There’s been some discussion on Facebook and Discord about the SSSR Labs Matrixarchate, which is a pretty nifty audio module for making drastic patching changes quickly using a patch matrix. It supports save and recall of settings to 60 different storage locations, and selection of saved patches via a trigger input or via the i2c bus.
There’s obviously been a lot of interest on whether this module can be used for video. The patch matrix is implemented with unbuffered I/O and a switch IC (the MT8816AE) that makes passive high-bandwidth connections, and indeed the datasheet for the IC calls out video as one of the intended applications.
Using the module in practice, with complex sources (e.g. live cameras) it appears to work more or less okay with video. That’s the good news.
The bad news is that it doesn’t pass the oscilloscope test and it will blur your crisp edges. I finally upgraded my home studio scope so I didn’t have to lug modules to the lab for testing. Here’s a Prismatic Ray square wave at max frequency (yellow) routed into a 4ms Buffered Mult and into Matrixarchate input 1, and the resulting wave appearing at output A (purple.)
Something else in the design must be attenuating higher frequency signals. That doesn’t mean it’s not useful, but now people can make an informed decision!
I also just got one of these, intending to primarily use it with video. I haven’t found edge blurring to be a problem visually so far and can’t see any difference between the waveforms with this patch (using Bridge as a buffered multiple), though I’m using a 20MHz analog scope.
One thing I thought would be cool, and seems to work quite well, is since LZX output impedances are all roughly the same, you can use this as a matrix mixer by shorting multiple outputs together – passively mixing like this can be nice because you can e.g. sum N signals with 0-1V swings and get an output with a 0-1V swing without worrying about clipping, as the “output” is the (weighted) average of all the “inputs”.
Are your probes set to 1X or 10X? I went through this same journey (using 1X probes, not using a buffered mult) and folks on Discord gave me great feedback on accurately measuring the PR square wave
But yes I agree, if the effects aren’t noticeable with your style and system then it’s quite useful!
I wish more activity happened here on the forum, it’s more asynchronous and a better knowledge record, but there have been some fun conversations so far.
Ah, yeah, that probably explains why these don’t look nearly so square. These aren’t scope probes at all, just minijack/RCA cables -> RCA/BNC adapters -> scope, no doubt a wild journey in impedance mismatch. I’ll patch something up tonight to do some more A/B testing with hard keys / video sources.
Yeah, it’s annoying that there isn’t an off-the-shelf euro solution for high-impedance probes (that I’ve found.) I just use bare 1/8” plug ends and clip a normal probe to the solder points.
I hooked up a vertical ramp -> keyer -> fader patch to A/B compare the horizontal square wave before and after the Matrixarchate, and yeah it’s definitely visible where the blur/shift kicks in but I think what little capture equipment I have is probably not capable of showing it. It’s like 0.5-1 pixel smearing on my display, I may never have noticed it if I hadn’t gone looking. A little vaseline on the lens.
Thanks for this testing. But im not sure about the conclusion here… ?
I was just about to order a kit when they come back in stock at Thonk. My plan was to use it primarily for vector rescan patches. Any of you out there that could test/explain to me what functional visual difference this small blurring would give in vector display use…
Since im not completely sure i want to blur my edges unintentionally. But on the other hand if its unnoticable, in most situations, it looks like a really practical tool.
Tried to follow the discussion on the Discord, but got lost (completely new there, just signed up - find it quite complex following both several facebook groups and pages, this forum, and now also this stream thingie… and get solid info oit of it. ) and didnt really find any conclusion.
What would the consensus be among all of you? To use it, or not to use it?
@luix If im not mistaken, you have one of these, and use it in your vector patches, right? Noticed anything undesireble?
The rolloff will probably not bother you at all if you’re working with non-synthesized/non-text video, smooth textures, or a glitch aesthetic. But if high contrast sharp vertical edges are important to you, you might not like it, especially if you’re going through the matrix more than once.
I put mine back in my rack, but I don’t use it on every patch.
I dont have it eh SSSR Matrixarchate, I think you are refering to @jjplano.
Anyway from the FB thread and from Javier’s testing, I can tell you that that Matrixarchate is capable of doing videobandwidth signal routings.
If you check the build manual/pcb layouts you will see that there are no OpAmps on it, this means there are not input/output buffering with audio bandwidth OpAmps (like TL072) which is very common in audio eurorack modules, that limit the bandwidth.
Anyway answering your question, this module should work perfectly both for vector graphics if you are doing oscillographics+lasers (audio bandwidth) stuff.
Thanks for the looking at datasheets for me. @luix Thought you had one from pms from the past… but anyway, think i will go buy one when it comes back in Stock. Thanks!
Thought you had one from pms from the past… but anyway, think i will go buy one when it comes back in Stock. Thanks!