Explicit Sax

Hope you got your Motif XF. The current stock is gone, gone, gone.

Comparing waveforms (Montage vs. XF) got me interested in the tenor sax samples and voices. The Yamaha MOX has the basic tenor sax samples (Med, Hard, Growl) while the Motif XF rounds out the set with Soft dynamic samples and falls. The XF (and MOXF) showcase the tenor sax in the “Tenor Max” preset voice.

Since I was curious to discover “what I’m missing,” I deconstructed four tenor sax patches on the MOX. Also, I compared the MOX voices against the Super Articulation tenor sax voices on the PSR-S950 arranger workstation in a listening test. The A/B test was enlightening as the MOX and S950 use the same waveforms — at least to my aging ears! The S950 triggers the samples using Super Articulation (SA) rules while the MOX triggers the samples using Expanded Articulation (XA) rules. Rules aside, you get to the same sonic place.

With XA, there are three main ways that samples are selected and triggered:

  1. Normal: Triggered when keys are played in the regular way.
  2. Legato: Triggered when Mono/Poly mode is Mono and keys are played in a legato manner, i.e. one or more keys are held while a new key is struck.
  3. AF1 and AF2: Triggered when either AF1 is ON, AF2 is ON or AF1/AF2 are both OFF and a key is struck. (The switch states are exclusive.)

See the Yamaha Synthesizer Paramater Manual for all the gory details. XA and SA differ in the amount of automated decision making made by the control software. SA is more automated and XA is more manual, giving the player more explicit control over articulations.

First up is the PRE5:008(A08)Tenor Dynamic AF1 voice. The AF1 and AF2 buttons are assigned in the following way:

    AF1: Mono/Poly mode
    AF2: FEG-D1

The AF1 and AF2 buttons do not affect sample selection in this voice other than putting the keyboard into Mono mode or Poly mode. Thanks to this simplification, it’s a good place to start ‘splaining.

The table below gives the waveform, key range, velocity range and volume level for each element in the patch.

    Elem#  Waveform            XA      Notes   Veloc  Level
    -----  ------------------  ------  ------  ------ -----
      1    Tenor Sax2 Med      Normal  C-2 G8  1   60   110
      2    Tenor Sax2 Med      Normal  C-2 G8  61  90   110
      3    Tenor Sax2 Med Of   Legato  C-2 G8  1   90    86
      4    Tenor Sax2 Hard     Normal  C-2 G8  91 127   120
      5    Tenor Sax2 Hard Of  Legato  C-2 G8  91 127    95
      6    Small Tabla Dom     Legato  C4  G8  1  127    52
      7    Small Tabla Dom     Legato  C-2 B3  1  127    78
      8    Sine                Legato  C-2 G8  1  127    78

The element levels are programmed to even out the perceived loudness across waveforms. Of course, there are many parameters for each element beyond what is shown in the table. For example, each dynamic level (velocity range) has its own filter and amplitude characteristics. There may even by a little velocity-sensitive pitch scoop at the beginning of a note!

The tenor sax waveforms (elements 1 to 5) cover the entire key range: C-2 to G8. The waveforms are assigned to different velocity ranges and are selected (and triggered) depending upon Normal or Legato playing gestures. The first element is triggered when a Normal (detached) gesture is detected and the key velocity (i.e., how hard the key is struck) is between 1 and 60 inclusive. The second element is triggered under the same conditions except the key velocity is between 61 and 90 inclusive. The AF1 button toggles between Mono and Poly mode — whether a legato gesture triggers a Legato element or Normal element.

You can see that only one of the first 5 elements is triggered at a time depending upon the combination of gesture, note range and velocity range. The Tenor Sax2 Med waveforms are played for quieter dynamic levels and the Tenor Sax2 Hard waveforms are played for the louder dynamic levels.

The Tenor Sax2 Med Of and Tenor Sax2 Hard Of waveforms are triggered by a Legato playing gesture. The “Of” in the waveform name means “Offset” and sample playback starts later in the waveform data, that is, skipping the attack part of the waveform. This eliminates the initial attack which is characteric of a sax playing detached notes.

Elements 6 to 8 are triggered only for Legato notes. These elements add a low-level “pop” at the beginning of each note. Think of this sound as a “connective tone” between notes. Tyros’s Super Articulation 2 technology (also known as “Articulated Element Modeling”) blends actual connective tones between notes, producing realistic articulations. The blending requires considerably more samples and processing power than the MOX or the S950.

The PRE5:009(A09) Tenor Soft Legato voice is a simplified version of the first patch. AF1 selects Mono and Poly modes. (AF2 is unassigned.) The patches use only the “Med” waveforms to achieve an overall softer timbre.

    Elem#  Waveform            XA      Notes   Veloc  Level
    -----  ------------------  ------  ------  ------ -----
      1    Tenor Sax2 Med      Normal  C-2 G8  1   70   110
      2    Tenor Sax2 Med Of   Legato  C-2 G8  1   80    99
      3    Tenor Sax2 Med      Normal  C-2 G8  71 127   110
      4    Tenor Sax2 Med Of   Legato  C-2 G8  81 127    99
      6    Small Tabla Dom     Legato  C4  G8  1  127    46
      7    Small Tabla Dom     Legato  C-2 B3  1  127    75
      8    Sine                Legato  C-2 G8  1  127    59

There are two dynamic levels (lower and higher velocity ranges) and two playing gestures (Normal and Legato) forming four combinations of dynamic level and gesture. Elements 6 to 8 implement a connective tone as previously described.

Life gets more interesting in the PRE5:0010(A10) Velo Growl Legato patch. AF1, again, selects Mono and Poly modes. (AF2 is unassigned.)

    Elem#  Waveform            XA      Notes   Veloc   Level
    -----  ------------------  ------  ------  ------- -----
      1    Tenor Sax2 Hard     Normal  C-2 G8  1    60   119
      2    Tenor Sax2 Med Of   Legato  C-2 G8  1    60    86
      3    Tenor Sax2 Growl    Normal  C-2 G8  61  127   125
      4    Tenor Sax2 Hard Of  Legato  C-2 G8  61  100   102
      5    Tenor Sax2 Growl Of Legato  C-2 G8  101 127    94
      6    Small Tabla Dom     Legato  C4  G8  1   127    52
      7    Small Tabla Dom     Legato  C-2 B3  1   127    78
      8    Sine                Legato  C-2 G8  1   127    78

There are roughly three dynamic levels:

  • Velocity 1 to 60: A hard attack is triggered for Normal notes and a soft attack, medium sax is triggered for Legato notes.
  • Velocity 61 to 100: A growl sax is triggered for Normal notes (up to velocity 127) and a soft attack, hard sax is triggered for Legato notes.
  • Velocity 101 to 127: A soft attack, growl sax is triggered for Legato notes.

This programming allows interesting one-hand control. Play soft to get a pure sax tone and play hard to get a growl. Play detached to get a harder attack and play legato to get a softer note attack (when Mono mode is selected via AF1).

The fourth and final patch is PRE5:011(A11) Tenor Growl AF1. The buttons are assigned in the following way:

    AF1: Mono/Poly mode and growl waveform
    AF2: Tenor Sax1 waveform

As you’ll see in the table below, the AF2 button selects the original Motif Tenor Sax1 waveform.

We again have two dynamic levels triggered by velocity ranges 1 to 100 and 101 to 127. Here, the assignable function buttons really come into play.

    Elem#  Waveform            XA        Notes   Veloc   Level
    -----  ------------------  --------  ------  ------- -----
      1    Tenor Sax2 Med      AllAFOff  C-2 G8  1   100   120
      2    Tenor Sax2 Hard     AllAFOff  C-2 G8  101 127   125
      3    Tenor Sax2 Growl    AF1 On    C-2 G8  1   127   127
      4    Tenor Sax2 Hard Of  Legato    C-2 G8  101 127   102
      5    Tenor Sax2 Hard Of  Legato    C-2 G8  1   100   102
      6    Tenor Sax1          AF2 On    C-2 G8  1   127   119

AF1 brings in a growl waveform (element 3) when it is turned ON. AF2 brings in an entirely different tenor sax waveform and tone (element 6) when it is turned ON. The first two elements play a pure tenor sax tone when all AF buttons are OFF. Elements 4 and 5 play a hard sax tone with a softer attack for legato playing gestures. You would be hard pressed to think about these combinations when actually playing — you just have to “go for it” intuitively, knowing that AF1 kicks in the growl.

Turning OFF AF1 while holding the key cuts off the note. Whether this is a bug or a feature is your’s to decide!

The effect programming in these four presets is not very adventurous. The effects are appropriate for a laid-back, mellow sound. Here’s a quick breakdown:

    Preset voice        Insert A   FX preset    Insert B    Dry/Wet
    -----------------  ----------  ---------  ------------  -------
    Tenor Dynamic AF1  VCM EQ 501    Flat     TempoCrosDly   D63>W
    Tenor Soft Legato  VCM EQ 501    Flat     TempoCrosDly   D59>W
    Velo Growl Legato  VCM EQ 501    Flat     TempoCrosDly   D54>W
    Tenor Growl AF1    VCM EQ 501    Flat     TempoCrosDly   D63>W

The Insert A effect is the VCM multi-band EQ. The EQ curve is flat, so the EQ is not coloring the sound at all. The Insert B effect is a tempo cross delay. The dry/wet mix is set conservatively (D54>W) or just plain off (D63>W). The system CHORUS effect is not applied and the system REVERB is a nice REV-X reverb.

The effect programming on the PSR-S950 is a little more exciting and adds a grittier sound for rock and R&B. The RockSax voice employs a distortion plus delay effect algorithm:


    Parameter       Value
    --------------  --------
    LCH Delay       250.0 ms
    RCH Delay       300.0 ms
    Delay FB Time   375.0 ms
    Delay FB Level  +16
    Delay Mix       50
    Dist Drive      10
    Dist Output     110
    Dist EQ Low     +3 dB
    Dist EQ Mid     +1 dB
    Dry/Wet         D40>W

Transporting this effect to the MOX, you could assign AMP SIMULATOR 2 to insert A. For insert B, you could stick with the tempo cross delay or you could program a fixed delay instead (e.g., DELAY L,R (STEREO)) using the parameters above. A third possibility is to use the MOX’s COMP DISTORTION DELAY algorithm which combines the distortion and delay into a single effect block.

The S950 GrowlSax voice uses a different distortion plus delay algorithm:


    Parameter       Value
    --------------  --------
    Overdrive       14%
    Device          Dist2
    Speaker         Combo
    Presence        6
    Output Level    98%
    Delay Time L    250.0 ms
    Delay Time R    250.0 ms
    Delay FB Time   500.0 ms
    Delay FB Level  +12
    Dry/Wet         D32>W
    Delay Mix       44
    FB High Dump    1.0

Programming options are similar. Set MOX insert A to AMP SIMULATOR 1 and either stay with the tempo cross delay for insert B, or set insert B to a fixed delay algorithm. Or, run everything through the MOX’s COMP DISTORTION DELAY algorithm. Tune the Dry/Wet mix to taste.

Hey, here’s a bonus — the effects for the S950 slapback guitar. This might sound good with a sax, too.


    Parameter       Value
    --------------  --------
    Overdrive       20%
    Device          Vintage
    Speaker         Twin
    Presence        14
    Output Level    66%
    Delay Time      16th/3
    Delay FB Level  +3
    L/R Diffusion   +10ms
    Lag             +0ms
    Dry/Wet         D<W63
    Delay Mix       127
    FB High Dump    1.0

In this case, go with AMP SIMULATOR 1 for MOX insert A. Use either the tempo cross delay for insert B or change insert B to the TEMPO DELAY STEREO algorithm.

Even though I’ve discussed voice and effects programming in the context of the MOX, these techniques all apply to the Motif XS, XF and MOXF, too.

If you would like to know more about Super Articulation voices, then please check out: SA and SA2: Is Motif up to the task? I also saved two informative posts from the Motifator forum about Super Articulation and Expanded Articulation.

Read about Motif XF (MOXF) “Tenor MAX” voice programming. Update: 18 May 2016.

Montage: New waveforms

Well, well. Interesting times, again. Yamaha have now released the Montage Reference Manual and the Data List Manual. Download them from your local support site.

At the same time, the Motif XF is being blown out. Not only have retailers dropped prices, Yamaha itself is saying “Sayonara” with a promotional rebate of its own. If you want a Motif XF, now is a terrific time to buy!

I started the decision making process last weekend by comparing the MOX waveforms against the Motif XF waveforms. To me, new waveforms represent true value — true sonic potential — over a keyboard’s predecessor. Unless MOXF owners want all of the bells and whistles of the Motif XF (e.g., big color display, on-borad sampling, sliders, version 1.5 Real Distortion effects, etc.), they already have the XF waveforms. MOX owners have the older Motif XS factory set, so they might be interested in upgrading to Motif XF. Here is a list of Motif XF waveforms that are not in the MOX:

    CF3 4 layer (vs. MOX 3 layer)
    Farfisa (Fr)
    Vox (Vx)
    Tango Accordian2
    Mussete Accordion
    Steirisch Accordion
    Jazz Guitar
    Pick Rndwound2
    Pick FlatWound
    Finger Rndwound
    Sect Strngs
    Tremolo Strings
    Live Pizzct
    Soft Trumpet
    Trumpet Vib
    Trumpet Shake
    French Horn Sft
    French Horn Med
    Soprano Sax3
    Alto Sax3
    Tenor Sax2 Soft
    Tenor Sax2 Falls
    Sax Breath

After looking over the list, frankly, I’m not motivated (bad pun) to buy an XF. My PSR-S950 does a great job covering these sounds. Plus, at 33.3 pounds (XF) vs. 15.4 pounds (MOX), a Motif XF is likely to remain in the studio, not at the gig.

The Yamaha Montage offers a bigger upgrade thanks to the large built-in waveform memory. Here is my first pass list of new Montage waveforms. I’ll leave it to you to comb through synth and percussion waveforms.

    CFX 9 layer
    S700 3 layer
    EP4 5 layer
    Rd Soft 5 layer
    Rd Hard 4 layer
    Rd73 5 layer
    Rd78 5 layer
    Rd KeyNoise
    Wr1 3 layer
    Wr2 4 layer
    Wr3 5 layer
    Wr KeyNoise
    Clav5 3 layer
    Clav KeyNoise
    CP80 5 layer
    CP80 KeyOff
    Motor Vibes
    Tonewheel1 Fast/Slow
    Tonewheel2 Fast/Slow
    Tonewheel3 Fast/Slow St
    Tonewheel4 Fast
    Tonewheel5 Fast
    Tonewheel6 Fast
    SctAcc Mussete
    Acc Key On/Off
    TC Cln Pick
    TC Cln Fing
    Acoustic2 (bass)
    Violin2 1st St
    Violin2 2nd St
    Viola2 St
    Cello2 St
    Celtic Violin
    US Strings
    Violins 1st
    Violins 2nd
    Trumpet 3
    Piccolo Tp
    Trombone 3
    Bass Trombone
    French Horn2
    BrassSect3 Acc/Doits/Shake/Falls
    Oboe4 NV/Stac
    Flute4 NV/Stac/Flutter
    Piccolo4 NV/Stac
    Low Whistle
    High Whistle
    Boys Choir
    Gospel Choir

Yamaha really upped the ante with new acoustic and electric piano samples. Yamaha have been promoting these improvements and rightfully so. I can’t wait to try these out. Jazzers will be glad to see the new vibraphone samples, too.

Tonewheel organ got a modest upgrade. I’ll reserve judgement until I can hear and play the Montage. The tonewheel samples have fast and slow variants, so the Leslie is probably sampled in. Not always a good sign, but, hey, I’m listening. A couple of more accordions round out the keyboard additions.

Guitars also got a modest upgrade. There are a few more acoustic guitars and two Telecaster variants (pick and finger). At this point, I must mention that all of the new waveforms have 3, 4, 5 or more layers and many articulations. So, even if the list looks short, the new voices should be quite rich and appealing.

Orchestral instruments got a major, major upgrade. As a liturgical musician who relies on these voices heavily, I’m excited. I called out only a few of the available articulations. Musicians who mock up orchestral scores or cover orchestral parts live should definitely take note of the Montage! Surprisingly, there aren’t new pipe organ waveforms. (Is an expansion pack in the works?)

Finally, there are a slew of choir and vocal samples from the Tyros 5. “Syllables” in the list above are all of the zillion duhs, doos, etc. ScatCycle includes the (infamous) scat syllables, but cycles through the syllables for variety. This is already a feature of the Tyros 5.

Given the boost in the orchestra department, I’m interested. I just wish that the Montage weighed about 20 pounds or less. Perhaps I need to wait for the MOXF follow-on in the light weight, mid-price category.

That’s it for now. I might have missed something during the first pass and will correct the list as I learn more about the Montage. At some point, I’ll take a look at Montage effects, too.

Read my initial review of the Montage8. Update: May 10, 2016.

All site content is Copyright © Paul J. Drongowski unless indicated otherwise.

PSR effects for electric piano (Part 1)

A common complaint about the electric pianos on the Yamaha PSR arranger workstations is their lack of “guts” or “grit.” The voice samples are reasonably good, but the effects programming is vanilla and way too polite, especially for rock and soul styles. Here is a table showing the default DSP effect for some of the electric piano voices in the PSR-S950:

    PSR-S950 voice  Category     Effect
    --------------  ----------   -----------------------
    SparkleStack    CHORUS       CHORUS3
    SweetDX         CHORUS       CHORUS3
    BalladDX        CHORUS       ENS DETUNE1
    DX Dynamics     CHORUS       CHORUS2
    BalladBells     CHORUS       CHORUS3
    SuitcaseEP      CHORUS       CELESTE2
    VintageEP       TREMOLO      EP TREMOLO    [DSP off]
    CP80            CHORUS       CHORUS3
    StageEP         CHORUS       CELESTE2
    SmoothTine      SPATIAL      EP AUTO PAN
    ElectricPiano   SPATIAL      EP AUTO PAN   [DSP off]
    Clavi           DISTORTION   DIST SOFT1
    WahClavi        WAH TCH/PDL  CLAVI TC.WAH
    PhaseClavi      PHASER       EP PHASER2

You can see that most of the voices use a chorus effect. In two cases, the DSP effect is turned off by default. (You need to turn it on using the [DSP] front panel button.) The Clavinet voices are a little more fun and use distortion, wah and phaser.

Chorus does not add much “heft” to a voice and it doesn’t add grit. Compression, mid-range boost (EQ) and overdrive are better choices when you need a punchy and/or grungy electric piano sound.

Let’s take a look at the effects programming for a few electric piano voices on the Yamaha MOX synthesizer workstation. The basic voices drive two insert effects connected in series:

    MOX voice             Insert A     Insert B
    --------------------  -----------  -----------
    Crunchy Comp          MltBndComp   CompDistDly
    Vintage Case          AmpSim 2     Auto Pan
    Chorus Hard           ClassicComp  SPX Chorus
    Drive EP AS1          AmpSim 2     Auto Pan
    Natural Wurli         AmpSim 1     Tremolo
    Wurli Distortion AS1  Tremolo      CompDistDly

On the MOX, every voice uses compression, amp simulation or distortion, even the voices employing the evergreen tremolo, pan and chorus effects.

At this point, PSR users tend to throw up their hands and say, “Well, that’s the Motif series!” and back away. Yamaha — bless them — share technology between workstation products. Quite often, you can find the equivalent PSR effect algorithm for an MOX (MOXF) or Motif algorithm.

Consider the MOX “AmpSim2” algorithm. This algorithm shares the same parameters as the PSR “DISTORTION AMP SIM2” algorithm. Here is a table showing the corresponence between MOX and PSR.

    MOX parameter  PSR parameter  MOX value
    -------------  -------------  ---------
    Preset         n/a            Stack1
    AmpType        AMP Type       Tube
    OverDr         Drive          16
    OutLvl         Output Level   70
    LPF            LPF Cutoff     6.3KHz
    Dry/Wet        Dry/Wet        D<W30

The parameter values given here are taken from the MOX “Drive EP AS1” voice. Bring up a PSR voice like “VintageEP,” edit its DSP effect and replace the tremolo effect with “AMP SIM2.” Plug in these values, listen and tweak!

My second example is taken from the MOX “Natural Wurli” voice. The MOX effect algorithm name is “Amp Sim1”. The equivalent PSR effect algorithm is “DISTORTION V_DIST WARM” and its siblings. Here is the equivalency table:

    MOX parameter  PSR parameter  MOX value
    -------------  -------------  ---------
    Preset         n/a            Stack2
    OverDr         Overdrive      2%
    Device         Device         Vintage tube
    Speaker        Speaker        Stack
    Presence       Presence       +10
    OutLvl         Output Level   53%
    Dry/Wet        Dry/Wet        D<W1

Again, change the PSR DSP effect to “V_DIST WARM” and plug in the values. Then, tweak away.

The final example is a multi-effect taken from the MOX “Wurli Distortion AS1” voice. The MOX effect algorithm is “CompDistDly” that is a compressor, distortion and delay effect chain. The equivalency table is:

    MOX parameter  PSR parameter         MOX value
    -------------  --------------------  ---------
    Preset         n/a                   Hard1
    OverDr         Overdrive             15%
    Device         Vin_tube              Vintage tube
    Speaker        Stack                 Stack
    Presence       Presence              +10
    DelayL         Delay Time L          307.3ms
    DelayR         Delay Time R          271.7ms
    FBTime         Delay Feedback Time   306.6ms
    FBLevel        Delay Feedback Level  +31
    FBHiDmp        Feedback High Dump    0.8
    OutLvl         Output Level          22%
    DlyMix         Delay Mix             0
    Compress       n/a                   -29dB
    Dry/Wet        Dry/Wet               D<W12

The almost equivalent PSR effect algorithm is “DISTORTION+ V_DST H+DLY”. The PSR algorithm is missing the compression component (parameter). If you want compression, then consider one of the other PSR distortion algorithms with mono delay.

Keep thinking “multi FX.” I’m going to visit the REAl DISTORTION multi FX algorithm in a future post.

Some of the MOX voices use VCM effects. I didn’t deconstruct the voices with VCM effects because my S950 doesn’t have them. However, if you have VCM effects, for heaven’s sake, use them!

Learn how to save your new creation in Editing and Saving PSR Effects (Part 2).

PSR effects for electric piano (Part 1)
Editing and saving PSR effects (Part 2)
Multi-effects for electric piano (Part 3)
Copy PSR DSP effects (part 4)

The SWP70 tone generator

As I mentioned in an earlier post, the Yamaha PSR-S770 and PSR-S970 arranger workstations have a new tone generator (TG) integrated circuit (IC) — the SWP70. (“SWP” stands for “Standard Wave Processor.”) The SWP70 is a new TG family in a long line of Yamaha tone generators. The SWP70 replaces the SWP51L, which has been the mainstay in recent generations of Tyros, upper range PSR, Motif, and MOX series workstations.

The SWP70 has much in common with the SWP51L, but also some very significant differences. The SWP70’s external clock crystal frequency is 22.5792 MHz versus 11.2896 MHz for the SWP51L. This funky looking clock rate is a multiple of 44,100 Hz:

    22.5792MHz = 44,100Hz * 512

Samples are transferred to the DAC, etc. at a multiple of 44,100 Hz (Fs). Thus, it makes sense to derive Fs and its multiples from the chip-level master clock. The higher crystal frequency and faster memory read clocks lead me to believe that the SWP70 is clocked twice as fast as the SWP51L.

I am comparing SWP characteristics as deployed in the S970 (SWP70) and the S950 (SWP51L) workstations. This keeps the basis of comparison even although many characteristics (clock rates, DSP RAM size) are the same in higher end models like Tyros 5 or Motif. Higher end models employ two SWPs in master/slave relationship and both SWPs share the same wave memory. For more information about the PSR-S970 internal design, look here.

Five interfaces are essentially the same as the SWP51L:

  1. CPU interface: Communicate with the Main CPU (e.g., Renesas SH7731) via the parallel CPU bus.
  2. Serial audio: Send/receive audio data to/from the DAC, audio ADCs, and main CPU.
  3. Clock interface: Synchronize serial audio data transfers (generate multiples of Fs).
  4. DSP SDRAM interface: Store working data for effect processing.
  5. EBUS interface: Receive controller data messages (e.g., pedal input, keyboard input, pitch bend, modulation, live knobs, etc.) from front panel processors.

The DSP SDRAM is the same size: 4Mx16bits (8MBytes). The SWP70 read clock is 95.9616 MHz, while the SWP51L read clock is 45.1584 MHz. This is more evidence for a higher internal clock frequency.

The Tyros 4, Tyros 5 and S950 have an auxiliary DSP processor for vocal harmony. The microphone analog-to-digital (ADC) converter is routed directly to the auxiliary processor. Prior to these models, the microphone ADC is connected to the tone generator. With the SWP70, the S970’s microphone ADC is once again routed to the SWP70 and the auxiliary processor disappears from the design. Thus, vocal harmony processing (fully or partially) is located in the SWP70. See my post about SSP1 and SSP2 for further details.

The biggest change is the wave memory interface.

A little history is in order. The SWP51L (and its ancestors) were designed in the era of mask programmable ROM. I contend that tone generation is memory bandwidth limited and the earlier interface design is driven by the need for speed. The SWP51L (due to its evolved history) has two independent wave memory channels (HIGH and LOW). Each channel has a parallel address bus (32 bits) and a parallel data bus (16 bits). The two channels account for over 100 pins. (System cost is proportional to pin count.) The user-installed, 512/1024MB flash DIMMs plug directly onto the two channels.

The SWP70 wave memory interface takes advantage of new NAND flash memory technology. The interface is described in US patent application 2014/0123835 and is covered by Japanese patent 2012-244002. I analyzed the US patent application in an earlier post.

The SWP70 retains the HIGH port and LOW port structure. Each port communicates with an 8Gbit Spansion S34ML08G101TFI000 NAND flash device. Address and data are both communicated over an 8-bit serialized bus. This technique substantially decreases pin count and the resulting board-/system-level costs. Smart work.

I did not anticipate, however, the introduction of a new parallel memory interface called “wave-work”. The wave work interface communicates with a 16Mx16bit (32MBytes) Winbond W9825G6JH-6 SDRAM. The read clock is 95.9616 MHz.

The purpose of the wave work SDRAM is revealed by US Patent 9,040,800. This patent discloses a compression algorithm that is compatible with serialized access to the wave memory. The wave work SDRAM is a cache for compressed samples. The characteristics of the Spansion memory device give us a clue as to why a cache is required:

    Block erase time               3.5ms    Horrible (relative to SDRAM)
    Write time                     200us    Terrible
    Random access read time         30us    Bad
    Sequential access read time     25ns    Very good

As the patent explains, two (or more) samples are required to perform the interpolation while pitch-shifting. If there is only one tone generation channel, access is paged sequential. However, random access is required when there are multiple tone generation channels. (The patent mentions 256 channels.) Each channel may be playing a different voice or a different multi-sample within the same voice. One simply cannot sustain high polyphony through random access alone. The cache speeds up access to recently used pages of uncompressed samples.

The wave work interface takes additional pins, thus adding to board- and system-level costs. The overall pin count is still lower when compared to SWP51L. The penalty must be paid in order to use contemporary NAND flash devices with a serialized bus. This is the price for catching the current (and future) memory technology curve.

A few SWP70-related printed circuit board (PCB) positions are unpopulated (i.e., IC not installed) in the PSR-S970. There is an unpopulated position for a second Winbond W9825G6JH-6 wave work SDRAM which would expand the wave work memory to 32Mx16bit (64MBytes). A larger cache would be needed to support additional tone generation channels. Perhaps only half of the tone generation channels are enabled in the mid-grade PSR-S970 workstation.

There is what appears to a second separate wave work interface that is completely unpopulated. The intended memory device is a Winbond W9825G6JH-6, which is consistent with the existing wave work interface.

The PSR-S970 also has a stubbed out interface that is similar to the DSP SDRAM interface. The existing DSP SDRAM signals are labeled “H” for HIGH while the unused interface is labeled “L” for LOW. Perhaps only half of the hardware DSP processors are enabled for the mid-grade S970, waiting to be activated in future high-end Tyros and Motif products.

I refer to future high end products by the names of the current product lines. Yamaha may choose to rebrand future products (e.g., the much-rumored “Montage” trademark).

The Spansion S34ML08G2 8-Gb NAND device is Open NAND Flash Interface (ONFI) 1.0 compliant. The S34ML08G2 device is a dual-die stack of two S34ML04G2 die. The 8-bit I/O bus is tri-state allowing expansion e.g., multiple memory devices sharing the same I/O bus and control signals with at most device enabled at any time. The SWP70 has additional chip select pins that would support this kind of expansion. The current expansion flash DIMMs will no longer be needed or used.

In this note, I concentrated on observations and fact, not speculation about future products. I’ll leave that fun for another day!

All site content is Copyright © Paul J. Drongowski unless indicated otherwise.

Clear the decks?

Yamaha have announced a truly stellar promotion to move Motif XF workstations. The Motif XF Fully Loaded expansion pack includes a FireWire expansion board, two FL1024M memory modules and an USB drive filled with content including Chick Corea’s Mark V electric piano. (See the promotions page at the Yamaha web site for additional details.)

Wow! This promotion really caught my attention and if ever there was a time to upgrade to an XF, it’s now.

Of course, this aggressive promotion could also mean that a new synthesizer workstation will be announced in the not-too-distant future. Winter NAMM 2016, perhaps? Old inventory has got to go!

After the Reface surprise, I’ve given up predicting specific product features, especially based upon a (rumored) product name. The word “Reface,” for example, means something completely different to a saxophone player and, yes, Yamaha manufacture saxophones and mouthpieces. 🙂 So, “Montage”, harumph. I am willing to predict, however, that the next high-end workstation will have a new member of the Standard Wave Processor (SWP) family — the hardware chip that underlies the tone generation infrastructure. (See Serial Memory and Tone Generation.) This is big step for Yamaha because the current SWP51L, for example, is used in everything from mid-range arrangers, to MOX/MOXF, to Motif, to Clavinova.

Just taking in the gestalt of Yamaha’s recent patent filings, they have been actively building their portfolio in at least three areas: human vocal processing and synthesis (VOCALOID), music analysis and combined MIDI/audio accompaniment.

VOCALOID has been a commercially successful software product. The tech has, by the way, some similarities to the “connective” capabilities of Articulated Element Modeling (AEM), known more broadly as “Super Articulation 2” on Tyros. VOCALOID requires frequency domain signal processing, so unless Yamaha have knocked down some real computational barriers, VOCALOID will probably remain a non-real time synthesis technique.

“Music analysis” is a broad area and a rather vague term. At a fundamental level, this area includes beat (tempo) detection and scale and harmony (chord) detection. I think we already see some of these results at work in the Yamaha Chord Tracker app. Chord Tracker analyzes an audio song. It detects the tempo and beats, and partitions the song into measures. Chord Tracker identifies the chord on each beat and displays a simplified “fake sheet” for the song. Chord Tracker can send the “fake sheet” to a compatible arranger keyboard for playback.

Music analysis also includes high-level analysis such as extracting the high level characteristics of a piece of music. This kind of analysis could allow a rough categorization and comparison between snippets of music (similarity index). We haven’t seen the fruits of this technology (yet), but one could imagine a tool that suggests an accompaniment based on what the musician plays or based upon an existing musical work. BTW, the word “musician” here includes guitarists, woodwind players, etc. and not just keyboardists. The world-wide market for non-keyboard instruments is bigger than the market for keyboard-based instruments. (Guitars alone outsell keyboards nearly 2 to 1 in the United States.)

The third main area of exploration and filings is combined MIDI/audio accompaniment. Up to this point, Motif arpeggios are MIDI-like phrases, not audio. Arranger workstation styles are MIDI (SMF in a Halloween costume). Neither product works with MIDI and audio phrases in a transparent way like the very successful Ableton Live. Yamaha’s patent filings disclose arpeggio- and/or style-like accompaniment using a mix of MIDI and audio phrases. Audio phrases are warped in time and pitch to match the current tempo and key scale.

Now, let’s throw these technologies into a bag and shake them around. Imagine a compositional assistant that analyzes a piece of music (recorded or played live), determines tempo, beats, chord changes and more, and automatically whips up an accompaniment or track. MIDI and audio phrases are selected from a library based upon a similarity index between the reference track and phrases in the library. If this is Yamaha’s vision, then double wow! The combination of these technologies would raise the level of music composition substantially from it’s tedious, point-and-click existence. It finesses the problem of listening to the phrases in the Motif/MOX arpeggio library, selecting the most applicable phrases and combining them. DigiTech TRIO is already sniffing around this territory.

Naturally, patents do not imply product. Therein lies the danger of making predictions.

Which brings me, finally, to US Patent 8,779,267 (July 15, 2014). If someone can explain this patent to me, thanks. The invention seems to analyze an incoming musical signal (using some heavy DSP), generate almost ultra-sonic (>18KHz) “control tones,” and produce a multi-timbral accompaniment or track. Amazing stuff.

The near ultra-sonic technique is already in use. The AliveCor Mobile ECG monitor uses ultrasonic tones to communicate with iPhone/iPad. The AliveCor doesn’t require power-sucking Bluetooth (and its emissions certification.) The monitor runs on a CR2016 battery. The downside, in the case of AliveCor, is that its monitor pad must be near the mobile device for reliable communication.

All site content is Copyright © Paul J. Drongowski unless otherwise indicated.

Crunchin’ da drums

In my last post, I discussed Motif/MOX eight zone (8Z) drum kits. The eight zone concept lets you assemble eight different percussion sounds into a custom kit. The waveforms are assigned to voice elements and are stretched/limited to eight different keyboard (MIDI note) zones. The Motif/MOX have matching arpeggios that work with the 8Z kits.

By the way, the 8Z drum kits were first introduced with the Motif XS. My notes on the 8Z kits and this note on effects apply to all later models including the Motif XF and MOXF.

If you have ever tried the percussion sounds alone without effects, the drum sounds are kind of “plain Jane” without a lot of impact. This post deconstructs a couple of effects which can be applied to break beats and other styles that require crunch and animation.

The first effect chain is taken from the Voice PRE8:060 “8Z Romps.” The voice has two insert effects connected in series. INSERT A is a Lo-Fi algorithm with the following parameters (effect preset “Max Lo-Fi”):

    #  Parameter              Value    Numeric
   --  ---------------------  -------  -------
    1  Sampling Freq Control  4.01KHz  (10)
    2  Word Length            93       (93)
    3  Output Gain            +7 dB    (14)
    4  LPF Cutoff             20.0KHz  (60)
    5  Filter Type            Radio    (2) 
    6  LPF Resonance          10.0     (100)
    7  Bit Assign             2        (2)
    8  Emphasis               On       (1)
   10  Dry/Wet                D<W63    (127)
   15  Input Mode             Stereo   (1)

The parameter number, name, values, etc. are taken from the MOX Data List. (See the section titled “Effect Parameter List” in the PDF file). The numeric values — given here in decimal — are what you need to program the effect through System Exclusive MIDI messages. More about this in a minute.

The Lo-Fi effect adds a lot of crunch and crush. But, wait! There’s more. The INSERT B effect is the AmpSim 1 amp simulator. Its parameters are:

    #  Parameter              Value    Numeric
   --  ---------------------  -------  -------
    1  Over Drive             54%      (54)
    2  Device                 dst1     (2)
    3  Speaker                Combo    (2)
    4  Presence               +10      (10)
    5  Output Level           34%      (34)
   10  Dry/Wet                D<W63    (127)

This is the “Beat Crunch” effect preset.

Please remember that my goal is to use the 8Z break beats in a PSR/Tyros style. In order to do accomplish this, I found the equivalent effects algorithms for the Yamaha PSR-S950 arranger workstation. Here are the equivalent algorithms:

    MOX            PSR-S950
    --------       -----------------------------
    Lo-Fi    --->  Lo-Fi DRUM1 (MSB:94 LSB:18)
    AmpSim 1 --->  V_DIST CRUNC (MSB:98 LSB:18 )

Unfortunately, the XG effects architecture supports at most one system-wide variation effect or one per-part insert effect. So, I decided to use the Lo-Fi algorithm because it seemed to provide most of the grit and nastiness that I was seeking.

It took a little detective work to find and match up the corresponding effect algorithms between the Motif/MOX and the PSR/Tyros. The effect type is enough to get into the same neighborhood. The rest of the sleuthing involves comparing the parameter lists in order to find the exact (or best) match. The MOX has Virtual Circuit Modeling (VCM) effects and the S950 does not. Therefore, you may not always be able to find an exact match.

With the S950 Data List in hand, I translated the effect parameters into the hexadecimal System Exclusive (SysEx) messages to configure the Lo-Fi effect on the PSR:

    F0 43 10 4C 02 01 40 5E 12 F7   Variation Type
    F0 43 10 4C 02 01 5A 01 F7      Variation Connection (SYSTEM)
    F0 43 10 4C 02 01 42 00 0A F7   PARAMETER 1 Sampling Freq Control (10)
    F0 43 10 4C 02 01 44 00 5D F7   PARAMETER 2 Word Length (93)
    F0 43 10 4C 02 01 46 00 0E F7   PARAMETER 3 Output Gain (14)
    F0 43 10 4C 02 01 48 00 3C F7   PARAMETER 4 LPF Cutoff (60)
    F0 43 10 4C 02 01 4A 00 02 F7   PARAMETER 5 Filter Type (2)
    F0 43 10 4C 02 01 4C 00 64 F7   PARAMETER 6 LPF Resonance (100)
    F0 43 10 4C 02 01 4E 00 02 F7   PARAMETER 7 Bit Assign (2)
    F0 43 10 4C 02 01 50 00 01 F7   PARAMETER 8 Emphasis (1)
    F0 43 10 4C 02 01 54 00 7F F7   PARAMETER 10 Dry/Wet (127)
    F0 43 10 4C 02 01 74 01 F7      PARAMETER 15 Stereo  (1)

I configured the effect as a system-wide variation effect such that multiple percussion parts may be sent to the effect. I inserted the SysEx messages into the set-up measure of the PSR style file using SONAR (my usual DAW/sequencer). Yow, the difference between the percussion sounds without and with this effect is like night and day!

The MOX insert effects are followed by a system-wide Tempo Cross Delay effect (effect preset “4beat Echo”). This effect adds a nice bit of animation to the overall sound. The MOX effect parameters are:

    #  Parameter              Value    Numeric
   --  ---------------------  -------  -------
    1  Delay Time L>R         4th      (11)
    2  Delay Time R>L         8th.     (10)
    3  Feedback Level         16       (80)
    4  Input Select           L        (0)
    5  Feedback High Dump     0.5      (5)
    6  Lag                    0ms      (64)
   10  Dry/Wet                D<W63    (127)
   13  EQ Low Frequency       250Hz    (22)
   14  EQ Low Gain            0dB      (64)
   15  EQ High Frequency      4.0KHz   (46)
   16  EQ High Gain           0dB      (64)

The equivalent S950 effect is TEMPO CROSS1 (MSB:22 LSB:0). I assigned this effect to the system-wide CHORUS block.

Here are the S950 (XG) SysEx messages to configure the delay effect in the CHORUS block:

    F0 43 10 4C 02 01 20 16 00 F7  Chorus Type TEMPO CROSS1
    F0 43 10 4C 02 01 22 0B F7     PARAMETER 1 Delay Time L<R     (11)
    F0 43 10 4C 02 01 23 0A F7     PARAMETER 2 Delay Time R<L     (10)
    F0 43 10 4C 02 01 24 50 F7     PARAMETER 3 Feedback Level     (80)
    F0 43 10 4C 02 01 25 00 F7     PARAMETER 4 Input Selection    (0)
    F0 43 10 4C 02 01 26 05 F7     PARAMETER 5 Feedback High Dump (5)
    F0 43 10 4C 02 01 27 40 F7     PARAMETER 6 Lag                (64)
    F0 43 10 4C 02 01 2B 7F F7     PARAMETER 10 Dry/Wet           (127)
    F0 43 10 4C 02 01 32 16 F7     PARAMETER 13 EQ Low Frequency  (22)
    F0 43 10 4C 02 01 33 40 F7     PARAMETER 14 EQ Low Gain       (64)
    F0 43 10 4C 02 01 34 2E F7     PARAMETER 15 EQ High Frequency (46)
    F0 43 10 4C 02 01 35 40 F7     PARAMETER 16 EQ High Gain      (64)

A little bit of delay on a busy drum part goes a long way. The send level (not shown here) is relatively low — just enough to add a little animation to the sound without creating a lot of clutter. It sounds OK, but I might adjust the send level dynamically and add more delay to exposed parts like the break while keeping the MAIN sections clean.

I hope this short effects clinic helps you out!