Fixing a Waldorf Blofeld’s jumpy encoders

Update, 25-Jan-2017: Post has been updated to reflect my recent insight that the “self-lubricating” encoders in the Blofeld use a contact oil which should be replaced to avoid abrasion and wear. My sincerest apologies!

I like my white Blofeld. A very versatile VA synth with a good programming concept, and I am simply smitten with its look and feel. Solid metal, not too bulky or too heavy though, a big friendly graphics display, and a very good keyboard, with very good aftertouch sensibility. Got it really cheap, too.

Waldorf Blofeld keyboard white

As with the E-MU ESI32 that needed a new backlight, a steel casing means solid build quality – for the mechanical parts. BTW: To take it apart, you have to remove all the screws on the synth’s underbelly – 18 of them, casing screws as well as the ones holding the keyboard. No need to remove the wooden side panels or the rubber feet though. Then, remove all screws holding the electronics board – one of them is under the power switch so it is necessary to remove that switch from the housing by applying gentle pressure from the inside. In short, be warned:  it’s a real pain in your lower backside to get the electronics dismounted, so if you do not have a good reason to remove Blofeld’s brain, just don’t.

Jumpy encoders being a very good reason, of course.

One thing that seems to plague the Blofeld, keyboard and desktop versions alike, is that the encoders are likely to behave eratically; they tend to become jumpy an unrealiable. Achim at stromeko.net, who has loads of insight into and experience with Waldorf synths and their tech, recommends soldering in buffer caps, but to make a long story short, I think you might be better off cleaning, lubricating, and bending rather than only soldering. I’d suggest addressing the mechanical problems first.

Jumpy? Greasy.

My Blofeld did indeed suffer misbehaving encoders. The two under the display were the worst, the ones that you use the most in tweaking sounds. This type of behaviour is not new to me; I fixed a Micro Q once, the Blofeld’s predecessor that had suffered from the same erratic encoder behaviour – after years of use. They still use the same kind of encoder – and these are easy to open and clean.

Once again, it proved that most problems in electronics are mechanical by nature – in my Blofeld, the encoders had been greased to ensure smooth operation, but a surplus of grease had seeped into the contacts. Update: Or so I thought – in fact it may be a special kind of grease to protect the contacts; you may use Kontakt-61 or a similar lubricant intended for contacts.

To fix the mechanical problems, carefully bend the pickup contacts on the rotor just the tinyiest bit higher to increase pressure After cleaning the encoder, and after re-applying the contact oil, encoder operation is now flawless.

The encoders are held together by four metal tongues from a very soft metal. It is very hard to bend it back into position so that it closes the encoder without play; a drop of hot glue holds the top in place and is easily removed if the encoder is due for maintenance again.

One final note: The main encoder left of the display is of a slightly different type where the metal tongues have to be straightened to pass through a hole in the top part of the housing. Extra care needed with that.

Resetting a Matrix-1000 with a new battery

03-2020: Two important updates: There is a much simpler way to reset the memory in the Matrix – just hold the ENTER key while switching it on. You may have to repeat that a couple of times to get a stuck Matrix unstuck. And if you consider changing the battery yourself, you can find a step-by-step description with video here.

Happy days! Bob Grieb just sent the newest iteration of his brilliant firmware rewrite for the Oberheim Matrix-1000 for testing – a firmware that breathes new life into the old 8-bit hardware by optimising critical routines for a couple of crucial parameters. Smooth real-time control, very musical; in my opinion, even better than GliGli’s great v1.16 hackI’ve described the differences in this Gearslutz post –  and, in the last couple of iterations, displaying the value of edited parameters.

(I used Modstep as a drum machine/sequencer on the iPad, and my Matrix-1000 control panel for TB Midi Stuffother than the new iPad editor, it does produce smooth parameter sweeps.)

Well, to change the firmware, you have to open the Matrix and exchange the firmware EPROM for a new one, and doing that, I’ve noticed that this machine was still equipped with its original battery. By lucky chance, I am the proud owner of two Matrixes, and the battery in this one has been doing fine – what kind of super battery did they use in these days, has been in service ever since 1989, and still producing fine 3.0 Volts of power – but I decided to exchange it anyway for a new CR-2032.

As you might know, the battery in the Matrix-1000 is soldered in with most machines, as it was customary with most synths from these days. I guess they never thought that they were building for the anoraks of the future. No problem, I came across battery holders with the same 20.5mm raster used in the Matrix – so no need for drilling, just a simple solder-and-replace job. While soldering, I bridged the backup battery voltage with an external power supply, and I even thought of desoldering the GND terminal first – the rationale behind this being that soldering pens are earthed, so by soldering the positive terminal first you might short out the battery. (Actually bollocks, but I did it anyway.) So I saved my precious memory settings while soldering in the battery holder.

Only to slide in the new battery the wrong way round.

You might not have realised – well, I never do – but the pad connector of a CR2032 is actually the GND terminal, and the housing is Vcc. And is labeled with a clearly visible “+” sign. Well, I put the battery in the wrong way round anyway, thereby effectively losing all my patch and memory settings.

Foto 10.01.16, 16 43 46

The battery in its new holder, now in the correct position: the plus terminal facing upwards

This is, of course, no big deal. I keep moving sound banks between my two Matrixes anyway, so I have pretty recent Sysex backups. Unfortunately, the unbuffered RAM chip lost just enough memory to put the machine into an undefined state – it would no longer boot beyond the init routine displaying the firmware version.

So: How do you factory-reset a Matrix-1000 synth?

I ran into this problem before when I equipped my other Matrix with a new CPU – as you can imagine, this gave me some really bad moments. But factory-resetting an M-1000 is simple:

  • Switch off the the M-1000, disconnect it from mains, open it.
  • Disconnect the battery. Leave it disconnected.
  • Switch the M-1000 on, draining its buffer capacitors. Leave it for a couple of seconds – the completely powerless RAM should be all FF’s now.
  • Connect the M-1000 to mains, and switch it on. It should start now.
  • Do a calibration run, just to be sure. (Navigate to Ext. Funct., select 7, Enter, select 2, Enter.)
  • Reconnect the battery.

Done. Now you may switch off the Matrix, or supply it with fresh patch data.

Foto 10.01.16, 16 44 19In case you may have wondered, there is a very simple and effective way to disconnect/reconnect the battery in a running machine: push a strip of paper between the battery and the battery holder’s terminal. Remove it to reconnect.

New code, explanations, schematics – Understanding the Matrix

Part of the new M-1000 schematic

Enter the Matrix – finally the R/W decoder is legible.

Looking to get the latest revision of Bob Grieb’s firmware rewrite mentioned below? Click here. 

As you may have seen in this blog, I really care about my old Matrix-1000 rack synths. A wonderful piece of retro technology from the final, post-DX7 phase of classic analog synthesizers. There have been attempts to improve on the old 6809 firmware code, most notably by Gligli, a French hacker best known for his SCI Prophet 600 hardware/firmware retrofit. His improved firmware, known as V1.16, introduced a couple of tricks:

  • It enabled NPRN control of the Matrix’ parameters by removing a small bug
  • It sped up the VCF parameter control by skipping seemingly unnecessary calculations
  • It told the synth to discard all parameter edits except the most recent one, thereby keeping the synth responsive.

This is a huge improvement and makes the Matrix feel and behave almost like a modern instrument. But it gets even better.

Matrix-6 project, Matrix-1000 upgrade

Bob Grieb has been analyzing the Matrix-6/1000 code for months. I guess you can say that these days, not even Marcus Ryle does understand the code as well as Bob does. Here is his explanation why the Matrix-6/1000 machines are not real-time responsive to parameter changes in the first place – it is the downside for the immense amont of real-time modulations the Matrix is capable of – 22 fixed modulation paths, 10 matrix modulation slots, 3 envelopes, 2 LFOs and 2 ramps. To implement that in software the programmers used a special technique; a pre-calculated memory area for each voice called the voice update stack. Quote:

This stack contains pointers to code, ptrs to variables, and some pre-computed values. Only pointers to the code needed to handle the enabled features are placed on the stack… This is a very fast and efficient way to update the voice cv’s.
A downside of this approach is that when parameters change, the stacks need to be updated for all six voices. Some parameter changes just affect one number on the stack, so that number can simply be changed very quickly. But some parameters can change the size of the stack. This is a problem, as the update values for that parameter may be in the middle of the stack.

This means moving around chunks of memory to make room for the updated parameters, and it has to be done for all six voices, which takes the ancient 8-bit, 2-MHz CPU a couple of milliseconds. When you turn an external VCF controller, all these parameter changes add up, and the machine freezes for a long, terrible moment, until it catches up. (Read Bob’s full description of the issue here.)

GliGli’s main trick is to tell the machine to discard anything but the last Sysex command. He also noticed that sometimes the stack is rebuilt although this is not technically necessary. And this is the road that Bob has been following. He rewrote parts of the firmware to handle a set of about a third of the parameters much, much faster – including VCF frequency and resonance, DCO PW and LFO control, and VCA level. Changing these parameters with an external controller will be smoother than with older firmware, others – increasing the effect of a modulator in the mod matrix – will still cause the machine to glitch.

Update, 2016: Now that you’ve made it this far, you’ll be glad to learn that Bob made his revised code available for Matrix-1000s and Matrix 6/Rs. You can get a firmware EPROM from him or, if you are in Europe, from me – just follow the links above. 

This was originally a project for the Matrix-6, but Bob ported it over to the Matrix-1000. In the process, he also redrew the schematics, so that after all these years, there is finally a legible circuit diagram for the M-1000 on the net. Incidentally, it prompted another guy to scan his printed schematic and send it to Bob, so that there are now not only one usable version but two. (Download link to ZIP archive here.)

Dead Novations Society

No, that Bass Station you are trying to play isn’t dead. It’s just gone… to an undefined state.

  
I had a Novation Bass Station that did not produce a single sound. Although I could confirm that MIDI was still working – the device was sending Key ON/OFF and CC messages – and the LFO LED was blinking and responded to parameter changes, the synth was mute. And some research on Ebay and on the net confirmed that there are lots of Bass Station owners with the very same problem.
Continue reading

No More Guru Meditation: Windows 7 Starter in Virtualbox zum Laufen bringen

Windows 7 in einer Virtualbox

Ach ja, mein alter 2007er MiniMac. Als Arbeitsrechner ist er inzwischen etwas zu alt, als Musikmaschine für den Proberaum reicht er aus. Fast: Die Einstell-Software für unser neues Audio-Interface setzt mindestens OS 10.8 voraus, und das kann das alte Maschinchen nicht mehr. Kein Problem, die Software gibt’s ja auch für Windows und darf ruhig in einer virtuellen Maschine laufen. Mit dem (für Privatanwender) kostenlosen Virtualbox sollte das kein Problem sein. Läuft auf dem Arbeitsplatz- und Proberaumrechner jeweils in Version 4.1.

Recovery-Disk mit Windows 7Meine Standard-Instanz mit Windows XP ist zum Glück einem Festplattencrash zum Opfer gefallen, da komme ich gar nicht erst in Versuchung. Nach einigem Suchen findet sich die Installations-CD für Windows 7 Starter – die Lizenz kann ich nutzen, weil auf dem betreffenden Rechner längst ein gekauftes Windows 8.1 läuft. Also auf dem Heimrechner, einem 2010er Mac mit OS X 10.9., eine virtuelle Maschine angelegt, die Recovery-CD gemountet und gestartet – läuft gut, bricht aber, sobald es ernst wird,  immer wieder mit einer Virtualbox-Fehlermeldung ab: einer “Guru Meditation”.

Spurensuche im Logfile: Irgendwas mit Grafik

Was sagt uns diese “Guru Meditation”, außer dass die Virtualbox-Programmierer Amiga-Vergangenheit haben? Oberflächliche Googelei in den Foren bringt außer Hinweisen auf einen längst beseitigten Bug nur den Tipp, sich doch mal näher mit dem Virtualbox-Logfile zu beschäftigen. Es residiert in einem Unterordner “Logs” der virtuellen Maschine auf der Mac-Festplatte.

00:00:09.307 !! {vgatext}
00:00:09.307 !!
00:00:09.307 Not in text mode!
00:00:09.307 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
00:00:10.307 Changing the VM state from 'RUNNING' to 'GURU_MEDITATION'.

Tatsächlich bringt der Blick in die Logdatei Aufschluss: Die virtuelle Maschine steigt in dem Moment aus, wenn das Windows-Installationsprogramm in den Grafikmodus gehen will. Ein Ansatzpunkt für weitere Versuche.

Nur ein Versuch, aber: im “Scale Mode” geht’s!

Die haben schließlich Erfolg: Die Installation lief durch, sobald ich die Maschine auf den so genannten “Scale Mode” umgestellt habe – da muss das Windows offenbar keinerlei Rücksicht nehmen auf die virtuellen Grafiktreiber, bzw. umgekehrt. Also: frühzeitig cmd-C drücken Jetzt läuft’s. Ich kann nur spekulieren, ob das tatsächlich der Grund ist, dass jetzt alles durchläuft, aber einen Versuch ist es wert. – die weiteren Einstellungen meiner VM: 64MB Grafikspeicher mit 2D- und 3D-Beschleunigung, PAE/NX aktiviert. Bisher keine weiteren Guru Meditations.

Ganz koscher ist die Grafikausgabe der VM immer noch nicht – trotz inzwischen installierter Gast-Treiber für Windows 7 klappt der Bildschirm-Refresh nicht immer, wenn man zum Beispiel in einen anderen Bildschirm-Modus umschaltet (was inzwischen funktioniert), beispielsweise in den “Seamless”-Modus, ist alles nur – schwarz. Lösung dafür: die VM unterbrechen und neu starten. Rechtsklick auf das Virtualbox-Fenster im Apple-Dock, “Beenden” wählen, “Save the Machine State” auswählen. Das beendet irritierenderweise nur die VM, nicht Virtualbox. Die Virtuelle Maschine danach neu wieder starten – alles gut.

Update: Keine Updates?

Ein Besuch der Virtualbox-Website brachte inzwischen die Gewissheit, dass meine Virtualbox-Version veraltet war – inzwischen sind die bei Version 5.0.2, bei mir lief noch irgendeine 4.1. Die Grafikausgabe bleibt trotzdem hakelig: die neu installierte V5.0.2 (die natürlich mindestens ein OS X 10.8 braucht) verweigerte die mühselig angelegte Win7-VM fortzusetzen – es half, (a) in den Grafikeinstellungen der VM einen Haken zu setzen, der”Nicht Skalierte HIDPI-Ausgabe” erzwingt, und (b) die VM über “Verwerfen” neu zu booten.

Securing Jenny’s power supply

Pretty straightforward, this one: Replace the JEN SX-1000‘s fixed power cable with an IEC socket. (Like weird German words? You’ll love this: In German, this thing is called a “Kaltgerätestecker”, more precisely, a “Kaltgeräte-Steckverbinderbuchse”, which translates to “Cold unit connector socket”. Don’t ask me.)

Jenny's backside

No problems here apart from cutting a hole for the socket; I used steel drills to mark the corners of the cutout and then cut the steel with my Dremel tool. Lots of metal dust but fast, and it did the job precisely.

A word on Jenny’s power supply: It’s extremely oldschool – a transformer and a diode bridge generating +/-18VDC, two 7812 regulators generating +12V and -12V, and another 7805 regulator generating the +5V supply rail from the 12V. If you would like your JEN to be a bit more eco-friendly I’d advise replacing those regulators by the 2931CT low-dropout type, but apart from that, there’s hardly any reason to look at the power supply – it’s rather solid and possibly not your primary concern if the synth does not work.

There is a true Oberheim Matrix editor for iPad now. And yes, it’s worth buying it.

If you have found this blog searching for the Oberheim Matrix-6/1000 synthesizer, you may already know that I still haven’t given up on breathing new life into hardware and software of this wonderful machine, and that I have made a controller template for the iPad. A controller, mind you, not a true editor – but a tool to control each parameter in a sound preset via a dedicated touch control, and pretty much without alternative.
Patch Touch app screenshot - all parameters of a sound on one page
No longer – there is a true Matrix editor app in the Store now, Patch Touch by Coffeeshopped, LLC. How does it compare? Is it worth the 15 30 Dollars or Euros? Chadwick, the guy behind Coffeeshopped, was so kind as to send me a download code for his app, and to comment on an early draft of my observations, so you’ll find my remarks updated with his comments here.
Continue reading

7 Apple-][-Spiele, die man WIRKLICH gespielt haben sollte

Es ist der “Editors Pick” im heutigen Wired-Newsletter: eine Liste mit Spielen für den Ur-Heimcomputer Apple II. Alle bei Archive.org hinterlegt und dort über den Online-Emulator spielbar wie damals. Perfekte Retro-Clickbait, zumal das Rezept so einfach zu übertragen war

Es gibt nur ein Problem mit dem Post: Die Liste ist Scheiße.  

Hat der eigentlich je einen Apple II gesehen, geschweige denn darauf gespielt? Ich sage das als jemand, der seinen Apple II wirklich geliebt hat – und sich sehr darüber wundert, wenn da mal fröhlich behauptet wird, der Apple hätte hochauflösend nur 2 Farben zur Verfügung gehabt. (Es waren fünf, und lasst mich jetzt nicht vom Farbraum-Bit anfangen, weil… naja, ich reagiere empfindlich auf so was.) Zugegeben, das sind Mäkeleien, die schlecht nach altem Geek riechen. Und über Geschmack soll man nicht streiten, schon gar nicht bei persönlichen Zusammenstellungen. Aber ich finde schon auffällig, dass auf der Liste fast ausnahmslos Portierungen von anderen Systemen stehen, und dann nicht mal gute – meiner unmaßgeblichen Meinung nach. 

Deswegen ohne weitere Vorrede eine Liste von sieben Spielen, von denen ich meine, dass man sie kennen sollte – Spiele, an denen mein Apple II und ich nicht vorbeikamen. Leider sind nicht alle bei archive.org. Aber zu finden sind sie trotzdem. 

(Und weil ich gerade nur iPad und Handynetz habe, habe ich keine Lust, Bilder zu suchen.)

1. Choplifter

Ich erinnere mich noch, wie ich das Spiel zum ersten Mal an einem Stand der Hobby Elektronik in Stuttgart gesehen habe und wie vom Donner gerührt war: So smooth, mit so bruchlosem Gameplay – das fühlte sich erstmals wirklich an wie ein Spiel am Automaten und nicht wie eine mangelhafte Kopie. Das Spiel war so gut, dass es auch den Sprung auf den PC gut überlebt hat – dort oder am C64 haben es wohl die meisten kennen gelernt. Aber es stammt vom Apple!

(Choplifter bei archive.org)

2. Dieses Spiel von Jordan Mechner

Lasst uns nicht über Prince of Persia sprechen. Ein nahezu perfektes Videospiel, gemessen an den Möglichkeiten seiner Zeit. Es hatte seinen Ursprung auf dem Apple II, auf dem Jordan Mechner schon ein paar ziemlich fiese Tricks einsetzen musste, damit die Kampf-Animationen so flüssig liefen – erst neulich hat er den Source-Code zufällig wiederbekommen und geriet selbst ein wenig ins Schwärmen – aber bekannt und berühmt geworden ist es auf dem PC. Deswegen lieber ein Wort über Jordan Mechners erstes Spiel: Karateka, ein Sidescroller mit nettem Kampfsystem, technisch und ästhetisch brilliant und mit ganz, ganz viel Charme.

(Karateka bei archive.org)

3. Aztec

Action-Adventure-Platformer. Hinter diesen drei dürren Worten versteckt sich ein Spiel, das den eben erwähnten und 7 Jahre später entstandenen Prince of Persia in vielem hinter sich lässt: Der an Indiana Jones angelehnte Held kämpft sich durch ein Atztekengrab, randvoll mit Schatzkisten, Fallen, Pantern, Schlangen, Krokodilen und anderen Nettigkeiten, bewaffet mit nichts als seiner Peitsche, seinem Revolver, einer Machete und ein paar Dynamitstangen. Ja, Dynamitstangen! Ein Riesenspaß, der es nie auf den PC geschafft hat. Schade. 

(Aztec bei Virtual Apple – Emulation über ein Browser-Plugin)

4. Bilestoad

Killerspiele? Ha! Zwei Gladiatoren stapfen durch eine futuristischen Arena, in der Flugscheiben und Teleportpunkte verteilt sind, und hauen mit der Axt aufeinander ein, bis ein Arm am Boden liegt (oder auch der Kopf). Große Blutflecken. Und zu allem düdelt “Für Elise”. Man sieht den Kämpfer in der Draufsicht und kontrolliert die Bewegungen, den Schwert- und den Schildarm. Ein merkwürdiges Spiel. Angeblich ist der Name eine Verballhornung von “Beilstod”. Das Ganze ist so abgedreht, dass nicht nur 15-Jährige daran eine Menge Spaß haben können. Und der Ruch des Verbotenen!

(Bilestoad bei Virtual Apple – Emulation über ein Browser-Plugin)

5. Apple Panic

Wenn schon Portierung, dann schräg. Diese Variante des Arcade-Klassikers Space Panic dreht sich um wild gewordene, naja, Äpfel. Die man einbuddeln und durch die Plattformen klopfen muss, um sie zu besiegen. Ein Spiel, das mir beibrachte, dass man auch die Generation meiner Eltern zum Zocken bringen kann. 

(Apple Panic bei archive.org)

6. Hitch Hiker’s Guide to the Galaxy

Dieses Spiel kann einen WAHNSINNIG machen. Zumal es ja zur Zeit seiner Entstehung noch kein Internet zum Nachgucken gab. Wie alle Adventure-Spiele von Infocom mit einem intelligenten Parser und ziemlich hinterhältigem Humor, aber ohne Grafik. Von Douglas Adams selbst mitgeschrieben. Übrigens war Douglas Adams nicht der einzige Bestseller-Autor, der ein ziemlich gutes Händchen bei der Umsetzung eines eigenen Buchs in ein Spiel hatte – auch Michael Crichton hat ein Adventure mitgeschrieben, das bemerkenswert ist. Auf andere Art. 

(Hitch Hiker’s Guide to the Galaxy, 30th Anniversary Version der BBC)

7. Pinball Construction Set

Flipper! Am Computer! Nicht nur spielen, sondern selber bauen. Mit virtueller, selbst verlöteter Logik. Okay, die Physikengine ist etwas unzureichend – der Ball kommt dank Rundungsfehlern nie wirklich zur Ruhe – und es bleibt das Grundproblem, dass Flipper sich auf einem Querformat-Bildschirm einfach nicht gut spielen – aber die unzähligen Möglichkeiten, etwas auszuprobieren! Und die üben einen unglaublichen Sog aus. Der Wert dieses Spiels, bei all seinen Fehlern, besteht darin, eins der ersten Sandbox-Games gewesen zu sein.

(Pinball Construction Set bei Virtual Apple – Emulation mit Browser-Plugin

Bonus: Dragon Maze

Das erste Spiel, das ich auf dem Apple II gespielt habe – kurz nach dem Weihnachtsabend, an dem ich die Kiste (gebraucht) von meinen Eltern bekam. Und das erste, das ich verändert habe. Dieses Spiel gab es als Listing – abgedruckt im “Red Book”, im Original-Handbuch zum Apple II (in dem sich auch seitenweise Source Code der internen Routinen des Apple fanden, und viele andere Geheimnisse). Es war in Basic geschrieben und passte in wenige Kilobyte. Okay, die Klötzchengrafik war eher bescheiden – aber es ließ die Wand zwischen Gamern und Codern ganz, ganz dünn und durchsichtig werden. Und viele sind durchgeschlüpft.  

Could you also patch an Alesis IO Dock II?

Is it possible to retrofit an Alesis IO Dock II with an internal USB hub, just like I did with my IO Dock 1? TL;DR: Some have tried and failed – seems like Alesis deliberately switched the code for this hack off. 

01

This photo was sent to me by Blek in the Czech Republic who asked that very same question. He has taken a look inside his IO Dock II and noted that it features an all-new PCB, so my original hack won’t work. And of course there is no guarantee that the prerequisite for the original hack is still implemented in the IO Dock II: the ability to function not only as a USB bus host for the iPad, but as a USB bus slave device, with the iPad working as the bus master.

The good news is: It is simple to give it a go, as Alesis took extra care to label the locations of the USB bus signals:

iodock2

So here is what you do to try it:

  1. Get a male-female pair of plugs matching the connector in question. Possibly a 12-pin version of these connectors, so as in the original hack, 2mm pin grid stripes could work. They are a bit hard to come by, but it is possible.
  2. Solder connections from male to female for all lines but 6 and 7 – the D- and D+ USB data signals.
  3. Get a suitable UBS2.0 hub – I used a Belkin F5U404; you might have to try a couple of hubs if that one does not work.
  4. Take the cable that is meant to connect the hub to the computer – it should have a standard USB plug on one end and a USB mini connector on the other end – and cut it in half. This is the only non-reversible action you are taking, but as it is easy to buy a replacement cable, there is not much harm done if it does not work.
  5. Take the cable half with the USB A-type plug, for connecting it with the computer. Solder the cable wires to the connector that goes into the IO Dock side as follows: Red (Vcc) -> pin 1, White (D-) -> pin 6, Green (D+) -> pin 7, Black (GND) -> pin 8.
  6. Take the other cable half with the USB mini-B plug for connecting to the hub and solder it to the iPad side, i.e. to the connector that is leading to the iPad connection cable. Once again, solder red to 1, white to 6, green to 7, and black to 8.
  7. Do some checking for connections and possible short-circuits. Believe me, it’s worth the effort.
  8. Unplug the IO Dock board connector for the iPad. Insert your freshly-made adapter.
  9. If it works, make a video of it. Become world famous. :)

Wiring diagram

Wiring the USB hub connector: Once again, the trick is having the iPad work as USB host rather than as a slave device, and the IO Dock as the slave rather than as the host. Use your hub’s connector cable, cut in half, and solder the color-coded wires to the connectors as shown.

Sorry for my rather artistic impression of the adapter, hope it gives you the right idea. Pin 1 is to the left, pin 12 to the right. IO Dock side is up, iPad connector side is down.

Just to be sure:

  • I’d strongly advise you to solder and try out the adapter rather than soldering any wires to your IO Dock. It is a good thing to keep that sort of stuff reversible. For this reason, don’t start dremeling before proving that it actually works…
  • …which I won’t guarantee you. Mind you, I don’t even own an IO Dock any more. If you start doing this, you should know what you are doing.
  • Please understand that the base for this hack is a feature that Alesis seems to have implemented deliberately into the first IO Dock (see Dan Radin’s comment): the ability to work as a USB slave to the iPad, in addition to normal operation, where the IO Dock works as a host for the iPad. If the IO Dock does no longer do that, you can try to rotate the USB hub, but that’s about it.
  • Please write me back with your experiences. Please don’t get on my tits with any attempts to make me do this hack for you, or repair your IO Dock if anything went wrong. (Oh my god – I just realize that bullshit warnings are obviously contagious.)

It’s worth giving it a try, isn’t it?

Thanks to Blek for allowing me to use his pictures.

Oberheim Matrix-6 source code file available

Update: Looking for the new, rewritten firmware? Info on how to get the latest version here

There is some (potentially) very good news for Matrix-6/6R owners hoping to get a firmware update – it has come a huge step closer. An extremely experienced engineer has just decided to put his annotated source code file for the Matrix-6 online – you will find it on his page at Oberheim Matrix 6 Firmware. Not the original sources from Oberheim, mind you – they are rumoured to have been lost when someone accidentally dropped the master source disk – but a very carefully annotated listing, reverse-engineered. The engineer who did this even spent the time to mark code that has been re-used in the Matrix-1000 firmware.

So what does that mean to you as a (potential) M6 owner?

The engineer has decided to abandon the project – he thinks that the performance problems of the Matrixes are a result of fundamental design decisions and would need too much effort to get around properly. But his code definitely improves the chances of doing something useful for the code. Someone with skill and spare time might even backport the M1000’s NRPN and matrix modulation Sysex commands into the M6 code.

BTW: My attempts at starting a documented source code file for the Matrix-1000 can be found here, with a hardware and software primer here. If I can find the time, I’ll try to backport a few of Bob’s insights into the M-1000 code.

With the amount of work waiting for me with Jen, I’m glad I don’t own a M-6…