Eurorack Video Modules

[laserpalace]

I've been thinking about getting a beauty case to house video modules for a while now. Outside of the Analogue Solutions VD-01, does anyone know of any other euro video modules. I think it would be great for live performance, this video has been pretty inspiring :

(obviously not looking for anything as complex as the Rutt Etra)

[thermionicjunky]

http://www.lzxindustries.net/

[giorgio]

http://www.lzxindustries.net/

[laserpalace]

Thanks for the link, these look pretty amazing! Now to sit and wait...

[numan7]

Mine should be arriving next week!!!!!!!!

Will throw together a quick demo and post to youTube and embed here the night they arrive... can hardly wait!!!

[giorgio]

i never heard prices on these modules? what price scale are they on?? i'd very much so like to get into this stuff as well, all the circuits over at the experimental television center look so rad. they have kool synths too from what I understand.

[MrDys]

Individual module prices have yet to be released. Lars is only selling complete systems (at a reasonable price--I'll let him chime in if he wants to say how much) for the first run.

[laserpalace]

that would be great numan7! I'm curious as to what these are going for as well, seriously considering building a euro just for video synthesis.

[lizlarsen]

Hi everyone -- nice to see this thread pop up!!

I'll let him chime in if he wants to say how much

These modules have much higher parts count and cost per chip than the average audio modules, but I am trying to keep the cost as low as absolutely possible because I want to get these modules out there.

I can give you a preliminary price list, but it's subject to change and we're not accepting new orders just yet:

Video Sync Generator + Color Video Encoder, $500
(every system needs one of each of these at a minimum)
Color Video Encoder, $350
Video Waveform Generator, $300
Voltage Interface I, $150
Triple Video Fader & Key Generator, $450
Video Blending Matrix, $550

All the modules sold so far have been sold in full system sets while we prepare to start doing larger runs. We'll have info on the next run soon. Eventually I'd like to be able to start stocking with the retailers, its just hard starting off releasing this many modules at once!

A "basic" system would be one of each module and 3-6 Video Waveform Generators. A "minimal" system would be Color Video Encoder, Video Sync Generator, and a couple Video Waveform Generators.

With the 7th module, a Timebase Corrected Input/RGB Decoder it will form the "core system", but there are many, many other processing, signal generator and some frame buffer/digital type modules in the works.

complex as the Rutt Etra

With upcoming Vector-to-Raster conversion modules, this type of synthesis is achievable, as well as scanimate type stuff, etc.

I'll be recording more demo videos and updating the website a lot soon, once these first systems are in the mail -- please let me know if there is something specific you'd like to see!

Cheers,
Lars

[lizlarsen]

Oh, and don't forget Dave Jones is planning to release his modules too! http://www.jonesvideo.com

He's just updated his website. Dave's system uses a different approach than ours, which means our system and his will complement each other very well, and both offer unique uses and workflows.

Looks like the year for modular video synthesizers!

[Cat-A-Tonic]

Analogue Solutions VD-01

I think you mean the MFB VD-01.

I like mine, but don't expect the beauty of a green-trace analog oscilloscope.
It is more like a Famicom 8-bit Nintendo oscilloscope.
This is fine if you like the low res. digital aesthetic.


Will there be any kind of slew, or smoothing agent in the LZX modules?

How about a high res. oscilloscope module in the LZX line?

[laserpalace]

so excited that these are being made, thank you Lars!

[lizlarsen]

Will there be any kind of slew, or smoothing agent in the LZX modules?

As far as analog circuits go, we've definitely talked about some video specific VCFs for video texturizing purposes. There is a module in the works for audiovisualization specifically, which includes envelope followers with adjustable decay to control the smoothness of the modulation. You could try using your EuroRack VCFs but those may blur the image more than you'd like, even at higher settings. Still could be some fun, there -- using the Triple Video Fader & Key Generator module to fade back and forth between blurred and non-blurred images.

As far as the frame buffer (digital frame manipulation) module goes, you should be able to achieve various types of video delay/streaking types of effects (and under voltage control.)

How about a high res. oscilloscope module in the LZX line?

There will be a Vector-to-Raster conversion module, that's all you need -- X, Y & Z inputs. This will run at full composite video resolution.

You can also have lots of fun doing waveform visualization by using audio-rate signals to modulate the pulse-width of a sync'ed Video VCO producing vertical bar(s).

[bubblesound]

Hi Lars - it's really great to see this finally coning out. i can't wait to play around with tommy's. congrats on this. i know it's been a LOT of hard work. the demos i've seen so far have been really great, a little quiet.... great job and best of luck with this.

[giorgio]

this stuff all sounds so amazing. i don't know a lot about it, but I am looking to get into more video art and this looking like the way to go for that real feel.


better start saving up now

[lizlarsen]

David -- thanks so much. It's been pretty much nonstop work on this since I last saw you in Texas! We'll get some audiovisualization-focused clips online soon. I need to get access to a bigger Euro system in order to really demonstrate some more complex synergistic patches! Hopefully these initial system owners are gonna go nuts with the clip uploading though, very excited to see what everyone does.

Maybe "voltage controlled" will be a new indie film subgenre by 2011.

[daverj]

My video synth modules aren't available yet, but should start being available later this summer. There will be a number of different modules based on the designs that I made in the 70s and 80s at the Experimental TV Center, and for other video artists over the years. Plus there will be modules based on new ideas that aren't at the TV Center (yet).

I haven't worked out prices yet, but video circuits are more complex than audio, and use more expensive parts, so expect most modules to be a bit higher priced than the average Euro audio module. They also use more power, so when ordering cases, always opt for the stronger power supplies.

I put up a system overview at http://www.jonesvideo.com a few days ago and hope to start adding new information there every week or two over the summer as the modules get closer to being ready.

When Lars contacted me a year and a half ago asking me how to build a video synth, my Euro video system had been under development for a couple of years but I hadn't settled on my final standard for video voltages, so didn't pass that info on to him.

Because of that when he decided to make commercial modules he ended up using voltages that aren't the same as mine. This means that some form of adapter module will be needed to pass video between his modules and mine. His use black and white 0 to 1 volt video signals (three jacks for RGB) while mine use bipolar +/-1 volt color video signals.

My system also uses stereo jacks with the "tip" of every input jack accepting standard +/-5 volt audio or control voltages and the "ring" accepting my +/-1 volt video signals. Using regular mono patch cables each input jack can be given standard signals from other Euro modules, or by using a stereo patch cable they can accept video from any of my video modules. So every input jack is a control voltage input or a video input, depending on which type of cable you use and what module you plug into it.

[lizlarsen]

Hey Dave --

The website update looks great, thanks for commenting on standards!

When Lars contacted me a year and a half ago asking me how to build a video synth

I wanted to comment on this, since it calls into question the history of the LZX project and I don't want anyone to be confused. I got into Synth DIY a few years ago primarily because I was frustrated by the lack of tools for working with analog video processing/synthesis. I have a B.A. in Film and am interested in using experimental technologies for filmmaking, and there seemed to be a lot of unexplored territory from the days of 70's video art.

By late 2008 I was collecting as much documentation and resources as I could find, and contacted Dave to see if his modules were still under development, or if he had any resources to share with an enthusiast. About this time I also had the good fortune of meeting up with Australian engineer Ed Leckie, who had been developing his own ideas and circuits -- after a few months of joint development we made the decision to take this from a DIY project to actually releasing modules and it's been nonstop ever since.

Because of that when he decided to make commercial modules he ended up using voltages that aren't the same as mine.

This isn't entirely true -- there are some fundamental differences between the two different standards which complement different patching workflows and philosophies. I don't believe either method is "better", they are just different and will lend themselves to different approaches, and the world will be better for having both!!

There are three main differences, and I'll point out our own philosophy on each --

1) Signal levels. In the LZX system, we did not want to make any sort of distinction between signal types (CV, Video, Audio, Logic, etc.) This cuts down the cost of the circuitry and makes the system less complicated to use. So all of our modules expect 0 to 1V signals. We offer an inexpensive, accurate Voltage Interface module that can transform anything from line level audio to 10Vpp CV whether AC or DC, to 0 to 1V signals. We'll be releasing other modules in the "Voltage Interface" series in response to user demand.

2) Bipolar vs. Unipolar signals. We decided on unipolar 0 to 1V signals instead of +/- bipolar signals because we feel it allows greater accuracy when mixing images together. In our system 0V = 0%, and 1V = 100%. If you're talking about black and white output video, that's 0V=Black and 1V=White. If you fade out this signal, you are fading to black. If you fade out a bipolar +/- signal, you are fading to gray. In order to subtract one image accurately from another, the 0 to 1V signal makes things much easier (in our opinion) and avoids adding a bunch of extra bias controls to the module. Our system also uses two types of inversions (see the Video Blending Matrix) -- in one mode, the inverted signal is subtracted from 1V (1-n) -- in this mode, you are "adding the inverted image" -- in the other mode there is no 1V bias (-n) so you are "subtracting the image". The fundamental difference between those two subtraction/inversion modes makes little difference when you're mixing audio, but with video it's huge -- and the reason for our choice to use 0-1V DC signals.

3) Triple Colorspace vs. Composite Color. This is probably the most important artistic difference between the Jones and LZX systems. In an NTSC/PAL video signal, you have luminance (brightness) information and chroma subcarrier (color) information being passed together. In Dave's system these are both passed throughout the system from module to module as a single signal. Within the LZX system there is no chroma subcarrier (color) information in any of the signals until the final output (Color Video Encoder module).

What this means is that in the LZX system, a full color image is represented by three separate signals. Typically these are Red, Green & Blue channels, but remember that these definitions are arbitrary until the signals are fed into the encoder -- and we'll also be releasing conversion modules for things like Hue, Saturation & Brightness colorspaces and other alternative colorspaces in the future. We felt that dealing directly with a triple colorspace in the system allowed with more flexible/powerful use of the modules and more routing power when feeding a multitude of signals into an image mix. Our processing modules (Triple Video Fader & Key Generator and Video Blending Matrix) implement a special triple-signal workflow with normalled jacks in columns. This means you can feed in full triple signals -- or you can just feed in a single signal and then you have separate triple controls for that one signal. This type of workflow was inspired by the Beck Direct synthesizer's "Color Chords" concept with cascaded matrix mixers and also the Sandin Image Processor's color encoder module.

So with the Jones system you may have less patching required to pass thru a full color image, and the ability for direct modulation of the color subcarrier that may open up different types of images. Likewise, having the image split up into triple colorspace opens up new possibilities as well.

Of course, Jones' system will probably also include processing modules for altering the RGB colorspace of a video signal -- and likewise we may eventually incorporate some post- or pre- processing modules that deal with the composite signal either before it gets encoded/decoded. But the fact that the internal workflow on each is different means they'll complement each other well.

Typically in an LZX system patch you configure your processing modules into a working environment to then insert signals into.

Some common patches would be...

"Sandin patch" -- two Video Blending Matrixes with outputs feeding A & B channels of a Triple Video Fader & Key Generator. This is similar to having three Sandin "Adder/Multipler" modules.

"Beck patch" -- two Video Blending Matrixes, with outputs of the first feeding top column inputs of the second. The first works as 3 independent signal mixers -- the second works as the "Color Chord" module to assign RGB levels for each output from the three input mixers.

"Colorizer patch" -- one Triple Video Fader with key outputs feeding top column inputs of a Video Blending Matrix -- this allows amplitude band adjustment of signals fed into the TVF, and then the VBM is used to assign RGB values to each amplitude band

Anyway, sorry for this long post. I hope this might help clear up some of the differences. Personally, I plan to have a cabinet of my own modules, and a cabinet of Dave's right next to it!

Cheers,
Lars

[daverj]

A lot of good info there Lars. It does help people to understand the differences in philosophy between the two systems.

A few details I'd like to add:

1) Signal Levels. My system also does not distinguish between CV, video, audio, etc.. Every input can be used with any of those signals. Every control input can accept video. Every video input can accept CV, etc.. The difference is that CV and audio can be directly patched in from other brands of Euro modules without needing a bank of converter modules to turn them into my video format. It does add extra circuitry in that there is a mixer/ converter built into every input. But when you go to use it you don't have to think about where the signal came from. Just patch it in.

2) Bipolar vs unipolar. While it makes mathematical sense to use unipolar for video, I found out years ago that I didn't find it as visually pleasing. If you start with a video signal that includes the full gamut from black to white and start to mix in a second one, unipolar pushes the first one brighter and brighter. When both are at full strength you are left with the bottom half of the mix. Middle grays are now pushed up to white and everything above middle gray is clipped off.

With bipolar as you mix in that second video the blacks get blacker and the whites get whiter. You are left with the middle portion of the mix, going from about 25% to 75% gray now filling black to white. Since most video images have their most interesting information in that middle gray area, the resulting image tends to be more pleasing, to me at least.

Having the bias controls on video processing modules are very useful. Unlike audio, video can look great with the blacks or whites clipped off. But without a bias control you lose the ability to adjust the brightness of the image to be able to see the range of grays or colors that are the most interesting to you.

There certainly are times where having to adjust the brightness after adjusting gains can get annoying, but having that ability adds extra possibilities too.

Another difference between the systems is that video never passes through a passive attenuator in my system. I found out long ago that the tiniest amount of noise in a pot adds a lot of ugly noise to an image as you turn it. After a couple of years of changing attenuator pots over and over to clean up the noise they added, I started making all of my video designs with built in VCAs for every attenuator. It is simple to filter the pot noise out when the pot is simply supplying a control voltage to the VCA. That makes the pots last a lot longer before the noise is noticeable.

Another advantage of using VCAs for all attenuators is that it is a simple matter for me to design it to fade an image to black, white, gray, or anything else.

3) Colorspace. You covered the differences between the systems here quite well. I'd just like to add that there are a couple of drawbacks to each system.

The drawback with composite color video signals is that each module adds a bit of delay to the signal (all modules do this, even audio ones, but it's so small you never notice). Since the composite color synth is adding the video syncs back on before you leave the system, those delays in each module cause the colors to shift on any video going through each module. The more modules you go through, the more the color shifts. There are hue controls for the whole system and for each input (when using TBC inputs) to compensate for the shifts. But it can be a pain to constantly be adjusting those after each patch change, and as patches get more complicated you can't compensate for all the mismatched delays. You don't have that issue with the triple color space system since each signal is really a black and white signal as it goes through the modules and only becomes color at the end when you encode them.

The drawback to the triple color system is that every processing module must happen in triplicate. To mix two (color) images you basically have three 2-input mixers in parallel. To key one image onto another requires a triple keyer. So besides having to patch 3 cables per video signal, the modules also need triple the circuitry in most cases. Of course some nice effects can be created by not processing all three signals together. Running any one of the three signals through a different set of modules gives all sorts of color effects, though understanding exactly what is happening and getting the exact look you want when doing that may not always be easy.

So while adding extra circuitry, it does also allow many image manipulations not possible in the composite color system. On the other hand the composite color system allows some manipulations not possible with triple colorspace. Triple colorspace is sort of like processing stereo audio in a modular. To maintain the stereo you need to run through two sets of identical modules and set everything the same. But you can also rip the stereo apart and send them through different paths.

So, as Lars says, the two systems are quite different, and have different advantages and disadvantages. The results from each system will look fairly different, and the user experience will be fairly different. The two systems do complement each other more than compete with each other. There are some complications combining the two systems because of the differences, but it will be possible.

The other difference of course is that if you have the money set aside you can buy an LZX system today, while you'll have to wait a couple more months to get your hands on mine.

(sorry about another long post to read)

[felixer]

(sorry about another long post to read)

no, not at all. this make for fascinating reading as far as i'm concerned. i come from the music side and have very little knowledge of video systems. like most people here, i guess. please continue to educate me/us ....

always wanted to add this side of things to my music and these are very exciting times indeed!

also love the way both of you are into making things work together instead of against each other. this is how it should be!

[lizlarsen]

Dave, those are all great points! I have a couple more comments, just to explain our reasoning behind the design decisions we made...

On converter/interface modules -- We were originally considering having double inputs or switches on our inputs for 10V or 1V signals, but decided it would clutter the system too much and add too much cost. Rather than passing this cost on to the user, which could be annoying for someone interested in a primarily video system, we decided to have the conversion modules, so that users can pick whatever they need for their setup. Also, sometimes the interface modules themselves provide useful output conditioning -- our "Audiovisualization Tools" module will offer several types of outputs (envelope followers, band-pass filtered outs, frequency-to-voltage) from a single audio input.

There are some drawbacks to the Jones stereo jack solution, although it's quite genius! For example, you can't mult a video signal, you still need a conversion module to run video out through non-video modules (or LZX modules), and you can't use stacking cables. The ease of use, however, of patching straight in, instead of through a conversion module is huge. But it's all gonna mean different things to different people, depending on their systems.

On bipolar signals and bias controls -- While the LZX modules respond to 0-to-1V CV ranges and that's the expected range for viewable video, all of our processing modules can process bipolar signals just as well. And with the switched mode attenuverters, it's like there are positive bias controls for "free". Some of our modules do include separate bias controls, such as the Color Video Encoder, and control inputs on the Video Waveform Generator.

And a general comment, philosophy-wise -- I come from a computer graphics/programming background -- so I'm used to seeing things like RGB and HSB sliders in Photoshop all the time -- so naturally the 0% to 100% triple colorspace workflow is what made the most sense to me, as well as things like mathematical blending modes like "Color Difference", (which is achievable with our Video Blending Matrix's Absolute outputs)

On VCAs -- It's worth noting that the Triple Colorspace signals without the chroma subcarrier are much less susceptible to pot noise than Composite Color. It also came down to cost -- with triple modules, that's a lot of VCAs! Our inputs are buffered with wideband video op-amps right as they enter the module, before the attenuverter, and quality components are used throughout.

In any case, it's apparent how both workflows will be interesting to use together! I can envision a rack of LZX modules focusing on tons of VCOs and blending matrixes, and two output encoders feeding a Jones system for further colorization/processing and blending with external color images from cameras, etc.

Maybe it's close to time to start a Video Synthesis subforum...?

[Animal]

It's great to see this and I'm very excited to start educating myself on video synthesis.... many new words for me to learn and understand.

As well, I will mention that Laserpalace's avatar is magnificently imperious and godlike; threatening yet strangely narcotic.
I like it.

[felixer]

As well, I will mention that Laserpalace's avatar is magnificently imperious and godlike; threatening yet strangely narcotic.
I like it.

like your avatar too. i can almost hear him scream: 'KRUPA, KRUPA'

[daverj]

On VCAs -- It's worth noting that the Triple Colorspace signals without the chroma subcarrier are much less susceptible to pot noise than Composite Color.

Back when I was changing pots on a regular schedule because pot noise was getting into the video, we only had black and white cameras at the TV Center. Of course the pots were being turned all day every day, so did get a lot of use. Time will tell. Also the quality and composition of the pots will affect their life.

It also came down to cost -- with triple modules, that's a lot of VCAs!

Very true. Having VCAs for every attenuator does increase the cost and size of the circuits. A definite drawback. But since most users aren't DIY folks, it can reduce how often they have to send modules in for pot replacement. A number of my video devices have been in daily use for 25 years (some for 35 years), so I've seen a lot of pots get replaced.

Our inputs are buffered with wideband video op-amps right as they enter the module, before the attenuverter, and quality components are used throughout.

I'm also using wideband low noise video amps and quality components, along with high frequency design and layout techniques. I also had my mini-jacks customized during manufacturing to double their life, as well as using pots with 5 times the normal life of the pots used on most Euro modules.

Maybe it's close to time to start a Video Synthesis subforum...?

The problem with specialized subforums is that a lot of people with a casual interest tend to not go into them. On the other hand there are maybe some people interested in video who haven't seen this thread because they don't use Euro. Like everything, there's reasons for it and against it.

Either way, the video party is about to start!

[daverj]

There are some drawbacks to the Jones stereo jack solution, although it's quite genius! For example, you can't mult a video signal, you still need a conversion module to run video out through non-video modules (or LZX modules), and you can't use stacking cables. The ease of use, however, of patching straight in, instead of through a conversion module is huge. But it's all gonna mean different things to different people, depending on their systems.

I forgot to comment on this in the previous post.

Even if I was using standard mini jacks for the video, passive mults wouldn't work. All of my video inputs are terminated with 75 ohms, like standard video, to reduce noise and distortion. So passive mults would double or triple terminate the signals.

I do have a module that is the equivalent of a buffered mult, which in reality is a mini video "distribution amp".

I haven't made a module yet to convert from video to +/-5 for non-video modules, though it's a pretty simple one to make and I will add it at some point. For the most part standard Euro modules can't handle even a fraction of the bandwidth of video.

My video oscillator does have two sets of output jacks. One set of Jones video jacks and another set of standard mono +/-5 minis to go to regular Euro modules. (though again, most of the frequencies of that oscillator are way beyond what standard Euro modules can handle). There are a couple of other modules with special +/-5 outputs meant to go directly to non-video modules.