And Another One Down...

This evening I verified that the Atari 130XE works great with the Liber809 board (thanks to Mike Rowen for graciously donating the system to me for testing).  Since the latest incarnation of the Liber809 is so compact, I was able to put the Atari 130XE together, case, screws and all.  Makes it look all natural!

You can see how it looks, and even watch it boot NitrOS-9 from DriveWire on this YouTube video.

Almost Ready...

The last 5 days or so have been spent putting together the board and testing.  Three trace cuts and three added wires later, the board is coming up on the Atari 800XL here on the bench.  The trace cuts and wires were needed to work around omissions in the latest board run (our fault, not the PCB manufacturer's).

Because of this extra post assembly work, I've decided against selling kits and will instead sell entire boards already put together.  This allows some quality control and assures me that what I'm sending out actually works.

There's still an issue with the board working the Atari XEGS, so we're trying to pin down the proper equations to deal with the timing issues surrounding that system.  Once that's in place, the boards will start going out the door.

They're Here....

The boards arrived this morning via FedEx and they look great.  I'm still waiting on another order of parts to arrive before I begin to assemble and test the board.  That should happen early next week.  Assuming all goes well, we'll start making kits available shortly thereafter.

Coming Together

The absence of a post over the last couple of weeks shouldn't be an indication that nothing is going on.

On the contrary, things are on the move.  A few days ago, we sent final artwork over to the PCB manufacturer and should be getting boards next week.  Parts have been ordered and I will be assembling a number of boards.  The preferred method of selling the Liber809, however, will be in kit form.

The product will be sold through www.cloud9tech.com.  A final price hasn't been set yet, but that should be set soon.  In the mean time, anyone interested in one of these boards, either in kit or in assembled form, can email me to be put on the list.

Progress On The Software Side

This evening I spent some time reworking the NitrOS-9 port.  The change in the Liber809 boot ROM made it necessary to do some modifications to the kernel to accommodate the new loading style.  But now all is good and we're booting NitrOS-9 Level 1 on the Atari 800XL.

Which brings me to Slor's excellent demo.  Fuji started life out as a blob of code loaded from the ROM load routines via DriveWire.  It was later modified to be position independent, but still had absolute references and self modifying code, neither of which are welcome in an OS-9 environment.

I spent some time with Slor tonight, reworking Fuji to actually run as an OS-9 executable.  It's not quite the good NitrOS-9 citizen that it should be, but with a bit more work, we will be able to run it from the shell, then quit it at our leisure and return back to the OS.

Finally, I'll be doing a KeyNote presentation on the project at the upcoming CoCoFest  (http://www.glensideccc.com).  I plan on making this a very thorough and detailed presentation and will post the notes when I'm done.

Back in the Saddle

As I mentioned in a previous post, work and other tasks have pretty much sidelined the Liber809 project.  However, I'm spending some time this weekend to rework the NitrOS-9 port based on several pieces of good news:

1. The Liber809 board had stopped working for some inexplicable reason.  As it turns out, a piece of solder flake had found its way onto the trace from the Maxim DS1100 that leads to the Q clock on the 6809, leaving it perpetually grounded.  Needless to say, I'm going with solder masked boards from now on.
2. The XEGS had timing issues with the firmware going into all-RAM mode.  The ANTIC was picking up writes to addresses outside of its range of $D000-$D7FF.  We have theories on this subtle timing problem, but reports from others that the 800XL and 1200XL worked fine in all-RAM mode has us thinking this is something unique to the XEGS.

Sloopy, a member of the AtariAge forums, sent over an Atari 800XL which I readily disassembled.  After some mucking around, the Liber809 board came up, and indeed worked well, even in all-RAM mode.  Slor's excellent Liber809 demo in the previous post came up just fine, sound and all.

With a solid, reliable Atari system running the Liber809 board, I can now focus on bringing up NitrOS-9 through DriveWire, which is how I envision the majority of folks will want to use the operating system.  So I've carved out some time this weekend to do just that.

Demo Time

Slor sent over a link to a YouTube video which shows an impressive demo of the Liber809 in his Atari 1200XL. Check it out here.

Thanks Slor!  Great job!

In other news, work has had me sidelined for a while on the project, but we are still working on the next revision of the board.  More news as it comes.