Just a reminder: if you're looking to follow me because of my interests in libre hardware and software hacking (including but not necessarily limited to #fpga, #Kestrel3 and #kestrelcomputerproject , #rc2014, and so on), you might also be interested in following my @vertigo account as well. I will slowly be migrating that way.
Thanks!
Briefly, what with all the concerns surrounding mastodon.social of late and the strongly expressed opinions on whether or not some should defederate with it, I'm hedging my bets and beginning the process of creating an alternate account at @vertigo .
Folks interested in following my progress with #Kestrel3 (more generically, #kestrelcomputerproject ), #vdcII (aka #vdc2), and other #fpga exploits should probably follow me there.
Another thought regarding #Kestrel3 : instead of Tripos, write a single address space subset of Plan 9. That way, when I can afford to support a PMMU, I can port a real Plan 9 system over, and I wouldn't have to completely rewrite all the utilities and documentation I'd've no doubt built up by then.
FML: why can't this be a simple decision?!
What I am thinking for the #Kestrel3 going forward is I should port Tripos (yes, that means system software will be written in BCPL, not C) to get *something* working.
Then, try to bring up a GUI environment within Tripos. It won't be compatible with GEM, but I envision it would share its essential features (its overall architecture, for example) along with some repairs to GEM's architectural deficiencies (8.3 filename limitation, single-click latency, etc.).
So, I looked at EmuTOS source code this morning. I don't think I'll be porting this software to the #Kestrel3 . Reasons include:
1. As I expected, it is tied tightly with the m68k architecture. And, that's OK; it does what it says on the tin. I just was hoping that some of the footwork for making it more portable has already been done.
2. There is no migration plan; nothing to tell me, "Start here." Tripos, however, does have at least *notes* to this effect, which is very helpful.
<p>I've completed the initial design for the printed circuit boards for the RC2014 computer.  This is my very first PCB design in KiCad, along with my very first 4-layer board, so I really have no idea what I'm doing.  That said, I did it anyway, and the first batch of printed circuit boards have been submitted to OSH Park.</p> <figure><img class="lazy" data-src="https://cdn.hackaday.io/images/9134751598667080661.png"></figure> <p>Nothing has been ordered yet, however.  I'm waiting to hear back from OSH's support team on one question I had after uploading the designs.  But, assuming everything checks out, I'll be placing an initial order for three circuit boards.</p> <p>The circuit boards will make use of a 14-pin DIP socket as a connector for a mezzanine circuit board.  This connector provides access to 11 digital I/Os on the FPGA, which is enough to drive an analog VGA port using discrete resistor DACs.  The pin-out of the J2 connector (as it's currently labeled on the PCB) is as follows</p> <pre class="hljs ruby"><span class="hljs-params">| Description |</span> Pin <span class="hljs-params">| Pin |</span> Description <span class="hljs-params">| |</span>-------------+-----+-----+-------------<span class="hljs-params">| |</span> R2 <span class="hljs-params">| 1 |</span> <span class="hljs-number">14</span> <span class="hljs-params">| R1
@vidak @theruran I have never actually run Plan 9 myself. I would be totally lost within it. However, as one of my personal projects is to build the #Kestrel3 home-brew computer and *eventually* equip it with a MMU, I'd like to port Plan 9 to it somehow, some way.
It seemed to be a task that was much simpler than porting a BSD or Linux environment, for instance.
The Amiga's Agnus chip was in part designed by Jay Minor's dog, Mitchie, according to ancient lore.
Plan 9's PR animal is a bunny.
Minix 3 has the best mascot ever: a raccoon.
Linux has penguins inside.
The #kestrel3 will be codesigned by . . . Big Chungus.
DJ Sundog
Vertigo