The end of 2017 had a few surprises in store for us which preempted work on Raiden. With that behind us, I’ve been able to resume the project and work on getting things back on-track.
I also had a chance to get my “badge” to operate their laser cutter which I’m planning to use to cut & engrave the (also redesigned) control panel.
Based on the fit of the print, I’m planning to drop one row of switches from the control panel. At first I was disappointed about this because let’s face it, panel switches are cool, but since the original design I’ve come up with an alternative to having an individual power switch for each node that is ultimately superior. I’ve also re-purposed the second switch from being a momentary “boot/shutdown” switch to toggling the LED array between displaying system status (boot, network, temperature warning and two user-definable indicators) and unix load.
I’m also considering exposing the two USB ports of the “head node” through the front panel and revealing an HDMI connector through the back. This would allow the computer to be used in a stand-alone fashion by connecting a keyboard, mouse & monitor. I’m on the fence about this for a couple reasons. One is that I’ve always thought of Raiden as a specialized machine tuned for high-performance computing applications and as such would be managed by an external workstation like most supercomputers.
I also don’t like what it does to the aesthetics.
On the other hand, I can see the value of making it a stand-alone machine especially given the developer-oriented audience of Raiden Mark II. While I’m concerned that putting a full-blown desktop on the head node might impact performance or otherwise “pollute” the system, I want to consider the possibility that by doing so, there may be users who are able to justify owning the machine if it doesn’t require an additional workstation to use it.