On May 15, 2012, at 2:09 PM, Dave McGuire wrote: Or even an /83, which will at least give you the PMI. - Dave

Just looked up "Megger" on WikiP and discovered that it's all part of the same group as AVO. The name of the company producing the gadged was changed to "Megger Group Ltd." only in 1987. Much more here: http://en.wikipedia.org/wiki/Megger_Group_Limited --Chuck

On 15 May 2012 at 20:00, Zane H. Healy wrote: "It is incorrect to use the word "Megger" as a verb, noun or in the plural form. [36] It is an infringement of the trademark if used in reference to a product not made by the Megger Group or its subsidiaries" So, I'm supposed to use it as what, an adjective or adverb? "This PDP-11 is Megger messed up!" Megger-doubtful, Chuck

Hi In any case, I used one like that to check coax until we got a TWR. I remember that we had to go out onto the field to disconnect the baluns or it would magnetize them. Dwight

On Wed, 16 May 2012, David Riley wrote: So what is the correct generic term for a MEGA-Ohm meter? Should "megger" have been permitted trademark registration?

Chuck Guzis wrote: If the National Association of Realtors really wanted to make sure that they protected the trademark properly they would require that, as well as using it as an adjective in all of their marketing materials.

* David Riley <fraveydank at gmail.com> [120516 21:52]: [..] Or a typewriter...

On Thu, 17 May 2012, Rob wrote: 'round here (and on both coasts), it's "scotchtape".

Well, when I talk about a Hoover, I mean the large concrete water-impounding bulwark. -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com -- I think, therefore I'm dangerous. ------------------------------------------

Well, I guess dams could be considered to have skirts. -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com -- Remember, Windows is not a virus. Viruses actually do something. -----------

Chuck Guzis wrote: That's just normal trademark stuff that everyone ignores. It's really not appropriate for the article, and should be removed. At most there should be a simple notice of the trademark.

I've just looked at a copy of 'Telephony' This is a serious textbook used by tleephone engieers over here back i nthe days of manual and Strowger (step-by-step) exchnages. In the chapeter 'Measuring apparatus and its use' there's a section headed 'PArt V -- The Megger and Bridge Megger'. It begins 'The Megger, which is the result of evolution from the original Evershed Ohmmeter consists of a direct current generator and moving double-coil instruemnt contained within a single case.' So misuse of this term is long-established (this book dates from 1932 I thin) and is not going to be easily reversed! -tony

I worked at a testing facility for testing high voltage interrupters (fuses, etc.) and we had a 13500 feed directly from a power plant. When we had to rewire the primary side of the facility, we'd pull a large high voltage breaker. Before reinserting it, it was always tested for megohm resistance before putting it back in circuit (open). A little resistance went a long way there. It was interesting having to use a Megger at one end, but for other blows of 6000 amp 600v fuses we had a Moh meter, which used a null meter to indicate the conductance. (hopefully I remembered all the right terms). Jim On 5/16/2012 12:40 PM, Tony Duell wrote:

AVO is of course an acronym for Amps, Volts, Ohms. The AVOmeter was one of the firrst multimeters in the UK (I think you can find them form the 1930s, if not earlier). Old hands over here use the term 'AVO' (or AVOmeter) generically to refer to a multimeter, even if not of that make. So if one fo them says 'Have yuo got an AVO?' and you hand him a Fluke 87, he'll be happy (although the company lawyers won't be :-)) AVO made other test gear too. Their valve testers (or 'Valve Characteristic Meters' are rather nice instruemnts, for example. I've also used AVO brand signal generators, bridges and transistor testers. -tony

Over here, the common term is 'multimeter'. DMM is also common, but not normally used orally (in other words, a bork or article might refer to a DMM, but if you want to borrow one, say, you'd say 'Can I borrow your multimeter'. DVM is only used for a pure digital voltmeter. And I've neer sene either DMM or DVM used for analogue instrumetns. VOM is not common over here. Most UK electorncis people woudl know wht it means, but would think of it as being an American term, like 'tube' in place of 'valve'. Incitentally, Iv'e sene 'VOM expanded to 'Volt, Ohm, Milliammeter'. Again, we'd recognise the term, but not use it normally. For obvious reasons, that would be called a 'valve voltmeter' over here, which I have never seen abreviated to VVM or anything like that. -tony

At 6:35 PM +0000 5/15/12, Benjamin Huntsman wrote: Save the money you'd spend on the /93 board & RAM, and get a SCSI Adapter (ideally one that will support a compatible CD-ROM drive in addition to hard drives and maybe even tape drives). You might also want to consider a BA123 rather than the BA23. Though if space is at a premium, you'll definitely want the BA23. Think about a nice DEC terminal (I like VT420's, others have their favorites). Also an LA75 is a nice small printer that uses the same ribbons as the Applewriter. Having a SCSI based VMS box can be useful in getting the PDP-11 media built. BTW, I'm still on a /73, but prices had dropped on /83 CPU's and RAM enough the last time I looked that I've seriously considered one. Also, have you looked at emulation? Both E11 and SIMH provide excellent emulation support. There are a bunch of others listed at my page on PDP-11 emulation, though it's horribly out of date. http://www.aracnet.com/~healyzh/pdp11emu.html Zane -- | Zane H. Healy | UNIX Systems Administrator | | healyzh at aracnet.com | OpenVMS Enthusiast | | | Photographer | +----------------------------------+----------------------------+ | My flickr Photostream | | http://www.flickr.com/photos/33848088 at N03/ |

On 5/15/2012 2:08 PM, Dave McGuire wrote: On that note does anyone have an extra they are looking to trade, sell, or hand over the cost of shipping ? I have a microPDP 11/73 in the original pedestal BA23 with no cards at all in it. Can it even be called a 11/73 with no cards ? I guess what I have is a BA23 pedestal with a cool MicroPDP 11/73 badge on it.

On 05/15/2012 08:10 PM, mc68010 wrote: I am mostly in "acquisition mode" nowadays, but I do have lots (and lots, and lots) of DEC boards. I can probably swap you a CPU at least. Whatcha got to swap? -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 05/16/2012 12:23 AM, Josh Dersch wrote: ...and while we're on THAT topic, I'm still looking for a power supply for a BA123. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 5/16/2012 12:21 AM, Paul Anderson wrote: I'm in the Seattle, WA area. I'm guessing that shipping costs would be prohibitive :). Thanks, Josh

On 5/16/2012 2:30 PM, Glen Slick wrote: Aww, thanks :). Nice! Also, I'm definitely going to have to drop by 3rtechnology's physical store sometime, just based on a couple of the pictures they have up. - josh

From: Ian King Sent: Wednesday, May 16, 2012 11:39 PM On 5/16/12 10:48 PM, "Josh Dersch" <derschjo at mail.msu.edu> wrote: [someone wrote] Actually, they've moved from where their map shows them (close to the museum) to a location several miles to the south. Fortunately, they're now across the street from Slim's Last Chance, where we can get a great bowl of chili and a beer. (Use the street view button on the Google Maps page and look for yourself.) Rich Alderson Vintage Computing Sr. Systems Engineer Vulcan, Inc. 505 5th Avenue S, Suite 900 Seattle, WA 98104 mailto:RichA at vulcan.com mailto:RichA at LivingComputerMuseum.org http://www.LivingComputerMuseum.org/

Hi! haha... why the 93? Because I pretty much woke up one day and thought to myself "I'd like to find a PDP", so I looked up which was the smallest/ fastest, and that's the 93. I'm new to the PDP's, so I really truly had no idea as to what the going rate is. I'm not opposed to a 73 or 83, but I'm pretty set on the BA23 pedestal as the ideal form factor, for space/weight reasons. As to what software, I figured I'd start with V7, Ultrix-11 or something like that. Hadn't really thought that far ahead though. Why the DEC hardware? What else is there? :) I've always liked DEC stuff, and don't usually consider non-DEC stuff for hobby purposes. After all, isn't most of this stuff just supposed to be for fun/curiosity's sake? Thanks! -Ben

And NONE taken. I definitely admit that I am just as addicted to software, specifically the RT-11 operating system and layered products, on the PDP-11 as many other individuals on this list are addicted to their own area of focus in hardware. And there probably is not that much of a difference. About the only aspect of software that might be substantially different is that there always seem to be additional enhancements along with bug fixes to provide a challenge in the software area. Perhaps one of the hardware addicts could help me understand something that is similar on the hardware side of the fence. For example, one recent software project is to enhance the KED / KEX RT-11 full screen editor to support more than 24 vertical lines which is all that were available in the early 1980s from DEC. When DEC produced the VT420, even that terminal has such a small screen that the 48 vertical lines result in very compressed characters. Now that I have available 24" monitors with software that supports up to 60 vertical lines, the challenge to use that capability seems irresistible - I guess because I am an addict PLUS the additional lines help make changing the software much easier. Since I now routinely use Ersatz-11 with a video card that supports 50 rows by 80 columns AND 44 rows by 132 columns, having a K4x.SAV program file which supports up to 60 vertical lines seems much more satisfying than having solved a Sudoku puzzle - although writing a program to solve Sudoku puzzles would also be a fun effort. Has anyone else any experience with PDP-11 software or hardware with terminals that support more than 24 vertical lines? Jerome Fine

Hi Mouse, I've been looking for books of the 4x4 ones fior quite a while, but usually just find a handful in the back of a book of 3x3s. Care to share any? Thank, Paul On Wed, May 16, 2012 at 12:09 PM, Mouse <mouse at rodents-montreal.org> wrote:

I didn't do a full complexity analysis on mine, because it uses heuristics that mean any human-solvable puzzle is solved in so close to zero time that it's not worth improving it. (Specifically, most of the time there is at least one cell that can be filled in only one way for some reason, and the program knows at least four different possible such reasons; I've rarely seen puzzles needing a search tree with more than a dozen or so nodes, and many need no search at all, ie, there is no backtracking.) /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

I'm not sure what you mean by "deterministic". But there are sudoku which have only one solution but for which that procedure does not work. I just ran my generator with the option that says to not exclude such cases and got (I've manually replaced three .s with -s and three more with *s; see below): +-------+-------+-------+ | 2 9 - | . . 1 | - 6 8 | | . 3 . | . . 2 | 4 - . | | . . 1 | . . . | . . . | +-------+-------+-------+ | 4 . * | . 6 9 | . . . | | . . 9 | . . . | 6 . . | | . . . | 2 5 . | . . 7 | +-------+-------+-------+ | . . . | . * . | 1 . . | | . . 6 | 3 * . | . 5 . | | 1 2 . | 8 . . | . 7 6 | +-------+-------+-------+ This has only one solution. However, if you just fill in forced cells (cells that can have been restricted to only one possibility by other numbers), you'll find that you get stuck after placing three numbers, the three spaces I've replaced with *s. Then my program resorts to trying possibilities and doing backtracking when it gets stuck. The three cells I've replaced with -s are the three cells where it's done this when it finds the solution. (They contain 4 in the leftmost one on the top row, 5 in the other one on the top row, and 9 in the one on the second row.) There's nothing nondeterministic here; the provided values do uniquely determine the values for all the other cells. The only thing that's interesting is _finding_ that unique solution. I find it interesting that his idea of "proper" sudoku includes no backtracking necessary, but does not mention symmetry - though all his sample puzzles do indeed have 180-degree rotational symmetry in the pattern of pre-filled cells - whereas I thought the symmetry was one of the things which made a proper sudoku but was unaware of any prohibition on backtracking. (And, yes, my program has no trouble with any of his examples, once the input format is converted, and resorts to tree search on only the one he doesn't say shouldn't need it.) /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Does good luck count? :) There probably is, but I think all such ways are equivalent, in the theory-of-computation sense, to backtracking, at least if you want all solutions rather than being satisfied with just one. Interesting question. I'm not sure how I'd go about computing that number. "Lots", I'm sure of that much, but just how many strikes me as difficult to work out. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

If and when that becomes a concern, I may do a complexity analysis and possibly rewrite the solver. :-) Heh. My generator works by starting with a completely blank grid. Then it runs the solver, slightly tweaked so that it bails as soon as it finds a second solution (rather than continuing to search for more). This can produce any of three results: `no solutions', `exactly one solution', or `multiple solutions'. If it returns `multiple solutions', then it drops in one, two, or four randomly-chosen numbers (how many depends on the symmetry setting, which can be set to produce no, 180-degree, or 90-degree symmetry, and, for odd sizes, on whether it picks the centre cell). If the solver then returns `no solutions', it removes them and tries again; if `multiple solutions', it leaves them there and loops, putting down more numbers. This is kept up until the solver returns `exactly one solution'. That's the basic generator. There are two flags: prune/noprune and fork/nofork. If the forking flag is set to nofork, then if the solver finds it has to do tree search, then `exactly one solution' gets is counted as if it were `multiple solutions'; if it's set to fork, then the code doesn't care whether the solver had to do tree search. If the prune flag is set to noprune, then the result is printed as soon as the above loop gets `exactly one solution'. If the prune flag is set to prune, then a postprocessing pass is performed: each number, pair of numbers, or quartet of numbers is tentatively removed and the solver run. This can return `multiple solutions' or `exactly one solution' (if it returns `no solution', the generator bugchecks, since, in the abstract, that can't happen). If it returns `multiple solutions', the tentatively removed number(s) go(es) back; if `exactly one solution', the removed number(s) stay(s) removed. This is repeated until every number, pair, or quartet has been tried, meaning that every number pair (or quarter or singleton) is necessary in the sense that removing it leads to a puzzle with multiple solutions. During this postprocessing pass, if the forking flag is set to nofork, then, again, one solution that required tree search is counted as if it were multiple solutions. This is computationally expensive. It turns out to be usably fast in practice; my program can crank out order-3 sudoku at some 20 to 25 per second on an 860MHz P-III, which is pretty modest hardware by today's standards. It gets much more expensive very fast as the size rises, though; generating a single order-4 on the same hardware rarely takes less than 5 seconds and can take multiple minutes if your luck is bad. (If it gets into a state where there are multiple solutions but it has to put down just the right number in just the right place to reduce it to only one solution, then it can sit there generating random numbers for a long time until it happens on the right ones. If I cared enough I might try to add code to improve performance in such cases - feeding back partial solution information from the solver to the number placement code would be one way to do this.) /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Sorry. :) It's all in C, or at least the nearly-but-not-quite-C most of my code is written in (it's C as extended by gcc with another extension of my own). It's not for any particular hardware, though it was not written with any particular thought given to systems where int is smaller than 32 bits, so it may not work well there. It is, however, a very unixy program in its I/O model; it is designed for use on a serial terminal or terminal emulator. I guess one out of three will have to do. :/ ftp.rodents-montreal.org:/mouse/games/sudoku/ contains what I have, in case you're interested. You likely want to look at /mouse/local/src/lcs-cvt/ as well, since I see on checking that the sudoku code does indeed use the extension of my own I mentioned above. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Yes, at least for some values of "online". I don't have a slick webpage, which is what "online" seems to mean for a lot of people. However, in this crowd, I don't expect that to be as true. I make most of my code available to the world in one form or another. In the case of the sudoku programs, what I think of as the primary distribution mechanism is a git repo which anyone who cares to is welcome to clone. It's at git://git.rodents-montreal.org/Mouse/sudoku. There's also an anonymous FTP directory, ftp.rodents-montreal.org:/mouse/games/sudoku, which is updated when I think of it (I just now did an update of it). The anoymous FTP area is also exported via HTTP (http://ftp.rodents-montreal.org/mouse/games/sudoku/), so it has a webpage of sorts, though it's hardly slick. As I think I mentioned upthread, you will need a compiler that supports at least a few gcc extensions to C, and lcs-cvt (ftp.rodents-montreal.org:/mouse/local/src/lcs-cvt/) for my own extension beyond that. (The stuff lcs-cvt handles, I have in the compiler, but I don't really expect everyone to patch their compilers the way I have just to build my sudoku programs. :) /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Rod Smallwood wrote: Those are nice, but a lot of us hobbyists are more interested in running real DEC gear than in replacing a DEC CPU with an aftermarket CPU.

That quote about the QED993 from the VC forum was from me. I was disappointed by the lack of compatibility and haven't done much with the one I have. I think Josh also bought one of those.

At 2:28 PM -0400 5/16/12, Ethan Dicks wrote: Now that's interesting, and well worth knowing. I've thought about something like the QED993 in the past, but if it won't run RSTS/E 10.1, it's broken for me as well. Never have gotten 2.11BSD running on any of my PDP-11's. My /73 has SCSI drives in removable trays for booting RT-11, RSX-11M, RSX-11M+ and RSTS/E. IIRC, RSTS/E is the one OS I haven't figured out how to install off of CD-ROM, it can be 'touchy'. As an example, you can install RSTS/E TK-50 first. Zane -- | Zane H. Healy | UNIX Systems Administrator | | healyzh at aracnet.com | OpenVMS Enthusiast | | | Photographer | +----------------------------------+----------------------------+ | My flickr Photostream | | http://www.flickr.com/photos/33848088 at N03/ |

On 5/16/2012 11:28 AM, Ethan Dicks wrote: I think Glen Slick may have written that. I have one as well, purchased before the above was discovered; it seemed too good to be true -- a faster-than-11/93 CPU (w/4mb onboard) for ~$100 on eBay. And it mostly was, alas. The slim documentation Glen managed to find describes it as being "J-11 compatible," I'm not sure how one could claim that given that it doesn't support the FP instruction set at all. Supposedly there's an add-on board that adds that support, but I imagine those are in short supply. Agreed. I haven't yet put the QED to good use (not quite sure what that would be at the moment -- anyone have suggestions for a good -11 OS (+ fun software) to run that doesn't use FP instructions (at all?). I've been thinking of recompiling all of 2.11BSD with FP emulation but that's a lot of work. Even the bootloader appears to use them. - Josh

So am I but not at $2000 for one board. Sooner or later the price will come down. Meanwhile I now have an 11/94 in which everything else is original DEC that runs. If you restored cars you would not let one sit rusting because one original item was not available and a more recent but alternative one would make it go. I have over twenty DEC systems and they all run. I have two (CMD) non DEC SCSI controllers driving DEC RH18-A 2 Gigabyte SCSI drives. Not pure DEC but very representative of what you would have found when the systems were current technology. I would of course prefer totally DEC systems (God knows I sold enough of them) However reality at the time the systems were current was that mixed vendor solutions were the norm. In fact Digital Field Service actively pursued maintenance contracts on third party items. Choose carefully - The Primrose Path or Reality Road Regards ? Rod Smallwood (Digital Equipment Coporation 1975 - 1985) ? ? -----Original Message----- From: cctech-bounces at classiccmp.org [mailto:cctech-bounces at classiccmp.org] On Behalf Of Eric Smith Sent: 16 May 2012 18:37 To: General Discussion: On-Topic and Off-Topic Posts Subject: Re: PDP-11/93 Rod Smallwood wrote: Those are nice, but a lot of us hobbyists are more interested in running real DEC gear than in replacing a DEC CPU with an aftermarket CPU.

On 05/15/2012 03:24 PM, Doc Shipley wrote: Wow, dirtbag alert! Hopefully he's long gone by now. I'll give you TEN KA650s, and a pile of memory, in trade for it. ;) -Dave -- Dave McGuire, AK4HZ New Kensington, PA