Over the course of the past year of development, a few new module designs were added to the Gen3 roster.
Most of them are simple utilities that expand the techniques possible with the initial set of five.
– Stencil is a dual luma hard key generator (extract flat regions from all those DSG3 shapes).
– Matte is a dual RGB offset generator (turn FKG3 into a colorizer). Our simplest module to date, but an important part of the system.
– Stairs is the update of our multi-stage wavefolder/rectifier Staircase (Expedition). It adds a 6th stage and offers an RGB/Luma input option.
– Sum/Dist is a quad summing amp + quad active mult/distribution amplifier (versatile signal routing for a growing system). It shares functionality with Bridge (Expedition)
– Proc is the new version of Triple Video Processor (Visionary) and Passage (Expedition).
– Color Wheel is the new version of Mapper + a Chroma rotator, and was developed in order to test one of the Chromagnon subassemblies in isolation from the rest of the hardware
I don’t know when we’ll be releasing these, but I wanted to share some progress and generate some discussion.
Here are some illustrations of the growing lineup, including the five initial modules.
I’m glad to see that passage is getting a spot in the new lineup it is such a workhorse module.
what kind of price point are you shooting for on the smaller modules? If it still isn’t clear no worries just wondering.
I’m pumped to have an offset generator in the LZX format. Would you care to expand on how this makes the FKG3 into a colorizer? Are we just talking a video into an input and a RGB out from matte into the other channel then just using a key based on the incoming video to replace that part with whatever color is generated by the offsets on matte?
with one matte you could have a triple passive external triple joystick module that could be hooked up for use within the system. This is something I’ve been thinking about for a while just a simple 1v generator for some more joysticks in the system. I already have my joysticks ready to go just need to find a case and solder up the I/O! I like the idea of using matte because then you also have the A+B outs to further influence the system.
Mapper(plus sanding function generator) was the cornerstone of our system for a long time and I was banking on an update coming out. I’m so glad to see it coming back.
The knob caps make the gen3 modules so much clearer at a glance with the filled and unfilled indicators.
are these modules going to have the fun naming conventions like the previous gen3 modules? Like ESG3, FKG3, DWO3 etc?
RE color wheel what happens when you have RGB and the YUV inputs patched at the same time?
what kind of price point are you shooting for on the smaller modules? If it still isn’t clear no worries just wondering.
We haven’t figured that out yet, but I expect the price-to-HP ratio will remain pretty consistent with the first five. Some of these will be lower due to a lack of video rate parts (like Matte.)
I’m pumped to have an offset generator in the LZX format. Would you care to expand on how this makes the FKG3 into a colorizer?
Matte can be patched as Foreground / Background colors to FKG3. If you patch your video source to FKG3’s key input, this creates a 2-band linear colorizer with both color controls and threshold controls.
You can then expand on that…
2x matte + 3x fkg3 is a 4-band linear colorizer
3x matte + 5x fkg3 is a 6-band linear colorizer (more than the Scanimate!)
You may want Sum/Dist as well in a rig like this, to distribute the key source to multiple key generators in parallel
We dropped the acronyms for these.
RE color wheel what happens when you have RGB and the YUV inputs patched at the same time?
The two will end up summing together before the colorspace modifications occur. So you can use RGB in, YUV in, or both.
Ah, but you can’t do RGB in, send YUV out for processing, and then put YUV back in for conversion back to RGB in a single Color Wheel? It would require two to do this (without feedback)?
Not a criticism per se, just trying to understand signal flow.
It is an HSL modulator with both RGB and YUV input options. So you could use it as a converter, but it’s quite a bit of circuit for that use case alone! An RGB/YUV cross converter is a good candidate for a future 8HP module. (In fact it may be a good idea to do both at the same release timing.)
Chromagnon of course is the big sibling to Color Wheel – combining multiple internal colorspace converters, an expanded vector processing core, an auxiliary modulation vector bus, etc.
8hp RGB → YUV & HSL and YUV or HSL → RGB cross converter could be even better!
That would be nice, but the HSL modifier is at least a 12HP module with a piggyback/core assembly (so maximum size stackup for gen3!) YUV/RGB converters can fit on 8HP without issues and are relatively inexpensive to implement.
Honestly not even sure how useful these cross-colorspace conversions would be for my workflow. But I know this issue was discussed a bit before and it did pique my interest as something with possibly interesting misuses (somewhat along the lines of how I like to “misuse” Castle ADC; though maybe that is interesting in its greater amount of abstraction).
I’m imagining that the UV inputs will be nice places to patch in ramps/shapes in concert with a separate video source into the RGB in. Chromagnon is set up in a similar way (you can patch into RGB and XYK at the same time.)
Yes, there is room for a lot of creative interpretation! And multiple paths to take as well – for example, SMX3 can be set up as any kind of YUV/RGB converter you want. The real benefit of Color Wheel is the same as Mapper – direct hue displacement on a single CV input allows for some colorizer looks and palettes that have a very different quality from RGB based sources.
Nice. I really like these. Looking forward to them.
Out of curiosity, how does Stencil work on RGB inputs? I’m assuming it converts the RGB input to a single luma channel then does its keying based on threshold – is this right? Or does it key the R,G,B separately then convert to luma? (Or does order of those operations not matter?) Or is it something else?
It’s the same core board as Color Wheel (which is also a Chromagnon sub assembly), so the affine function is still coming – the Color Wheel implementation allows us to test the assembly’s performance more closely as it relates to Chromagnon itself, which is our primary focus.
I’d assume these are all projects for after the release of pending projects?
A few of these designs are already released in the sense that they are finished, validated and production ready – they are all small projects in comparison to the first five, Chromagnon or TBC2 – but as far as when they will enter production or go up for sale, they are lower priority than anything else we have going on.
Color Wheel uses the ART3 core assembly, as does Chromagnon – we opted for Color Wheel’s implementation first, as it is more closely related to the Chromagnon’s implementation for it. So the affine module will come after. These new modules are all in the category of “what happened along the way” during development of Chromagnon and the first five.
Any idea when a “system reference node” as discussed in a different thread might come down the pipeline?
That’s high priority for the Gen3 modular system, but work needs to start after the ESG3/DSG3 releases, since the system reference (or any ramp gen powered module) will need to derive from the released version of those projects.
How does the Matte module connect to the FKG3?
RGB A/B out to FG/BG input. Solid color mattes/fields are a fixture of any video mixer (selectable as a source), so these are patchable color field generators in their usage context. But those static bias controls can be patched anywhere you need a variable voltage, too.
FWIW, just thought I’d add that I appreciate the work that seems to have been put into providing density of functionality and patching, with lots of ins/outs, while still maintaining relatively well-spaced manual controls, in the Gen3 modules thus far.
