Nerdseq Tracker Video Output Adapter

I appreciate that this is a bit niche, but xor-electronics have just released a video expander for the Nerdseq sequencer.
This simply mirrors the sequencer’s onboard screen through a HDMI or Composite connection, so I guess it would be a bit like a live coding display.
I know that there are a few Nerdseq users here. Looks totally cool.


Yes, it’s finally out :slight_smile: Mine should arrive soon. If you watch the latest firmware video there’s a teaser of some video synth magic… at least some patterned output that is not just a mirror of the tracker screen. Thomas hasn’t commented much on how far he plans to push that feature, but there’s at least something in there waiting to be discovered when the new firmware is released next week and the video output modules show up.


This is great news! Cheers for creating this thread @Marizu :+1:t3:
I’ve a mate with the NerdSeq, I’ll blag him into buying one or offer to pay half if he’s not super enthused by the idea.


I hope it has some classic demoscene effects… Plasmas, fires, raster bars, rotating cubes, greetings scrollers, etc, etc… That’d just seem so appropriate for a tracker module.

I’m sure they considered it, at least. Hopefully they implement some.


I absolutely love the Nerdseq! So if the video expander would offer video synthesis I’d be all over it!


Mine has arrived and it’s cool as hell.
I’ve been putting it through the Cortex and into patches.

The guy that is Xor-electronics has a long history of doing chiptune and demoscene music. He is interested in putting some kind of visualiser into the video adapter. The things that he wants to do usually end up in Nerdseq, but it can take a while because, as far as I know, he is the only developer.

I’m assuming that he’ll put some demoscene effects in there at some point, but I’m also interested in having something that can augment my video synthesis rig. I’m kind of wondering about whether it would be possible to use some of the Nerdseq’s sample memory to define a wavetable oscillator with different samples on the X and Y dimensions (I find the single sampling input in Diver to be a bit limiting at times).

Does anybody else have any ideas for generators/unusual ramp ideas/other things that would be useful in an LZX environment?


Thomas from NerdSeq is awesome. He has a lot more in store too. My boyfriend is good friends with Thomas and they hung out last weekend. I can’t reveal anything specific but keep an eye on Nerdseq.


That’s a hot tip. Thanks! I’ve been waiting for this expander.


Please take anything I say on this message with a grain of salt, this info is unofficial and subjet (very likely) to change.

Wanted to chime in on some backstage info I have about this expander. Many weeks ago I got contacted by Thomas about this project, and how could be added to a LZX setup. And we have a cool long chat and brainstorming for the video expander module. The main idea of Thomas is to have 3 modes on this module which uses a PI compute module to render the hdmi/composite video outputs.

  1. First mode is about screen mirroring, so you can program your tracks in your nerdseq on any screen/Projector, this feature is by far the most usefull because its very confortable to write long sequences/compositions plus you can use a USB keyboard instead of the keys on the NerdSeq. Thomas has used this mode in the past on several places like this tutorials or some demos he did on superbooth with a HUGE NerdSeq. AFAIK this mode works flawlessly and is stable. Actually having the screen clonning with your track for projection during shows is a common practice on the demoscene.

2 . Second mode is about sprite loading, titler and lissajous, this mode is a simple way to do some banners and graphics and should be stable also on the current firmware version, here are some demos of this mode

  1. the third mode is the “video synth” mode, which is the most ambitious because like most of you know, video is way more complex and fast than audio, thus, more difficult to generate and mangle. This mode is experimental and unstable for now, but should have more features in the future. There are some really hard software/hardware limitations that need to be tackled by Thomas but I’m sure he will eventually get some amazing on this! The demo he show me abous about some scrolling ramps which is simple and beautiful enough for me to do some simple backgrounds.

The “video synth” mode probably never get super complex (like a big LZX system on a single mode) because thats a lot of digital power for rendering complex compositions and animations but it will be usefull at least for me, so if you have a NerdSeq consider grabbing the expander I think is great. Thomas is a amazing person and wizkid of embedded programming systems… he writes his own bootloaders and hacks the hardware (like most demosceners) to the extreme he deserves to be supported if you can do so.


This is amazing. I was pumped for this expander for a while and just got it in my rack the other day. I haven’t gotten into playing with the modes and Thomas has been fairly quiet on his vision beyond screen mirroring. What you just showed could be incredible… having nerdseq manage basic graphics that correspond and are saved to tracks, then processing the output with other modules, mixing, and having the modulation of whatever is going on in the video side also being sent out to other modules… amazing, precise, coordinated, saved states, even basic saved patches could turn into crazy stuff once run through a larger system. Having a sort of preset base graphic for a track and combining that with everything else downstream… this is so much better than I expected and totally what I secretly hoped could happen. :slight_smile:


Let me add some more first hand information to this all :slight_smile:

Basically the screen cloning/mirroring is together with the USB Keyboard support the only official functionality this expander supports for now and the future (as far as I can see). All other features are unofficial, unsupported and probably unstable except for if I state this differently at some point.

To start with some background information. Though I’ve been kind of part of the demoscene since the 90’s, I always only contributed with music and some silly drawings. I never coded for any demos and I never coded for graphics/animations either. So I got to slow down your hopes a bit that this is going to be a cool demo maker with rotating cubes etc. I don’t even know myself if I can do this at some point.
However, I had (and still have) plans to add some nice video sequencing and synthesis stuff into the expander. They are already some cool results but also some problems on the way which forced me to change plans. To keep it simple, with extensive drawing there is a chance that one or more video commands from the NerdSEQ are getting skipped. And this is nothing I can sell as ‘full featured’ at any point. I think you can understand this. That doesn’t mean that you won’t be able to make cool things with it. More like if there is a probability of 90% on the video commands. (I think NerdSEQ users know what i mean).
Anyway, I will work on this to solve this problem and add features on the way just like you know it from the NerdSEQ firmware itself. But it takes time as I am also busy with other features and expanders. And no promises here.

What you can already do right now (which is also unfinished here and there and is only a small set of what I tested and experimented with already) → When you set a track to be a video track, then you get some video functions:
Mode → Clone / Sequence. Sequence is here the only current mode to make ‘animations’.
Color → Change the color of text or the ‘pens’.
Set X/Y → Set position to X/Y for text, shapes or pens
Line → Draw line x/y
Circle → draw circle with radius (Filled and not filled) to X/Y
Rectangle → Draw rectangle …
Bar → (filled rectangle)
Fill → Fill screen with color
SFNT → Set font (with backgroundcolor or transparent)
Text → Draw text ( 4 letters max now, since they are 2 command columns you can for now draw 8 letters per step per track)
Surf → Add or Sub color from surface. (could be used to generate a fade to color/fade out effect)
And some others which are not implemented. Upper functions are also only partly working as I changed some of these on the fly.
A fun part is to use the automators because they have also video destinations on a video track:
Pen X. Pen Y, Circle X, Circle Y, Color Red/Green/Blue, Bar X, Bar Y.
If you make a LFO for PEN X and another for PEN Y and use other LFO’s for the color, then you can create nice Lissajous for example (or draw the LFO waveforms etc). Using the SURF command to let it fade out… Interresting results come when you set the LFO speed for Pen X and Y to very high speeds and keep the speeds very close to each other (like only 1 - 20 numbers away), then you get nice patterns causes by the lower sample rate of the LFO’s.
There is already some room to experiment. But as i said, highly experimental and unfinished.

And to be honest, I never plan to replace any of the cool video synthesizers. So never expect a full fledged video synthesizer.

That is what I can say for now. I will dive deeper into it in the next weeks.


There’s nothing better than straight talking truth from the (metaphorical) horse’s mouth :clap:t3:
Welcome aboard @firestARTer :wave:t3:
It’s great to read your clarifications & the NerdSeq video output adapter most definitely still sounds nuts :+1:t3:


Holy fuck. This expander is awesome. I see they’re sold out now, so huge thanks for the tip. I’d been waiting on the video expander to fully explore the Nerdseq, because previously it was painful to be hunched over my rack while dedicating the time needed to fully learn and program it. This module solves all of that, and now the Nerdseq can be central to my studio. The whole damn game is changed for me with the HDMI output.


Yeah, I’ve been loving it for all of the same reasons.
I’ve also been merging it with the video synth output using a combination of the Cadet Hard keyer and the Visual Cortex.