figure 4. Spectral phase view of the stretch test, mid notes, normalized to -1dB to increase clarity. High end is similar, low end is more stretched. Stretch group transitions are audible over the low and mid notes.


Figure 5. Waveform view of the entire stretch test (chromatic walk up the keyboard), left channel on top, right on bottom. Highlighted region indicates a lower stretch group with a right panning issue, you can also see indications of a left panning issue with some of the higher stretch groups.


Figure 6. Spectral frequency view of the layer test. Five unblended velocity layers.


Figure 7. Spectral phase view of the layer test, compressed 20:1 to bring out detail. The most audible layer transition is the loudest one on the right, which looks and sounds kind of tacked-on to me.

--------------------
- Fatar Numa Piano -
--------------------
FILE & SETUP:
- dpbsd_v2.0_fatar_numa_piano.mp3
- This is the first acoustic piano patch "Grand Piano" (firmware update pack 1.0).
- MIDI sequenced with Anvil Studio, laptop mic input, audio recorded & MP3 via Reaper.
- Recorded by "voxpops".
PROS:
- Notes have fairly long decay, though lowest notes could go a bit longer.
- Attack and loop sample lengths are long (as these things go) with nice crossfades.
- This is an unblended 5 layer sample set.
- Visible layer switch @ vel=24,40,62,104.
CONS:
- No pedal sympathetic resonance.
- No key sympathetic resonance.
- Fails the silent replay test @ pedal up (note damps).
- Fails the quick partial damping test (damps @ first pedal up).
- No response to half pedaling.
- Attack sample lengths are (C1:C8): 4.9(?),6.2,4.5,4.2,4.9,3.2,1.3,? seconds.
- Loop sample lengths are (C1:C8): ?,2.5,2.7,2.1,1.6,?,?,? seconds.
- Low & mid stretch group transitions are audible.
- Stretch distances: 5,3,4,5,5,6,4,3(x18),2 = 36 groups
- All layer transitions audible, particularly the highest.
- Fourth stretch group from the bottom [A1:C#2] audibly panned to the right.
- Many of the top stretch groups sound panned somewhat left.
- No obvious pedal up/down "loom of strings" or key up "clunk" sounds.
OTHER:
- Passes/fails the late partial damping test (no partial damping at all but late pedal is OK).
- Notes played @ vel=1 produce no sound.
- Dynamic range 59dB (vel=1:127).
- Dampered | undampered transition: F#6 | G6.
- MP3 levels: peak @ -0.5 dB, noise floor below -66 dB, ~5% DC offset.
- Date reviewed: 2011-09-06

Thanks for the review dewster.

voxpops, to clarify, is firmware update pack 1.0 the most recent available, and the update that made a considerable improvement to the sound?

Cheers,
James
x