Some machines had both sorts of port -- my Zenith Z90 has 3 serial ports , 2 of one flavour, one of the other. I can't remember which way round, but they are clearly labelled on the cxse. In theory a male connector should be used for a DTE, a female conenctor for a DCE. But don't bet on it. One of the worst 'serial sex offenders' (as I tend to call manufacturers who get this wrong) that I've come across was HP who routinely fitted DB25 sockets wired as DTEs. I find one of those little plug-to-socket adpaters with LEDs monitoring the important lines ot be very useful here If you plug it into an unknown port and connect nothing else to it, then at least one LED is likely to come on [1[ Be seeing which line(s) are driven you can make a good guess as to whether it's a DTE or DCE [1] OK, this fails if the serial port is powered from the handshake lines. Or if it expects to see an active line before powering up the output buffers. Such cases are rare but not unheard-of The other way is to find the RS232 buffers inside the device (for classic hardware these are likely to be separate ICs) and trace a few pins of the conencotr to them. If you find that pin 2 is connected to the output of a driver and pin 3 is conencted to the input of a receiver it's a good bet the prot is wired as DTE. As for the jumpers, the time I came acorss this was the HP82164 HPIL-RS232 interfce. There's a jumper plug you can fit either way round, one way gives DTE wiring, the other gives DCE -- in theory. Actually, it messes about with soem of the handshake lines, jumpering them together, etc. I have found the only sane thign to do is to set the jumper to DTE and leave it alone, Then wire up external adapters as needed. One of the few machiens to make this easy was the Tatung Einstein. It used a quincuncial 5 pin DIN plug, like the BBC micro, for the RS232 port. Unlike the Beeb, it was wired sensibly. If you plug it in upside odwn it swaps TxD with RxD and RTS with CTS. Just what you want. -tony

It is a normal-siced circular DIN connnector, the same size as normal audio DIN connectors. It has 5 pins, arraged as a square with one i nthe middle -- a quincunx, hence my name for it. I have also seen it called a '5 pin DIN plug Form C', a '270 degree 5 pin DIN plug' and a '360 degree 5 pin DIN plug' The other 2 5 pin DIN plugs, i nthe same size hosuing are : 'Form A' or '180 Degree' -- the one used for nost audio connections, IBM PC (5150, etc) keyboard ports, all DIN sockets on a TRS-80 Model 1, etc. This is the one the 3 pin plug will fit, and the one that will go into a 7 pin socket. 'Form B' r '240 degree' -- the one that was used for original CoCo Joysticks. And for the PSU/remote control conenctor on old Philips cassette recorders. This one will go into a 6 pin socket. The standard socket does, indeed, have 2 grooves, adn there is the nromal 'bump' in the housing to engage with a groove. There is also a slot cut in the metal housing at a positon 90 degrees from the locator groove. The plug was originally (?) used for headphones. The wiring was to put the left chanel on one pair of diagonally opposite pins, the right channel on the other. The centre pin wasn't used (or maybe screening). Turnign the plug over simply reveres the pahse of both channels, which has no audible effect. I belive the outer housing of the plug operated a switch in the socket to cut out the speakers when the headphoens were plugged in. If you turned the plug over, the notch in the housing lined up with the switch actuator which therefore was not operated. So the speakers stayed operational. What I do not know is why nobody ever made a socket for this plug with only one notch so that it couldn't be fitted upside-down. -tony

"Wordstar on an 4.077 MHz 8088 could keep up with my typing; WinWord under Windoze on a 300 MHz PII can't." --Seth Breidbart (That it cites a PII-300 as the "modern" machine gives some idea how old the quote is.) In computers, it does seem to go that way. "Every program grows until it is able to read mail." I suspect it's largely due to Moore's Law: hardware has gotten so incredibly faster that it's now not only feasible but economic to put a general-purpose computer running a general-purpose OS into mobile phone. We tend to not notice it, because we've gotten so used to it. But computers have grown more powerful to an extent that is completely unmatched in any other field of human endeavour I can think of. If athletes' performance had increased correspondingly, track-&-field would be dealing with severe relativistic effects. If workers' wealth production had kept that sort of pace, we'd all be working five or ten seconds a year and still living like J. Paul Getty. If a factory turning out, say, cheap kids' toys produced one defective product per billion, that would be unbelievably good. A computer that mis-executed one instruction per billion wouldn't even last long enough to finish booting. If it could be made to with as little effort as a modern computer can be made to do "anything"? Then, hell yes! /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

I forgot to add that we had one college vice president who would: create a short memo (such as a room change for a meeting) add letterhead, signature print it on a color printer SCAN IT! send it as an attachment to an email with subject: FYI with body text: Read the attachment The letterhead included a horizontal rule. We puzzled whether there was ANY other easy way to get a couple of pixel stair step accross the page. Meanwhile, IT was begging everybody to please stop "opening attachments Previously, he (and the rest of the upper administration) had upgraded their 386 machines to 486 due to the urgent need (based on existence of something shinier), at a cost of well over a thousand dollars each. We begged them to let our lab have the 386 machines to upgrade some of our 286s. They refused, because they were getting a trade-in of $150-$200 per machine! Oh, well. I'm retired now! Don't have to deal with them much any more. And, I can get by just fine with this POS Dell laptop until I resume any woprthwhile projects. -- Grumpy Ol' Fred cisin at xenosoft.com http://www.xenosoft.com/FPUIB

On Sun, 22 Sep 2013, Fred Cisin wrote: And offered a full-duplex audio path. The conversational experience on a cell phone (or even a lot of IP telephony platforms) is horrid. The original Bell Labs researchers must be turning over in their graves... --

On Mon, Sep 23, 2013 at 12:37:58AM -0400, Mouse wrote: [...] Or it's a reasonable engineering trade-off given how little the phone functionality is expected to be used compared to everything else. I still have memories of the bad old days of carrying two boxes around, 0845 numbers, 9600bps transmission, 300-1,500ms latency caused by the gaffer-taping of incompatible communication technologies together, and hoping that the IR beam wouldn't be broken. It is thus rather pleasing that somebody has finally seen sense and integrated the cellular modem into my PDA. One day I may even forgive them for also integrating a voice calling feature at the same time, although I'll concede it's occasionally useful for ordering a pizza. I think the only real complaint is that somebody in the marketing department hadn't had enough Bolivian marching powder that morning, and boringly called this remarkable device a "phone".

On 25/09/2013 07:21, Mouse wrote: This debate is as hold as the hills. I see the same argument mentioned everywhere, I hear it with "software bloat" and cars with computers but , and I guess its the same in the US as in the UK. No one will admit to buying a phone or a piece of software or a car or a TV because its simple to use. No we buy it because its "feature rich". It been a while but look at any comparison list in any "buyers" magazine and it will contain grids of features and folks buy them because of the more features. I remember when I was working supporting Microsoft Exchange to PROFS/SNADS connectivity tools in the UK, around 1997. There was a tick box next to Outlook saying you could use Word to compose your Exchange e-mails. At the time Lotus Notes was the main competition and Gartner was predicting a 50/50 split between Notes and Exchange. So as Notes didn't have that "tick box" lotus came out with a "Feature" that allowed you to compose e-mails in Notes with Word.It was totally useless as you could only read the mails with Word and I don't know any one that used it, but it filled in the tick box! Madness.... I guess it goes back to this mis-quote :- "You can fool most of the people some of the time... which is just long enough to get elected... But I would disagree with "it is difficult", it isn't!... expensive perhaps but difficult no! In the UK simple mobiles are widely available. I have a brand new Nokia C1-02 as provided by my employer. It sits in my pocket for about a week till it beeps as its battery is getting low and it wants feeding. Worse than a cat. Its revued here:- http://reviews.cnet.co.uk/mobile-phones/nokia-c1-02-review-50003257/ but note it gets 3.5 out of 5, and Nokia can't make money out it and have sold out to Microsoft. Why , well in the UK networks subsidise most phones. Some one who buys a simple phone is usually some one who isn't going to make many calls, or spend on data. They have it for essential use. Some one who has say a Sony Experia T (my personal phone) will probably buy extra services like data, so the network makes money, the phone maker gets their phones subsidised by the networks and make money, and we fall for it every time. I make more calls per month on the Experia than I do in a year on the Nokia... By the way I don't think there is any difference in call quality between either phone. The Experia T is fine. As for VOIP much dep[ends on the "last leg". My son lives in an apartment block in Salt Lake City and when he "Facetimes" my wife, Ipad to Ipad (Skype for apple users btw) it works fine if he leaves the apparment with shared broadband and goes to the local cofee shop. Full Duplex voice seems fine given the path length and propogation delays.. (analogue voice east pondia to west pondia never was good for duplex because of the echo cancellation required)... By the way if you want a real simple easy to use phone "Doro" are also good and available in the US... http://www.dorousa.us/ but again pricey, and in many ways the on-screen dialling on the Experia is as easy to use as as the buttons on the Doro my dad had till he passed away last Christmas... Dave G4UGM -- Dave Wade G4UGM Illegitimi Non Carborundum

On 25 September 2013 11:21, Mouse <mouse at rodents-montreal.org> wrote: Maybe you want this: http://web.media.mit.edu/~mellis/cellphone/index.html Though it may not have the required quality... Acoustic design is _hard_. Maybe stick to your Nokia. (or look into picking up an old Motorola F3: despite Wikipedia's claims, I remember it being available in Canada at one time) Joachim -- Joachim Thiemann :: http://jthiem.bitbucket.org :: http://signalsprocessed.blogspot.com

On Fri, Sep 27, 2013 at 07:46:16AM -0400, TeoZ wrote: Er, analogue systems can't vary the bandwidth based on demand, so it doesn't get better if nobody else is using it. It's a circuit-switch network, not packet-switched! Your nostalgia is getting the better of you if you think that was "perfect sound". The thing was limited to about 2.5kHz, and there was all of the digital chatter from the in-band signalling. About the only great thing about it was that it existed at all.

On 25 September 2013 14:52, Peter Corlett <abuse at cabal.org.uk> wrote: Exactly. This. Whenever I hear some Luddite complaining that you just can't get low-end plain phones any more, I think to myself "well you just aren't looking hard enough, then." The low end is currently getting /lower./ Q.v. http://www.bunniestudios.com/blog/?p=3040 That's a GSM phone with Bluetooth sold /at a profit/ for US$12, no SIM, no bundle, nothing. The phone market is exploding across the developing world in the last few years, and lots of them are dumbphones or feature phones. Nokia's Asha S40 device & the like are what was keeping it afloat since the Windows Phone debacle. But look at it this way. Moore's Law means the components are getting cheaper all the time. What is a cellphone? Well, the minimum components are: * a radio capable of doing send/receive of voice & data (modern cellular networks are part- or all-digital) * a data entry device for dialling * a display, for seeing who you're dialling, that can show icons for signal/battery strength * a microphone * a speaker * some kind of CPU to manage the above, run the comms stack * some local storage for the comms stack etc. * a battery to power it Very desirable and cheap: * a headphone socket so you can use a headset * possibly even a link to a wireless headset So you have a pocket computer with human-usable input and output of sound and graphics, a local processor and both nonvolatile and volatile storage. Why would you /not/ make that the most powerful capable computer you could, if it did not raise the price significantly? You'd be foolish /not/ to. The more powerful it is, the easier you can make it to use - it can have a richer UI and so on. It can store a phonebook, always a useful addon for any user. It can be a calculator, a handy device that still sells in the hundreds of millions. It can send and receive texts, which are fantastically useful - they make phones usable by the deaf, and in noisy environments, and they're very cheap (free in most of the world, except the USA). Add this stuff, you make your device more appealing in the market. You'll sell more. So you would be stupid /not/ to. Now, we're at the point where for $8-$10 you can have all this - Bluetooth included, and some decoration and a colour display. That is the /absolute bottom end/ where there's little room for competitive edge. But double the price - to a mere $25 or so, still cheap and affordable even in the developing world - and you can add games, a camera, music playback and make your device much more desirable. That's still about as cheap as a big supplier with fixed costs could go, factoring in packaging etc. So don't ask why you can't get devices without this stuff. Ask why you still /can/ get devices without it if you shop around. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Mon, 23 Sep 2013, Peter Corlett wrote: Or, that SOME people really don't CARE about telephone sound quality. Some jerk called peddling long distance service (or SOMETHING, couldn't make out what he was saying); told him to call back on an AT&T line. Saw an early demo of VOIP. I asked Q: How is the sound quality? A: "REALLY REALLY GOOD!" Q: How good? What's the frequency response? A: "It's really really good!" Q: Is it better, or worse, than a regular phone? A: "It's really really good!" back and forth a few more times. The booth bimbo would not acknowledge that he had no idea how to quantify sound quality, neither numerically, nor comparatively. My hearing is BAD. If it were an actual ski slope, I would not risk it. I can manage barely adequately on my amplified phones, and almost adequately on a regular landline. But, degrade that signal another few percent, and I can't even use it. I have a cellphone. It lives under the seat of my car. in an emergency, I can bellow into it, and MAYBE make out YES/NO responses. How is the current sound quality? Is it "really really good"?

On 09/23/2013 12:16 PM, Peter Corlett wrote: Call it idle speculation, but what's the highest speed a consumer-grade audio modem can manage over VoIP, state of the art or otherwise? Has anyone tried this? --Chuck

On 09/23/2013 01:18 PM, Mouse wrote: I'd suspect only 28.8K or perhaps 33.6K, but not beyond that. As I understand it, 56K is a function of the telco 64K sampling and if you're doing 56K answer, it's done with the assistance of the local CO--i.e., no consumer-grade modem will operate at 56 in answer mode. --Chuck

On Mon, Sep 23, 2013 at 12:40:17PM -0700, Chuck Guzis wrote: [...] In theory, the top speed of your modem. A VoIP ATA is after all doing exactly the same job as the line card in your local telephone exchange, except without miles of manky old copper in the way. Practice, of course, is another matter. I know, because I tried it. It was far too ambitious a project as the calling leg was GSM, and it also went over a relatively slow ADSL line before being terminated on a modem that had already had a very hard life. But hey, we like a good challenge :) The outgoing GSM leg was 9600bps CSD, although the mobile telco converted that into a V.34 call to my VoIP number for no obvious reason. V.34 wouldn't work at all: it'd just squeal away negotiating lower speeds until giving up. I fiddled with a whole load of S registers to disable modulation schemes that might be latency-sensitive, and eventually got it to sort of hold a connection at 7200bps. Had I *two* modems, I would have expected it to just work across my LAN.

Some of them do, but nothing requires using such compression algorithms! Indeed, nothing requires using compression of any sort. $DAYJOB's VoI offerings, for example, are normally configured to not compress (except to the extent that u-law sample representation can be viewed as compression). /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

This is so true. One of my employers is a VoI[%] provider. For the most part, we don't sell the el-cheapo phones; we have enough call quality troubles as it is. I think the cheapest phone we sell is something like $75, our cost. And, provided the VoI system isn't misbehaving (eg, dropping RTP packets), sound quality is excellent. Since the D->A and A->D conversions happen closer to the human user than with POTS, there is less analog-layer degradation. The phones we sell even have decent sound quality when running full-duplex handsfree, which can't be easy. One way or another, you end up paying: in money, sound quality, or whatever. [%] VoIP is not new. Voice has been carried over IP on private links for a long time. What's new (well, relatively new) is use of VoIP over the public Internet - which is why I call it VoI. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On 9/24/13 5:51 PM, Steven Hirsch wrote: Calling BS on this. In practice, the delay/bufferbloat, crap echo cancellation and lossy compression w/ all the in-use codecs sound like ass compared to a clean AMPS or POTS line. Specifications wise, the audio bandwidth would be better (because of the lossy compression), and if you could get the delay down to something lower than transatlantic,it _should_ sound better, but to find this in the real world is rare. Intra organization (pbx) calls vs. voip -> crappy clec -> sip -> over public internet.

On Tue, Sep 24, 2013 at 06:51:13PM -0400, Steven Hirsch wrote: [...] Yes. The send and receive channels are coded and transmitted independently. This is all in the digital domain and is a solved problem with standard solutions. The hard bit is the analogue side, which is not so amenable to corner-cutting, and bad designs can cause audio to leak from speaker to microphone. An unsophisticated echo-canceller trying to paper over this bad design will effectively render it half-duplex. So don't buy one of those cheap and nasty handsets. You get what you pay for.