Now (mostly) retired, I'm pursing electronics and computing just for the fun of it! I'm a computer scientist and engineer who has worked for AMD, Hewlett Packard and Siemens. I also taught hardware and software development at Case Western Reserve University, Tufts University and Princeton. Hopefully, you will find the information on this site to be helpful. Educators and students are particularly welcome!
Korg have announced the Nautilus AT music workstation. As the name suggests, Nautilus AT 88 and 61 have aftertouch. For some reason, the Nautilus-73 doesn’t get an AT version.
Well, of course, you can read all about it on Korg’s web site. Two points to be made here.
First, I was smacked by the first two statements on the Nautilus AT page: “NAUTILUS is KORG’s flagship workstation. The successor to the wildly popular KRONOS, …” Yep, Kronos is dead, long live Kronos. No point in pining away for a successor as Nautilus AT is it. No point wishing that Elway will return and put an end to Russell Wilson, either.
Kronos was an interesting build, being based upon a commodity Intel Atom motherboard. The weight and heat dissipation of the Kronos demonstrated the limitations of such an approach — essentially putting a mini desktop computer into a box with a keyboard. The Raspberry Pi-based models (e.g., Wavestate, OPSIX) are technologically more viable.
Second, Korg are finally doing what I’ve wished for a long time — upgrade your existing keyboard instead of discarding it:
Existing owners of 61 and 88 key NAUTILUS* need not miss out. KORG is rolling out an upgrade service that updates both the hardware and software of your keyboard, transforming your NAUTILUS into a NAUTILUS AT. For more information, and pricing of the upgrade service where you are, contact the KORG customer service team in your territory.
Let’s face it, not that much changes inside most New! Improved! synths. Usually the digital logic board is a new design, but the keyboard, display and other peripherals are largely the same.
Instead of dumping the old synth into a landfill, why not upgrade the electronics (or keybed) in the old platform?
Available exclusively from Korg UK to customers in the UK and Ireland, the service includes a hardware and software upgrade by a Korg service engineer alongside the collection and return of your Nautilus.
Must be nice to live on a small island. 🙂 I don’t think USA folk will get pick-up and return by a friendly Korg rep. The price quoted is £429 or about $560 USD depending upon currency fluctuations. If the hardware mod involves changing out the keybed, that’s a pretty reasonable charge.
Yamaha? Roland? Casio? Kawai? Nord? Are you watching? We are.
That is an impressive list of new product announcements. Clearly, Yamaha’s engineering and manufacturing teams were quite busy during the pandemic and global slow-down.
Yamaha have not yet updated all of its regional Web sites. If you can’t find the new Piaggero models on the USA site, check the European site.
Yesterday, Yamaha announced the second generation CSP series 200 digital pianos! I’m awaiting a P-515 successor — shouldn’t be long now that Yamaha have updated virtually every other digital piano offering.
Yamaha are reinventing the home digital piano. Yeah, every manufacturer offers a range of models from value-oriented entry pianos to mid-life crisis parlor toys. 🙂 The Yamaha CSP series are an interesting take on home players. The CSP front panel has exactly one button — the ubiquitous Yamaha FUNCTION button. The button is a minimalist’s gateway to the internal functions within, if you want them.
By minimizing front panel controls, customers aren’t confronted by a panel full of buttons, lights, knobs, sliders and other off-putting (and confusing) gizmos. A player can walk up to a CSP, turn it on and play a decent digital representation of an acoustic piano, both touch and sound. CSPs aren’t stripped down and have all the good stuff like CFX, Bösendorfer Imperial, Virtual Resonance Modeling (VRM) and GrandTouch.
A CSP piano really comes alive through the Smart Pianist app. In fact, the Smart Pianist app is required in order to make the most of the CSP’s capabilities. The CSP is designed for people who want to learn piano and have fun doing it. A matrix of LEDs above the keys create a waterfall display showing when and where to strike the keys. Smart Pianist knows the score [pun] and controls all of this. When the player is ready for reading, Smart Pianist displays the score. Thanks to Chord Tracker and other software smarts, Smart Pianist can generate a score from audio. Thus, Smart Pianist and CSP gamify the experience of learning piano.
Play-along is an important aspect of the CSP approach. I don’t know about you, but I have the most fun playing along with other people, a backing track or creating my own backing through auto-accompaniment. Yes, the CSP has auto-accompaniment with a zillion styles.
The new CSP models are CSP-295GP, CSP-295, CSP-275 and CSP-255. Judging from the initial pricing, I expect these models to replace the CSP-150 and CSP-170.
I read through the data list PDF. The new CSPs have more voices and styles than the Yamaha DGX-670. In terms of voices, drum kits, styles and chord recognition, you have all the main elements of a Yamaha mid-level arranger (PSR-SX700 and PSR-SX900). In this respect, the new CSP models are a terrific value and, frankly, I’m jealous! I doubt if the forthcoming P-515 successor will have such outstanding non-piano voices.
If you own an arranger keyboard, you’re already set although you probably don’t have Bösendorfer, GrandTouch, VRM, streaming LED lights or Smart Pianist. You might think that the CSP is lacking for style control buttons and such, but please consider this notion — Yamaha are evolving the CSP and other digital pianos to be “adaptive instruments.” An adaptive instrument follows chords across the entire keyboard (i.e., AI Full Keyboard fingering) and changes style sections depending upon your playing strength and the number of notes you play (i.e., Adaptive Styles). Brilliant! The instrument should and can follow the player instead of the other way around. Who needs buttons?
The Yamaha P-S500 has many features in common with the CSP series including the streaming light note display. The P-S500 puts it all into a quasi-portable form factor. I would jump at the P-S500 except for its Graded Hammer Standard (GHS) keyboard. After playing P-515 (NWX action), I just can’t return to GHS. I need to switch to a for-real Petrof acoustic grand at church and I don’t want to compromise. [Stubborn me!]
I’ve collected links to my remarks about CSP Gen 1, DGX-670, etc.:
Last week, Yamaha announced two new members of the P-series portable pianos: Yamaha P-145 and P-225. The Australians got to roll out new models first this time around. The P-145 replaces the old P-45 and the P-225 replaces the old P-125.
Both pianos feature a new slim, minimalist design with forward-firing internal speakers. Yamaha must have observed Casio’s success with slender slabs and decided to join the party. The slim design reminds me of my first electric “piano”, the Crumar Roadrunner. 🙂
The dawn of Yamaha P-225 — touch the GHC monolith!
Unlike the Roadrunner, the new P-models sound darned good. The P-225 features the CFX grand and the P-145 features the CFIIIS. The CFIIIS samples are tried and true, having finally trickled down to the entry level. The P-225 CFX is warmer and has more depth (to my ears) than the CFIIIS. The P-225 also has VRM Lite and key-off samples. Polyphony is 256 (P-225) versus 64 (P-145). The P-225 is further enhanced by “Wall EQ”, Intelligent Acoustic Control (IAC), and Stereophonic Optimizer.
Speaking of outputs, both have two stereo headphone jacks. The P-225 adds left and right AUX OUT jacks. Amplifier output is 7 Watts per channel for both models. The P-145 has two oval 12cm by 8cm speakers. The P-225 has two oval 12cm by 8cm speakers and adds a 5cm high frequency driver.
The P-225 has 24 voices to the P-145’s 10 voices. The voices cover the usual range of Yamaha tones: acoustic piano, electric piano, pipe organ, strings and pads. However, things do get interesting! There are actually four new models: the 143/145 pair and the 223/225 pair.
The 143/145 have two grand pianos, a tines (Rhodes) electric piano, an FM (DX) electric piano, strings, principal pipe organ and tutti pipe organ. The 143 has two harpsichords and vibraphone. The 145 has one harpsichord, accordian and Di Zi. Di Zi is a Chinese transverse flute. So, depending upon your region, you will be getting either the 143 or the 145!
The 223/225 pair have different voice groups and voices, too:
P-223 has an “Others” voice group and 225 has a “CLV/VIB” group. Guzeng is Chinese zither. Strange that 145 has a flute and the 223 has a zither. Yamaha have clearly targeted models for Asia and models for “Western” regions. Gotta make a buck somewhere…
User interface on both models is the minimum. Neither have a display. Settings are made by holding down front panel buttons (METRONOME+RHYTHM) and striking the appropriate key. There is some feedback like a voice announcing “on” or “off”. Frankly, I can’t see doing this at a gig, especially a church gig when absolute silence is expected. OK, for home use.
The P-225 has a bunch of other extra features over the P-145: Bluetooth, recording, split voices, etc. Good news — Yamaha brought back the USB audio interface functionality that went missing in the P-125A. Both models have Smart Pianist and Rec’N’Share support.
The big news is the new Yamaha Graded Hammer Compact (GHC) action. GHC replaces the Graded Hammer Standard (GHS) action in the previous models. The reduced front-to-back depth of GHC allowed Yamaha to design and deliver a slimmer slab piano. (Front-firing speakers help reduce depth, too.) Since none of us have played GHC (as yet), it’s impossible to comment. However, I wonder if we will see GHC in other Yamaha products like synths or arrangers?
My heart and prayers go out to family, friends and other folks in Ludlow, Cavendish, Perkinsville and Weathersfield VT. I just spent several days in Ludlow and the surrounding area.
Today, Ludlow and the region have been devastated by heavy rain and flooding. The area was struck hard by Hurricane Irene (August 2011). The steep canyons channel rain water, which quickly rises to flood level.
I drove Vermont 131 between Ludlow and Perkinsville quite frequently over the last few days. Today, the road is flooded out and blocked in many locations.
Please keep these communities in your prayers and send assistance.
It should go without saying — the Leslie rotary speaker sound is critical to getting a good drawbar organ sound.
The Montage/MODX and Genos/PSR product lines have two rotary speaker effects in common:
MODX effect name Genos effect name ---------------- ----------------- Rotary Speaker 1 ROTARY SPEAKER 1 Rotary Speaker 2 REAL ROTARY
Yamaha’s naming scheme often makes it difficult to match up algorithms across product lines. Rest assured, however, the algorithms (and code) are the same. If you need to double check yourself, simply match up the effect parameters as listed in the Data List PDFs.
It’s like the engineers deliberately chose bad, meaningless variable names. Throw preset names on top of the algorithm (effect type) names and things get really confusing!
Rest assured, a lot of the information discussed in this post can be applied to Yamaha synthesizer and arranger products alike.
The real thing: mechanical
Ain’t nothin’ like the real thing, baby. A real Leslie speaker moves air and throws it around in a 3D space. A “2D” stereo simulation will never do. Further, rotary speaker simulators model a mic’ed up speaker putting strong energy surges into the left and right channels.
We all know that a Leslie speaker, like the 122, has a rotating horn and rotor. Yamaha sometimes refer to the rotor as a “woofer”, so keep that in mind when reading through effect parameters!
The horn and rotor each have a motor. The motors have two speeds when turning: slow and fast. A pulley and belt system transfers rotation from a motor to the horn or rotor. The horn motor has three different sized-pulleys: small, middle, and large. The belt is usually around the middle pulley, AKA the factory setting. If the belt is around the small pulley, the rotor turns slower. Put the belt around the large pulley and the rotor turns faster (relative to the factory setting, of course.)
You should keep these speeds in mind when tweaking parameters, if you want authenticity. Please note that the rotor turns more slowly than the horn.
I gave the nominal speeds in both rotations per minute (RPM) and cycles per second (Hertz). Some algorithms need RPM and other algorithms need Hertz. Here are conversion formulas:
RPM = Hertz * 60 Hertz = RPM / 60
The pulley and belt system causes even more fun. The belt is flexible and slips around the pulley. Belt tension and wear determine slip. Tension (slip) has a greater effect on acceleration (change from slow to fast) than deacceleration (change from fast to slow). If you want authenticity, acceleration time should be shorter than deacceleration. In other words, the pick-up rate is higher than the slow-down rate.
The real thing: electronics
The Leslie 122 tweeter and woofer are driven by a three tube power amplifier through a cross-over network. The classic Leslie crossover frequency is 800 Hertz. The crossover filter is not super steep and there is definite frequency bleed beyond the crossover frequency.
The power amp consists of a 12AU7A tube driving dual 6550 power tubes. The 12AU7A belongs to a family of nine pin, twin triode tubes. The chief difference between family members is the gain factor. The 12AU7A has a gain factor of 20 while the more powerful 12AX7 — the most familiar member of the family — has a gain factor of 100.
The power amp has a gain knob. At about 70%, the power amp starts to distort. Oh, never, ever go past 70%. Ever. 🙂
The sims
With that background in mind, let’s take a look at the Yamaha MODX rotary speaker effect algorithms.
Yamaha MODX Rotary Speaker 1
Historically, “Rotary Speaker 1” came first. The following table summarizes the Genos parameter values for the “Dual Rotary Speaker Bright” and “Dual Rotary Speaker Warm” presets:
# Parameter Bright Warm -- -------------------- -------- ---------------- 1 Woofer Speed Slow 40.2rpm 40.2rpm 0.67Hz 2 Horn Speed Slow 45.6rpm 45.6rpm 0.76Hz 3 Woofer Speed Fast 383.4rpm 363.6rpm 6.06Hz 4 Horn Speed Fast 403.8rpm 403.8rpm 6.73Hz 5 Slow-Fast Time Woofer 39 45 6 Slow-Fast Time Horn 7 7 7 Drive Low 35 31 8 Drive High 37 36 9 Low/High Balance L<H4 L13>H 10 11 EQ Low Frequency 100Hz 100Hz 12 EQ Low Gain +8dB +8dB 13 EQ High Frequency 1.0kHz 1.0kHz 14 EQ High Gain -3dB -4dB 15 Mic L-R Angle 177deg 177deg 16 Speed Control Slow Slow
Feel free to borrow the Genos settings for MODX (and vice versa).
The horn and woofer speeds are ballpark with respect to the factory settings. If there is one major complaint with this algorithm, it’s the relatively weak drive effect. Increasing drive does not add distortion. On MODX, be prepared to couple “Rotary Speaker 1” with an amp simulator (e.g., STEREO SMALL or MULTI FX).
Obviously, there are a lot of parameters to tweak: microphone angle, equalization, rotor and horn balance (low/high balance). Imagine yourself as a studio engineer mic’ing up a real Leslie.
Just for grins, the following table summarizes rotary speaker parameters for four MODX presets:
# Parameter Basic Horn Mic Light Heavy RTR -- -------------------- -------- -------- -------- --------- 1 Rotor Speed Slow 0.88Hz 0.80Hz 0.88Hz 1.01Hz 2 Horn Speed Slow 1.30Hz 0.72Hz 1.30Hz 0.93Hz 3 Rotor Speed 6.06Hz 6.06Hz 6.06Hz 6.06Hz 4 Horn Speed Fast 7.07Hz 6.73Hz 7.07Hz 6.73Hz 5 Slow-Fast Time Rotor 40 40 40 64 6 Slow-Fast Time Horn 13 12 13 33 7 Drive Rotor 52 37 26 28 8 Drive Horn 31 29 21 22 9 Rotor/Horn Balance RH 10 11 EQ Low Frequency 1.0KHz 1.0KHz 1.0KHz 1.0KHz 12 EQ Low Gain 0.0db -1db 0.0db 0.0db 13 EQ High Frequency 2.0KHz 2.0KHz 1.0KHz 2.0KHz 14 EQ High Gain 0.0db +1db 0.0db 1.0db 15 Mic L-R Angle 150deg 177deg 180deg 30deg 16 Speed Control Slow Slow Slow Slow
Enough starting points yet? Please don’t be afraid to tweak, listen and evaluate. It’s only bits and if you don’t like what you’re hearing, then throw the bits way away. Ever wonder why people spend so much time in the studio? That’s why!
Yamaha MODX Rotary Speaker 2
The MODX “Rotary Speaker 2” algorithm is the newer algorithm. It first appeared in Montage and Genos, and it’s been trickling down to mid- and low-range Yamaha products. In Genos-land, it’s known as “REAL ROTARY”. This algorithm provides control over both acceleration and deacceleration times and drive level/tone.
Here are settings from Genos for three drawbar organ voices:
# Parameter Jazz Gospel Rock -- ------------------- -------- -------- -------- 1 Speed Control Slow Fast Fast 2 Drive 2.0 4.0 10.0 3 Tone 8.2 10.0 10.0 4 Low/High Balance L<H10 L<H9 L<H9 5 Output Level 100 100 100 6 Mic L-R Angle 180deg 120deg 120deg 7 Input Level +6dB +6dB +6dB 8 Modulation Intensity 0 63 63 9 10 11 Slow-Fast Time Horn 1.19 1.13 1.13 12 Fast-Slow Time Horn 0.86 0.97 0.97 13 Woofer Speed Slow 43.5rpm 43.5rpm 43.5rpm 14 Horn Speed Slow 47.3rpm 47.3rpm 47.3rpm 15 Woofer Speed Fast 413.8rpm 403.7rpm 413.8rpm 16 Horn Speed Fast 474.4rpm 464.3rpm 464.3rpm 17 Slow-Fast Time Rotor ? ? ? 18 Fast-Slow Time Rotor ? ? ?
Unlike “Rotary Speaker 1”, REAL ROTARY brings overdrive. Turn up the drive to add distortion. Tone sweeps from darker to brighter.
No doubt, you noticed values missing in the last two rows. The Genos user interface supports only 16 effect parameters! [Genos engineers need to fix this limitation.] Your guess is as good as mine — maybe 1.22 and 1.86?
Other blog posts about Yamaha MODX drawbar organ sound design:
Let’s take a brief look at the electronics inside the Kawai MP11 digital piano.
The Kawai MP11 is a pro-level 88-key stage piano with Grand Feel wooden key action and triple sensor detection. The MP11 has 40 voices total including Kawai concert and studio acoustic pianos, electric pianos and the usual motley crew of non-piano voices. Acoustic pianos are enhanced by Kawai’s Harmonic Imaging™ XL technology. 23 different kinds of effect can be applied. Electric piano gets special treatment with five amp simulators: Suitcase, Marshall stack, Jazz Chorus, Bassman and Leslie. The MP11 Virtual Technician allows control of tonal character, stereo width, string resonance, etc.
The MP11 has pitch bend and mod wheels, key set-ups (zones), assignable knobs, and many other features that make it suitable as a controller. It also has sought-after features like built-in power supply (no wall wart), 5-pin MIDI, XLR audio OUT.
That’s one respectable stage piano!
Having crawled through many service manuals before, the internals are not too surprising. There is one main digital electronics board surrounded by several peripheral boards for the LCD (128×64), USB interface, front panel gizmos, audio amplification, and various external connections. There are separate boards for handling AC main power.
Kawaii MP11 block diagram [Kawai service manual]
I’m most interested in the on-board compute and, thus, I will focus on the main logic board. Like other high-end digital synthesis products (e.g., Yamaha Montage), there are multiple hardware processors. The MP11 has three main processors:
Main CPU (MPU): Renesas SH7203
Tone Generator (TG): unspecified by the service manual
Effects (DSP): unspecified
Unfortunately, the Kawai service manual does not provide a detailed schematic or electrical parts list. Thus, the TG and DSP type are unspecified by the service manual. The DAC and ADC are unspecified, too. A fourth processor, Renesas 8-bit MPU M38K07M4, handles USB to HOST duties.
The MPU is the well-known Renesas SH7203, a 32-bit microcontroller incorporating an SH2A FPU RISC core. There are two external clocks: 48MHz and 16.5MHz. The 48MHz oscillator is probably the USB clock. The 16.5MHz clock is multiplied internally according to the software-configured frequency control register. The SH7203 can operate up to 200MHz internally.
The MPU has two 32 MByte SDRAMs (type unspecified), for 64 MBytes total. The MPU shares a 4 MByte flash ROM with the TG and DSP processors. Data moves in 16-bit words.
An arty, notional Kawai K023-FP IC [Kawai]
The tone generator, TG, has a 33.8688MHz external clock. The clock is an even multiple of 44.100kHz, the outgoing sample frequency. I suspect that the DAC and ADC employ the standard I2S serial communication protocol for digital audio. Thus, TG has 768 clock cycles to knock out a single N-bit sample to go to the DSP effect unit or the DAC.
Tear down videos such as the Kawai CN23 disassembly reveal Kawai proprietary tone generation integrated circuits (IC). The IC in the CN23, for example, is a massive, 208-pin beast!
TG has a small working SRAM (512MB, 8-bit data path, type unspecified). TG has two dedicated flash ROM devices (type unspecified) for TG program and waveform (sample) storage: 256 MBytes and 64 MBytes. Waveform data moves on 16-bit paths.
This MP11 repair video shows a large custom IC: Kawai K023-FP. There is another custom IC on the main logic board. Which one is sh7203? TG? DSP? It’s clear that Kawai design their own ICs.
The DSP unit has a relatively leisurely 6MHz clock. The DSP unit receives digital audio from the TG and returns it to TG, DSP does not have a direct connection to the DAC.
The MP11 has a LINE IN for external audio. The LINE IN is buffered and then converted to digital audio by the ADC (type unspecified). The digital audio stream goes to TG where it is mixed digitally with synthesized audio.
Armed with other tear down videos and resources on the Web, I’ll try to identify these unspecified MP11 components.
Main CPU: Renesas 72030W200FP (package QFN-240, 240 pins)
MPU SDRAM: Elite Semiconductor Memory Technology ESMT M12L25616A-6T 256Mbit SDRAM organized as 16M x 16-bit words
TG: Kawai K023-FP identified in another Kawai service manual as “Sound Source/Touch LSI”
Flash ROM: Generic NOR flash organized as N x 16-bit words (e.g., Spansion S29GL01GP12TFI020); Flash ROM ICs have Kawai labels identifying pre-loaded content
TG SDRAM: Infineon CY7C1049DV33 4Mbit fast async SRAM organized as 512K x 8-bit bytes
DSP: Texas Instruments TMS320VC5507 Fixed-point Digital Signal Processor
As to DAC, Kawai have used the Asahi Kasei AKM 4490EQ 32-bit DAC and Texas Instruments PCM1795 in previous products. I wonder how Kawai have coped while AKM recover from their factory fire? Perhaps they are using the ESS Technology ES9028Q2M 32-bit DAC instead?
If you see a photo of the Kawai KEP-339 main board, you are probably looking at the top side only. Please keep in mind that the bottom side is also populated. You will find the TI TMS320 DSP on the bottom side.
Folks are always surprised by the modest compute resources in digital keyboards. They expect to see giga Hertz clock speeds, gobs of memory and solid state storage devices (SSD). Synth engines are so-called embedded systems, not desktop workstations or laptops. The digital electronics need to run cool without heavy heatsinks and fans. In order to achieve power/heat goals, clock rates are low and basic memory interfaces are employed instead of PC busses such as SATA.
Long-term readers know that my mind will eventually turn to tear-downs and electronics. So it goes with digital pianos.
Thanks to the Piano World Forum, I found two links of interest to deep divers.
Inside Yamaha P-515
First up is a long-ish Yamaha P-515 repair video. I’ve got to say, tearing down a P-515 is not for the faint-hearted. We owe Late Night DIY much gratitude for taking the challenge.
I’d loved to include some screensnaps here, but Late Night’s camera work is a little shaky. (Dude shot the video after a late night gig.) Here is a list of things to watch for:
5:14: Particle board base.
6:57: The AJACK board with external connectors.
7:51: The keybed.
8:17: Rubber contact strips.
9:47: Contact PCB traces.
11:10: Key removal.
24:11: Speaker box.
25:16: AJACK, again. DM board is underneath.
27:07: Top of speaker box.
These camera shots should give you a pretty good tour of the insides without going through the whole disassembly and repair narrative.
Disassembly tips: Take pictures while ripping things apart. Mark holes on printed circuit boards. Do not put screws and other hardware in speaker cones!
A few observations. First, look at the size of the particle board base! The wood base provides strength and rigidity for the NWX keybed. The base must weight at least ten pounds. Think about that when hefting your P-515 around.
The AJACK printed circuit board (PCB) has all of the MIDI and audio connectors. The main PCB — the digital logic (DM) mainboard — is underneath the AJACK. I dearly wish we had pictures (top and bottom) of the DM mainboard.
The rubber contact strips beneath the keys have three nubby buttons for each key. The nubby buttons make contact with the PCB contact traces (those squiggly squares). This is the so-called “triple sensor.” Other than three contacts instead of two, the basic hardware tech ain’t different from digital synths. The hardware “sensor” isn’t that sophisticated as all of the smarts are in the scanning software.
The speaker box is cool. I wouldn’t be surprised to see it inside of P-S500, too. Yamaha reuses components to cut costs and spare parts inventory.
The speaker box and NWX key designs are probably patented. I am surprised that the Piano World folks haven’t exploited the U.S. PTO patent database…
More than a look, this is one of the most gutsy tear-downs that I’ve seen. Guy buys an expensive CLP-685 and literally takes it apart out of curiosity. I usually play with my toys before taking them apart and I rarely take apart anything over $100 USD retail. That’s some Chinese chutzpah!
This article has a lot of terrific pictures (some a little blurry, unfortunately). There are good close-up pictures of the tan power supply boards and the DMH main digital electronics board.
Yamaha CLP-685 digital piano main board (DMH)
I wish the DMH board legends were legible and I wish that we had a picture of the other side, too. Plain as day is a Yamaha SWX09 (YH621A0) integrated circuit. The SWX09 integrates the main CPU, tone generation and digital effects onto a single chip. I’ll be thinking about this more in days to come, especially the “empty” board real estate. The SWX09 has nearby companions. The Winbond W9864G6KH-6 4M by 16-bit 166MHz SDRAM is DSP working memory for the SWX09. The two labeled ICs are wave ROM. The rest of the components are mounted on the bottom of the board. We see only the components mounted on the top side of the board.
A Yamaha GrandTouch key
The Piano World Forum folk are positively obsessed with key design and action. Yamaha describe the CLP-685 keybed thus:
I see a lot of plastic, a metal weight, and a little decorative wood. I’m not hung up on wood vs. plastic as even Einstein would say “Dead mass is dead mass.” If it plays good, why worry? The only concern I have about anything is long-term durability (e.g., those rubber contact strips and dirt).
Hope other musical electronics nerds have found this exposition interesting. 🙂
Yes, I played one! The pandemic has let up to the point where I can drag my old bod to music stores, again. Fortunately, I haven’t worn out my welcome with local shop owners and clerks (yet).
I don’t intend to run down the DGX-670 features in detail. The Yamaha DGX-670 has been on the market for two+ years and you can find all the details on the Yamaha Web page. Yamaha position the DGX as a “portable grand piano,” part of the “P” line of digital piano products.
And, as of late, my primary interest is in a digital piano that will help me to sharpen my piano skills. I want those skills to translate to the acoustic piano (Petrof acoustic grand) at church.
In that regard, the DGX-670 does not disappoint. The action is Graded Hammer Standard (GHS) and the main piano multi-sample is Yamaha CFX. The 670 has Yamaha’s Virtual Resonance Modeling (VRM) which “reproduces the complicated interaction between both string and soundboard resonance.” The CFX sounds very good through the in-built amplification and speaker system: 2 x 6 Watt amplifier, 2 x (12cm + 5cm) speakers. If I have one immediate slam on the DGX-670, its front panel legends are difficult to read in poor light (black model).
Yamaha DGX-670 digital piano decked out in white
In short, the DGX-670 has me asking, “Do I really need to move up to the Yamaha P-515 digital piano?” This question is more complicated than it sounds, especially when you roll the Yamaha CK88 into the analysis!
The DGX-670 street price is $850 USD versus $1,600 for the P-515. The store which hosted my adventure was prepared to discount the DGX-670 to $750 — that’s half of a P-515. That’s half of a Yamaha CK88, too ($1,500 MAP).
It comes down to personal musical goals — and desire. 🙂 Let’s take each alternative separately.
The P-515 has the Natural Wood X action with escapement (NWX) and the glorious Bösendorfer Imperial. That’s a lot to argue against. The P-515 sound system is more capable: 2 x (15W + 5W biamplified) amplifier, and 2 x (12cm x 6cm) woofer and 2 x 2.5cm dome. The DGX-670 has a very nice 4.3 inch (480 × 272 dots) LCD display that adds a splash of color to Yamaha’s drab black. The P-515 display has less resolution (128 x 64 dots) and is monochrome. [Yamaha should adopt the color display for the P-515 successor].
Those P-515 characteristics would be easy to live with. My biggest beef with the P-515, however, is the paucity and lower quality of its non-piano (AP or EP) sounds. The DGX-670 outstrips the P-515 for non-piano sounds. I quickly compared the DGX-670 voices against my once-beloved, owned and played PSR-S950. The DGX-670 is (roughly) a PSR-S950 without the Organ Flutes drawbar organ. I know these voices and would be very happy to have them in my piano instrument.
As to auto-accompaniment, the DGX-670 is “arranger lite.” It has styles and chord recognition. The styles are now represented in the Style File Format Guitar Edition (SFF GE) form. SFF GE makes the DGX-670 compatible with styles from the mid- and upper-end Yamaha arranger keyboards. A virtual ocean of styles are available at the PSR Tutorial web site as well as a growing community of DGX-670 players on the PSR Tutorial forum.
Yamaha do not say much about P-515 auto-accompaniment other than it’s bass plus drum, and that it follows chords. What is P-515’s chord recognition technique? Is it similar to full keyboard, A.I. recognition? Wish I knew more about this aspect of the P-515…
Regular visitors know that I test drove the Yamaha CK88, too. The CK88 is positioned in Yamaha’s stage keyboard product line. The sound system is comparable to the DGX-670: 2 x 6 Watt amplifier and 2 x (12cm x 6cm) speakers. The CK88 has the lower resolution (128×64 dots) monochrome display.
Piano-wise, the CK88 has the CFX multi-sample and GHS keybed. No VRM. No Bösendorfer. The CKs are well-equipped for drawbar and pipe organ. (The pipe organ multi-samples originated on Genos™.) The DGX-670 — and P-515, for that matter — are relatively deficient in the drawbar and pipe organ department. The CKs have a good selection of other non-piano voices. No Super Articulation. The CKs do not have auto-accompaniment of any kind.
So, if a player doesn’t care about drawbar organ and wants VRM, why not DGX-670 at half the price of a CK88?
At this point, I would be remiss to not mention slab weight:
Yamaha’s engineers did a remarkable job of slimming the CK88 down to 28.8 pounds. The respectable GHS-equipped P-125 is 26 pounds and I doubt if Yamaha can design a lighter, robust, 88-key GHS keyboard.
I neglected the whole issue of connectivity and app support. Your mileage will vary. The DGX-670 does not have 5-pin DIN MIDI. That might be a deal-breaker for some folks. Smart Pianist does not presently support CK88. Look to Yamaha Soundmondo, instead.
I didn’t work the Yamaha CP88 into the analysis. After its updates, the CP88 is a virtual library of pianos! It’s also $2,600 USD (MAP), $1,000 higher than the three digital pianos in my analysis.
Finally, why did I enjoy playing the DGX-670 GHS and not so much the CK88. Yamaha swears that the GHS action is the same everywhere. If there is an external factor, perhaps it is the insane way trial pianos are set-up in stores? Often, I feel like a contortionist. (Forget reading sheet music in some settings, too.) Playing piano is, after all, a physical act.
I’m trying to recreate the experience of playing the new Yamaha CK keyboards with MODX. A bird in the hand is worth two in the bush…
I received a few comments and ideas that are worth passing along. All of the basic principles and sonic DNA are applicable to Genos and PSR, I might add.
First up, what makes a good overdrive or distortion? My colleague Uli is pursuing a similar investigation and he is using an Electro Harmonix Lester K as a reference. Using a reference sound is a super idea.
So, I dug into my old bag of tricks — literally. I have four different references on hand:
ART Tube MP
Electro Harmonix Lester K
Behringer GDI21 guitar preamp
Fender Super Champ xD modeling guitar amp
As yet, I haven’t tried the Super Champ xD because it is a solid state and tube hybrid. I will need to find a way to isolate the effect of the 12AX7/6V6 power stage. That experiment will likely require high volume and there’s only so much my poor ears can take! [Be sure to protect your ears.]
I bought the ART Tube MP mic preamp a zillion years ago. Tube MPs are still inexpensive going for roughly $100 USD. I used the Tube MP to warm up the sound of my old Nord Electro 2, trying to file the edge off of its digital sound.
The Tube MP is really intended as a mic preamp having a 12AX7 vacuum tube gain stage. If you crank up the gain, you can hit the 12AX7 pretty hard and get gobs of distortion. The Tube MP distortion gradually increases with gain and is warm and smooth. By “smooth”, I mean distortion which is not grainy or fizzy. To me, grainy distortion has an uneven clickiness to it, like a fast-ticking geiger counter instead of an old analog TV receiving the sounds of the universe. [Really.]
That said, the Lester K overdrive — a digital simulation — is not bad. The Lester K overdrive gets good reviews and deservedly so. The Lester K overdrive has a pleasing smoothness.
The Behringer GDI21 is sometimes disparaged as a copy of the Tech21 SansAmp. Compare schematics and you will find differences. The GDI21 uses a pair of JFETs to simulate a tube preamp (12AX7).
Both GDI21 and SansAmp emulate the input stage and power amp effects of a guitar amp. There are lots of variations to dial in, but you pretty much get three flavors: Fender Tweed, Mesa/Boogie high-gain and British lead. Maybe the GDI21 is authentic on guitar, but I didn’t care for its sound when applied to drawbar organ. The Tube MP and Lester K are better references, IMHO.
Given all that, what is my favorite MODX (Genos) amp sim (or whatever) for overdrive? Here are my top three picks:
SMALL STEREO (Preset: Overdrive)
US COMBO (Preset: Rich clean)
BRITISH LEAD (Preset: Dirty)
Judging from its distortion parameters, the SMALL STEREO is part of the MULTI FX pedal board chain. Thus, if you want to slap pedal effects on a drawbar organ, MULTI FX is a good way to go, giving you a pedal or two for free along with the amp sim. Uli is experimenting with chorus, too, and is giving MULTI FX a try.
The presets mentioned above are just starting points. Check out other settings in Part 3.
I grew up with the sound of 1960s Fender and that’s probably why I have a fondness for the US COMBO. I always wanted a Fender Twin Reverb (or Bassman) as a kid. Well, THAT dream never came true. [Might as well add a Vox Continental to the list of broken dreams, too.] 🙂 Twin and Bassman amps were the weapons of choice for combo organ back in the day.
The BRITISH LEAD conjures the sound of Wakeman, Emerson and others. Nuf said.
Yamaha have announced the Stagepas 100/100BTR Portable PA System. The 100BTR model has a built-in lithium-ion battery pack. It’s a good size and weight for busking or stage monitoring. The straight stuff:
100W Class-D amplifier
6.5″ woofer plus 1.4″ tweeter
Frequency response: 70 Hz – 20 kHz
Coverage area: 90º H x 90º V
3 channel mixer, each with level control
2 band EQ (low/high shelving)
2 x mono MIC/LINE inputs (Neutrik) plus 1 3.5mm LINE input
