Re: out-of-mainstream minis
On Sat, Jul 4, 2015 at 6:42 PM, Diane Bruce wrote: > On Sat, Jul 04, 2015 at 12:33:05PM -0400, Mouse wrote: >> > The problems revolve around the fact that instructions cannot be >> > properly restarted on the 68000. Not enough context is saved. >> > [...] >> > (The tricks done by those who did fix this consists of having a >> > second processor which gets interrupted when you get a page fault, >> > and the second processor do all the work related to the page fault, >> > while the primary processor just stalls until the memory is >> > available, at which point it can continue. There is no limits to how >> > long the CPU can wait for memory to return data on a read.) >> >> I recall hearing of a company that build a machine with two 68000s, one >> running one instruction behind the other. When the leading processor >> got a page fault, hardware interrupted the lagging processor (which had >> not yet encountered the faulting instruction) and there's a dance where >> the two processors switch roles, allowing useful page faults. >> >> Perhaps such a thing existed. Perhaps my informant was misled - it >> sounds like a plausible corruption of what you describe. Perhaps my >> own memory has bitrotted. But it sounds to me as though it certainly >> _could_ work. >> >> /~\ The ASCII Mouse >> \ / Ribbon Campaign >> X Against HTML mo...@rodents-montreal.org >> / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B >> > > Both Apollo and SUN did this. The clocks were two phase so one ran > behind the other. It was a hack. Sun did not do this. The Sun-1 CPU was indeed a 68000 to which was coupled a custom MMU in discreet logic. It didn't handle demand paging. Sun soon switched to the 68010. -Jon
Re: EPROM image for Xylogics 472 (multibus Pertec controller)?
Hello, Sorry to revive this old discussion but I'm having problems with getting a Xylogics 472 board recognized when probing the Multibus. I'm hoping that maybe it's bitrot in the EPROM since that would be an easy fix. I dumped my current one (180-001-957 rev A I think it says, the label is pretty scratched up...) and at least it's not all ones, but without something to compare it with, it's pretty difficult to say if it's damaged... Would anyone out there be kind of enough to provide with me with another image? -Jon On Fri, Dec 19, 2014 at 8:19 PM, Josh Dersch wrote: > Thanks, Al (and everyone else). Looks like I'm all set here. > > - Josh > > On Thu, Dec 18, 2014 at 2:03 PM, Al Kossow wrote: >> >> On 12/18/14 1:23 PM, r.stricklin wrote: >> >>> >>> On Dec 18, 2014, at 12:37 PM, Earl Baugh wrote: >>> >>> I'll try to check tonight when I get home... spreadsheet says I have 8 >>>> Xylogics SMD controllers... have to see if any is an XY472. >>>> >>> >>> The XY472 is a Pertec-interface 1/2" tape controller. Not SMD. >>> >>> ok >>> bear. >>> >>> >>> >> I have that and a XY450 dumped >> >> >> >>
Re: Computers in Election Vigils - take two
On 10/09/2015 09:38 AM, Paul Koning wrote: Very convenient for those who run the government that runs the election process. In the state where I live the setup (by law, as I recall) is a nice hybrid. Paper (mark sense) ballots, scanned by machine. But anyone can look at a ballot and see what it says, and you can recount them by hand if necessary. So the security of the counting machines is not actually critical because they aren't the final authority. Here in Missouri, we have touch screen machines that print a paper roll with human-readable vote info PLUS 2D bar codes for fast scanning, and the voter can watch the paper scroll by to verify the human-readable data agrees with their votes. I think it puts time and date on the paper as well, so somebody can't run off a bunch of phony votes before or after the election hours. The paper is like cash register tapes. the first vote info is all electronic, but if there is a recount, the paper tapes can be examined. Any voter can also opt for mark sense ballots. Jon
Re: Computers in Election Vigils - take two
On 10/10/2015 10:44 AM, Paul Koning wrote: That's not the real problem. The real problem is that you had no way to be sure, no way to verify, that the machine was recording your vote and would accurately report it later. It might just as easily report numbers that someone had told it to report, not connected to any reality. How would you know? If anyone were to question this, how would you prove that the count is honest? An election official in Ohio, I think, not an IT guy at all, just somebody who knew how to open files, etc, played around with a touch screen machine at his precinct. I THINK it was the original Diebold machine, but I could be wrong on that. I think he plugged in a USB cable or something, and found the vote totals were just an open file on a memory card, and could be opened and edited with standard Windows tools like notepad!He went public with it, and it caused a pretty large furor over this blatant lack of security. The manufacturer of the machine had told the state that all files were encrypted, you had to log on with a password, etc. etc. and it was all lies. You should be able to search for articles in the press about this. Jon
Re: Q-bus I/O project
On 10/12/2015 10:30 AM, Paul Koning wrote: On Oct 11, 2015, at 12:53 AM, John Wilson wrote: ... but I'd rather go RoHS. I would recommend against that. Not unless you are trying to create a commercial product where you *must* be RoHS to conform to the requirements of the bureaucrats. Use real solder -- the job will be much easier and the result more reliable. Real solder is still available, including solder paste. I asked one of the technicians at work about this stuff a while ago -- they know all about this question since they have to use RoHS when doing rework on modern designs. And what I just said is what they told me (i.e., don't, unless you are forced to). They also told me that real solder works just fine on lead-free parts. I tried that and can confirm this is correct. ... I was hoping for the Pick 'n' Paste machine to come along and save me from having to do all that by hand but that project seems to have faded away. Oh seriously! I've been hoping one of the homebrew SMT assembly robot projects would "take" too, at a reasonable cost. Making prototypes is bad enough ... but what if I get it working and then I'd want to make dozens of these things, at 4+ hours each? I would think a pick & place robot would be a fairly straightforward derivative of a 3d printer. But is that really needed? Placing the parts isn't all that terrible. The trouble is the soldering. I have read (and posted here in the past) a nice article (in German) on the use of a toaster oven with some clever temperature control as an IR reflow soldering machine. I haven't tried this yet, but it sounds like a good scheme, and would allow the use of BGA parts at least in moderate sizes. At least one of the moderate cost small volume PCB fab shops will deliver solder paste screens along with the finished boards if you ask for it. I have moved over almost entirely to lead-free. Yes, there is more hassle, but I ship boards all over the world, so I do have to be compliant. At first, I had HUGE problems, but it got better with more experience, and as the makers of solder got it all figured out. I use SAC305 solder, which wets better and has a shinier finish than pure tin solders. I have a 1600 Lb. Philips CSM84 pick and place in my basement. I can do one-offs by hand, as long as the parts aren't too small, but for more than a couple boards, I really DO want the P&P machine. But, it is not just a 3D printer. it has 3 heads with different nozzle sizes for different parts, two of them have centering jaws that center the part on the nozzle after it is picked up. it also has a sort-of camera to pick up fiducial marks on the board for alignment, and a centering station that centers large parts like FPGA chips. The big difference is that it has a huge amount of recovery software to detect mis-picks (by vacuum level) and try to readjust the part by cycling the jaws, if that doesn't work it dumps the part and tries another one. The toaster oven reflow soldering is amazing. I use a ramp and soak controller with a thermocouple poked into a through hole in the PC board for temperature reference. Jon
Re: VT52s, VT61s lots of DEC and DG keyboards- return trip through Maine, MA, NY, PA, OH, IN to IL
On 10/12/2015 10:32 PM, Nigel Williams wrote: On Tue, Oct 13, 2015 at 2:16 PM, Paul Anderson wrote: I picked up a few of the VT52s today and they looked pretty good. Paul, if you come across any VT52s that have the built-in screen printer could you take some pictures please. Has anyone ever seen one? I had an idea it used a silvered-paper and burned it off? or am I mis-remembering. I stripped some VT-52s (or possibly VT-55s) a LONG time ago. (They were in a dumpster and not complete at the time, so don't yell sacrilege!) But, I do remember the printer. Yes, I believe it did burn off the silver to make a black mark. Jon
Re: Source for 3M Bumpons
On 10/12/2015 11:13 PM, Ali wrote: Hello all, I am trying to replace some rubber feet on a couple of items and it appears the OEM used 3M Bumpons. I've been trying to find a source that will sell small quantities but all I can find is one box minimum (which is a couple of thousand!). I am willing to buy as much as a whole sheet but prefer less if possible. Looking for models: Digi-Key sells rubber feet in small quantities. Just type "rubber feet" into their search engine, and you get 216 items. Most are sheets with multiple feet that can be peeled off a backing. You can choose styles (round, square) height and width. Jon
Re: PDP-12 Restoration at the RICM (tony duell)
On 10/13/2015 07:32 PM, Michael Thompson wrote: Tracks 1 & 2 look fine. We swapped the probes for track 2 & 3 just to make sure that it wasn't a logic analyzer problem. A 'scope connected to the differential signals shows the same track 3 glitches. The glitches were present with the TU55 and the TU56 tape drives. We used different control and data cables for the TU55 and TU56. We swapped the G882 modules between tracks, and from the TC01 in the PDP-8/I and did not see a change. We swapped the W603 module in slot F16, the R107 in F12, and the W512 in F13. The LTR WRITE ON L signal is inactive and stable. We tried several tapes, including ones formatted and written by DEC and the glitches are present on track 3. OK, so the drive must have a head amplifier and a comparator (sometimes called a slicer in tape technology) to convert the amplified analog signal to a digital one. Have you swapped the read amps and comparator cards? (Sounds like yes, above.) Have you looked at the analog signals? it is possible a defect in the tape head is causing the read signal to be different than what should be seen. Looking at the amplified analog signal might give some hint at the behavior. (Don't bother trying to look at the raw tape head signal, it will be really tiny.) I'm not sure where the read amps and slicer are, whether they are in the control or the drive. (Seems like the read amp HAS to be in the drive, though, for noise immunity reasons.) Jon
Re: VT52s, VT61s lots of DEC and DG keyboards- return trip through Maine, MA, NY, PA, OH, IN to IL
On Oct 13, 2015, at 1:22 PM, tony duell wrote: The Versatec electrostatic plotters are not the same as the VT52 printer, they are yet another process. WIth those the paper passes between a set of electrodes that build up a charge image on the paper. I beleive the paper is specially treated to make it more resistive so the charge doesn't leak away too quickly, and there is Yes, I had a bunch of Versatec 1200A's with the Tektronix hard copy feature. the Versatec was the greatest graphics printer until laser printers came out, then they became instant boat anchors. Here's the process. There is a double-sided PC board that touches the face of the paper end-on, so the traces just come to the end of the board and make contact with the paper. On the 1200A, that was a 200 DPI printer, so each side of the PC board had 100 traces/inch, and they were interleaved, so you got to paint 200 raster lines/inch along the axis of the paper. The back side of the paper had wide electrodes that defined zones. One of these backplate electrodes was charged at a time to the opposite polarity of the front electrodes. I seem to remember there were +800, -200 and -800 V power supplies. The raster line was written about one inch at a time across the page, then the next backplate was charged and the next inch was written, etc. Once the whole line was written electrostatically on the paper, a stepper motor advanced the paper and the next line was written. About an inch from the writing electrodes, there was a toner applicator that produced a fountain of this hydrocarbon-smelling solventy stuff with the carbon toner suspended in it. The charge on the paper would attract the toner particles, and when the solvent dried (assisted by a blower) it pretty well stuck to the paper. The paper had this awful chalky feel on the print side, the toner smelled like printer's ink, and when it was working really well, the paper came out gray with fairly decent print. But, it was FAST!!! It could print at about 1000 LPM in print mode, and if your computer could feed it, it could plot images (black and white only) at better than a page every 10 seconds or so. So, it could actually run faster than most of today's laser printers - although the print quality, of course, was WAY worse. And, with the Tek hardcopy board, it could hardcopy a Tek storage tube terminal in less than 30 seconds. I still have some Versatec printed output, as I ran one here at my house for a couple years, from my MicroVAX. We did have a TEK hard copy unit before the Versatecs. That was a pretty awful unit. it had a line-scan CRT with a fiber optic faceplate that exposed the image onto thermal-developing silver paper-film that rolled past the CRT. It also made bad smells, and the paper came out brown with dark brown images on them. In normal fluorescent lighting, the hard copies started turning totally brown after just a day or two. Also, the silver paper was QUITE expensive, maybe close to a Dollar a page or something, even back in the 1970's. Jon
Re: VT52s, VT61s lots of DEC and DG keyboards- return trip through Maine, MA, NY, PA, OH, IN to IL
On 10/14/2015 08:55 AM, tony duell wrote: Was it a stepper motor? I am sure mine uses a permanent magnet DC motor. I do remember that the paper feed roller is in 2 parts with a differential gear between them. Yes, absolutely, on the Versatec 1200A. I put those motors in a milling machine. Big, round case stepper motors, with a ghastly resistor-transistor drive. writing electrodes, there was a toner applicator that produced a fountain of this hydrocarbon-smelling solventy stuff with the carbon toner suspended in it. The charge on It's called the 'toner fountain' in the manuals, but it actually works below atmospheric pressure. The results are that (a) the paper is sucked down onto the toner fountain and (b) if the paper is torn or runs out you don't get toner sprayed all over the machine room. The 1200A did not have any mechanism for negative pressure that I know of. The fountain was in the middle, then there was a larger, rectangular region around it that returned the fluid to the container. You could activate the pump while the lid was open, with a button on the machine. The toner would not overflow even with the paper away from the fountain. Now, there was ONE way to make it spill. If you opened or closed the lid while the drain space around the fountain was still filled, the drying blower would spray some of the toner. So, when the paper tore or some other mishap occurred, you had to wait 10 seconds or so before opening the cover. The toner is circulated by a little electromangnetic pump. The toner system tends to block, I found that what we call 'white spirit' was a suitable solvent to unblock it. One time I tried the old 'suck it and see' method to get the pump valves working and found that the toner tastes horrible! How could you do that??? Just the smell of the stuff should have been adequate warning. We did have a TEK hard copy unit before the Versatecs. That was a pretty awful unit. it had a line-scan CRT with a fiber optic faceplate that exposed the image onto thermal-developing silver paper-film that rolled past the CRT. It also made bad smells, and the paper came out brown with dark brown images on them. In normal fluorescent lighting, the hard copies started turning totally brown after just a day or two. Also, the silver paper was QUITE expensive, maybe close to a Dollar a page or something, even back in the 1970's. I don't have a Tektronix hard copy unit (one of the few oddball printers I've not managed to obtain) but I am told that the paper goes off with time, and that it unlikely there's any useable paper left for them ;-( The fact that the paper turned brown within a day or so just sitting on the desk makes me think there would certainly be no paper remaining usable for this printer. The same type of system was used for some years after for making medical quick copies on ultrasound and similar machines. Usually, these were on about 5" wide paper, though. Jon
Re: VT52s, VT61s lots of DEC and DG keyboards- return trip through Maine, MA, NY, PA, OH, IN to IL
On 10/14/2015 09:00 AM, Paul Koning wrote: On Oct 13, 2015, at 11:27 PM, Jon Elson wrote: ... Yes, I had a bunch of Versatec 1200A's with the Tektronix hard copy feature. the Versatec was the greatest graphics printer until laser printers came out, then they became instant boat anchors. Here's the process. ... And, with the Tek hardcopy board, it could hardcopy a Tek storage tube terminal in less than 30 seconds. Ok, but for screen hardcopy, the question is how the image on the screen is read out so it can be fed to the printer. For the Tek storage tube case, that presumably is done by scanning the screen with the beam, which will produce the stored image in the same way as a TV camera tube works. But how would you do that in a plasma panel? Actually, the plasma panels used a bistable effect, that was how the screen memory worked. Each pixel would stay lit once the discharge was struck. The drive electronics could scan a pixel and tell whether it was lit or not by the voltage across the pixel. Jon
Re: New logo: Vintage Computer Federation
On 10/14/2015 09:15 AM, Rod Smallwood wrote: OK so if we agree there are three classes computer Namely Micro,Mini, and Mainframe. It follows that there must be three classes of vintage computer. We dont need patches with pictures but it should say what type of system we major in For example Rod's Retro Restorations - IBM360 Are you actually restoring a 360? I'd sure like to hear more if you are working on this! Jon
Re: VT52s, VT61s lots of DEC and DG keyboards- return trip through Maine, MA, NY, PA, OH, IN to IL
On 10/14/2015 11:46 AM, tony duell wrote: [Versatec paper feed motors] Yes, absolutely, on the Versatec 1200A. I put those motors in a milling machine. Big, round case stepper motors, with a ghastly resistor-transistor drive. You've got me worried now... I have V80. Actually, it's ICL-badged, and has a GPIB adapter board at the back connected to the normal parallel port [1]. I was prepared to swear the motor in that is a normal DC motor, but perhaps I have misremembered it. I really don't want to strip it down tonight, and finding the manuals [2] Is an even bigger job. Versatec made a lot of stuff, and went through a whole bunch of iterations. But, a stepper seems MOST logical to advance the paper every time the line has been written. A stepper is great to start and stop on a dime, and otherwise stay perfectly in synch with intermittent data from the computer. The motor drive is inside the hinged top of the printer, but no telling if the V80 was built the same way. Yes, that's like the V80, but I am pretty sure on that machine the 'fountain' is on the suction side of the pump, that is it sucks from the outer region. I can absolutely say the 1200A did not work that way, as you could manually start the fountain pump with the cover open. Now that I think about it more, I don't think you were supposed to be able to do that, but you could push a relay manually to do it. Without the paper there, it could overflow if you let the pump run too long. When there is paper present this causes toner to flow from the bottle up through the middle bit, then back round the outside to the pump and back to the bottle. Again I might be mis-remembering things... That makes some sense, as it makes it pretty impossible to cause a spill. But, it probably takes a lot longer to get the fountain started that way. How could you do that??? Just the smell of the stuff should have been adequate warning. Got any better ways to unclog the pump? We never, ever had the pump clog. Sometimes we needed to clean gunk out of the fountain area, but the pumps were never a problem. The pump ran filled with the toner, and was below toner level in the bottle. Sometimes if you hadn't printed in a while, the first page would come out partially blank, but then it would get going and work fine. Not sure if the V80 worked the same way, it sounds like maybe it was allowed to drip down and go dry when not printing. That might have been the cause of the clogging. Jon
Re: New logo: Vintage Computer Federation
On 10/14/2015 12:34 PM, Rod Smallwood wrote: Sorry old chap just an example. I'm a old DEC guy. My biggest system is a VAX I think the 360 was back in the days when they rented every thing so not much was left behind Mind you I would not turn down a racks worth of AS400 In the EARLY days of the 360, that was true, but in later days many people owned their machines. Washington University rented their 360/50, but then bought a used 360/65, and then added two used 370/145s. Despite the number of 360s made, there are REALLY few left, and I'm not sure anybody has any that run. (Of course, with the SLT modules, spares would be a REAL problem.) Yup, for sentimental reasons, I still have my MicroVAX-II here. Jon
Re: Vintage Computer IBM1130
On 10/14/2015 01:07 PM, ben wrote: On 10/14/2015 9:05 AM, Jon Elson wrote: On 10/14/2015 09:15 AM, Rod Smallwood wrote: OK so if we agree there are three classes computer Namely Micro,Mini, and Mainframe. It follows that there must be three classes of vintage computer. We dont need patches with pictures but it should say what type of system we major in For example Rod's Retro Restorations - IBM360 Are you actually restoring a 360? I'd sure like to hear more if you are working on this! Carl has a IBM 1130 http://rescue1130.blogspot.ca/ Fascinating! And, of course, with discrete transistors, it should not be that hard to keep the electronics running. The mechanicals look like a pretty major repair project, though! Jon
Re: Vintage Computer IBM1130
On 10/14/2015 10:32 PM, ben wrote: On 10/14/2015 9:21 PM, Jon Elson wrote: Carl has a IBM 1130 http://rescue1130.blogspot.ca/ Fascinating! And, of course, with discrete transistors, it should not be that hard to keep the electronics running. The mechanicals look like a pretty major repair project, though! NO TRANSISTORS ... IBM's special logic for 1965. Yeah, I guess I'm getting it confused with the 1620. The 1130 and 1800 are supposed to be mostly the same, and the 1800 is definitely SLT. Well, that may be a real challenge, then. SLT is simple logic, but the art of chip passivation was not totally figured out back then, and so the chips degraded from moisture and oxygen exposure. Do you have any idea what shape the CPU is in? Have you powered it on? Jon
Re: nut bars? ;)
On 10/15/2015 03:36 PM, Jay West wrote: HAH! The DG documentation does call them "Nut Bars" :P "threaded rail" seems to be something different according to google, but it yields "threaded insert" which is close if not exact. Still googling... There were some systems, like Vector configurable card cages and VME systems that had extruded rails with nuts that could slide to where you wanted them. These had real nuts (either hex or square) slipped into the rails. Boards or connectors could be screwed into the nuts. Threaded rail sounds to me like the threads are permanently drilled and tapped to specific positions in the rail. Jon
Re: Oddball floppies for trade - 8", HS (outer edge), weird cutout
On 10/18/2015 08:00 PM, Chuck Guzis wrote: IBM occupied a large segment of the market and was a favorite target for lots of followers. If you just observe the passage of Memorex through the history of acquisitions and being acquired and sold, it's truly amazing that Memorex still exists--as a brand of Imation. As 3rd party vendors in the IBM plug-compatible arena went, Memorex was pretty large. We had an IBM shop at Washington University, but after the CPUs, practically the whole room went to Memorex-supported gear. They get 3330-compatible drives, later upgraded to 3350-compatible, and maybe later upgraded again. They used Memorex 3674 controllers (they had 2 of them). They had at least one Memorex 1270 com controller. They had two 370/145s with AMS/Intersil 1 Meg memory units, installed and supported by Memorex. I can't recall specifics, but I think there were other memorex units there, too. Jon
Re: Has anyone hear of the Computer History Archives Project?
On 10/20/2015 05:39 PM, Chris Elmquist wrote: I watched the IBM 360 one (17 minutes) it had some GREAT pictures of making SLT modules and core planes. They had a machine in the pictures that I didn't recognize. It had a black front panel and a long row of address dials, no switches. It might have been a stripped-down version of what eventually became the 360/22, /25, or maybe /30, but it didn't seem to really just be a version of one of those just in black. The same machine is shown a number of times at the end of the video. Anybody know? Jon
Re: 360 mockups Re: Has anyone hear of the Computer History Archives Project?
On 10/21/2015 12:42 AM, ste...@malikoff.com wrote: Great film. I'm sure there are a few mockups in there. For instance, towards the end at 17:23 there is a mockup of a console with a slim white/black masthead. I have a high-res copy of that photo, and here is a the lower right corner detail enlarged: http://www.surfacezero.com/g503/data/500/IBM_System_360_mockup_enlarged.jpg You can see a chip out of the console panel that looks like particle board, and there are no cutouts for the button 'inserts' - they appear glued on. Also, the numbered dials on the address switches look painted on, the 5 lights (system, manual, wait, test, load) between the interrupt and load keys appear to just be painted circles, so that one is clearly a hollywood-style mockup. (By the way - I'm looking for a red INTERRUPT insert, red STOP insert and pearl POWER ON insert for my 360/40 console, if anyone can help :) Steve. I believe these were just commercial Microswitch lighted pushbuttons. You may still be able to get colored buttons for them and engrave or label the legend. Jon
Re: 360 mockups Re: Has anyone hear of the Computer History Archives Project?
On 10/22/2015 07:13 AM, ste...@malikoff.com wrote: Steve wrote: (By the way - I'm looking for a red INTERRUPT insert, red STOP insert and pearl POWER ON insert for my 360/40 console, if anyone can help :) Is this for a mockup or a real 360/40 console? Regards, Peter Coghlan Oh, it's quite real :) I have some ideas on machining the inserts from blocks of ABS plastic, incuding the curved depression on the front. I have some red ABS that is the right colour but not thick enough so I'll pay another visit to the plastics shop to get something more suitable. Just another of my gunnadoos... I have a box of Honeywell pushbutton inserts from various machines scrapped over the years. I have a couple translucent red ones that are real deep, and look a lot like they would fit the switches on a 360. They are .980" wide, .749 high and .693" deep. (24.9 mm W, 19 mm H and 17.6 mm deep) They have a groove on the rear top and bottom that the clip of the switch grabs them by. They are not ABS, as they are translucent. I also have 2 each, green and white versions of these same items. I don't know if these will fit your console switches, but if they will, you are welcome to them. I also have a bunch of the two-piece buttons that are more typical of the Honeywell switches, they have a colored insert and a clear snap-on cover. You can put a piece of film between these to create a legend. Jon
Re: Previous message: KIM Uno /PiDP-11 plans...
On 10/23/2015 09:57 AM, Noel Chiappa wrote: I tried looking through the catalog at that site, to see if I could find the SPST version, but no luck. (Although maybe I'm just a klutz about that site - it wasn't easy for me to use.) Maybe your parts-search-fu is better than mine? You might try digging on Digi-Key, they have a very good search engine, and you may be able to select qualifiers that lead you right to the part you are looking for. Jon
Re: Retro Reproduction.
On 10/23/2015 01:57 PM, Paul Koning wrote: On Oct 23, 2015, at 2:48 PM, ben wrote: ... It does seem mechanical stuff is dieing art. I don't know about that. Yes, there is an open-source project that provides a CNC machine control for mill, lathes, hexapods and robots. That is LinuxCNC. It started as a NIST project to see what was so complicated in a CNC control, but then moved out of NIST and has been extended greatly. it is also now being used for 3D printers. I sell some hardware for motion control based on this software, and have sold over 300 controller-interfaces. Some other outfits likely have sold a lot more than that. Most of this is going into hobby shops and small businesses that maybe grew out of a hobby shop. Also, I am a member of an Atlas lathe group and the Sheldon lathe group, they both have a fairly large membership. The Atlas lathe group is closing in on 6000 members! Jon
Re: Retro Reproduction.
On 10/23/2015 04:32 PM, geneb wrote: On Fri, 23 Oct 2015, Jon Elson wrote: I sell some hardware for motion control based on this software, and have sold over 300 controller-interfaces. A popular one seems to be using MachineKit on a Beagle Bone Black Yes, my latest product is the CRAMPS board for machinekit and BBB. Jon
Re: Retro Reproduction.
On 10/24/2015 08:01 PM, geneb wrote: I wouldn't mind finding step-by-step instructions on how to refit a 2000 model year MaxNC 10 open loop mill. The controller it has now is parallel port based and the control software is DOS only. Well, this is a stepper-driven machine? Are the stepper drives in good condition, and are they good units? Stepper drives have come a long way since 2000. Gecko stepper drives are pretty amazing. You can run stepper drivers from the parallel port, but the performance may be less that optimum. I make a board that generates steps in hardware that gives more accurate step timing at higher speeds, and also allows closed-loop operation if you fit encoders to the machine. This is all supported by the LinuxCNC program, which is pretty fantastic. I use it here in my shop to run a Bridgeport mill with my own servo drives. I also have a desktop mill that I built mostly for taking to shows that uses my controllers and drives. You can check out the LinuxCNC.org web site, and see what others are doing in the retrofit area. Jon
Re: Oddball question: really small terminals
On 10/24/2015 08:54 PM, Mike Ross wrote: For reasons too abstruse to explain in detail I'm on the lookout for terminals that are, physically, really small - especially serial and coax 3270, and possibly twinax 5250. I had a thing that was apparently used in France and maybe Canada for telephone directory lookup and similar uses. I believe it was called a Minitel. It had a TINY CRT screen, maybe 5 x 7". It used a standard 40-pin microprocessor and other standard parts. I junked mine because it would randomly reset every minute or so, and I didn't really have a use for it. I think mine was smaller than the one in the Wikipedia article, but it did look similar to that. There are some on eBay, but they are all from France and Belgium. Interface was pure serial RS-232. Jon
Re: Oddball question: really small terminals
On 10/25/2015 05:27 PM, Mike Ross wrote: A pull-out keyboard in the base? I have one too. Would be perfect if it worked, and I had half a dozen of them! No, mine did NOT have a pull-out or fold-down kbd, it was all one piece, like a micro-miniaturized ADM3 or similar terminal. Possibly it WAS the same as the picture on the Wikipedia article, I couldn't tell the scale from that pic. Jon
Re: Oddball question: really small terminals
On 10/25/2015 08:32 PM, william degnan wrote: On Oct 25, 2015 8:54 PM, "Jon Elson" wrote: On 10/25/2015 05:27 PM, Mike Ross wrote: A pull-out keyboard in the base? I have one too. Would be perfect if it worked, and I had half a dozen of them! No, mine did NOT have a pull-out or fold-down kbd, it was all one piece, like a micro-miniaturized ADM3 or similar terminal. Possibly it WAS the same as the picture on the Wikipedia article, I couldn't tell the scale from that pic. Jon Something like the iXO handheld terminal? http://vintagecomputer.net/iXO/iXO_telecomputing-system.jpg No, it WAS a Minitel. The Wikipedia picture gives nothing for scale, but there is a picture of one on a Google pictures collection that shows it on a desk with a regular keyboard and LCD screen, and you can see how small the thing really was. Jon
Re: PDP 8 panels. Feedback
On 10/26/2015 11:38 PM, Chuck Guzis wrote: On 10/26/2015 08:54 PM, wulfman wrote: To effectively drill in plastics you need to run the drill press on the highest speed you can and use a freshly sharpened drill bit. If this is Perspex/Plexiglas, I've had great results with a good sharp Forstner bit in my drill press at medium (say 750 RPM) speed and a not-too aggressive feed.. No melting, just lots of crumbly shavings. I've done this with sizes down to about 1/4", but no smaller. When you're almost through the material, turn it over and complete the hole from the back side. Very clean edges, with no chips at all The ultimate way to drill holes in Plexi is with an end mill. It can make a slight chipping when it punches through the back, so you either need a backstop material or lighten up the feed a bit just before it goes through. (This of course requires a center-cutting end mill, I prefer 4-flute for this.) Even better than just plunging an end mill is to use a CNC mill and a cutter smaller than the required hole diameter. You use a pocket boring routine, set up for "climb milling" and after working through the material in several depth steps just less then the desired diameter, does a finish pass at full depth to bring it to final hole size. This takes longer to describe than it actually takes the machine to perform, and leaves beautiful holes. Jon
Re: PDP 8 panels. Feedback
On 10/27/2015 11:59 AM, Chuck Guzis wrote: On 10/27/2015 09:33 AM, Jon Elson wrote: The ultimate way to drill holes in Plexi is with an end mill. It can make a slight chipping when it punches through the back, so you either need a backstop material or lighten up the feed a bit just before it goes through. (This of course requires a center-cutting end mill, I prefer 4-flute for this.) Sure, but I'm assuming that the original poster doesn't have a Bridgeport mill at his disposal. But a drill press is quite within the budget of most hobbyists. I'd think that if you already owned a vertical mill, you'd know how to work with plastics. You CAN use an end mill in a drill press if you are careful. You get no self-centering, so the work must be **SECURELY** clamped, as the cutter may try to walk. But, you'll get perfectly round, smooth holes. Twist drills tend to wander and make triangular and slightly helical holes. Obviously, you can only do this with small size end mills in a drill press. Jon
Re: The last fix for a "All Shook Up" 33
On 10/27/2015 05:23 PM, Eric Christopherson wrote: I'm curious: Does the static from the peanuts noticeably affect electronics? Have you ever had them cling to your arms or whatever? Yes, definitely. That's why they have green and pink peanuts, those are supposed to be anti-static. Jon
SCO Integra
Hi All, I ran across the following while cleaning: A set of "draft" manuals for SCO Integra database software. It looks to be a complete manual set in 4 hardbound volumes. Also in the pile are 4 sequentially labeled 5 1/4" floppy disks that have "Dewitt" written on them. I don't have the resources to check the contents of the disks. There is a single page color promotional flyer for SCO Integra. Are these of interest to anyone? Here's a picture: https://www.flickr.com/photos/jja572/albums/72157660431177871 Available for beer money and shipping from 53714. -Jon
HP Vectra stuff
All, Also while cleaning I found a box of HP Vectra docs & a couple disks. Docs: Getting Started With the HP BASIC Controller HP 24540B & HP 24541B - Installation Guide 35743 HP Enhanced Graphics Display Installation Guide HP Enhanced Graphics Adapter User's Manual Disks: VECTRA PC UTILITIES AND DRIVERS ENHANCED GRAPHICS ADAPTER - UTILITY DISC All of this in a HP 82301A BASIC Language box. Again, beer money & shipping from 53714. -Jon
Re: Up for Auction: Memory from the First Computer in Space
On 10/28/2015 06:32 PM, Brent Hilpert wrote: Not a proof in entirety of the claim, but from a ref and looking at the closeup pics from the auction website, it is an unusual form of core memory where the cores have two holes through them, like a blocky figure 8, apparently an aspect of a technique to achieve non-destructive readout. This is quite unusual and would go some ways to showing a provenance to the Gemini project. That scheme is also known as BiAx. The docs with my Honeywell Alert machine described the same thing. As far as I can tell, the EMM core memory unit that came with it is NOT BiAx, however. Jon
Re: Up for Auction: Memory from the First Computer in Space
On 10/28/2015 11:48 PM, Brent Hilpert wrote: Very interesting to hear of another scheme, but it's not clear whether it applies to the Gemini auction memory. The BiAX scheme shows cores with the holes (apertures as they're called in the business) perpendicular to each other. In contrast, the Gemini auction cores have two apertures with the same orientation (a figure 8). Yes, but I'm pretty sure the concepts are related. The remanent flux in the non-volatile side of the core affects the flux hysteresis in the volatile side, so when you flip the flux polarity on the volatile side, you can see some effect caused by the non-volatile side. - Another topic: what is a Honeywell Alert machine? Searches are just bringing up some current-day "Honeywell Instant Alert" messaging system. It was a 24-bit DTL machine designed for the X-15 project, and then used in a number of other projects. It is about 25 Lbs (without memory), draws 25 A at 5 V, and runs off a 3 MHz clock. It is constructed of 6 multilayer PC boards with ceramic flat packs on both sides. The boards are attached to a motherboard with flexible PCB, so there are no connectors inside the machine. They chose this for reliability, it makes it REALLY hard to work on. There are cold plates between the boards that conduct heat down to the baseplate. The instruction set was apparently modeled after some Honeywell business machine of the time. It has no floating point hardware, but DOES have hardware multiply and divide. I have gotten it to run, sort of. Since I have no memory for it, I can jam op codes into the data in plug and observe the speed at which the memory address ripples. Jon
Re: Up for Auction: Memory from the First Computer in Space
On 10/28/2015 11:48 PM, Brent Hilpert wrote: Another topic: what is a Honeywell Alert machine? Searches are just bringing up some current-day "Honeywell Instant Alert" messaging system. Oh, one other thing, the Alert is supposed to be the second computer ever manufactured with integrated circuits. The first was the Apollo Guidance Computer. Jon
Re: Emulex QD32
On 10/30/2015 09:42 AM, Christian Corti wrote: Hi all, I'm trying to set up a MicroVAX II that has an SMD disk attached to an Emulex QD32 controller. I need to test and/or format the disk and so I'm looking for images of the Emulex diagnostic floppies (should be RX50 AFAIK). Those found at http://www.headcrashers.org/comp/rx50/ boot, but I have not the faintest idea of how to start anything in there; even a "DIR" How about HELP or even "?" I seem to recall they have a help menu in the program. Also, you log shows MicroVAX I, so it is possible the program won't support a Q-bus board on a MicroVAX-II. Jon
Re: Emulex QD32
On 10/30/2015 12:04 PM, Christian Corti wrote: On Fri, 30 Oct 2015, Jon Elson wrote: Also, you log shows MicroVAX I, so it is possible the program won't support a Q-bus board on a MicroVAX-II. Oh, that was a typo. It actually says this: B DUA0 2..1..0.. Emulex VAX Monitor V1.2 MicroVAX II 16-SEP-1985 09:00:00 OK, that makes much more sense with the 1985 date. Can't imagine anybody using a MicroVAX I for more than a couple months after the UV-II came out. Jon
Re: Lurker no more...
On 10/30/2015 12:27 PM, couryho...@aol.com wrote: Welcome Corey! how did you get the upside down font in the email sig? Ed# In a message dated 10/30/2015 9:59:06 A.M. US Mountain Standard Time, appleco...@optonline.net writes: Well I finally setup a separate email address so I can receive individual messages from cctalk instead of the daily digest which really doesn't lend itself to posting back. Just wanted to say hi. Cheers, Corey corey cohen uǝɥoɔ ʎǝɹoɔ That last line prints out in show message source like this : u=C7=9D=C9=A5o=C9=94 =CA=8E=C7=9D=C9=B9o=C9=94 Jon
Re: Gov. & the machine(s) we love
On 2015-Oct-30, at 7:25 PM, william degnan wrote: So chances are that various government programs helped at the margins, but were not drivers. (Think I got that attribution right...) Well, one BIG thing that moved LSI forward was the VHSIC project and others at DARPA. They funded a lot of work on computer programs to synthesize digital logic (VHDL), analyze chip performance (SPICE) and also MOSIS, which allowed people to experiment with designs and share the cost of masks and wafer fab. SPICE has been there from the very beginning, VHDL and MOSIS came later, after digital ICs were already in production, but certainly were a huge help as chips became more complicated. There are probably other projects in this vein. Jon
Re: Honeywell Alert / Re: Up for Auction: Memory from the First Computer in Space
On 10/31/2015 01:40 AM, Brent Hilpert wrote: Very neat, too bad the memory is gone. Is building a new memory with modern stuff a consideration (enough documentation .. )? it is not gone. I have a VERY strong belief that the EMM memory box was made for the Honeywell Alert, but I have no docs on it, and one of the 4 memory blocks was disassembled, probably to attempt a repair on the electronics. Without pinout or even knowing what voltages were used, pretty hard to do much with it. Yes, I could build an SRAM board that would hook to it, and a console interface, but I really don't see the point. Jon
Re: Front panels
On 10/31/2015 03:11 PM, Charles Anthony wrote: What I don't know how to do is drive 500 or so LEDs. I am guessing a bunch of shift registers, but I've pretty much reached my design limits. I need some guidance on locating and understanding the technology to run that many LEDs. it really isn't that complicated. The simplest might be a byte shift register, ie. a bunch of octal D-FFs like the 74HC374. Given a byte-wide group of GPIOs on the Beagle Bone, you could send out 63 8-bit words with one additional GPIO to act as a clock for the FFs. The LEDs could be driven directly from the FF outputs with a resistor. With the current generation of high-efficiency LEDs 10 mA would be plenty of current, and so the FF outputs would still be close enough the specs to drive the next stage. One downside of this scheme is if the serial transmission was slow, you'd see a blink each time the Bone sent a new light pattern. If you want to get more complicated, you could have one HC374 for the shift register and one HC374 as the latch. You'd shift all 63 bytes through the byte-shift register, pulsing the byte clock 63 times, and then pulse the latch clock once to latch all the 5xx bits of light info into the latch register, which would allow the LEDs to be updated without any flash as the shift reg is being shifted. Now, another way to do this is with multiplexing. You could maybe have 8 64-bit words that loaded to a small RAM, and the RAM is scanned to load data to banks of 64 LEDs. This reduces the number of drivers to, say, 64 cathode drivers and 8 high-current anode drivers, but complicates the rest of the thing a fair bit. It will also cause the whole panel of LEDs to flicker at the multiplexing rate, which could be annoying when you flick your eyes across the panel. Jon
Re: RMS [was Re: Terminology [was RE: IBM z890]]
On 11/04/2015 10:42 AM, Fred Cisin wrote: On Tue, 3 Nov 2015, Chuck Guzis wrote: Cool. How about "label" versus VTOC versus "folder" versus "directory" versus "FNT"...versus...? So many names... Was it considered "copyright infringement" to use the same terminology as somebody else? Like "standards" everybody had a unique one of their own. But, IBM had a WHOLE dictionary full of IBM-specific terms for everything. memory was storage (or main storage) disks were direct access storage devices (DASD) tapes were sequential access the register set of the 360 was local store (or local storage) microcode was in control store and, there is a LOT more of this that escapes me at the moment. Jon
Re: DECtape reliability?
On 11/12/2015 07:24 AM, Tom Moss wrote: I've had very mixed (about 50/50) success with 9-track, but after reading a bit about DECtape it looks like they should still be holding up nicely. Anyone care to share their experiences? Some guys at Washington University restored a LINC that had been stored in somebody's garage for decades. They mostly just had to reform the power supply capacitors. Then, they got some LAP-6 tapes that had been in somebody's office and it booted right up! These were tapes that likely had been sitting for 30 years. LINCtape is essentially identical to DECtape. I think the directory format is different, but I think the block format is the same. I think the tapes are wound in reverse order and the order of increasing block numbers is reversed. Jon
Re: [cctalk] was: Could someone make the list do the thing in the subject line?
On 11/18/2015 04:05 AM, simon wrote: Hello all, unfortunately a lot of people *hate* things already. As I am subscribed to a lot of lists (about 10), this is the only one not using a [header] in the subject lines. I use Thunderbird, and the filter is set for : To: contains cctalk To: contains cctech cc: contains cctalk This seems to catch all variants of the messages and put them in their folder. Jon
Re: Altera Quartus - was Re: Project Oberon and OberonStation (resend)
On 11/23/2015 09:59 PM, Toby Thain wrote: Although not always. It's gratifying to see Altera Quartus runs on Linux. Xilinx Ise also has run on Linux for at least 5 years now. The webpack version is free, and supports up to fairly large devices. Jon
Re: Replacing MicroVAX II PSU With a Modern PSU
On 11/24/2015 05:52 PM, Mark Wickens wrote: I haven't looked into this at all and I suspect it's probably quite tricky indeed. I did look around a while back drew a blank. I built up a Micro-VAX II system out of boards, backplanes and assorted junk. I made my own power supply. The power supply is actually easy, just +5 and +12, with a touch of -12 for serial I/O. The only tricky thing is the power-OK and reset logic, which really wasn't all that tricky, either. I had a power and thermal safety control panel on it repurposed from the original 3rd party IBM mainframe memory box that would shut it down if there was an abnormal voltage or cooling failure. Jon
Re: Replacing MicroVAX II PSU With a Modern PSU
On 11/24/2015 05:52 PM, Mark Wickens wrote: 4. The most difficult bit, I suspect, would be the PSU LTC signal , which I believe is some kind of clock. I don't know what the spec of the signal is, but I will get a scope on a working one to see (NB don't want to risk a working PSU on this machine in case it was a problem with the machine itself that caused the first PSU to fail, I don't mind sacrificing a modern PSU if need be). The line time clock is a vestige of the Q-bus LSI-11 heritage. As far as I know, there is a reserved bus signal, but it was NOT implemented on the UVAX-II system. (I don't have my print set anymore.) Jon
Re: Oberon and the OberonStation (retro-style FPGA computing)
On 11/24/2015 07:34 PM, Paul Koning wrote: Judging a language without even looking at it seems rather odd. Pascal is largely obsolete now, but I still appreciate it -- of the 40 or so languages I know, there are only two where I could go from no knowledge at all to having a working program of significant size in one week: Pascal and Python. I mostly LOVED Pascal, and years ago it was my language of choice for most things. I wrote some significant programs in it, first Borland's Turbo Pascal, later DEC Pascal on the Vax. I recently revived my biggest Pascal program, which I had run in Windows 2K and XP, but had not maintained in over a decade. the Linux Free Pascal Compiler came out, which was aimed specifically at compiling Turbo and DEC pascal variants. This program converted Gerber photoplot files to a raster format for my laser photoplotter. I was able to get the original program working with minimal effort directly under Linux. It runs a lot faster, too. My only complaints with Pascal were the I/O was a bit clunky and slow, and the stupid / vs. div for real vs. integer divides. Otherwise, I really thought it was great. Jon
Re: Replacing MicroVAX II PSU With a Modern PSU
On 11/25/2015 01:34 AM, Robert Jarratt wrote: -Original Message- From: cctalk [mailto:cctalk-boun...@classiccmp.org] On Behalf Of Jon Elson Sent: 25 November 2015 02:05 To: gene...@classiccmp.org; discuss...@classiccmp.org:On-Topic and Off- Topic Posts Subject: Re: Replacing MicroVAX II PSU With a Modern PSU On 11/24/2015 05:52 PM, Mark Wickens wrote: I haven't looked into this at all and I suspect it's probably quite tricky indeed. I did look around a while back drew a blank. I built up a Micro-VAX II system out of boards, backplanes and assorted junk. I made my own power supply. The power supply is actually easy, just +5 and +12, with a touch of -12 for serial I/O. The only tricky thing is the power-OK and reset logic, which really wasn't all that tricky, either. I had a power and thermal safety control panel on it repurposed from the original 3rd party IBM mainframe memory box that would shut it down if there was an abnormal voltage or cooling failure. Jon Thanks Jon. I think standard PC PSUs only do +5 and +12, so how did you do -12? I did not use PC power supplies, I used a couple industrial power supplies. I built this system in 1986. But, generic PC power supplies DO have a little bit of -12 V for serial cards. The standard PCI connector has a -12 V pin. PCIe seems to have dropped the -12V supply. For P OK and DC OK I found this: http://home.windstream.net/engdahl/powerup_reset_circuit.htm, but just for test purposes I assume I could connect the +5 direct to P OK and DC OK. BPOK needs to pulse low after power is stable, BDCOK should be high when power is OK. Or, at least, that's what I got from the drawing you linked to. Jon
Re: Oberon and the OberonStation (retro-style FPGA computing)
On 11/25/2015 09:01 AM, geneb wrote: On Tue, 24 Nov 2015, Paul Koning wrote: Pascal is largely obsolete now, but I still appreciate it -- of the 40 or so languages I know, there are only two where I could go from no knowledge at all to having a working program of significant size in one week: Pascal and Python. Um, no. Check out Delphi, Lazarus and FPC. No where near "obsolete". I was really pleased with FPC! Are people actually creating new projects with it? I was grateful to discover FPC so I could update my photoplotter program and move away from the Turbo Pascal / Windows XP executable I have been using. But, I haven't used it for any new projects. (Yet.) Jon
Re: Replacing MicroVAX II PSU With a Modern PSU
On 11/25/2015 11:01 AM, Robert Jarratt wrote: My reading of the Qbus spec is that P OK and DC OK are active high. That diagram tells me that both control signals are driven the same way because the two circuits are identical except for the timing caps. When the power is OK the output of the inverters is low, so the transistors are off, presumably allowing the signals to float high. When the power is not OK, the inverters are high, turning on the transistors and shorting the signal to ground. But, I am no expert, and could have interpreted those circuits all wrong. Regards Rob Yup, I read the circuit the same way. Jon
Re: DEC H745 regulator common failure modes?
On 11/26/2015 11:13 PM, Pontus Pihlgren wrote: I can only say that they fail often. We have replaced lots in our 11/70. Luckily we have had spares. Sometimes the failure has been rather catastrofic with blown components. I think these modules run quite hot since discoloured pcb:s are common. My experience with an 11/45 that ran 24/7 was the AMP Mate-n-Lok connectors would get a poor contact, heat up and go into thermal runaway, melting the plastic housing and totally frying the contacts. I had to repair these at least 5 times while we had that machine. I never actually had a regulator itself go bad, it was always the contacts. Maybe we just got lucky. (This was also about 1977 - 1980, so parts have probably deteriorated a lot more since then.) Jon
Re: Memory Voltage on MicroVAX II
On 12/05/2015 09:51 AM, Robert Jarratt wrote: I finally fixed my H7864 PSU so I can now run my rtVAX 1000. However, I think the machine is damaging memory boards. I checked the ripple and 5V looks OK, but 12V looks suspicious. Is the 12V supply used by the memory? I don't have my print set anymore, but I highly doubt it. I think the +/- 12 V is only used by serial comm boards, and possibly by some graphics boards. I would not be surprised if the regulation of the 12 V supplies is not very precise. Jon
Re: Memory Voltage on MicroVAX II
On 12/05/2015 12:29 PM, Robert Jarratt wrote: -Original Message- From: cctalk [mailto:cctalk-boun...@classiccmp.org] On Behalf Of Jon Elson Sent: 05 December 2015 18:03 To: gene...@classiccmp.org; discuss...@classiccmp.org:On-Topic and Off- Topic Posts Subject: Re: Memory Voltage on MicroVAX II On 12/05/2015 09:51 AM, Robert Jarratt wrote: I finally fixed my H7864 PSU so I can now run my rtVAX 1000. However, I think the machine is damaging memory boards. I checked the ripple and 5V looks OK, but 12V looks suspicious. Is the 12V supply used by the memory? I don't have my print set anymore, but I highly doubt it. I think the +/- 12 V is only used by serial comm boards, and possibly by some graphics boards. I would not be surprised if the regulation of the 12 V supplies is not very precise. Jon There seems to be a spike on the 12V supply, surely that isn't good, even if it doesn't affect the memory? How BIG a spike? In a lot of these power supplies, they don't regulate the "auxiliary" voltages. So, whatever power is needed to be sent to the main voltage, the aux. windings just tap off some of that energy. If the flux produces the right voltage on the +5V (with so many turns on that winding) then they put a few more turns on the 12 V windings, and assume it should track FAIRLY well. But, maybe during the power-up surge, charging all the caps on the +5 network, the +12 might surge a bit. I built my own power system on my uVAX-II system, and had separate supplies for +5 and +12 (mostly for the disk drives). One day the 12 V supply went haywire, and I had inadvertently disabled the crowbar circuit. It went up to 22 V and damaged the disk drive and blew caps and the RS-232 driver on the CPU board. The disk drive worked for another week, and then totally croaked. I ordered the parts for the serial transmitter and got it fixed fairly easily. (He he, glad I had that print set!) So, I can say that if the spike gets close to 22 V, that would be real bad. Jon
Re: Memory Voltage on MicroVAX II
On 12/05/2015 01:49 PM, Robert Jarratt wrote: As the 5V seems fine, the ripple seemed to be about 20mV (although I am going to check again), I do wonder what could be causing the memory modules to appear to be failing. I am hoping that re-seating will cure it. Regards Rob I ran a uVAX-II for 21 years here in my house. It was HOT STUFF when I first got it in 1986! By 2007, it was the slowest computer in the house. It ran continuously during that period (at the end, it was only running a home environmental monitoring program, or I would have shut it down earlier.) Anyway, after some years of flawless operation, I started getting crashes every couple months. When it would hang, I would power down and re-seat all the boards. It seems like it was usually a failure of one of the grant chains (either interrupt or DMA) and the disk controller would not be able to transfer. Every once in a while I'd pull all the boards and vacuum out the backplane and gently vacuum off the boards. That sort of helped, maybe. The external UVII memory also had ribbon cables across the boards. Rough handling of these cables can cause intermittents. Jon
Re: IDE knowledge anyone?
On 12/06/2015 02:57 PM, Oliver Lehmann wrote: Hi, I've built a Harddisk-Controller-Emulator for my system which accesses a IDE (PATA) harddisk with an ATMega in PIO mode. It works like a charm except for one WD harddisk. The harddisk itself works fine with MS-DOS 6.22 and FreeBSD but refuses to work with my ATMega. On reading or writing a sector, right after the command is issued, the error bit is set in the status register, and the error register indicates an ABRT. I don't have any specific knowledge of this, but it has to be a timing problem. Something about the timing of the command signals is just different enough that the drive thinks this is wrong. Sorry I can't give more specific help. Jon
Re: VAX 4000-500 PSU Overload?
On 12/10/2015 04:30 PM, Robert Jarratt wrote: My VAX4000-500 will no longer power up, with the PSU starting up and then immediately shutting down. I suspect a possible short somewhere. I have measured the resistance of the load presented to the PSU by connecting probes to the backplane sockets used to power the machine. The odd one is the 5V load. With all the boards in and drives inserted I measure a resistance of about 4R. As I pulled out boards, drives and fans, it gradually crept up to 6R. So with nothing connected to the backplane I get a 6R load across the 5V supply. Most likely there is a failed capacitor somewhere. Look over the boards for small Tantalum caps. They may not show any outward signs of failure. Finding the bad cap on a board with dozens of them in parallel is a nightmare! You should be able to safely power the machine with only a couple boards at a time to find which one is bad. And, of course, it could be the power supply itself! Jon
Re: [cctalk] Re: TOP POSTING
On 12/12/2015 07:22 AM, Mike wrote: The one question I do have for the older gentlemen on here is what in the world did the computers without a screen to look at do? Now I know about the tape, cassette tape's and even the paper with the hole punches in them but what kind of applications were they use for? Mathematics or? ? ? Well, in the 1960's lots of companies ran their payroll, billing and general ledger on IBM 1401 and similar machines. IBM sold 12000 of them, and then there were additional partially compatible machines in the same line (1410, 1460). Many 1401's were used in the beginning entirely from cards, or cards and tape, with no disk. Certainly, some people used them for math problems, as well. I know our local phone company had a 1401 in their planning department, and they used it to compile statistical info telling them what equipment they needed to plan in the future. Our University had a 1401 for accounting, and a 7094 for the rest of the University's needs. Then, in 1969 they upgraded the 7094 to a 360/50. None of these machines had "screens". All programming was done on punch cards, most of the data was prepared on punch cards, although some came on tape. Then, if you were a user with privileges, you could get some tiny amount of disk space to hold data for a while. Sometime later, maybe around 1971 they got some IBM 2741 terminals which were Selectric typewriters connected to an interface. Later they got some IBM 2260's, which were Zenith 9" TV sets and a keyboard connected to an interface box in the machine room. Very primitive, but very interactive, great for quick program editing and submission. Jon
Re: Display-less computing was Re: TOP POSTING
On 12/12/2015 08:18 AM, tulsamike3...@gmail.com wrote: So did you have to learn how to read the punch hole cards also or did the punch hole cards go into the computer and than printed out the data on the fan fold paper also was it in code or just plane English? You COULD read the holes, if you really HAD to. Keypunches printed the alphanumeric form on the top edge of the cards. if you punched a deck of cards on the CPU's card punch, there was no printing. If it was an "object deck" ie. binary code, you would never "interpret" the deck. But, if it had something that might be human readable, there was a machine called an interpreter, and it would type the symbols on the top of the card for you. Jon
Re: TOP POSTING
On 12/12/2015 09:07 AM, Gene Buckle wrote: [*] A power outage on Wednesday killed the power supply in my main mail exchanger that I read email on via Alpine. F*ck Dell and their proprietary power supplies and connectors! All my Dell Optiplex boxes (that's the commercial grade models) have standard power connectors. All the Dell boxes do have proprietary shape/size power supplies, but in a pinch you could hang one outside the box. Jon
Re: Display-less computing
On 12/12/2015 11:17 AM, William Donzelli wrote: Sometime later, maybe around 1971 they got some IBM 2741 terminals which were Selectric typewriters connected to an interface. Later they got some IBM 2260's, which were Zenith 9" TV sets and a keyboard connected to an interface box in the machine room. Very primitive, but very interactive, great for quick program editing and submission. Were 2260s really Zenith products inside? I certainly could believe it, judging from the tube lineup. Someday I would like to see one up close and personal. Yes, they were hideous hacks of a cheap TV set. The VHF tuner was still in there, as it was needed to complete the series filament chain. They had 5 tubes in series, with 26-volt filaments, running off 120 V AC. I'm PRETTY sure the brand was Zenith, but this is from about a 43-year old memory of pulling the cover off one. All it had was a keypunch-style keyboard and the TV, all the electronics were in the machine room. Jon
Re: Display-less computing
On 12/12/2015 11:17 AM, William Donzelli wrote: Sometime later, maybe around 1971 they got some IBM 2741 terminals which were Selectric typewriters connected to an interface. Later they got some IBM 2260's, which were Zenith 9" TV sets and a keyboard connected to an interface box in the machine room. Very primitive, but very interactive, great for quick program editing and submission. Were 2260s really Zenith products inside? I certainly could believe it, judging from the tube lineup. Someday I would like to see one up close and personal. Yes, they were hideous hacks of a cheap TV set. The VHF tuner was still in there, as it was needed to complete the series filament chain. They had 5 tubes in series, with 26-volt filaments, running off 120 V AC. I'm PRETTY sure the brand was Zenith, but this is from about a 43-year old memory of pulling the cover off one. All it had was a keypunch-style keyboard and the TV, all the electronics were in the machine room. Jon
Re: IBM 2260 acoustic delay line
On 12/12/2015 09:13 PM, Eric Christopherson wrote: I'm reading about those terminals and find it just fascinating how they used acoustic delay line memory to remember the pixels. But I have lots of questions: 1. Did the cables connecting the 2260s to the display controller actually contain the delay lines themselves, over the whole length; or were the delay lines just inside the controller and then some electronic signal was sent out to the terminals? The delay lines were little coils of steel wire in a housing. Probably some 30 feet or so wound up in a spool about 9" diameter. See these for some examples. The 2nd one might actually be a 2260 delay line. https://upload.wikimedia.org/wikipedia/commons/e/ef/Torsion_wire_delay_line.jpg http://www.glennsmuseum.com/ibm/med/med_ibm_wire_memory.jpg I actually have a bunch of 2260 cables here, a couple coaxes for sync and video, and about 15 wires. 2. I would think that the wave travelling along the delay line would weaken over time. How was it refreshed? You read the output through a read amplifier, squared it up to a digital signal, and re-launched it onto the wire. if you wanted to change the info, you switched a multiplexer and inserted the new data instead of recirculating the old character. 3. What kind of speed could be acheived, and did this depend on the number of connected terminals? It depended on the 360 CPU, which were not all that fast. But, WAY faster than just about anything at that time. I'm not sure what you really mean by "speed". You could alter the characters on the page you were viewing, and then transmit that to the computer. This all happened at IBM 360 channel speed, so quite fast. (Actually, it was almost certainly running on the multiplexer channel, so each character transferred caused a channel request to put it in a buffer, then when 4 were in the buffer, the channel would cause a memory transfer. (4 bytes was the memory word width for an IBM 360/50.) So, that all happened in milliseconds. Still, this was lightning-fast for that time (1972 or so). Since the OTHER way to do programming was editing decks of punch cards with a keypunch, then submitting it for batch processing and getting your error messages back 4 - 8 hours later, it was REALLY a step forward. I wasn't in any courses where I was actually allowed to use the 2260, though. Jon
Re: IBM 2260 acoustic delay line
On 12/12/2015 09:37 PM, William Donzelli wrote: The IBM 2848, the control unit for the 2260 terminals, contained mercury (!) delay line for the video memory. There may have been some compensation for the transmission to the terminal, but I have have not seen the technical details - I think it was not a concern, probably. I don't know where you got this information. I can find a NUMBER of references that say it was a wiresonic delay line. (Due to wiresonic being a trademark, IBM used a different name for it.) But, most certainly, IBM did not have barrels of mercury in mainframe computer rooms in the 1960's. Page 2-29 has the description of the magnetostrictive delay line, see : http://www.bitsavers.org/pdf/ibm/2260/Y27-2046-3_2260_2848_FETOM_Mar68.pdf Jon
Re: Display-less computing was Re: TOP POSTING
On 12/12/2015 06:13 PM, Eric Christopherson wrote: On Sat, Dec 12, 2015, Fred Cisin wrote: you could feed the cards through an INTERPRETER, which printed the card content on the card. [snip] For many years, I kept around a plug-board labelled "COBOL INTERPRETER", just to prove that a COBOL interpreter was possible :-) Are you using "interpreter" in two senses here, or just one? That is to say, I'm not sure if you're saying the "COBOL interpreter" was just a program that printed COBOL source on a punched card, or if you mean it actually ran the program. I'm pretty sure this was a joke based on the meaning of the word interpreter. If he could have actually run an arbitrary COBOL program, with named variables and conditional branching on a plugboard-based accounting machine, I think IBM would have been VERY interested in hiring him! (Or possibly having him dumped in a nearby body of water with concrete overshoes.) Jon
Re: Webster WQSMD/04 Qbus SMD Controller
On 2015-12-13 1:40 AM, Robert Jarratt wrote: I picked up some Qbus cards yesterday. They seem to be board set for a MicroVAX II. However, one of the cards was, to me at least, a bit unusual. It was made by a company called Webster, and it appears to be a controller for SMD disks. I was not familiar with SMD disks and had to look them up. I suspect this might be a little out of the ordinary, and, possibly, an odd combination for a small Qbus system to access such a physically large type of disk. Were MicroVAX IIs used much with such disks? Is this a bit of an unusual find? Not completely unusual. Our company had a pair of Microvax IIs, in rack mount form, that shared a pair of SMD drives in another rack mount chassis. The cards used to interface to the 8" drives were EMULEX QD33 MSCP-compatible SMD/SMD-E disk controllers. I would have one of these up and running right now, except that the power supply in the drive chassis died the last time I tried it and haven't had time to fix it. I believe the drives are from fujitsu. I believe the drives were mainly used as NFS mounts for other UNIX boxes. -Jon
Re: What did computers without screens do?
On 12/14/2015 06:31 PM, Mike wrote: What would you do with a home no screen computer? I mean what could be done with one that would benefit your work / hobby. I mean NO DISREPECT by asking this question. Well, that would severely limit things. But, it could still be useful. I have one app that runs on my home everything-server, that doesn't have a normal screen. It takes temperature, humidity and energy consumption measurements and records them to a file every 15 seconds. it also puts time, date, temp and humidity on some 20-character LCD screens around the house. So, that is SORT OF a screen, but not in the traditional sense of an interactive terminal. My laser photoplotter has no screen or other HMI connected to it, just a network cable. So, I log into it from another computer to make artwork films. A lot of people have 3D printers with no screen at all, or a very limited LCD screen that lets them set a few options and shows the percentage progress of the build. Oh, for the lowest possible case, I have a computer-controlled air compressor. This is an AVR chip which manages starting/stopping the motor and loading/unloading the compressor valves. (This is called automatic dual-mode control, when the tank is full, it unloads the compressor and leaves the motor running for a minute to see if there is more demand for air before shutting it off.) Jon
Re: Decisions you regret Was: Mystery IC: Allen Bradley 314B102
On 12/16/2015 11:01 AM, Jay Jaeger wrote: Anywho, I was looking at a couple of 19" racks containing an odd computer of some sort. Had this funny square keyboard, and what looked like LINCTapes to me. Looked kinda "home brew", using DEC Flip Chips. Well a couple of years later I saw a photo of a LINC, and then it was "head slap" time - I realized I had passed up a LINC. Could have had it for $25. I fear it was probably scrapped. Sigh. A Classic LINC used "system building blocks", generally single-sided boards with an aluminum frame around the board, and a single-row 22-pin connector that was a separate piece, not a card-edge connector. The little keyboard on the Classic LINC was made by Soroban, and it was indeed funny. Each keystroke locked the keyboard, and when the program picked up the character from the buffer, the keyboard unlocked. The delay was often heard, as LAP-6 spent 99% of the time refreshing the screen. If it was real flip-chip modules with the little molded plastic handle, that would have been a LINC-8 or PDP-12. Jon
Re: IBM Mainframe terminal stuff
On 12/18/2015 08:40 PM, William Donzelli wrote: Most (all?) 029/129s have a chipboard top. Also, you S/3 should, right? The only bit of classic computer stuff I have ever seen with REAL wood on it was a Computervision system during college. The cabinets has strips of oak on the front. Quite nice. I'm pretty sure the desktop on the system 360 and 370 machines were also chipboard with laminate. 129 keypunches were certainly chipboard with laminate, but I think the 029s were steel, with a laminate applied to it. Jon
Re: IBM Mainframe terminal stuff
On 12/18/2015 09:27 PM, William Donzelli wrote: I have a spare 029 that got wet, so the desk portion expanded. Definitely chipboard. Sorry, I was thinking of the 026! Jon
Re: VAX 4000-500 PSU Overload?
On 12/19/2015 04:56 AM, Robert Jarratt wrote: Actually, I may have just narrowed it down a bit. I connected a 12V DC bulb to the -12 output, in parallel to a 15R resistor, and the PSU shuts down immediately. The 12V bulb has a 21W rating. If my calculations are correct, that means the bulb draws 1.75A. The 15R resistor would draw 0.8A. So a total of 2.5A, which is well under the spec of 4A, but causes the PSU to shutdown. Using the 5W filament of the bulb does not cause the PSU to shut down. Is there a flaw in my calculations, or am I really not overloading the -12V supply? Just to recap I had a 15R resistor and a 12V DC 21W bulb in parallel, on the -12V output. Cold light bulbs draw WAY more current than when the filament is hot. Not sure if a 21 W 12V bulb could draw 20 A cold, but it might. Jon
Re: Ferroresonant transformers and 3278
On 12/21/2015 09:03 PM, William Donzelli wrote: If you can get a rotary one, those are really nice - just wasteful and loud. With proper maintenance they last forever, can take a beating, and do not give waveshape issues that cheap solid state units can have. And, maybe most importantly, you can make one yourself. But considering the mix of 50 and 60 Hz stuff you likely have by now (that is what you get for moving!), spending some decent money on a real VFD might be worth it. I might think a cheap VFD may give ferroresonant iron fits with all those extra harmonics. You can't run electronics with VFDs designed to run motors, only. They put out PWM chopped square waves at 300+ Volts. A motor's winding inductance smooths that out to a proper current waveform, and it only causes a little extra eddy current losses. But, typical transformers will have real fits with that kind of waveform. There are "frequency changers" made by Elgar and others that will do the job right, but they will cost a REAL bundle of cash! (Also known as frequency converters.) It may be possible to retune the resonant circuit of the constant voltage transformer by adding capacitance in parallel to the existing capacitor. Jon
Re: Ferroresonant transformers and 3278
On 12/22/2015 10:25 AM, William Donzelli wrote: Another thought - with scrap transformer prices in a crapper right now. you might also be able to pick up a really beefy isolation transformer for spare change, in order to clean up a VFD. Generally, normal industrial grade 50 Hz transformers will not care too much if you use them at 60 Hz. An isolation transformer will NOT "clean up" a VFD's output. The output of a VFD is NOT some dirty sine wave, it is a train of 400 V PWM pulses with perhaps a 15 KHz carrier frequency. It could be filtered, but would take a filter built out of a bunch of HUGE inductors and capacitors. It might actually be a decent solution for keeping museum-quality gear running in an unaltered state, but it isn't something you could whip up in a couple hours in your garage. Jon
Re: Nuts & Volts ESR Meter
On 2015-12-23 7:18 AM, Antonio Carlini wrote: Can someone with the digital edition clarify whether it really is a PDF or not? If you look at the online version, there is a "download" link to the .PDF in the upper right corner. It downloads and reads just fine. Thanks Jay! -Jon
Re: Ferroresonant transformers and 3278
tis 2015-12-22 klockan 16:45 +1300 skrev Mike Ross: Finagle's law says 90% of my 3-phase big iron was acquired in the USA and expects 3-phase 60Hz 208V... :-( Many machines which did NOT have a ferroresonant transformer could be run from 50 or 60 Hz with no problem. The one place it was a problem would be getting the disk spindles turning at the right speed. Matching the available voltage would be the other detail, but a standard voltage adjusting transformer would do it fairly simply. But, the ferroresonant transformers needed the right frequency. Jon
Re: XY11 Manual, Anyone?
On 12/23/2015 10:56 AM, couryho...@aol.com wrote: it appears the pen kit for our plotter got LIFTED (aka stolen, ripped off... etc... bummer..) before we had it glassed in living the glassed in display a areas if anyone has a calcomp box with the solenoid and pens that is extra to their needs please let us know I have a bag of original, unused Calcomp pen nibs. These are basically identical to Rapidograph pen nibs. These worked with my Calcomp 1076 plotter. I got rid of the rest of the stuff when I scrapped that plotter. (It was a HUGE beast.) So, all the holders, adapters, etc. are gone. The rest of the stuff was all badly gummed-up. Jon
Re: PDP-12 Restoration at the RICM
On 12/24/2015 01:22 PM, Jay Jaeger wrote: On 12/24/2015 9:59 AM, Michael Thompson wrote: We have been able to fix all types of broken flip-chips. Sourcing the components is sometimes a challenge. The Germanium transistors for the TU20 on the PDP-9 were hard to find. I remember replacing a germanium (at least I think it was) transistor on an SMS card with an ordinary silicon transistor in a 7094-II floating point unit back around 1974. Luckily, that worked fine, though for a museum I imagine one would prefer to use the "real thing". ;) Yes, I overhauled an old HP digital frequency synthesizer that was all built out of PNP Germanium transistors (no ICs.) I substituted the first one with a VHF Silicon transistor and did some tests on the bias, etc. and was pleased to find it was a total drop-in replacement. I replaced over 10 of them in that unit, and they all worked flawlessly. I can imagine some difficult circuits where you couldn't get away with this, maybe a magnetic read amp or a timing circuit or something, but I think in most cases a Silicon transistor will work well. Jon
Re: HP 16702A logic analyzer failed Ethernet (was Re: dumping Western Digital Microms)
On 12/24/2015 01:57 PM, Glen Slick wrote: On Wed, Dec 23, 2015 at 11:24 PM, Eric Smith wrote: Then in the progress list, it shows: Configure LAN interfaces [FAIL] * Check LAN Status .. [N/A ] [...] Configure HP Ethernet interfaces . [ OK ] Configure HP 100BT interfaces [N/A ] Configure LAN interfaces . [ OK ] Network hardware not usable - Is the network connected? I think the Ethernet ports on this beast are dead. :-( If this uses an AUI cable to an Ethernet tap, it is possible the diags don't work without something connected to the AUI cable. Just totally guessing! Jon
Re: Merry Christmas
Yes, I also want to wish everybody, and Jay, especially, a merry Christmas and a happy new year! Jon
Re: The KGB, the Computer, and Me
On 12/29/2015 06:24 PM, Ray Arachelian wrote: On 12/28/2015 04:45 AM, Rod Smallwood wrote: I now expect to get a long list of weveseenits. Indeed! I remember seeing this on live TV many years ago. You can find it on youtube now: https://www.youtube.com/watch?v=EcKxaq1FTac The book is really good too: http://www.amazon.com/Cuckoos-Egg-Tracking-Computer-Espionage/dp/1416507787/ The book is totally amazing! I've rarely read anything by a scientist that is quite so well-written! Of course, I was interested in the topic, but Cliff Stoll writes very well. it reads like a thriller. Jon
Re: 10 forgotten wonders of 1980s homes
On 12/30/2015 12:39 PM, Guy Sotomayor wrote: We just dropped our POTS service last month. Our phone now is an OOMA VoIP Office system to get the features we wanted/needed. I haven't tried any rotary dial phones on it since I don't have any. We need the landline phone to call when the cable is out. We need the VOIP phone to call when the copper line is out (which is more often). Where I'm at, I can't get DSL. Right now I'm dependent upon a dedicated microwave link to my iSP. I'm getting between 20-35Mb/s symmetric (used to be 10Mb/s but it looks like they upgraded the link recently). I'm paying $$$'s but since I work from home I need the bandwidth (I'm currently using about 800GB/mo). I had to put in a point-to-point WiFi link to get internet to the house from my shop (where the microwave dish is). I can't get DSL here anymore, the lines are WAY too bad for that. I used to have DSL way back when. I wish I could get symmetric cable speeds here, 30/30 would be FANTASTIC, as I run a web server here. I live with 60/5, but rumors are that we will get upgraded to 100/10 sometime fairly soon. Jon
Re: 10 forgotten wonders of 1980s homes
On 01/01/2016 01:52 PM, Chuck Guzis wrote: I'm amazed that "caller ID" costs CL anything, but it's one of the highest-priced add-ons. It clearly costs them NOTHING! It is software on the switch, and the hardware needed to support it is present on every subscriber line. They charge for it because people are willing to pay for it. Jon
Re: Building a PC - then & now
On 01/13/2016 04:29 PM, Murray McCullough wrote: I was reading in a dated magazine article on the "freedom to build(a PC)": Well you can't build phone; can't build a car; can't build a refrigerator; can't build a TV. Do we have the freedom to build a computer? We did in the earliest days of the PC- the 8-bit era. Heck, that's all one could do! It was limited and is to this day. AMD vs INTEL control what we can do. Has anything really changed? Hmmm, well. I can build a phone (not a cell phone, of course, but a land-line phone is relatively simple.) Some people have definitely built cars, and gotten them licensed. I have built things that are essentially refrigerators. We have one of those ice cream makers that takes about a ton of ice cubes to make one quart of ice cream. I hacked up some refrigeration components to chill the brine, which gets re-used. Works great! My insane friend Walter turned his Tek RM35 scope into a TV, and watched TV on a 5" green screen while he was in college. Hmm, Walter also cloned a Data General Nova with piles of TTL chips. Probably very little by AMD or Intel in there, either. Jon
Re: Archiving CP/M 2.2 Source Code Programs to a PC (Fat or NTFS media)
On 01/15/2016 08:09 AM, Robo58 wrote: Hi Folks, I have many diskettes worth of CP/M 2.2 assembler source code and programs that I'd like to archive in the PC environment. I'm worried that my media is degrading and I want to move it before it's too late. The media is mostly 8" SD or DD, there are also some 5.25" HD diskettes too. I have the original hardware and can view the media and run the programs. I'm looking for suggestions on how to move it to the PC environment. Got a console serial port on the CP/M system? You should be able to use a program like Kermit to suck up the files. Probably log a file list, edit the file list to only have source files, and then you can get that into Kermit to make the CP/M system type out the files, and Kermit will log them to individual files. I've never done this, but I think kermit was specially designed to make this easy to do. I'm guessing a few other terminal programs have the same sort of capability to run scripts to send commands to a foreign OS and log the result. Jon
Re: My last word on building computers!
On 01/15/2016 12:33 PM, Murray McCullough wrote: I’m not sure to what degree one can/wish to build there own car. If one puts their mind to it; then anything is possible. I’m sure this applies only to die-hard builders and not representative of the ‘average’ guy/gal. A guy in France built a 1/3 scale Ferarri roadster. He made EVERYTHING himself. Dashboard instruments, tires, ignition coils and spark plugs, and on and on. I saw it at the NAMES show in 2004, I think. Totally awesome. And, it took him 12 years! Jon
Re: My last word on building computers!
On 01/15/2016 09:44 PM, ben wrote: On 1/15/2016 6:59 PM, Jon Elson wrote: A guy in France built a 1/3 scale Ferarri roadster. He made EVERYTHING himself. Dashboard instruments, tires, ignition coils and spark plugs, and on and on. I saw it at the NAMES show in 2004, I think. Totally awesome. And, it took him 12 years! Smaller cars well never sell. :-) Right, he claims to have it on his mantelpiece! (Must have a really big mantel!) For the effort he went through, he could have built a full-size car and driven it! While I can appreciate these incredible models, I don't quite "GET" the model building mania. Jon
Re: My last word on building computers!
On 01/15/2016 11:25 PM, Fred Cisin wrote: A guy in France built a 1/3 scale Ferarri roadster. He made EVERYTHING himself. Dashboard instruments, tires, ignition coils and spark plugs, and on and on. I saw it at the NAMES show in 2004, I think. Totally awesome. And, it took him 12 years! On Fri, 15 Jan 2016, ben wrote: Smaller cars well never sell. :-) That's why they never made another one. Oh, and it is not a static model. The engine runs (spewing clouds of run-in oil and making a HUGE noise. If you prop up the rear end, you can go through the gears and spin the tires. Jon
Re: TU58 problems
On 01/16/2016 11:44 AM, tony duell wrote: So I get that a bit should take around 41us. And a bit starts with a rising edge, the position of the falling edge (recorded at 1/4 or 3/4 of the bit time) determines whether it's a 0 or 1. No, not correct for 800 BPI (NRZI). That is somewhat how 1600/3200 PE tape works. Well, sometimes that's what I see on the LogicDart. Sometimes I see a 1:1 square wave with a period of 40-odd us. 800 BPI (NRZI) only has transitions on a data track when there is a 1 recorded. There will always be at least one 1 recorded in any byte, as the odd parity track will have a 1 for an all-zero data byte. So, a track that has all 1's in consecutive bytes will have your ~40us square wave. When the head is porperly aligned for skew, there will be a skew test point that should have an approximate square wave. When the skew is bad, this will have a wave with stepped sides, showing where individual bits came in with a time skew. If a track has a run of zeroes, then that data track will have no transitions in it for that interval. At the end of the data block, there will be a 2 character-time gap, then the CRC and LRCC will be recorded. The CRC is computed with a shift register with XOR gates, the LRCC is produced by just clearing the NRZI register. it is a longitudinal parity of all the 1's in that data track. If a track has an odd number of 1's in it (including the CRC) then there will be a transition in that bit of the LRCC. Jon
Re: TU58 problems
On 01/16/2016 02:31 PM, tony duell wrote: 800 BPI (NRZI) only has transitions on a data track when As I understand it the TU58 is not NRZI. WHOOPS! I Somehow read that as one of the 1/2 9-track tape drives! Jon
Re: Non-baking cure for sticky shed? (being serious for a moment)
On 01/18/2016 04:39 AM, Dave G4UGM wrote: I know that tapes have different coatings, so some are ferrous and other chrome based, but what about the backing and "glue" that holds the two together? There was a tape sold by a few manufacturers (I think 3M's was called Black Watch) that had a coarse black matte finish on the back side. A problem with some drives which had fast rewind speeds was that air was trapped in the tape wind, and then once a bunch more layers of tape was wound on the reel, the tape wrap would cinch and fold over a piece of tape. This matte finish was supposed to cure that, and I think it did. But, that finish was not as durable as the magnetic coating, and would start to shed all over the drive, leaving stuff on the data surface. I think about 5 years after the stuff came out, most data centers searched out and trashed all these tapes. Jon
Re: Non-baking cure for sticky shed? (being serious for a moment)
On 01/18/2016 10:56 AM, j...@cimmeri.com wrote: On 1/18/2016 11:33 AM, Jon Elson wrote: There was a tape sold by a few manufacturers (I think 3M's was called Black Watch) that had a coarse black matte finish on the back side. A problem with some drives which had fast rewind speeds was that air was trapped in the tape wind, and then once a bunch more layers of tape was wound on the reel, the tape wrap would cinch and fold over a piece of tape. This matte finish was supposed to cure that, and I think it did. But, that finish was not as durable as the magnetic coating, and would start to shed all over the drive, leaving stuff on the data surface. I think about 5 years after the stuff came out, most data centers searched out and trashed all these tapes. Jon Interesting. I just happen to have a 3M Black Watch tape right here next to me that I'm using to test a Mark 300 tape cleaning machine. The tape so far is fine, not shedding anything front or back. - J.
Re: Non-baking cure for sticky shed? (being serious for a moment)
On 01/18/2016 10:56 AM, j...@cimmeri.com wrote: Interesting. I just happen to have a 3M Black Watch tape right here next to me that I'm using to test a Mark 300 tape cleaning machine. The tape so far is fine, not shedding anything front or back. Sorry about previous reply, hit wrong button. Well, I don't know if it was the Black Watch that was the problem, or somebody else's version. We started with spring-arm drives, which could not keep constant tension on the tape. You'd hear this HORRIBLE scrunch noise at various times when the tape was being processed, which was the layers of tape sliding over each other. Usually didn't cause problems, but on occasion it could be so bad that the drive would lose tension and shut down in the middle of the reel. Vacuum column drives didn't seem to have much of this problem. I think it was the jerk of tension when the drive reversed to reread a bad block that started the slipping on the spring arm drives. Longevity of tapes was very spotty. Some expensive, name-brand tapes just disintegrated after a few years. Others have held up amazingly well. Last year I read in some 1993 backups of my MicroVAX system with no trouble at all. I did have to clean the tape head after every tape, but that wasn't greatly different from when the tapes were new. Jon
Re: Non-baking cure for sticky shed? (being serious for a moment)
On 01/18/2016 11:53 AM, j...@cimmeri.com wrote: On 1/18/2016 12:46 PM, Jon Elson wrote: Last year I read in some 1993 backups of my MicroVAX system with no trouble at all. I did have to clean the tape head after every tape, but that wasn't greatly different from when the tapes were new. What media were those backups on (eg. TK70)? 9 track, 6250 BPI, recorded on CDC Keystone (92185) streaming drives, and read back on the same type. I built an interface so I could read the tapes in on a Linux system, and then used some publicly available software to unpack the BACKUP formatted images to native Linux files. The Pertec formatted tape interface is quite simple. My interface is pretty simple, though, and is controlled through the parallel port, so the tapes are processed at 25 IPS start/stop mode. If I were going to do it now, I'd use a Beagle Bone, and it probably would stream at 75 IPS. Jon
Re: M. Minsky - AI & Classic Computing
On 01/27/2016 11:24 AM, Murray McCullough wrote: I learned today of the passing of a true computing visionary, Marvin Minsky He of artificial intelligence fame. We in the classic computing fraternity, and computing in general, can enjoy our ‘hobby’ because of his work. Yup, sad day! Jon
Re: IBM 3101-12 ASCII terminal - need fuse holder
On 01/28/2016 10:28 PM, Ethan Dicks wrote: Hello, all, I was just gifted with an IBM 3101-12 ASCII terminal that happens to be missing the fuse and fuse holder. Unlike a lot of 1960s and 1970s gear, it's not round. It's square. Is this a standard IBM thing from the 70s/80s? Anyone know where I could get one? It seems to snap in and probably fell out at some point under its previous owner. I'd open up the case and see if you can find out who made the whole fuse holder assembly. Lots of outfits made "fancy" fuse holders that were different entirely for visual appeal. Or, just browse through the Digi-Key or other catalogs and try to find a fuse holder that fits the opening in the case. I'm sure it was NOT a custom IBM part. Also, I found only a little info on it from Googling. Later IBM ASCII terminals emulated ANSI command or Wyse-50 or something. I couldn't find anything on the 3101. Is it a glass TTY or does it respond to any cursor positioning, etc. commands? It definitely has cursor positioning commands. I ran one on a CP/M system, and added in the cursor and other commands on a configurable editor and used it for some time. Jon
Re: BA23 fan noise
On 01/31/2016 08:42 PM, Mark G. Thomas wrote: Does anyone have any suggestions for figuring out how much airflow I actually need, and achieving it with either stock fans at further reduced voltage, or some kind of replacements? I don't need an accurate solution, or something with complicated compensation for varying temperature, just something quieter, moving less air, but still enough air. I may have cooked some peripheral boards in my homebrew uVAX cabinet. The KA630 seems pretty robust, it ran for 21 years continuously under these conditions. I got an EBM motorized impeller and made up my own plenum. The AC motor ran it WAY too fast, so I cobbled an 8" floppy brushless motor onto the original impeller, and I could adjust the speed. My main test was to let it run an hour and see if the air coming out seemed too warm. This was pretty unscientific. The only boards that croaked were Dilog and such 3rd party boards. I did have a thermal safety system that would cut power if the cooling failed (which it never did). I later got a tangential blower for the expansion backplane, and it was a good deal quieter, but maybe didn't move enough air. Jon
Re: Mystery system
On 02/04/2016 07:48 PM, Mike Loewen wrote: I'm trying to identify a system which appeared in "The Killer Elite" (1975), with a room full of tape drives and a couple of terminals: http://q7.neurotica.com/killer-1.png http://q7.neurotica.com/killer-2.png http://q7.neurotica.com/killer-6.png http://q7.neurotica.com/killer-9.png The drives appear to be IBM 3420s, but with an additional box on top, labeled "SMS". The system itself doesn't appear in any of the shots. Some sort of IBM 370, perhaps? I'm thinking they might be 2420's. They have a "9" label on the head cover. This is to distinguish 9-track from 7-track drives, I don't think they put those labels on 3420 drives. The drive addresses are 5xx, meaning they are attached to channel 5 on the CPU. That is not impossible on a 360, but few 360's could handle that many channels. The cluster of 3270's also suggests a 370 system, although they certainly could be used on 360's. The SMS boxes displayed the volume label of the tape to mount. Mounting the wrong tape was a BIG problem in large systems, so a number of vendors came up with these sorts of schemes to try to reduce those errors. Jon
Re: Mystery system
On 02/04/2016 08:57 PM, William Donzelli wrote: Some models of 3420s were 7 track. Yes, I suppose there was a need in some installations to handle old, archival tapes. But, I certainly never saw one. Most sites got rid of their 7-tk stuff as soon as they could. Jon
Re: Mystery system
On 02/04/2016 11:22 PM, Chuck Guzis wrote: On 02/04/2016 08:25 PM, Jon Elson wrote: On 02/04/2016 08:57 PM, William Donzelli wrote: Some models of 3420s were 7 track. Yes, I suppose there was a need in some installations to handle old, archival tapes. But, I certainly never saw one. Most sites got rid of their 7-tk stuff as soon as they could. Some 6-bit systems hung around for a very long time. E.g. CDC CYBER 170/180. 7-track drives for those never really went out of fashion until well into the 1980s. Here's a 1981 CW article that specifically mentions the CDC 667 in the drive inventory of McDonnell Douglas automation.DP center: http://bit.ly/23MFZOb I was there on a tour, a few years earlier. I think they still had the model 195 at the time. A vast facility filled with big blue boxes. The basement had so many 415 Hz UPS's that you had to stick your fingers in your ears. Jon
