and I've been putting some effort into building one of my own and learning the tricks of the trade.
Believe it or not, this is more than just indirectly related to this project... In fact, it's been part of the plan for some time, now.
For sdramThing4.5 to progress much, beyond just the CKE test (and actually, even for it, really) it's about time I have a PCB, rather than the ratsnest of wires it's currently using.
This is the only backside picture I have currently, from version 3.0... 4.5 is in a box, currently, wherein I added a clock-divider, multiplexers, and oneshots... nevermind the sample/repeat I/O... and you can imagine these antennae are the least of the current worries:
(at least these are length-matched!)Then, with sdramThingZero, there's plans to add buffers/latches to make an 8-bit (or maybe 16) interface with the host, be it an RPi's GPIOs, an AVR, an FT2232H, or a friggin' Parallel-Port. So, there are a lot more "wires" to run, and possibly a few different configurations, as well...
And, again, having decent trace-length-matching and better shielding would probably help the speed-limit problem (30MHz, so far) *dramatically*. And even more than that, probably, would be replacing the 74HC parts with something a bit newer/faster (and most-likely in surface-mount form)...
So, you can imagine, being able to print PCBs at home would make prototype-progress significantly easier. Nevermind, it might've been nice to actually have separate entities for the various versions... I've only done two hand-wired boards in all these versions.
(Wherein I *just now,* after already beginning the build, realize a big flaw in this plan... being that the DVD drives' sleds have something like 2in of travel, and I was planning to build a system with two drive-sleds for X and Y... D'oh! Maybe I'll use the one axis (so far) of my #Big Ol' Modular CNC and make boards up to 12in x 2in hahahahaah. OK, I clearly forgot to think about some things...)