A modern reinvention of the MIDI router by Conductive Labs
Latest Updates from Our Project:
A brief update on MRCC enclosure design (warning: this is a sausage making post)
over 4 years ago
– Thu, Dec 19, 2019 at 12:54:52 AM
If you are not interested to see how the sausage is made, you can skip this one. ;-)
We've been tossing around ideas for how to do the rack ears. We put some connectors on the right end of the enclosure, so those cables can easily go into a rack. Not decided if the power button will stay on the side, but we don't expect it would be used often. We just don't want it to be easily pressed by accident.
So with some stuff on the end, we need access to it while it's in a rack. We also want to keep the connector profile low so as not to defeat the vertical connector model.
We are currently leaning toward using an extruded enclosure. Extruded aluminum provides a strong, cool looking and cost effective box. It also simplifies the assembly by providing internal PCB rails which eliminates some stand-offs for board mounting.
For desktop use, this is an example of what the end plate would look like.
For rack mounting, we need access to those connectors on the end, and a place for cables to route into the rack.
Another example of a rack ear concept design we like. That cable hanger is pretty handy!
Let us know what you think! We'll keep refining the design and share the results when we are getting close to the perfect solution.
All the best,
Steve and Darryl
MIDI performance baselining
over 4 years ago
– Wed, Dec 18, 2019 at 11:51:53 PM
The Kickstarter for MRCC is soon coming to a close, and we want to thank our contributors for your support! We will continue to offer regular and transparent updates as we progress on development. We feel people taking a risk with their money on a Kickstarter project deserve to know what's truly happening on the project. We'll not exaggerate or hide the challenges that are inevitable with a complex hardware product like MRCC. Having said that, it is coming together nicely so far!
We've done some MIDI performance testing, just to get an idea of where MRCC is currently with respect to latency and jitter. We have not done anything special to optimize performance yet, but the base line will tell us how we are doing over time as the code develops, where we need to optimize, and what impact filters and other MIDI modifiers have on the system.
In this case we were using a logic analyzer to capture the data. The logic analyzer doesn't give us picosecond timing like a good scope can, but it's great for capturing many outputs at once. We connected the analyzer probes directly to the MIDI 5 pin DIN input optocoupler outputs, then looped 2 outputs back to inputs to capture the data. The NDLR was generating MIDI clock and messages.
The image below shows a typical latency for a MIDI Note On message.
MIDI Note On latency
This image shows a typical MIDI Clock jitter. There are two outputs being captured and there is no latency between outputs.
MIDI Clock jitter
The graph below shows a section of a 2880 sample size of both clock and MIDI Note On and Off data, captured with the logic analyzer. This shows us the minimum and maximum jitter over a larger period of time with real-life MIDI data. So far the results look promising. The numbers are in microseconds (millionth of a second).
Jitter over time
This isn't the final word on MRCC performance, it's just a snapshot in time that we can use to see how MRCC is improving or degrading with further development. The results are exactly what we'd expect, so far so good.
