[cctalk] Re: Inline Serial Device?
> On Nov 13, 2022, at 9:09 AM, Paul Koning via cctalk > wrote: > > [EXTERNAL EMAIL] > > > >> On Nov 12, 2022, at 1:08 PM, Anders Nelson via cctalk >> wrote: >> >> I bet NN/AI would be helpful with data recovery - if we can model certain >> common failure modes with those old drive heads we could infer what the >> data should have been... > > NN maybe, I need to understand those better. I see they are now a building > block for OCR. > > AI, not so clear. In my view, AI is a catch-all term for "software whose > properties are unknown and probably unknowable". A computer, including one > that executes AI softwware, is a math processing engine, so in principle its > behavior is fully defined by its design and by the software in it. But when > you do AI in which "learning" is part of the scheme, the resulting behavior > is in fact unknown and undefined. > > For some applications that may be ok. OCR doesn't suffer materially from > occasional random errors, since it has errors anyway from the nature of its > input. But, for example, I shudder at the notion of AI in safety-critical > applications (like autopilots for aircraft, or worse yet for cars). A safety > critical application implemented in a manner that precludes the existence of > a specification is a fundanmentally insane notion. > > paul > Paul, not a fan of AI myself. But, I feel constrained to point out that the alterative to "AI in safety-critical applications” often is “a minimum-wage employee in a safety-critical application” which may or may not be an improvement. Agreed that AI is fundamentally not absolutely predictable - but neither are people. For problems complex enough to require either in a safety-critical decision-making loop, it may resolve down to a question of either 1) trusting the statistics (AI driving is maybe already *statistically* safer than human driving), 2) desiging the whole system in such a manner as to be tolerant of decision-making faults, or 3) Not doing the dangerous activity because it’s not monitorable. I would say our current road and automobile system doesn’t satisfy any of those criteria, FWIW. For problems simple enough to write closed-form, formally-verifiable software to handle, I *definitely* agree that is the way to go. - Mark
[cctalk] Re: Inline Serial Device?
> On Nov 13, 2022, at 2:31 PM, Cameron Kaiser via cctalk > wrote: > >> AI, not so clear. In my view, AI is a catch-all term for "software whose >> properties are unknown and probably unknowable". > > Someone recently on Hacker News talked about the possibility of neural net > models to translate code for other architectures. The best response to this > idea described it as a "turbo SIGILL generator." That's a good description. > The mental image of a CPU ramming into a silicon brick wall, reversing and > doing it again, over and over, possibly infinitely (the halting problem) comes > irresistibly to mind. I see all too many "inventions" that can be summarized as "use prior art solution X to well known problem Y by executing X in an AI (or "machine learning") system." Sigh. paul
[cctalk] Re: Inline Serial Device?
> AI, not so clear. In my view, AI is a catch-all term for "software whose > properties are unknown and probably unknowable". Someone recently on Hacker News talked about the possibility of neural net models to translate code for other architectures. The best response to this idea described it as a "turbo SIGILL generator." The mental image of a CPU ramming into a silicon brick wall, reversing and doing it again, over and over, possibly infinitely (the halting problem) comes irresistibly to mind. -- personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckai...@floodgap.com -- The earth is like a tiny grain of sand, only a lot heavier and bigger. -
[cctalk] Re: Inline Serial Device?
IMHO, AI is bull Martin -Original Message- From: Paul Koning via cctalk [mailto:cctalk@classiccmp.org] Sent: 13 November 2022 15:10 To: cctalk@classiccmp.org Cc: Paul Koning Subject: [cctalk] Re: Inline Serial Device? > On Nov 12, 2022, at 1:08 PM, Anders Nelson via cctalk > wrote: > > I bet NN/AI would be helpful with data recovery - if we can model > certain common failure modes with those old drive heads we could infer > what the data should have been... NN maybe, I need to understand those better. I see they are now a building block for OCR. AI, not so clear. In my view, AI is a catch-all term for "software whose properties are unknown and probably unknowable". A computer, including one that executes AI softwware, is a math processing engine, so in principle its behavior is fully defined by its design and by the software in it. But when you do AI in which "learning" is part of the scheme, the resulting behavior is in fact unknown and undefined. For some applications that may be ok. OCR doesn't suffer materially from occasional random errors, since it has errors anyway from the nature of its input. But, for example, I shudder at the notion of AI in safety-critical applications (like autopilots for aircraft, or worse yet for cars). A safety critical application implemented in a manner that precludes the existence of a specification is a fundanmentally insane notion. paul
[cctalk] Re: Inline Serial Device?
> On Nov 12, 2022, at 1:08 PM, Anders Nelson via cctalk > wrote: > > I bet NN/AI would be helpful with data recovery - if we can model certain > common failure modes with those old drive heads we could infer what the > data should have been... NN maybe, I need to understand those better. I see they are now a building block for OCR. AI, not so clear. In my view, AI is a catch-all term for "software whose properties are unknown and probably unknowable". A computer, including one that executes AI softwware, is a math processing engine, so in principle its behavior is fully defined by its design and by the software in it. But when you do AI in which "learning" is part of the scheme, the resulting behavior is in fact unknown and undefined. For some applications that may be ok. OCR doesn't suffer materially from occasional random errors, since it has errors anyway from the nature of its input. But, for example, I shudder at the notion of AI in safety-critical applications (like autopilots for aircraft, or worse yet for cars). A safety critical application implemented in a manner that precludes the existence of a specification is a fundanmentally insane notion. paul
[cctalk] Re: Inline Serial Device?
On Sun, Nov 13, 2022 at 9:54 AM Robin Downs via cctalk wrote: > > Hi, > Actually, I built exactly this many years ago (1990s) to operate a cash draw > for dumb terminals on Unix systems, used on counters as point of sale > devices... > > The existing solution used a processor, ram, rom, double sided board etc and > was too expensive, so I designed a replacement with a real UART and a finite > state machine consisting of a EPROM and 8 bit latch that simply monitored the > RS232 data passively and when the appropriate character sequence was matched, > it triggered the solenoid to open the cash draw. I am now thinking of totally crazy ways to detect a serial character. OK, a Model 33 Teletype with the right option in the stunt box is trivial. One odd idea is to detect the start bit and then generate the chracter bit-serially at the right baud rate. XOR that with the bitstream. Start with a flag ff set, at the middle of each bit-time, clear the flag if the bitstream and generated bit differ. At the end of the character time if the flag is still set, it's a match, Has the advantage of only needing a single-bit comparison not 7 or 8. > > It decoded a long 14 character code sequence easily and reliably and used 5 > chips in total on a smaller single sided board. > > Nowadays, a small microcontroller is the obvious way to go for cost and ease > of development. Cost, probably. Ease of development, it depends on who you are. I reckon I could solder up a suitable circuit using TTL only (i.e. not using a dumb UART which would simplify things a lot) in less time that it would take me to write the firmware. I am not a programmer. I tried the Arduino boards once and got fed up with a lack of proper printable documentation, no formal language specification, etc. -tony
[cctalk] Re: Inline Serial Device?
Hi, Actually, I built exactly this many years ago (1990s) to operate a cash draw for dumb terminals on Unix systems, used on counters as point of sale devices... The existing solution used a processor, ram, rom, double sided board etc and was too expensive, so I designed a replacement with a real UART and a finite state machine consisting of a EPROM and 8 bit latch that simply monitored the RS232 data passively and when the appropriate character sequence was matched, it triggered the solenoid to open the cash draw. It decoded a long 14 character code sequence easily and reliably and used 5 chips in total on a smaller single sided board. Nowadays, a small microcontroller is the obvious way to go for cost and ease of development. An easy way to implements is to use a small box to contain the two DB25 connectors and simply tap into the receive data line and run a short cable to the monitor circuit, probably built in and powered off what you want to operate... Regards, Robin Downs -Original Message- From: W2HX via cctalk Sent: 11 November 2022 21:16 To: General Discussion: On-Topic and Off-Topic Posts Cc: W2HX Subject: [cctalk] Inline Serial Device? Hello all, I am looking for a device that sits transparently in an RS-232 serial line and upon seeing a particular code go over the serial line ((or sequence of codes) will actual a relay (or a transistor). Something with two DB25s or DE9s and is configurable to what code will trigger the output? Some kind of box? Does anyone know of such a thing? I guess it could be cobbled up with a microcontroller, but hoping to just get something "off the shelf." Thank you 73 Eugene W2HX Subscribe to my Youtube Channel: https://www.youtube.com/@w2hx
[cctalk] Re: Inline Serial Device?
On Sat, 12 Nov 2022, Peter Corlett wrote: Farnell Nederland is quoting me ?1.06 (+21% VAT) for the cheapest brand of A quick search reveals that a single NE555 costs 0.25 Euros at Reichelt, *including* VAT. I'm sure they can be had much much cheaper in quantities. Christian
[cctalk] Re: Inline Serial Device?
On 11/12/22 17:23, Sellam Abraham via cctalk wrote: > I recommend the DEADBEEF dish. Are you certain that you don't mean EFBEADDE? On the STAR, DEADBEEF was the dead code for a call to page and un-pageable (i.e. kernel resident) page. There were other DEAD codes, such as DEADCACA, etc. All of which caused the system to throw up its hands and die. Looks cool in a dump with 64 bit words. --Chuck]
[cctalk] Re: Inline Serial Device?
On Sat, Nov 12, 2022 at 8:23 PM Sellam Abraham via cctalk wrote: > I recommend the DEADBEEF dish. FEED FACE DEAD BEEF -ethan
[cctalk] Re: Inline Serial Device?
On 11/12/22 15:41, Glen Slick via cctalk wrote: > I went to an Endian restaurant once. I was disappointed. I wanted > something little, but everything on the menu was big. In one, dessert comes first and finishes with salad.
[cctalk] Re: Inline Serial Device?
I recommend the DEADBEEF dish. Sellam On Sat, Nov 12, 2022 at 4:41 PM Warner Losh via cctalk < cctalk@classiccmp.org> wrote: > On Sat, Nov 12, 2022, 4:41 PM Glen Slick via cctalk > > wrote: > > > On Sat, Nov 12, 2022 at 12:36 PM Chuck Guzis via cctalk > > wrote: > > > > > > On 11/12/22 11:56, ben via cctalk wrote: > > > > On 2022-11-12 12:33 p.m., Chuck Guzis via cctalk wrote: > > > > > > > Watch out for Indians. :) > > > > Ben. > > > > > > Nonsense--they have some very fine restaurants here. > > > > > > --Chuck > > > > I went to an Endian restaurant once. I was disappointed. I wanted > > something little, but everything on the menu was big. > > > > I'm never quite sure which way to read the menu there, but the food is > good. > > Warner > > > >
[cctalk] Re: Inline Serial Device?
On Sat, Nov 12, 2022, 4:41 PM Glen Slick via cctalk wrote: > On Sat, Nov 12, 2022 at 12:36 PM Chuck Guzis via cctalk > wrote: > > > > On 11/12/22 11:56, ben via cctalk wrote: > > > On 2022-11-12 12:33 p.m., Chuck Guzis via cctalk wrote: > > > > > Watch out for Indians. :) > > > Ben. > > > > Nonsense--they have some very fine restaurants here. > > > > --Chuck > > I went to an Endian restaurant once. I was disappointed. I wanted > something little, but everything on the menu was big. > I'm never quite sure which way to read the menu there, but the food is good. Warner >
[cctalk] Re: Inline Serial Device?
On Fri, Nov 11, 2022 at 4:46 PM Bill Gunshannon via cctalk wrote: > If they do it probably costs about $500. Blackbox is not the company > they used to be. That sounds like exactly the kind of company they used to be :P I just wanted to mention that back when we did POS systems, we had cash drawers that were operated off of an RS232 port not by looking at the data stream but simply by triggering off the DTR line. The drawer contained basically a capacitor and a solenoid. So if you have control over the serial port on the "commanding" end, that sort of thing can be an option for triggering physical devices. G.
[cctalk] Re: Inline Serial Device?
On Sat, Nov 12, 2022 at 12:36 PM Chuck Guzis via cctalk wrote: > > On 11/12/22 11:56, ben via cctalk wrote: > > On 2022-11-12 12:33 p.m., Chuck Guzis via cctalk wrote: > > > Watch out for Indians. :) > > Ben. > > Nonsense--they have some very fine restaurants here. > > --Chuck I went to an Endian restaurant once. I was disappointed. I wanted something little, but everything on the menu was big.
[cctalk] Re: Inline Serial Device?
Nice one! LOL! m On Sat, Nov 12, 2022 at 3:36 PM Chuck Guzis via cctalk < cctalk@classiccmp.org> wrote: > On 11/12/22 11:56, ben via cctalk wrote: > > On 2022-11-12 12:33 p.m., Chuck Guzis via cctalk wrote: > > > Watch out for Indians. :) > > Ben. > > Nonsense--they have some very fine restaurants here. > > --Chuck > > >
[cctalk] Re: Inline Serial Device?
On 11/12/22 11:56, ben via cctalk wrote: > On 2022-11-12 12:33 p.m., Chuck Guzis via cctalk wrote: > Watch out for Indians. :) > Ben. Nonsense--they have some very fine restaurants here. --Chuck
[cctalk] Re: Inline Serial Device?
> Can the magic control codes pass through to the remote device or does the > "black box" have to eat them? According to a manual I found on the internet, it has three modes. One mode where the "arming" character and the subsequent port selection character are not passed through, one where it does pass through and one where there must be a (settable) delay between the arming and port selection characters. I looked at a picture of the main board of one of these black box COS units (code operated switch) and it appears there are two UARTS back to back. Between the UARTs where the data is parallel, there is an 8-bit comparator (74HCT688E) that checks the data flowing to an 8 position dip switch that lets you set the arming character. All sounds simple. But I must admit, if I were to build something similar, I wouldn’t bother with two back to back uarts, an MCU would certainly be the way to go. 73 Eugene W2HX Subscribe to my Youtube Channel: https://www.youtube.com/@w2hx -Original Message- From: Mike Stein via cctalk Sent: Saturday, November 12, 2022 3:09 PM To: General Discussion: On-Topic and Off-Topic Posts Cc: Mike Stein Subject: [cctalk] Re: Inline Serial Device? Can the magic control codes pass through to the remote device or does the "black box" have to eat them? Makes quite a difference... m On Fri, Nov 11, 2022 at 5:14 PM W2HX via cctalk wrote: > Thanks to everyone who responded. And thanks to Wayne for the black > box suggestion. I have actually been thinking about that very product. > What it does is switch a master port to one of several other ports. > Not far from what I'm seeking. I was thinking of buying one of these > and hacking it to trigger a relay instead of a change of port. But > before I went down that path, I thought I'd ask if anyone knew something more > fit for purpose. > Seems not and either this black box unit with a hack or a mcu might be > my options. > > Thanks all! > > > 73 Eugene W2HX > Subscribe to my Youtube Channel: https://www.youtube.com/@w2hx > > -Original Message- > From: Wayne S via cctalk > Sent: Friday, November 11, 2022 5:08 PM > To: General Discussion: On-Topic and Off-Topic Posts < > cctalk@classiccmp.org> > Cc: Wayne S > Subject: [cctalk] Re: Inline Serial Device? > > Something like this… > > > https://www.ebay.com/itm/Black-Box-Corporation-Modified-SWED98174-Cos- > II-Code-Operated-Serial-Switcher-/165759564735?mkcid=16&mkevt=1&_trksi > d=p2349624.m46890.l49286&mkrid=711-127632-2357-0 > > > Sent from my iPhone > > On Nov 11, 2022, at 14:00, Fred Cisin via cctalk > > wrote: > > On Fri, 11 Nov 2022, Chuck Guzis via cctalk wrote: > I keep a big jar around that's full of blue- and black pill MCU boards, > as well as a few of the more capable STM32F4 and F7 boards.Nowadays, > everything seems to look like a job for an MCU. > > > The MCU has replaced the hammer! > . . . "To a man with a HAMMER|(big jar of blue and black pill MCUs), > everything looks like a NAIL|(job for an MCU)" > > > >
[cctalk] Re: Inline Serial Device?
Can the magic control codes pass through to the remote device or does the "black box" have to eat them? Makes quite a difference... m On Fri, Nov 11, 2022 at 5:14 PM W2HX via cctalk wrote: > Thanks to everyone who responded. And thanks to Wayne for the black box > suggestion. I have actually been thinking about that very product. What it > does is switch a master port to one of several other ports. Not far from > what I'm seeking. I was thinking of buying one of these and hacking it to > trigger a relay instead of a change of port. But before I went down that > path, I thought I'd ask if anyone knew something more fit for purpose. > Seems not and either this black box unit with a hack or a mcu might be my > options. > > Thanks all! > > > 73 Eugene W2HX > Subscribe to my Youtube Channel: https://www.youtube.com/@w2hx > > -Original Message- > From: Wayne S via cctalk > Sent: Friday, November 11, 2022 5:08 PM > To: General Discussion: On-Topic and Off-Topic Posts < > cctalk@classiccmp.org> > Cc: Wayne S > Subject: [cctalk] Re: Inline Serial Device? > > Something like this… > > > https://www.ebay.com/itm/Black-Box-Corporation-Modified-SWED98174-Cos-II-Code-Operated-Serial-Switcher-/165759564735?mkcid=16&mkevt=1&_trksid=p2349624.m46890.l49286&mkrid=711-127632-2357-0 > > > Sent from my iPhone > > On Nov 11, 2022, at 14:00, Fred Cisin via cctalk > wrote: > > On Fri, 11 Nov 2022, Chuck Guzis via cctalk wrote: > I keep a big jar around that's full of blue- and black pill MCU boards, > as well as a few of the more capable STM32F4 and F7 boards.Nowadays, > everything seems to look like a job for an MCU. > > > The MCU has replaced the hammer! > . . . "To a man with a HAMMER|(big jar of blue and black pill MCUs), > everything looks like a NAIL|(job for an MCU)" > > > >
[cctalk] Re: Inline Serial Device?
On 2022-11-12 12:33 p.m., Chuck Guzis via cctalk wrote: On 11/12/22 10:59, ben via cctalk wrote: I do too, but don't welcome the market droids that make you upgrade all the time. I am putting a projection TV in the living room, and needed to use all this HDMI or wire less crap. What ever happened to 75 video and 150? ohm terminated audio. What's all this 75 ohm stuff? 300 ohm twinlead. And the separate UHF down-converter to feed your VHF teevee. I'm still trying to figure out how my TV broadcast preview never displays any of the commercial advertisements, but I have to watch them when they're fullscreen. It's a conspiracy, I tell you... Now, if you'll pardon me, I have to go to the telegraph office to send a TWX. --Chuck Watch out for Indians. :) Ben.
[cctalk] Re: Inline Serial Device?
A surprisingly complex project ;-) For another alternative, the ever prolific Geoff Graham and co. have developed an amazingly versatile BASIC interpreter for the Pi Pico: https://geoffg.net/picomite.html m On Fri, Nov 11, 2022 at 9:06 PM Cameron Kaiser via cctalk < cctalk@classiccmp.org> wrote: > >>> I am looking for a device that sits transparently in an RS-232 serial > line > >>> and upon seeing a particular code go over the serial line ((or > sequence of > >>> codes) will actual a relay (or a transistor). Something with two DB25s > or > >>> DE9s and is configurable to what code will trigger the output? Some > kind of > >>> box? > >> > >> not that it's easy but a raspberry pi could be set up to watch the > serial > >> line. > > > > Or even cheaper, and Arduino uno > > I second the Arduino recommendation. I have a Power Mac G4 with a serial > dongle > that drives an Arduino Nano-based IR blaster. It sends serial commands to > it > and the blaster transmits a signal to the room air conditioner. Should be > easy > to adapt the GPIO pins to a relay. Arduino programming and interfacing is > pretty straightforward. > > > https://oldvcr.blogspot.com/2022/10/ir-controlling-new-air-conditioner-in.html > > -- > personal: > http://www.cameronkaiser.com/ -- > Cameron Kaiser * Floodgap Systems * www.floodgap.com * > ckai...@floodgap.com > -- The more corrupt the state, the more numerous the laws. -- Tacitus > - > >
[cctalk] Re: Inline Serial Device?
On 11/12/22 10:59, ben via cctalk wrote: > I do too, but don't welcome the market droids that make you upgrade all > the time. > I am putting a projection TV in the living room, and needed to use all > this HDMI or wire less crap. What ever happened to 75 video and 150? ohm > terminated audio. What's all this 75 ohm stuff? 300 ohm twinlead. And the separate UHF down-converter to feed your VHF teevee. I'm still trying to figure out how my TV broadcast preview never displays any of the commercial advertisements, but I have to watch them when they're fullscreen. It's a conspiracy, I tell you... Now, if you'll pardon me, I have to go to the telegraph office to send a TWX. --Chuck
[cctalk] Re: Inline Serial Device?
On 2022-11-12 11:05 a.m., Chuck Guzis via cctalk wrote: On 11/12/22 09:41, Ethan Dicks via cctalk wrote: Yeah, the 555 is extremely simple and is well known and is fairly cheap, simple MCUs are simple (and cheap) even if they aren't 100% deterministic like a chip with 20-30 transistors. There's economic advantage in flexibility. There's also efficiency in mass-produced numbers. There are several Chinese MCUs that go for less than USD$0.10 in low quantities. I think the bottom end a couple of years ago was about USD$0.03. At that price point, you have to wonder if some of that isn't the packaging (tape reel) cost. There are far more MCUs made today than 555s, if that's any indication. And some of the newest ones feature neural net hardware (e.g. MAX78000). Do that with your 555s! I, for one, welcome our robot overlords. --Chuck I do too, but don't welcome the market droids that make you upgrade all the time. I am putting a projection TV in the living room, and needed to use all this HDMI or wire less crap. What ever happened to 75 video and 150? ohm terminated audio. Ben. PS At least my new computer still has a COM port.
[cctalk] Re: Inline Serial Device?
On 2022-11-12 9:08 a.m., Tony Duell via cctalk wrote: On Sat, Nov 12, 2022 at 11:59 AM Antonio Carlini via cctalk wrote: Surely a microcontroller is just a 555 with a few extra transistors? For suitable (large) values of 'few'? Actually I can think of many differences... Firstly, the full equivalent circuit of the 555 is in the datasheet. So I can predict how it should behave under all conditions (there are many things you can do with a 555 besides astables and monostables). I have never seen an equivalent circuit, or a gate level description of a microcontroller. All 555s are the same. If it fails I can replace it. Microcontrollers cease to be the same once they are prgrammed. If a microcontroller fails then I'm stuck. I won't be able to get the firmware I would argue that 555s are a lot more reliable than microcontrollers. And have a much longer life than the time to bitrot of most microcontroller flash memories It's a lot easier to test a 555 than it is to test a microcontroller. 555s do not have illegal internal states they can get into. Microcontrollers almost always do. Hence the need for watchdog timers which IMHO are a kludge, Another tool in the box, just that it happens to be very cheap. Cheap != good They have their uses. But like many tools they can be misused and often are. -tony They often now have huge development software now days, that may tie you to a specific computer platform. Pal's I am using for example, 16v8's require win-cupl and can't be adapted for original FORTRAN V software. Ben.
[cctalk] Re: Inline Serial Device?
I bet NN/AI would be helpful with data recovery - if we can model certain common failure modes with those old drive heads we could infer what the data should have been... -- Anders Nelson On Sat, Nov 12, 2022 at 1:05 PM Chuck Guzis via cctalk < cctalk@classiccmp.org> wrote: > On 11/12/22 09:41, Ethan Dicks via cctalk wrote: > > > > > Yeah, the 555 is extremely simple and is well known and is fairly > > cheap, simple MCUs are simple (and cheap) even if they aren't 100% > > deterministic like a chip with 20-30 transistors. There's economic > > advantage in flexibility. > > There's also efficiency in mass-produced numbers. There are several > Chinese MCUs that go for less than USD$0.10 in low quantities. I think > the bottom end a couple of years ago was about USD$0.03. At that price > point, you have to wonder if some of that isn't the packaging (tape > reel) cost. > > There are far more MCUs made today than 555s, if that's any indication. > > And some of the newest ones feature neural net hardware (e.g. MAX78000). > Do that with your 555s! > > I, for one, welcome our robot overlords. > > --Chuck > > >
[cctalk] Re: Inline Serial Device?
On 11/12/22 09:41, Ethan Dicks via cctalk wrote: > > Yeah, the 555 is extremely simple and is well known and is fairly > cheap, simple MCUs are simple (and cheap) even if they aren't 100% > deterministic like a chip with 20-30 transistors. There's economic > advantage in flexibility. There's also efficiency in mass-produced numbers. There are several Chinese MCUs that go for less than USD$0.10 in low quantities. I think the bottom end a couple of years ago was about USD$0.03. At that price point, you have to wonder if some of that isn't the packaging (tape reel) cost. There are far more MCUs made today than 555s, if that's any indication. And some of the newest ones feature neural net hardware (e.g. MAX78000). Do that with your 555s! I, for one, welcome our robot overlords. --Chuck
[cctalk] Re: Inline Serial Device?
On Sat, Nov 12, 2022 at 12:18 PM Chuck Guzis via cctalk wrote: > On 11/12/22 02:28, Tony Duell via cctalk wrote: > > ... This is the sort of > > thing I'd do with a couple of transistors or an NE555 depending on > > which turned up in the junk box first. > > One thing that a small MCU has over a 555 is that it can be programmed > once and you can be assured of its frequency stability. No fooling with > pots and caps to get the thing to work the way you'd like. Yes, that, plus since many products have a hardware team and a firmware/software team, the hardware can be designed to general requirements and sent out for manufacture while the software team has time to write the firmware (and make changes long after the hardware is set). One of the first times I encountered this was stripping some old emergency exit lights for parts c. 2008. The switching supply had an 8-pin PIC for the oscillator instead of a 555. Yes, a 555 could have done it, but the PIC didn't need any external components to set the frequency, components that can drift with age, and components that take up board space. Even if the 555 and MCU were identical in cost for the IC, the MCU was cheaper because of the smaller footprint. Additionally, the designers had some flexibility. To change the frequency with a 555 after manufacture is an expensive proposition. With an MCU, if it's flashable in-circuit (clip or possibly programming pads near/at the MCU), then one can change the behavior without melting any metal or purchasing components. While one may never need to change the frequency of a SMPSU oscillator after initial design, there are plenty of products where it's handy that the hardware guys can say "here's an output that can go from 1/10Hz to 20Khz - what do you want it to be?" and not worry about design limitations, just set the frequency in the firmware and it does that. You can build some generic hardware (X inputs, Y outputs with Z mA current drive) and fine tune things later, or have variations on what the inputs mean and not have to change the PCB. Yeah, the 555 is extremely simple and is well known and is fairly cheap, simple MCUs are simple (and cheap) even if they aren't 100% deterministic like a chip with 20-30 transistors. There's economic advantage in flexibility. -ethan
[cctalk] Re: Inline Serial Device?
On Sat, Nov 12, 2022 at 5:18 PM Chuck Guzis via cctalk wrote: > MCUs make a lot of very complex stuff simple. A tape controller that > would have required a couple of large PCBs now can be put on a > filing-card size pcb--and most of that is connectors. You've hit a raw nerve there. I've recently being doing battle with an old-ish tape controller board that has a microcontroller with internal ROM and 3 ASICs amongst other things on it. I'd much prefer a cardcage of boards containing simple components. I do not understand this desire to miniaturise everything. > Almost anything > electronic sold today has an MCU in it--even a lithium cell. And that is why I hate modern electronics and buy very little of it. > > One thing that a small MCU has over a 555 is that it can be programmed > once and you can be assured of its frequency stability. No fooling with > pots and caps to get the thing to work the way you'd like. Now wait a second. I've not come across a simple microcontroller with a crystal in the same package. If you're going to use an external crystal, then I can do that too, with a couple of divider ICs. If you use the internal clock option of the microcontroller, it can drift. If I use R's and C's on a 555 I can choose ones with the stability, temperature coefficient, etc that I need. > > I'm certain you'd be tickled to see your beloved HP 9800 series box > re-imagined in TO5 germanium point-contact transistors and relays. Isn't that called an HP9100.Much the same functionality as an HP9810, but discrete transistors. Some of them are germanium (albeit junction ones). And yes I do love it. -tony >
[cctalk] Re: Inline Serial Device?
I can probably build firmware to this for you and assemble a list of OTS parts you can connect together in a day or two - what string do you want to trigger on? Parameters, 8N1? RTS/CTS? Voltage at RS-232 or TTL level? OTS parts cost maybe $30? Source: am firmware engineer at Peloton (and still employed, wheee). -- Anders Nelson On Sat, Nov 12, 2022 at 12:27 PM Peter Corlett via cctalk < cctalk@classiccmp.org> wrote: > On Sat, Nov 12, 2022 at 10:28:09AM +, Tony Duell via cctalk wrote: > [...] > > The other day I saw a product with a flashing LED, the flash rate was set > > with a knob. Yes, a microcontroller with a pot connected to an analogue > > input and LED hung off an output port. This is the sort of thing I'd do > > with a couple of transistors or an NE555 depending on which turned up in > > the junk box first. > > Farnell Nederland is quoting me €1.06 (+21% VAT) for the cheapest brand of > 555 in stock. Their search won't let me find the cheapest microcontroller > without drilling down further, but an 8 pin AVR is €0.88. That's single > item > quantities in DIP packaging, as is typical for small home projects. The 555 > will also need a capacitor for its RC timer circuit which is another few > tens of cents. And that's why people use microcontrollers to blink LEDs. > > The MCU in the Pi Pico is also well under a euro if you buy a reel of 3,400 > of them. That's probably a few too many for an average hobbyist :) > >
[cctalk] Re: Inline Serial Device?
> Farnell Nederland is quoting me €1.06 (+21% VAT) for the cheapest brand of > 555 in stock. Their search won't let me find the cheapest microcontroller > without drilling down further, but an 8 pin AVR is €0.88. Just checked RS components which is the supplier I normally use. Assuming I want a through-hole DIP device (not surface mount), I can get a TI NE555 for 28.4p if I buy 50 at a time. That is not a large quantity even for a hobbyist. -tony
[cctalk] Re: Inline Serial Device?
On Sat, Nov 12, 2022 at 10:28:09AM +, Tony Duell via cctalk wrote: [...] > The other day I saw a product with a flashing LED, the flash rate was set > with a knob. Yes, a microcontroller with a pot connected to an analogue > input and LED hung off an output port. This is the sort of thing I'd do > with a couple of transistors or an NE555 depending on which turned up in > the junk box first. Farnell Nederland is quoting me €1.06 (+21% VAT) for the cheapest brand of 555 in stock. Their search won't let me find the cheapest microcontroller without drilling down further, but an 8 pin AVR is €0.88. That's single item quantities in DIP packaging, as is typical for small home projects. The 555 will also need a capacitor for its RC timer circuit which is another few tens of cents. And that's why people use microcontrollers to blink LEDs. The MCU in the Pi Pico is also well under a euro if you buy a reel of 3,400 of them. That's probably a few too many for an average hobbyist :)
[cctalk] Re: Inline Serial Device?
On 11/12/22 02:28, Tony Duell via cctalk wrote: > The other day I saw a product with a flashing LED, the flash rate was > set with a knob. Yes, a microcontroller with a pot connected to an > analogue input and LED hung off an output port. This is the sort of > thing I'd do with a couple of transistors or an NE555 depending on > which turned up in the junk box first. It's inevitable evolution--why fight it? Without that, we'd all be nematodes or bacteria. MCUs make a lot of very complex stuff simple. A tape controller that would have required a couple of large PCBs now can be put on a filing-card size pcb--and most of that is connectors. Almost anything electronic sold today has an MCU in it--even a lithium cell. One thing that a small MCU has over a 555 is that it can be programmed once and you can be assured of its frequency stability. No fooling with pots and caps to get the thing to work the way you'd like. Signal processing is fairly easy when a commodity MCU has a fast-enough ADC and lots of memory? You can write your own realtime FFT software for it with no problem. Why write a CRC calculation routine when there's dedicated hardware to do that? If anything, a modern medium-scale MCU can be so packed with peripherals and features that the reference manual outlining them can run to a couple thousand pages. Productions yields have drastically improved since Hans Kamenzind spun the 555. I'm sure that he would approve of MCUs, were he still alive. I'm certain you'd be tickled to see your beloved HP 9800 series box re-imagined in TO5 germanium point-contact transistors and relays. --Chuck
[cctalk] Re: Inline Serial Device?
On 2022-11-12 11:08, Tony Duell via cctalk wrote: On Sat, Nov 12, 2022 at 11:59 AM Antonio Carlini via cctalk wrote: Surely a microcontroller is just a 555 with a few extra transistors? For suitable (large) values of 'few'? Actually I can think of many differences... Firstly, the full equivalent circuit of the 555 is in the datasheet. So I can predict how it should behave under all conditions (there are many things you can do with a 555 besides astables and monostables). I have never seen an equivalent circuit, or a gate level description of a microcontroller. And if you like to "tune" your ne555, you can do it on your own: https://www.adafruit.com/product/1526 :)
[cctalk] Re: Inline Serial Device?
On Sat, Nov 12, 2022 at 11:59 AM Antonio Carlini via cctalk wrote: > > Surely a microcontroller is just a 555 with a few extra transistors? For suitable (large) values of 'few'? Actually I can think of many differences... Firstly, the full equivalent circuit of the 555 is in the datasheet. So I can predict how it should behave under all conditions (there are many things you can do with a 555 besides astables and monostables). I have never seen an equivalent circuit, or a gate level description of a microcontroller. All 555s are the same. If it fails I can replace it. Microcontrollers cease to be the same once they are prgrammed. If a microcontroller fails then I'm stuck. I won't be able to get the firmware I would argue that 555s are a lot more reliable than microcontrollers. And have a much longer life than the time to bitrot of most microcontroller flash memories It's a lot easier to test a 555 than it is to test a microcontroller. 555s do not have illegal internal states they can get into. Microcontrollers almost always do. Hence the need for watchdog timers which IMHO are a kludge, > Another tool in the box, just that it happens to be very cheap. Cheap != good They have their uses. But like many tools they can be misused and often are. -tony
[cctalk] Re: Inline Serial Device?
On 12/11/2022 10:28, Tony Duell via cctalk wrote: I thought it was well-known that nothing can be designed without at least one microcontroller. The other day I saw a product with a flashing LED, the flash rate was set with a knob. Yes, a microcontroller with a pot connected to an analogue input and LED hung off an output port. This is the sort of thing I'd do with a couple of transistors or an NE555 depending on which turned up in the junk box first. $deity I hate modern electronics. Surely a microcontroller is just a 555 with a few extra transistors? Another tool in the box, just that it happens to be very cheap. You should check out Usagi Electric on youtube: https://www.youtube.com/c/Nakazoto/videos, he's putting together a valve-based recreation of 1-bit processor (the MC14500B). He makes his own PCBs too :-) Antonio -- Antonio Carlini anto...@acarlini.com
[cctalk] Re: Inline Serial Device?
On Fri, Nov 11, 2022 at 10:00 PM Fred Cisin via cctalk wrote: > The MCU has replaced the hammer! > . . . "To a man with a HAMMER|(big jar of blue and black pill MCUs), > everything looks like a NAIL|(job for an MCU)" I thought it was well-known that nothing can be designed without at least one microcontroller. The other day I saw a product with a flashing LED, the flash rate was set with a knob. Yes, a microcontroller with a pot connected to an analogue input and LED hung off an output port. This is the sort of thing I'd do with a couple of transistors or an NE555 depending on which turned up in the junk box first. $deity I hate modern electronics. -tony
[cctalk] Re: Inline Serial Device?
On Fri, Nov 11, 2022 at 9:24 PM Bill Degnan via cctalk wrote: > > not that it's easy but a raspberry pi could be set up to watch the serial > line. So you're suggesting that it takes more components to detect what character my computer has sent than there are in the rest of the computer? -tony
[cctalk] Re: Inline Serial Device?
>>> I am looking for a device that sits transparently in an RS-232 serial line >>> and upon seeing a particular code go over the serial line ((or sequence of >>> codes) will actual a relay (or a transistor). Something with two DB25s or >>> DE9s and is configurable to what code will trigger the output? Some kind of >>> box? >> >> not that it's easy but a raspberry pi could be set up to watch the serial >> line. > > Or even cheaper, and Arduino uno I second the Arduino recommendation. I have a Power Mac G4 with a serial dongle that drives an Arduino Nano-based IR blaster. It sends serial commands to it and the blaster transmits a signal to the room air conditioner. Should be easy to adapt the GPIO pins to a relay. Arduino programming and interfacing is pretty straightforward. https://oldvcr.blogspot.com/2022/10/ir-controlling-new-air-conditioner-in.html -- personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckai...@floodgap.com -- The more corrupt the state, the more numerous the laws. -- Tacitus -
[cctalk] Re: Inline Serial Device?
On 11/11/22 17:07, Wayne S via cctalk wrote: Something like this… https://www.ebay.com/itm/Black-Box-Corporation-Modified-SWED98174-Cos-II-Code-Operated-Serial-Switcher-/165759564735?mkcid=16&mkevt=1&_trksid=p2349624.m46890.l49286&mkrid=711-127632-2357-0 Looks like some kind of serial port switcher. Not really what W2HX was looking for. bill
[cctalk] Re: Inline Serial Device?
On 11/11/22 16:52, Wayne S via cctalk wrote: Black box may have something. They had something similar where you could call in via phone and activate a switch to reboot a computer. If they do it probably costs about $500. Blackbox is not the company they used to be. (From someone who still has a lot of things floating around here from them like short haul modems!!) bill
[cctalk] Re: Inline Serial Device?
On 11/11/22 16:24, Bill Degnan via cctalk wrote: not that it's easy but a raspberry pi could be set up to watch the serial line. Probably overkill. I think it could be done with an arduino. bill
[cctalk] Re: Inline Serial Device?
Thanks to everyone who responded. And thanks to Wayne for the black box suggestion. I have actually been thinking about that very product. What it does is switch a master port to one of several other ports. Not far from what I'm seeking. I was thinking of buying one of these and hacking it to trigger a relay instead of a change of port. But before I went down that path, I thought I'd ask if anyone knew something more fit for purpose. Seems not and either this black box unit with a hack or a mcu might be my options. Thanks all! 73 Eugene W2HX Subscribe to my Youtube Channel: https://www.youtube.com/@w2hx -Original Message- From: Wayne S via cctalk Sent: Friday, November 11, 2022 5:08 PM To: General Discussion: On-Topic and Off-Topic Posts Cc: Wayne S Subject: [cctalk] Re: Inline Serial Device? Something like this… https://www.ebay.com/itm/Black-Box-Corporation-Modified-SWED98174-Cos-II-Code-Operated-Serial-Switcher-/165759564735?mkcid=16&mkevt=1&_trksid=p2349624.m46890.l49286&mkrid=711-127632-2357-0 Sent from my iPhone On Nov 11, 2022, at 14:00, Fred Cisin via cctalk wrote: On Fri, 11 Nov 2022, Chuck Guzis via cctalk wrote: I keep a big jar around that's full of blue- and black pill MCU boards, as well as a few of the more capable STM32F4 and F7 boards.Nowadays, everything seems to look like a job for an MCU. The MCU has replaced the hammer! . . . "To a man with a HAMMER|(big jar of blue and black pill MCUs), everything looks like a NAIL|(job for an MCU)"
[cctalk] Re: Inline Serial Device?
On 11/11/22 14:00, Fred Cisin via cctalk wrote: > On Fri, 11 Nov 2022, Chuck Guzis via cctalk wrote: >> I keep a big jar around that's full of blue- and black pill MCU boards, >> as well as a few of the more capable STM32F4 and F7 boards. Nowadays, >> everything seems to look like a job for an MCU. > > > The MCU has replaced the hammer! > . . . "To a man with a HAMMER|(big jar of blue and black pill MCUs), > everything looks like a NAIL|(job for an MCU)" Probably true--there are commodity light bulbs and flashlights with MCUs in them. The price of the Chinese OTP ones has dropped to a few cents--less than a package of chewing gum. "Silicon is cheap" is the motto of our times. --Chuck
[cctalk] Re: Inline Serial Device?
Something like this… https://www.ebay.com/itm/Black-Box-Corporation-Modified-SWED98174-Cos-II-Code-Operated-Serial-Switcher-/165759564735?mkcid=16&mkevt=1&_trksid=p2349624.m46890.l49286&mkrid=711-127632-2357-0 Sent from my iPhone On Nov 11, 2022, at 14:00, Fred Cisin via cctalk wrote: On Fri, 11 Nov 2022, Chuck Guzis via cctalk wrote: I keep a big jar around that's full of blue- and black pill MCU boards, as well as a few of the more capable STM32F4 and F7 boards.Nowadays, everything seems to look like a job for an MCU. The MCU has replaced the hammer! . . . "To a man with a HAMMER|(big jar of blue and black pill MCUs), everything looks like a NAIL|(job for an MCU)"
[cctalk] Re: Inline Serial Device?
On Fri, 11 Nov 2022, Chuck Guzis via cctalk wrote: I keep a big jar around that's full of blue- and black pill MCU boards, as well as a few of the more capable STM32F4 and F7 boards.Nowadays, everything seems to look like a job for an MCU. The MCU has replaced the hammer! . . . "To a man with a HAMMER|(big jar of blue and black pill MCUs), everything looks like a NAIL|(job for an MCU)"
[cctalk] Re: Inline Serial Device?
Black box may have something. They had something similar where you could call in via phone and activate a switch to reboot a computer. Sent from my iPhone > On Nov 11, 2022, at 13:40, Fred Cisin via cctalk > wrote: > > Or even cheaper, a dumpstered/e-waste PC > >> On Fri, 11 Nov 2022, Nigel Johnson Ham via cctalk wrote: >> >> Or even cheaper, and Arduino uno >> >> cheers, >> >> Nigel >> >> >> Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU >> Amateur Radio, the origin of the open-source concept! >> Skype: TILBURY2591 >> >> >>> On 2022-11-11 16:24, Bill Degnan via cctalk wrote: >>> not that it's easy but a raspberry pi could be set up to watch the serial >>> line. >>> On Fri, Nov 11, 2022 at 4:16 PM W2HX via cctalk >>> wrote: Hello all, I am looking for a device that sits transparently in an RS-232 serial line and upon seeing a particular code go over the serial line ((or sequence of codes) will actual a relay (or a transistor). Something with two DB25s or DE9s and is configurable to what code will trigger the output? Some kind of box? Does anyone know of such a thing? I guess it could be cobbled up with a microcontroller, but hoping to just get something "off the shelf." Thank you 73 Eugene W2HX Subscribe to my Youtube Channel:https://www.youtube.com/@w2hx
[cctalk] Re: Inline Serial Device?
I suspect that a cheap minimal MCU could not only do the job, but do it by scavenging power from the RS232 signal lines (like old serial mice did). It's really surprising how miserly these things can be with power requirements nowadays. I keep a big jar around that's full of blue- and black pill MCU boards, as well as a few of the more capable STM32F4 and F7 boards.Nowadays, everything seems to look like a job for an MCU. I recall that a couple of years ago, either ED or EDN had an open question about whether it was preferable to replace NE555s with small MCUs--the parts price between the two being negligible. --Chuck
[cctalk] Re: Inline Serial Device?
Or even cheaper, a dumpstered/e-waste PC On Fri, 11 Nov 2022, Nigel Johnson Ham via cctalk wrote: Or even cheaper, and Arduino uno cheers, Nigel Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU Amateur Radio, the origin of the open-source concept! Skype: TILBURY2591 On 2022-11-11 16:24, Bill Degnan via cctalk wrote: not that it's easy but a raspberry pi could be set up to watch the serial line. On Fri, Nov 11, 2022 at 4:16 PM W2HX via cctalk wrote: Hello all, I am looking for a device that sits transparently in an RS-232 serial line and upon seeing a particular code go over the serial line ((or sequence of codes) will actual a relay (or a transistor). Something with two DB25s or DE9s and is configurable to what code will trigger the output? Some kind of box? Does anyone know of such a thing? I guess it could be cobbled up with a microcontroller, but hoping to just get something "off the shelf." Thank you 73 Eugene W2HX Subscribe to my Youtube Channel:https://www.youtube.com/@w2hx
[cctalk] Re: Inline Serial Device?
Perhaps cheaper still, a Raspberry Pico. Those go for $4, which even for the smaller Arduinos is hard to beat. paul > On Nov 11, 2022, at 4:29 PM, Nigel Johnson Ham via cctalk > wrote: > > Or even cheaper, and Arduino uno > > cheers, > > Nigel > > > Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU > Amateur Radio, the origin of the open-source concept! > Skype: TILBURY2591 > > > On 2022-11-11 16:24, Bill Degnan via cctalk wrote: >> not that it's easy but a raspberry pi could be set up to watch the serial >> line. >> >> On Fri, Nov 11, 2022 at 4:16 PM W2HX via cctalk >> wrote: >> >>> Hello all, >>> >>> I am looking for a device that sits transparently in an RS-232 serial line >>> and upon seeing a particular code go over the serial line ((or sequence of >>> codes) will actual a relay (or a transistor). Something with two DB25s or >>> DE9s and is configurable to what code will trigger the output? Some kind of >>> box? >>> >>> Does anyone know of such a thing? I guess it could be cobbled up with a >>> microcontroller, but hoping to just get something "off the shelf." >>> Thank you >>> >>> 73 Eugene W2HX >>> Subscribe to my Youtube Channel:https://www.youtube.com/@w2hx >>> >>> >>> >>>
[cctalk] Re: Inline Serial Device?
Something quite similar, . . . (probably a hassle to switch over) I had a KVM switch, that permitted two computers to share one keyboard mouse, monitor, and speakers. DE15, two USB ports and a headphone jack (IOGear GCS632U?) It apparently watched for a hotkey from the keyboard to activate switching. I replaced it with one that had a physical pushbutton for switching (would have preferred a DT switch, with visible positions, so that you know which one is selected, not bouncing back and forth until it's what you want) On Fri, 11 Nov 2022, W2HX via cctalk wrote: Hello all, I am looking for a device that sits transparently in an RS-232 serial line and upon seeing a particular code go over the serial line ((or sequence of codes) will actual a relay (or a transistor). Something with two DB25s or DE9s and is configurable to what code will trigger the output? Some kind of box? Does anyone know of such a thing? I guess it could be cobbled up with a microcontroller, but hoping to just get something "off the shelf." Thank you 73 Eugene W2HX Subscribe to my Youtube Channel: https://www.youtube.com/@w2hx
[cctalk] Re: Inline Serial Device?
Or even cheaper, and Arduino uno cheers, Nigel Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU Amateur Radio, the origin of the open-source concept! Skype: TILBURY2591 On 2022-11-11 16:24, Bill Degnan via cctalk wrote: not that it's easy but a raspberry pi could be set up to watch the serial line. On Fri, Nov 11, 2022 at 4:16 PM W2HX via cctalk wrote: Hello all, I am looking for a device that sits transparently in an RS-232 serial line and upon seeing a particular code go over the serial line ((or sequence of codes) will actual a relay (or a transistor). Something with two DB25s or DE9s and is configurable to what code will trigger the output? Some kind of box? Does anyone know of such a thing? I guess it could be cobbled up with a microcontroller, but hoping to just get something "off the shelf." Thank you 73 Eugene W2HX Subscribe to my Youtube Channel:https://www.youtube.com/@w2hx
[cctalk] Re: Inline Serial Device?
not that it's easy but a raspberry pi could be set up to watch the serial line. On Fri, Nov 11, 2022 at 4:16 PM W2HX via cctalk wrote: > Hello all, > > I am looking for a device that sits transparently in an RS-232 serial line > and upon seeing a particular code go over the serial line ((or sequence of > codes) will actual a relay (or a transistor). Something with two DB25s or > DE9s and is configurable to what code will trigger the output? Some kind of > box? > > Does anyone know of such a thing? I guess it could be cobbled up with a > microcontroller, but hoping to just get something "off the shelf." > Thank you > > 73 Eugene W2HX > Subscribe to my Youtube Channel: https://www.youtube.com/@w2hx > > > > >
