unibone disk images naming convention -- _34

24 views
Skip to first unread message

Jay Jaeger

unread,
Jan 27, 2025, 3:13:46 PMJan 27
to UniBone
I note that several of the image names (and the associated scripts) have _34 in them (and one with _84).

Presumably the _34 means targetted for a PDP-11/34 configuration.

But what about those with no such _## designation?  what is the difference?  Are those for, say, machines with no memory management?  Or ????

I am asking because my 3 unibones will likely end up in a /24, a /34 and a /45 (or somtimes a /40 or /20 or /05) [to say nothing of a couple of QBones on order.  ;)]

JRJ

Mark Matlock

unread,
Jan 27, 2025, 4:35:38 PMJan 27
to Jay Jaeger, UniBone
Jay,
    I can comment a bit on the RSX disk images on the UniBone. I was involved in some of the beta testing of the UniBone when Joerge Hoppe was finalizing the software for it. A number of other people (some of which I’m sure I’m missing) were also involved including Angelo Papenhoff,  Josh Dersch and Johannes Thelen.

    Most of the initial testing was done with XXDP and RT-11. Later as more testing proceeded, I sent Joerg a RSX11M V4.1 disk image for a PDP-11/35 (11/40) system with default CSR and Vectors supplied by Simh. It includes a console, 4 RL02s, a RX01 dual floppy, a line printer and card reader along with a DZ11 8 line serial terminals. 

   When I got my UniBone (Jan. 2019), I tried it in a PDP-11/84 and under RSX11M+ we found some bugs during DMA, when a different device made an interrupt on a MSCP disk emulation (written by Josh Dersch).  I sent UniProbe bus traces to Josh and Joerg that helped them fix the bugs. So the 11/84 image is from that time, I think. Others who were UNIX gurus supplied some of the UNIX disk images.

   When the QBone came along similar work was done, chasing down bugs in Block Mode DMA and then eventually merging the QBone and UniBone code bases. So the SD card image supplied by Joerg has an interesting legacy of all these disk images including microVAX BSD 4.3 etc.

   There was no intention of supplying a complete portfolio of disk images since the actual hardware that one might have is so varied. The key idea was use Simh to build the disk image you want then sftp it to the BeagleBone. At this point I should also mention that Simh can append data that needs to be removed with auto zap or just use Open Simh to the disk image before you send it to the Beagle Bone.

   This contrasts to the PiDP-11/70 where the hardware was the same and a larger number of curated OS disk images could be assembled and improved over time.

   I’ve run UniBones in an 11/05 and the 11/84. I have an 11/24 but it has that same power supply as the 11/44 and that PSU needs repair. The baseline version of RSX11M should boot fine on your /24, /34 or /45. RT-11 SJ would work well on the /20 or /05. 

   Also remember that the UniBone can load paper tape images directly into RAM for things like paper tape basic. Lastly, Macro-11 listings when assembled to absolute address can also be loaded making it easy to code on bare metal.

Best,
Mark Matlock

--
You received this message because you are subscribed to the Google Groups "UniBone" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unibone+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/unibone/a69ca69c-0fca-4660-bf61-20e6168a6731n%40googlegroups.com.

Jay Jaeger

unread,
Jan 28, 2025, 10:14:21 AMJan 28
to UniBone
Thanks for the info, but it really doesn't answer my original question:  what the difference is between those images without the _34 suffix and those with the suffix, whose name is otherwise the same.  

I'm not asking, per se, which will run on my systems.  I have run my Unibones in an 11/24 and an 11/34 over the past few years.

(The UNIX images, aside from V1, I have actual running copies of on real hardware, and it is possible I was the source for the Mini-UNIX image, a friend having obtained it back in the 1970s, and I have his tape - though there is one other source for Mini UNIX that I have encountered: I provided mine to PUPS/tuhs, but the one on bitsavers is from a tape read in by Al Kossow.)

JRJ

Joerg Hoppe

unread,
Jan 29, 2025, 1:25:20 AMJan 29
to uni...@googlegroups.com
Hi,
Thanks for the info, but it really doesn't answer my original question:  what the difference is between those images without the _34 suffix and those with the suffix, whose name is otherwise the same. 

Sorry for the confusion.
The job here is to handle the whole matrix of <all PDP11 models> x <all possible OSses> correctly.

The model suffix like "_34" means: 

that image works on an 11/34, for other machines it is either not running or just wasn' tested.

From my experience, some OSes like XXDP or RT11 run on almost every PDP-11 hardware unmodified,
so these images may have names whithout a machine code.

Other OSses  need a precises SYSGEN for the hardware, or they run on a whole class of PDP-11s (like: "on all J11s").

If an image appears with and without a "machine-suffix", that may show the transition from "specific" to "universal".
Can you give an example?

kind regards,

Joerg



Jay Jaeger

unread,
Jan 29, 2025, 8:29:38 AMJan 29
to UniBone
I guess the only ones where this really happened where I wondered are for RT11:

rt11v5.5.rl02 vs. rt11v5.5_34.rl02
rt11v5.5_games.rl02 vs rt11v5.5_games_34.rl02

(And of course the .cmd and .sh files that refer to them)

JRJ

Jay Jaeger

unread,
Jan 29, 2025, 7:23:31 PMJan 29
to UniBone
"  I’ve run UniBones in an 11/05 and the 11/84. I have an 11/24 but it has that same power supply as the 11/44 and that PSU needs repair. The baseline version of RSX11M should boot fine on your /24, /34 or /45. RT-11 SJ would work well on the /20 or /05.  "

As of yesterday, so does my 11/24 power supply, or rather, the power control box.  It had a spectacular failure of a 0.1uf 270v capacitor that goes from hot to neutral.  Tripped the power control box breaker, and flipping that back on generated some fireworks.  I did that enough times that now that capacitor is now an open.  ;)  Pieces scattered inside the box.  (The power control box has three of those capacitors, and I will probably replace them all.)

On Monday, January 27, 2025 at 3:35:38 PM UTC-6 ma...@rsx11m.com wrote:

Mark Matlock

unread,
Jan 29, 2025, 9:53:54 PMJan 29
to Jay Jaeger, UniBone
    Sounds like some exciting fireworks! I’ve seen the BA23 RIFA caps blow but not (yet) one of the power control boxes. I’d be interested to hear where you find a good capacitor part number to use for a replacement.

On Jan 29, 2025, at 6:23 PM, Jay Jaeger <cub...@gmail.com> wrote:



Jay Jaeger

unread,
Jan 29, 2025, 11:35:41 PMJan 29
to UniBone
Any 0.1uf (100nf) capacitor, 275V or greater, rated X1 or X2 should be fine - better than the original.  (The X1 and X2 mean that the capacitor is supposed to fail open, rather than shorted, like the original did.) This is just a powerline noise filter, in parallel with a MOV across the mains (hot to neutral).

I found some on Amazon.

JRJ

Reply all
Reply to author
Forward
0 new messages