The Crumar GDS Restoration Blog...

[BrianK]

SO thankful you're doing such excellent (and diligent) work on these. Without question, the right person for the job. I'm glad present situations allowed me to ship you the newly-found GDS from California, and to have enough work going to pay for this (necessary) restoration. It is true that the dumb things (like keyboard plastic and slide rubber) can be more of a bear to deal with than major CPU issues!

I'm UBER-excited for the new disc substitute; (1) back ups of all the discs I have is amazing (2) making copies to use for the KS system owner is wonderful! (3) Having ready and fast access to otherwise unwieldy disc libraries is a godsend. Four feet in the past, one in the present - sounds good.

I hope to be able to spend a good amount of time (the hard part) on learning the system. I have goals to try some new sounds that are unlike those found on the traditional GDS/Synergy presets; longer, evolving tones, more bizarre and less-typical sounds. Even with an older modular synthesizer, we've been stuck in the same old sound boxes for years. The concept of synthesis is to have new vistas of sound. Easy in concept, hard to realize and be musical.

I love hearing your demos, very tasteful and beautiful. Sonically, I think the Con Brio (next up to restore?) has clearer A/D going, especially in bells and HF. But I prefer the Synergy/GDS to the Synclav II.

Hope to see you in person soon! Thanks for the posts!

[HideawayStudio]

@Brian - Its truly been an honour to have had the opportunity to work on two out of the three only known operational examples of this amazing instrument and to have done my bit to immortalise the ground breaking software that ran on it in a modern format.

I have made extremely good progress rebuilding Brian's GDS keybed after receiving 40 rapid prototyped replacement key actuator spigots from Laser Lines here in the UK. They did a brilliant job 3d modelling an original part and 3d printing new parts in ABS on a Fused Deposition Modelling machine (FDM).

I have spent the last couple of days fitting nearly 30 replacement actuators, painstakingly polishing the removed key contacts and busbars, lubricating the torsion springs and reassembling the keybed.

The newly printed key actuator spigots were drilled to accept their self tapping fixing screws. The fact that these parts can even accept such a screw without shattering really is a testament to modern rapid prototyping technologies!



All of the broken keys were fitted with new actuator spigots resulting in a perfectly level keyboard. After lubrication there really is no difference in feel between the original working keys and the new replacements.



The two contact pcb assemblies were then refitted to back of the keybed.



The 60 removed silver plated key contact wires and twin busbars were black with 35 years of tarnish and were gently returned to life with metal polish.. a seriously tedious task!



The bottom bus bar was then threaded back through the carriers and the key contact wires soldered back into place one by one using the top busbar rod threaded through the jigging holes (a nice design touch!).

The contacts were then aligned one by one to close onto the bottom busbar. There are two busbars because the 32 oscillator GDS is velocity sensitive (and really makes fantastic use of it!!)

Incidentally, for those who don't know how velocity sensitivity is implemented, its actually a pretty clever but very simple concept with almost all examples working by timing how long it takes for the key contact to change over from its normally closed busbar position (key up) to its open busbar position (key down). Making use of accurate timers determines how quickly the note was actuated and hence how hard it was struck.

The top busbar was then reinserted.



The key contacts were then aligned to ensure each key actuation switched over from normally closed to open and back using a continuity meter. Each contact wire has an integral torsion spring.

The reassembled GDS keybed in all its glory!!

[HideawayStudio]

Richard James kindly sent me a picture recently taken by a friend of his of The Bell Labs Synthesizer aka Alles Machine aka The Blue Monster as it stands now.

Sadly it doesn't appear to be complete internally (I like the blue wire trying to escape from under the computer keyboard!) but still its nice to see the controls much closer up...

[madtheory]

Woah. That's beautiful. Richard D James?

[HideawayStudio]

Woah. That's beautiful. Richard D James?

That's D one..

[madtheory]

Would be amazing to see that brought back to life. Does it have exactly the same functionality as the GDS?

[HideawayStudio]

Would be amazing to see that brought back to life. Does it have exactly the same functionality as the GDS?

Its more the other way round.. the GDS and Synergy were a subset of the Alles Machine in many ways.

The Alles Machine was significantly ahead of its time and had a completely open control architecture defined by the programs running on the LSI-11 host computer. It was able to run everything from Max Mathews and Laurie Spiegel's real-time compositional interactive algorithms, early music sequencers and even Hal Alles real time voice pitch shifter!

All of which was extremely impressive and ahead of its time but without any user interface and the total reliance of coding directly in C and FORTRAN to achieve anything it was never going to have much of a life outside of the lab. That said, it clearly was an important development tool that helped spawn a number of groundbreaking concepts in digital audio and signal processing.

[madtheory]

Does anyone have a copy of the code/ discs for the Alles Machine? Software preservation is something that interests me greatly. In my own little way, I'm currently working on acquiring the original long unlooped samples for the Yamaha TX16W library. The cellos are to die for. Violins and violas are rather special too. Sound like a Mellotron in pin stripes, if that makes sense! We're hoping that (a) the discs are still where they were stored years ago and (b) they still work! Fingers crossed...

I guess you're getting the info from Bell Labs patents? Would you be so kind as to post some links? That would be awesome.

[MITJ]

The only video i know where one can both see and hear the machine! Great piece by Laurie!
Would be marvelous if it could be restored into working condition.


The Computer music journal no4 1977 about the synth.
http://retiary.org/ls/obsolete_systems/ ... h_1977.pdf

Halles had two patents for the oscillator as far as i know and some years ago a japanese gentleman
named "pochi" ported the DKI SYNERGY II+ into an Altera FPGA, unfortunately he later took down the
page, sad because he had published all the code to compile a working unit with Midi including original
sound set.
http://www.matrixsynth.com/2006/12/poch ... nergy.html

Richard James kindly sent me a picture recently taken by a friend of his of The Bell Labs Synthesizer aka Alles Machine aka The Blue Monster as it stands now.
Sadly it doesn't appear to be complete internally (I like the blue wire trying to escape from under the computer keyboard!) but still its nice to see the controls much closer up...

[madtheory]

Thanks for the links. Now that I remember it, CMJ were doing a retro of all of the computer synths in academia. I must see if they've done the Alles machine. Unless I get distracted by all of the other weird and wonderful synths in the series

[HideawayStudio]

I guess you're getting the info from Bell Labs patents? Would you be so kind as to post some links? That would be awesome.

I had the honor of Hal personally sending me copies of the patents he filed at Bell Labs and been very fortunate to have had the kind assistance of most of the surviving members of the original development teams for all three machines. This has included key players Hal Alles and Stoney Stockwell. Without their help this project would have been considerably more difficult. I also had the pleasure of exchanging several emails with Laurie Spiegel and managed to track down and help reunite members of the team who it turned out hadn't heard from each other in many years. What I can also say is that several have written to me to say they really miss the late Max Mathews who has been universally missed by everyone who worked with him and clearly there was a great deal of fondness and respect for him between friends and colleagues. Max in many ways is the grandfather of computer based music software design and theory.

[HideawayStudio]

This evening I managed to setup both complete GDS systems together in the workshop for the first time following restoration - all hooked up to the test PA just begging for someone to perform a duet on these two monsters!!

Here is the complete KS GDS system including the original ADDS serial terminal supplied with it by MTI in 1979 in all its glory. I spent today testing her following reassembly having receiving the metalwork from the paint shop after powder coating.



The two systems will need several hours of soak testing now before they can be signed off as ready to go.

[HideawayStudio]

...and in the other corner - Brian's GDS finally all back together and running....



So I wonder when was the last time two of these beasts stood side by side in the same room!??



To the best of my knowledge this literally represents 2 out of the 3 known operational examples in the world!

[madtheory]

That's an amazing team. And I don't think we'd have MIDI or DAWs without Max Mathews.

The two systems will need several hours of soak testing now before they can be signed off as ready to go.

You just want to play with them

[AlanC3]

Quite aside from getting those old machines working again - the 3D printing of replacement keyboard parts was particularly intriguing - you've done a really beautiful job of refurbishing them. I just went back to the first pictures of Klaus Shulze's GDS looking rather tired to compare with how it looks now and, well,

So... will your next project be restoring the Alles Machine?