ersatzplanet wrote:Are there options to wire internally to the normalized connectors of those jacks?
Those particular jacks are Tini-Jax without break contacts. We could fit break contacts, but they can only be guaranteed to break with a Tini-Jax plug and then that will probably jam in your other modules. There is no jack that can be guaranteed to work with all the varieties of 3.5mm plugs out there because there is no standard to refer to. What often happens is that the contact does not completely break and you end up with a short between the normalled signal and the one being plugged in.
However, if you want direct wiring you can go in to the DB25 on the pcb which conforms to the Tascam standard. All of our modules use this method which is how we can offer different connector options and it also makes it more serviceable. All the inputs are differential and inversion is almost free by adding a DPDT "phase change" switch.
If I had one of these, I would hardwire it to modules in my rig. Would it be possible to make a smaller and cheaper version with no jacks?
No jacks, yes. Smaller, a little. Cheaper, a custom panel would be a bit more, but we could do one with your module names on for about Ã‚Â£100 or let you do your own and just supply the OEM works. It would be cheaper to have a standard panel and convert the jack holes to pin parks. We can supply crimped DB25 looms to save you soldering up DB25 plugs.
More specifically, the Analogue Solutions matrices are passive,
whereas the Hinton matrices are buffered.
An NxN matrix with shorting pins
gives you no more that two rows of N jacks and some mults do--a waste of a matrix.
Our pins are precision resistor pins and our NxN matrix is equivalent to N N-input precision adders.
The buffered matrix has all the benefits that a buffered multiple + an active mixer have.
No signal droop.
There will be no droop by adding more pins, even if you route one input to all outputs, but you will lose 1% of the signal if you are coming from a module with 1k output resistors. This is a problem with all buffered mults, they do not make up prior losses. For these situations our Trimmer module may be set to precisely compensate for drops earlier in the chain.
The biggest caveat with all the pin matrices is the expense of the pins.
At 10 British Pounds Sterling they are still overpriced IMO.
That is even more expensive than a TipTop Stackable,
yet with a much simpler machining process.
That's not an equivalent function. EMS charge Ã‚Â£8 for their pins which are nowhere near the same quality.
I understand the niche manufacturing quantity issue,
but there must be some way to bring the price down on those pins.
There is only one manufacturer of these pins and they set the price and a MOQ. If this becomes a popular product we can negotiate a better price based on quantity. Same with the matrices. We are going to manufacture this in batches, so if you don't mind waiting a little while we can aggregate orders and pass the savings back.
A pin matrix allows you to fit more connections in a smaller space,
but I would much much rather have the instantaneous rerouting action of a button matrix like the SwitchMix.
Button matrices have the potential to make the modular more of a cohesive, live-playable instrument IMO.
It's horses for courses really. A switch matrix gets too big as the size goes up and costs a lot more as all the hardware is there all the time and it goes up geometrically with the area. A pin matrix is better for larger sizes as the cost only goes up by the length of the busbars and you can choose how many pins you need, certainly less than one for every hole.
To this end, I'm building a 4MS BendMatrix.
Beware that these are not the same thing. Like you pointed out above they are like a passive matrix except that the switches are FETs and have an on resistance.