- vcf-midatlantic - lists.vcfed.org

Re: Digest, Vol 25, Issue 21, Message 4 "HAASOScope"
by SVCatITC＠aol.com 21 Nov '17

21 Nov '17

Thanks Very Much Herb for the technical lecture. Us old farts will agree with you on the analog scope versatility, and I also agree with the functionality of the new "finagled" stuff out there now. The latter makes capturing and documenting the triggered event (for later analysis) easy, rather than the Polaroid coupled cameras to the old Tektronix 548 fire breathing monsters' screen of the day. This feature is great for documenting screen shots in reports to customers. I find though that you have to sometimes "help" the new technology by "presetting" the "window" where you might want to look, otherwise trigger thresholds might lose the event. You may not know exactly where to find the "glitch" in the "haystack". >From my experience, analog gives you the "widescreen view" of the disturbance and you can then "zoom" in on the area of interest. The 20 MHz luggable that I bring into the museum does not have the bandwidth to check at what frequency the TR20 power supplies are oscillating at (as an example), however as you well noted, you do not necessarily care about the exact quantity to know you have to kill the "screaming" and make a quiet supply for analog work. The analog scope sets up easy for a quick GO/NOGO check. As always, many thanks for the thought provoking insight. Bill Inderrieden In a message dated 11/21/2017 12:04:17 P.M. Eastern Standard Time, Herb Johnson writes: Date: Mon, 20 Nov 2017 13:45:42 -0500 From: Herb Johnson <hjohnson(a)retrotechnology.info> To: vcf-midatlantic <vcf-midatlantic(a)lists.vintagecomputerfederation.org> Subject: [vcf-midatlantic] Open Hardware DSO - 100 MHz/250 Msa/sec - haasoscope Message-ID: <2320de34-08fc-86ff-33dd-8635edf667cd(a)retrotechnology.info> Content-Type: text/plain; charset=utf-8; format=flowed I agree with my old friend and colleague, mostly. But I'd put stronger emphasis on detecting "glitches". I prefer analog scopes, because they can show in some fashion glitches even above their bandwidth. The following lecture will be technical. This matters to me, as this is work I've done for decades. "Bandwidth as twice sampling rate" is fine for detection of repeated events. It's not fine for reconstructing the signal, or finding one-off events. Digital sampling limits aren't the same as the bandwidth rolloff of an analog oscilloscope amplifier (or probe). Many 20th C. brand-name analog oscilloscopes gave (still give!) good performance well past their nameplate bandwidth. also: scope trigger circuits have to respond to events far above bandwidth! Triggering is very important. A "glitch" is a signal where a logic state changes, not because of a change in logic condition, but because of a timing event such as "propagation" or a "triggering delay". These are very short pulses, on the order of several nanoseconds in TTL circuits. Or, they may manifest as nanosecond delays in signals. Power supply lines carry glitches, caused by literal "spikes" in DC power to logic-switching circuits; large currents plus fast changes produce L di/dt or C dv/dt signals. These matter, because they can trigger logic events. If you can't see the trigger, you don't know "why" you have such events. Old TTL designs, sometimes generated glitches due to poor design. Aging components may cause glitches, by increasing capacitance, or sagging DC voltages. Dead or absent bypass caps, fail to filter out power-supply-line glitches. Look at schematics of the IMSAI front-panel. They actually CREATE "glitches" with a series capacitor to turn a logic-level change into a triggering pulse! Ugly ugly ugly design. All that said, any scope beats none. Looking at logic-level signal activity (any activity) is a go / no-go test for logic chips. Many faults can be diagnosed with a 20Mhz analog oscilloscope and probe. (Probe bandwidth matters. Not today's lecture.) But some faults are design problems that finally came due, or non-logic component failures, or "tired" chips. TTL logic responds to signals and events of several nanosecond duration - that's hundreds of megahertz "bandwidth" technically. and even a 20Mhz scope, will trigger on much faster events - that at least "detects" them. and you can guess if an analog "bump" is actually a pulse above your bandwidth. End of technical lecture. In my 20th century view, a $100ish old toolbox-sized HP or Tek 100Mhz analog oscilloscope, will show more information about pre 1990's computer chips, than most users can appreciate. But a $300 "100Mhz sampling" digital oscilloscope you can carry in your pocket, likely performs little better than a 30-40Mhz no-brand-I-remember analog oscilloscope that you leave behind at a hamfest. Will a 1Ghz modern DSO do much better? (shrug) I lack that experience, ask an owner, faster is better but.... And a DSO is conveniently tablet-sized, and likely does other things too. So the answer for you is: what's in your wallet? Buy BOTH if you can afford it, choose otherwise. Herb "race condition" Johnson retrotechnology.com

1 0

Re: [vcf-midatlantic] Cobol on micro
by Jeffrey Jonas 21 Nov '17

21 Nov '17

> anyone use apple /// cobol, used it back when it was current? I did not but I think my experience with CP/M C and PL/1 are relevant. 1) I used Whitesmith's "C" compiler on CP/M in 1982. It was very slow and almost painful to use - so many phases with constant floppy disk access. And it was not quite the full language or environment. A friend gave me a substitute printf() with more options than the standard library. 2) I used PL/I "F" on IBM mainframes in high school (early 80s). The PL/I subset for CP/M was just too weak to use for much. And like the "C" compiler, was horribly s l o w to compile and link. -- jeff jonas

1 0

Open Hardware DSO - 100 MHz/250 Msa/sec - haasoscope
by Herb Johnson 20 Nov '17

20 Nov '17

David Gesswein: > It all depends on that you want to do with it. This is written assuming you > are troubleshooting old computers. > > You need sufficient bandwidth > to view your signal of interest and sufficient sample rate to make it look > reasonable. Sampling twice the frequency of a sine wave captures the > information but it doesn't look like [the original signal]. > > I think the sample rate will be the biggest limitation. You will need to > see the character of signals you will want to look at. > > For troubleshooting old equipment the degradation of the signal edges > is normally not as big an issue. The faults are normally not that subtle > but I have run across glitches on signals causing trouble that the low sample > rate can hide. I agree with my old friend and colleague, mostly. But I'd put stronger emphasis on detecting "glitches". I prefer analog scopes, because they can show in some fashion glitches even above their bandwidth. The following lecture will be technical. This matters to me, as this is work I've done for decades. "Bandwidth as twice sampling rate" is fine for detection of repeated events. It's not fine for reconstructing the signal, or finding one-off events. Digital sampling limits aren't the same as the bandwidth rolloff of an analog oscilloscope amplifier (or probe). Many 20th C. brand-name analog oscilloscopes gave (still give!) good performance well past their nameplate bandwidth. also: scope trigger circuits have to respond to events far above bandwidth! Triggering is very important. A "glitch" is a signal where a logic state changes, not because of a change in logic condition, but because of a timing event such as "propagation" or a "triggering delay". These are very short pulses, on the order of several nanoseconds in TTL circuits. Or, they may manifest as nanosecond delays in signals. Power supply lines carry glitches, caused by literal "spikes" in DC power to logic-switching circuits; large currents plus fast changes produce L di/dt or C dv/dt signals. These matter, because they can trigger logic events. If you can't see the trigger, you don't know "why" you have such events. Old TTL designs, sometimes generated glitches due to poor design. Aging components may cause glitches, by increasing capacitance, or sagging DC voltages. Dead or absent bypass caps, fail to filter out power-supply-line glitches. Look at schematics of the IMSAI front-panel. They actually CREATE "glitches" with a series capacitor to turn a logic-level change into a triggering pulse! Ugly ugly ugly design. All that said, any scope beats none. Looking at logic-level signal activity (any activity) is a go / no-go test for logic chips. Many faults can be diagnosed with a 20Mhz analog oscilloscope and probe. (Probe bandwidth matters. Not today's lecture.) But some faults are design problems that finally came due, or non-logic component failures, or "tired" chips. TTL logic responds to signals and events of several nanosecond duration - that's hundreds of megahertz "bandwidth" technically. and even a 20Mhz scope, will trigger on much faster events - that at least "detects" them. and you can guess if an analog "bump" is actually a pulse above your bandwidth. End of technical lecture. In my 20th century view, a $100ish old toolbox-sized HP or Tek 100Mhz analog oscilloscope, will show more information about pre 1990's computer chips, than most users can appreciate. But a $300 "100Mhz sampling" digital oscilloscope you can carry in your pocket, likely performs little better than a 30-40Mhz no-brand-I-remember analog oscilloscope that you leave behind at a hamfest. Will a 1Ghz modern DSO do much better? (shrug) I lack that experience, ask an owner, faster is better but.... And a DSO is conveniently tablet-sized, and likely does other things too. So the answer for you is: what's in your wallet? Buy BOTH if you can afford it, choose otherwise. Herb "race condition" Johnson retrotechnology.com -- Herbert R. Johnson, New Jersey in the USA http://www.retrotechnology.com OR .net preserve, recover, restore 1970's computing email: hjohnson AT retrotechnology DOT com or try later herbjohnson AT retrotechnology DOT info

1 0

Museum report.
by Evan Koblentz 20 Nov '17

20 Nov '17

Adam and I were at the museum from 10:30 this morning to 6ish tonight. John H. and his g/f Jen came for a tour along with Stefan who donated slide rules. (He estimates there are 1,000 in the collection!) We gave them the full tour until lunchtime. In the afternoon, Adam did warehouse work and I greeted museum visitors. After closing, we replaced many of the books in the museum workshop with different ones that are more interesting. Also did a good amount of cleaning and planning for some modest exhibit changes this winter/spring. Evan Koblentz Director, Vintage Computer Federation evan(a)vcfed.org / (646) 546-9999 www.vcfed.org @vcfederation

1 0

Open Hardware DSO - 100 MHz/250 Msa/sec - haasoscope
by J. Alexander Jacocks 20 Nov '17

20 Nov '17

There's a newish project on Crowdsuppy that I thought the group might find interesting: https://www.crowdsupply.com/andy-haas/haasoscope It's an open hardware/open source DSO, based on an Altera MAX10 FPGA. It seems to be much more capable than the DSO Nano, and the other mini scopes that I have seen. What does anyone think? - Alex

2 2

Multibus
by Mike Loewen 19 Nov '17

19 Nov '17

Is anyone here into Multibus systems? I have two Zendex ZX-905 prototyping cards, and an Electronic Solutions 8000D158AW extender card, free for the cost of shipping. Contact me off-list. Mike Loewen mloewen(a)cpumagic.scol.pa.us Old Technology http://q7.neurotica.com/Oldtech/

1 1

apple /// cobol
by william degnan 19 Nov '17

19 Nov '17

anyone use apple /// cobol, used it back when it was current? I have seen this and the Commodore Waterloo Cobol for the SuperPet. Never met anyone who actually used it. I used to use Cobol from a terminal connected to who knows what on the other end, or when I worked at IBM I ran Cobol jobs from JCL/TSO. But Apple /// ? Bill

1 0

Degnan Festivus Trade/for sale/wanted thread (festivus or otherwise)
by william degnan 17 Nov '17

17 Nov '17

---Available (sale / trade) --- I have a lot of extra Digital / DEC stuff available at the moment. VAX 4000-60 VAXStation 3100 RK06 Alignment Cartridge RK06 cartridges RL01 Cartridges DEC Rainbow parts UNIBUS Regulators (need minor repairs) XT era Keyboards (not IBM) IBM Terminal keyboards (with a RJ11 type jack) VT 100 with a bad flyback(?) (no keyboard) VT 220 with a logic issue (no keyboard) Non DEC-branded 11/44-era A/D type cards http://vintagecomputer.net/browse_thread.cfm?id=554 --- Wanted --- - DecMate I 8" OS/8 disk(s), and if anyone has a spare 37-25 pin RX02 cable. If you make me a disk copy I can return an RX01 or RX02 diskettes. - UNIBUS RK05 controller backplane cards (to acquire or borrow). I have a complete backplane and cards but I'd like some known-working cards to compare and contrast with, or maybe someone with these would like to pop over to help me out some time. - Bootable RL02 stand-alone RSX-11M compatible with an 11/40 that has no EIS card. It needs to be an older version. I need to keep trying but so far my attempts boot and then crash, I must be missing a step. I can trade for an RL02 disk or two for your trouble More items wanted (holy grails and others): http://vintagecomputer.net/browse_thread.cfm?id=3 Contact me privately if it's a business question/offer. I will bring items to festivus if there is interest. Not looking to make full ebay retail prices, only looking to get back what I put into them. Bill

1 0

Possible FS: Anyone Interested in GRiD PalmPad 2360s?
by systems_glitch 17 Nov '17

17 Nov '17

I've got: * Three handheld units, one with battery damage * Two chargers * One case I don't know too much about these, other than they're 286s and run off of a PCMCIA disk. I don't think I have the proper charging cables, it looks like the power supply uses a much different output than the PalmPad takes. If someone is super interested in these (I know GRiD stuff is often sought after) I'll bring them to Festivus and you can take a look. I could be easily persuaded to sell or trade one or all of these -- they seem like *very* cool little devices, but I've got enough other projects going on at the moment. Thanks, Jonathan

3 3

Festivus Shopping List
by Jeffrey Brace 17 Nov '17

17 Nov '17

Some have asked what they can bring. I've made a Festivus shopping list. Sign up of you want to bring something from the list: Go to the shopping list tab: https://docs.google.com/spreadsheets/d/1odmPxfBK9BrR-NDZIx4_c-V2lZpt_-vU-3f…

2 3