Wellom programmers (not willem). Look for VP280, VP290, VP380. These are **excellent** programmers for the price.

Can something be cheaper than this? http://www.ebay.com/itm/Wellon-VP190-VP-190EEprom-Flash-MCU-Programmer-USB-… I had the VP280. Only sold it because of the Beeprog (waaaaaaaaaaaaaay better, but almost 10x more expensive). But it is such a good performer, I'd buy another in a blink of my eye. --- Enviado do meu Motorola PT550 Meu site: http://www.tabalabs.com.br ----- Original Message ----- From: "Chuck Guzis" <cclist at sydex.com> To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Sunday, April 29, 2012 1:22 PM Subject: Re: EPROM and EEPROM Programmers

AFAIK this is not true. Why don't you download the updated software from elnec site and try? :oP

Ops... --- Enviado do meu Motorola PT550 Meu site: http://www.tabalabs.com.br ----- Original Message ----- From: "Philip Pemberton" <classiccmp at philpem.me.uk> To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Sunday, April 29, 2012 3:38 PM Subject: Re: EPROM and EEPROM Programmers

Willem. satisfied also board port You can also count me in as a satisfied Willem user, I use the Willem for more modern components and I have a Data I/O 29B for the more vintage parts like Bipolar PROMS. Together with some conversion sockets I can program almost everything . -Rik

It seems that there are USB to DB25 devices available so you can plug the Willem into a USB port on a modern PC that does not have a parallel port. Has anyone tried this? Regards Rob

I concur. I bought one of these dirt cheap ones from China too. Forget the documentation and go for what's written on the circuit board. It worked for almost everything I wanted to do (Mostly 27xx work). I say almost everything because for some reason it couldn't cope with 2716 CMOS EPROMS? (standard 2716 was fine). Point is, I need to burn EPROMS only once it a blue moon and it WAS dirt cheap. If burning EPROMs was a major activity, I would have gone for more expensive and better support. It has its faults but hey, it cost next to nothing. Horses for courses..... Tez On Mon, Apr 30, 2012 at 7:55 AM, Alexandre Souza - Listas < pu1bzz.listas at gmail.com> wrote:

In article <alpine.DEB.2.00.1204291543140.3236 at duo>, Steven Hirsch <snhirsch at gmail.com> writes: They don't say how the device connects to the PC. Is it serial port, parallel port or USB? -- "The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline> The Computer Graphics Museum <http://computergraphicsmuseum.org> The Terminals Wiki <http://terminals.classiccmp.org> Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

On Sun, 29 Apr 2012, Richard wrote: It connects to the parallel port. --

On Apr 30, 2012, at 10:50 AM, Keith Monahan wrote: Oh, lord, here we go again. USB is great for what it's great at, but it does have a nasty habit of unnecessarily complicating simple things. They could probably do a relatively simple hack and put an FT245 (USB->parallel FIFO) chip on there that would work with some modifications, but then to make it work on Windows, you have to make it install a driver (regardless of whether you're using libusb or not). It really does a good job of mucking up the simple things when you can bang something together that "just works" on a parallel port, and it's aimed towards people who can justify hanging onto some legacy equipment if they want to burn some EPROMs. I like USB well enough, especially for the more complex things that it's good at (and the higher speed is certainly a plus), but I very much resent how much higher the barrier to entry is for a USB peripheral versus a simple serial/parallel port one. And yes, I've had to write the code on both ends (host and device) for both USB and "legacy" devices. It was industry standard when they wrote it, as you mentioned below. I should point out that lots of PC motherboards still incorporate at least a single serial port not because they expect people to be using serially-attached modems, but because it's still the best way to debug Windows drivers (having done so in the past) because USB is one-way and 1394 is decidedly an outlier in the PC world. - Dave

In article <536ABADE-E886-416A-AF19-BBC1A00EFD18 at gmail.com>, David Riley <fraveydank at gmail.com> writes: Since when? -- "The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline> The Computer Graphics Museum <http://computergraphicsmuseum.org> The Terminals Wiki <http://terminals.classiccmp.org> Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

On Apr 30, 2012, at 2:03 PM, Cameron Kaiser wrote: Yes, I suppose "asymmetric" would have been a better term. It's not peer-to-peer like serial ports or 1394 are, which is what would make it nearly useless for kernel debug; that's before you get to issues of protocol stack, which 1394 gets away with by using DMA on the controller, which is questionable from a security standpoint but great for kernel debug. - Dave

In article <4F9E7874.11091.8A5EE9 at cclist.sydex.com>, "Chuck Guzis" <cclist at sydex.com> writes: None of this is making sense to me. Please clarify what you mean by these terms in this context: "polled nature" "asymmetric" "half-duplex" ...and it would help that instead of just rephrasing in another euphamism, you would say exactly what it is about USB that matters here. -- "The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline> The Computer Graphics Museum <http://computergraphicsmuseum.org> The Terminals Wiki <http://terminals.classiccmp.org> Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

On Apr 30, 2012, at 4:06 PM, Richard wrote: Sure. To be clear, what I meant initially by "one way" is that USB transfers occur when the host requests them (i.e. a device cannot say "Here, host, have some data!", but rather the host has to poll the particular device looking for data). That's what I meant by "asymmetric" (it's not a bus where all devices have equal standing as targets or initiators, like SCSI or PCI or Ethernet or 1394), and what someone else (Cameron?) meant by its "polled nature". Half-duplex is generally true of USB, but it's not what I was talking about. There's only one (differential) data line in the USB bus, so communication is necessarily half-duplex, but that's not really material to the problem I was getting at. USB 3.0 may have extra lines in its socket (I haven't really looked) to enable full-duplex communication, but I don't know what that entails since I've never looked into USB 3.0. There are "host to host" USB cables available (probably the same devices that Glen Slick was talking about), but they require an active device in the middle to store and forward the data from both hosts as if they were devices on the other end. They were used for file transfer a long time ago when the serial port was going away and people were bemoaning the loss of the null-modem cable transfer; I recall someone made a homebrew PlayStation2 app which could use that to load homebrew code over USB instead of burning lots of discs (this was before the Ethernet attachment for the PS2 was popular; the "slim" ones have it built in, so now Ethernet is the medium of choice). - Dave

On 30 Apr 2012 at 14:06, Richard wrote: Okay (although this could have been resolved with a simple lookup on USB protocol). The third term should be obvious. Data can flow in only one direction at any given time over the interface. USB 3.0 uses a pair of connectors to get around this, but I'm not aware of any USB 3.0 full-duplex EPROM programmers. "Asymmetric" refers to adherence to a master-slave topology; a member on the bus is either a controller or a slave. What ties this with "polled nature" is that there is no way for a slave to initiate a transfer--all conversations are initiated by the controller. In other words, conversation between slave and controller is "speak only when spoken to". Thus, to determine the status of an operation, the host must periodically poll the slave asking "Is it soup yet?". I hope this clears things up. --Chuck

Richard wrote: ...simple or not, I for my self think that classic interfaces are better fit MY needs. I've finished to build an Programmer for CY7C291 and Am27C291 Chips yesterday and it has a simple AtMega644P CPU on it. This way I have enough RAM to buffer the ROM contents + some extra for the Program on the device. Could have build it with an FTDI in it, but what for? I can connect that device to old Computers I'm using, starting with some CP/M gear (mostly Robotron), an Zilog S8000 compatible System (EAW P8000) with some Zeus Unix Clone on it (WEGA) and even an PDP11 or a VAX. Some Terminal Emulator is sufficient to do useful work on it. Software is mostly the same (excluding the Program algorithms itself) to an previously build 2708 Programmer that I've build since almost none of the actual Programmers support them anymore. Next one is for 1702As. I don't need shiny gadgets with USB, most modern gear has JTAG/Flash anyway.. I have an Conitec GALEP3 to, it has a parallel port. Modern programmers in them... who needs that? Regards, Holm -- Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, Freiberger Stra?e 42, 09600 Obersch?na, USt-Id: DE253710583 www.tsht.de, info at tsht.de, Fax +49 3731 74200, Mobil: 0172 8790 741

Chuck Guzis wrote: Not only that, I think there are some opensource libs that make an USB device out of a plain Atmega w/o some hardware USB Support like on chip devices or FTDI and I have 90% of the flash ROM left over free. (It's because of Atmels politics to not making Devices with enough RAM in them, you get more Flash than you need and only 4 Kbytes of RAM) I'm simply don't need USB in that class of Hardware. But there is another Problem raising in the meantime, I should change the Motherboard from my PeeCee this year I think. There are some Capacitors on the Voltage converters blowing up AGAIN and performance wise it isn't that good anymore altough it is running with FreeBSD 365 day/year. I have to buy somehing next time that don't has the features I want. (Have an PCI COM Card additionally to the onbord COM ports in there already). But this changes nothing about that what I think cwabout the Promer interface. Regards, Holm -- Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, Freiberger Stra?e 42, 09600 Obersch?na, USt-Id: DE253710583 www.tsht.de, info at tsht.de, Fax +49 3731 74200, Mobil: 0172 8790 741

David Riley wrote: You don't intend? :-) Simply do a "kldload uftdi.ko" *grin* Regards, Holm -- Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, Freiberger Stra?e 42, 09600 Obersch?na, USt-Id: DE253710583 www.tsht.de, info at tsht.de, Fax +49 3731 74200, Mobil: 0172 8790 741

On 1 May 2012 at 19:52, Tony Duell wrote: Not as far as I've read the standard. "Speak only when spoken to" appears to be the rule at least in 1.x and 2.x. USB OTG is a little different in that some devices can assume either the guise of a controller or a slave, but one still can't have multiple controllers in a USB topology as far I know--which may well be dated by now.. --Chuck

David Riley wrote: At connection time it is fixed, not negotiated, based on which kind of USB plug (A or B) is inserted into the OTG entity's AB receptacle. This is sensed by the ID pin, which is grounded in a USB A plug, and unconnected or pulled to Vbus in a USB B plug. There are three main cases: 1) OTG entity with AB connector is connected to a host. Cable is A (host) to B (OTG entity). The OTG entity senses via the ID pin that a B plug is inserted into its AB connector, and enters device role. 2) OTG entity with AB connector is conntected to a device. Cable is A (OTG entity) to B (device). The OTG entity sense via the ID pin that an A plug is inserted into its AB connector, and enters host role. 3) Two OTG entities with AB connectors are connected together. Cable is A to B. The OTG entity sensing via the ID pin that an A plug is inserted enters host role. The OTG entity sensing via the ID pin that a B plug is inserted enters device mode. Note that the user could plug the cable between the two devices in either orientation, and that determines which OTG entity is configured as host and which as device. In case 3 *only*, Host Negotiation Protocol (HNP) can be used to allow the OTG entity in the device role to request that the roles be reversed. Note that A-A cables or B-B cables are NOT used even for OTG, and are not allowable cable configurations per the USB standard. Some stores actually sell passive A male to A male cables, which are completely useless, and I've never been able to figure out why they are made or why someone would buy one. There are active A male to A male cables, which are used for file transfer between two hosts, but these actually contain electronics that implements device role with respect to each host. Plugging a passive A male to A male cable into two hosts will not accomplish anything useful, and could even damage the hosts. On the other hand, A male to A female or B male to B female cables are at least useful as extenders, though not permitted by the standard. It is intended for and supports a lot more than plugging two like devices together. For instance, it can be used to connect a camera to a printer. The main limitation is that an embedded host usually is only able to control a limited selection of devices, which comprise that host's Targeted Peripheral List (TPL). Unlike a PC, a camera doesn't usually support the user installing printer drivers, so it probably can only support specific printer models.

On 05/02/2012 02:44 PM, Tony Duell wrote: It's got an entirely different target application set. Sure it's "serial", but then so is Ethernet. (mostly) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

At 01:33 PM 5/3/2012, Tony Duell wrote: Yeah, that's exactly what I've always thought was wrong with USB. Why can't it travel backwards in time? - John

On May 1, 2012, at 2:52 PM, Tony Duell wrote: I suppose it depends on what you mean. The USB host is constantly polling the devices; when the devices have a message for the host, it gets sent then. There's no out-of-band signaling like an IRQ, if that's what you mean. Devices just have to wait their turn. It's like Token Ring in a lot of ways. - Dave

Actually, this is the main reason I destest USB (OK, it's over-complicated for the sort of things I do which is bad enough, but the restrictions on what you cvan connect togetehr are even worse). it strikes me that one of the main moans about RS232 was that there were straight cables and null-modem cables and this confused people. If it had been agereed that all 'hosts' -- that is computers -- would be wired as DTEs, and all 'slaves' -- modems, printers, etc -- would be wired as DCEs, there would bne no problem. The problem arrose because you could link 'hosts' (computers) directly togther with RS232. USB seems to avoid this my simply making it impossible to direclty link 2 hosts. I ds not see that as an improvement!. -tony

Actually, thats' 2 more resons in favour of RS232 for me... Firstly, USB _requires- a microcontroller really. The protocol is too complciated to build from hardwired logic. Asynchronous RS232 can by done using a microcontrolelr (either ther a built-in UART or bit-banged) _or have more computer power in the I/O chip than in the rest of the machine :-) Secondly, the fact that RS32 does not use TTL (or 3.3V logic) levels means that most microcontrolelrs/FPGAs/etc (if not all) could not drive it direcrlly. For a proper implementation of an RS232 interface you had to have some sort of buffer, like a MAX232 chip. And that would bear the brunt of any zaps on the interfce connector [1]. Such a chip is a lot easier to get and replace than a programmed microcontrolker for a device that you've never seen before. I guess I don;'t want reliability... I'd rather hace somethign that fails mroe often when the parts that fail are easy to get and replace than soemthign that fails evey 5 years where I have major problems getting the IC> [1] Although not always. The serial port in my HP9817 was defective when I got the machine and I found that not only were the buffers damaged, so was the 8250 serial chipe (yes, an 8250 in a 68K machine. Odd...). I am not sure what had happened to the machine. Not the ethernet I use.... All cables are srraight and have one screen and one cetnre core. -tony

I haev never understood this love of making thigs as small as possible. OK, I don;t want a pocket calcualtor that actually takes up a 19" rack cabinet, but for soemthign that I don't have to carry about, I don't care if it's a half-rack-width thing, a 3U rack module, a 6U rack module. etc. In fact things that are too smal lare often difficult to use because the cotnrols are too close together. Iam not even going to mention repairing them.. In a differnet context, I much prefer the handling of a Nikon F camera over an Olympus OM series, for similar reasons. Things can be too small to handle easily. What's an 8350? Do you mean Z8530 (the serial chip used in older Macs), ro is this some other device I've not come across? The HP9000 serial ports (both built-in and on plug-in cards) did use 8250s -- the chip used in the IBM PC -- for some odd reason. Well, there was a nice microprocessor-controlled serail card that used a Z80A CPU and a Z80A_SIO serial chip, but then the serial chip wasn't on the 68K bus. The HP Integral used a 68000-family serial chip, the number of which I;'ve forgotten for the moment, but can trivially look up. -tony

I don't know whether anything analogous is true of USB, but in RS-232 there is a spec that (to paraphrase) any pins or combinations of pins may be shorted to one another or any voltage source within a specified range (probably -25 to +25 or some such) indefinitely without harm. This sort of robustness is very nice to have in an external-world interface. Heh. Well pointed out. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

So, it's not 'universal' by any reasonable definition. It _is_ serial. It is mopt definitely not a bus. I guess one word right out of 3 is OK these days :-( As I've said before, I don't the point of a 'one size fits all' interface. Differnt peripherals have different requirement,s itdoesnt' seem unreasonable that you plug them into different conenctors (in the same way that on an audio amplifier there are differnet input sockets for line-level signals (output of a radio tuner, tape recorder or CD player), microphone signals, pickup cartirdge signals, and so on. The thibg I reall find pointless, ocme to think of it are USB keyboards and mice. I think it is reasonable to assume that a modern PC is going to be connected to a keyboard and a mouse -- just one of each. So having a speciifc conenctor for each of them doesn't seem to be a big problem. Why over-complicate things? -tony

On 03/05/12 2:39 PM, Tony Duell wrote: Because it's clearly simpler to have a single bus and lose the specialised connectors. Which PCs eventually did (long after Apple did). Additionally, isn't USB more electrically forgiving than PS/2? --Toby

On Thu, 3 May 2012, Eric Smith wrote: And, are therefore now the defacto standard for charger connectors! With what sort of amperage limits? 'course now maybe I can get rid of a few totes full of wall-warts, with hundreds of different coaxial barrel connectors.

On 4 May 2012 at 20:36, Tony Duell wrote: Well, ethernet (PoE) is going that way, but in a bigger sense (ca. 44V at 350 ma.). Most modern PCs will continue to supply power to the USB connectors even when "off" (soft "off"), so mobile devices can be charged even when the system isn't being used. And USB as a cheap source of power makes all of those really cool devices--shavers, foot warmers, fans, etc. possible. --Chuck

Thinking about it, most of my machines have power available on the ethernet conenctor. That's the AUI connector, of course :-) I am sure there are more efficient ways to do this. Do does a bench supply... Anyway, I dnon't particularly want to sahve near my computer. Shaving produces a lot of dust (consisting of small particles of hair, of ocurse), which I'd rahter didn't end up in my systems. And with only 2.5W availalbe (after negotiation, 5V at 500mA), I don't think any of the heaters are going to do a lot. -tony

On Sat, 5 May 2012, Chuck Guzis wrote: Sure, they sell it as a toy. now. But would it work with upgraded hardware? How much computing resources do they require?

There are already, um, devices of a personal nature available. You can easily guess which industry. -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com -- Time is an illusion. Lunch time, doubly so. -- Douglas Adams ---------------

On Sat, 5 May 2012, Chuck Guzis wrote: How many USB ports does the Toto Washlet have?

Sure, but you usually can't get a bench supply for under the price of a fast food burger, while you usually can get a USB "charger". I think it's pretty nifty to be able to get 2.5W and 5W 5V power supplies that inexpensively. Of course, they're also not of the same quality as a bench supply. I wouldn't use them for anything that I can't easily repair.

On Thu, 3 May 2012, Toby Thain wrote: Are you SURE that you want to call it a "BUS"??!? Knowing the difference between a keyboard and a mouse was stretching the limits? They had different shaped connectors. They color coded them! Eventually, they just gave up and added an extrea microcontroller for each, just so that people didn't have to insert the plugs into the jacks that fit.

Provided it has, or can have, substantial intelligence and fairly-high-speed interfaces built in (high speed compared to, say, serial, at any rate). Provided it doesn't need to initiate sends to the host, or is non-real-time enough, and the host is sufficiently overmuscled to have spare cycles enough, that making the host poll is acceptable. Probably provided a few other things, too, which I might know if I knew USB better. True. Not worth the costs in my opinion, not for most applications at any rate. (Not that I have any evidence anyone cares about my opinion.) /~\ 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 4 May 2012 at 11:04, Fred Cisin wrote: Nothing that can't be easily solved with a hammer and a little elbow- grease. If you break one, there's always duct tape. --Chuck

On 04/05/12 3:38 PM, Tony Duell wrote: You've already asked that. NOW imagine that a whole bunch of unforeseen devices - printers, cameras, touch sensitive tablets, disks, modems, e-book readers, 'pen' drives, portable audio players, ... - arrives. Are they meant to use PS/2? Are they supposed to have individual connectors? USB is no more a silly idea than RS-232 is. --Toby

On Fri, May 04, 2012 at 04:47:52PM -0700, Fred Cisin wrote: One standard (the "universal" part in UBS) protocol - and cabling setup - for a lots of things. Mass storage (from USB sticks to memory card readers to harddisks), serial interfaces (the CuBox is recently aquired has Micro- USB for the console connection), media interfaces like sound and video (video cameras, TV sticks). Even phones (Android Debug Bridge runs over USB as well). And if you run out of ports, just stick on an USB hub (of course, there are still bandwidth limits). Also, just three + one very different connectors: standard, mini and micro plus the host connector. And these days, USB also helps clean up the mobile phone charger mess: an increasing number of new phones just use USB with a micro USB connector instead of every phone needing its very own wallwart with a very special connector. Kind regards, Alex. -- "Opportunity is missed by most people because it is dressed in overalls and looks like work." -- Thomas A. Edison

On Sat, May 05, 2012 at 11:56:35AM -0400, Mouse wrote: That is someone playing silly games. If you remember, name & shame please (manufacturer & device) because this nonsense deserves being blacklisted. The whole damn _point_ of switching to micro USB as the charging connector is to get rid of the custom chargers and be able to use generic ones ... Kind regards, Alex. -- "Opportunity is missed by most people because it is dressed in overalls and looks like work." -- Thomas A. Edison

I don't remember, but I can find out. That makes sense - provided it scales back to the standard max if the magic indicator isn't present. Refusing to charge at all, I can't see that as anything but customer lock-in. Mouse

Eric Smith wrote: If you have it on and hooked up to a 1A USB charger (port or standard USB charger) while it won't charge, it won't really discharge (much ) either - so it isn't exactly lying - it is NOT charging if you are using it, but asleep or off it will charge (faster if completely off). John :-#)#

On 05/05/2012 23:58, Alexander Schreiber wrote: OK, "after negotiation" -- but if the iPhone doesn't get any negotiation how does it know what it can draw? Should it assume the full 500mA, or might it be a low-power charger that won't even do that? Some laptops will give 5V but only at a few mA when asleep, some chargers give substantially less than 500mA. Hence the resistor network. In fact an iPhone /will/ negotiate, but if the negotiation fails it relies on the cruder simpler hardware. I'm not suggesting it's perfect -- I've found plenty of situations where it fails (like with some docking units) -- but I can see some logic to it. -- Pete Peter Turnbull Network Manager University of York

On 5 May 2012 at 10:26, Eric Smith wrote: History beginning to repeat itself? 25 pin D-sub connectors were the most common type, could be obtained from a large number of sources, had a standard description, and, due to volume, were inexpensive when compared to other proprietary connectors. So, you use them for everything. As far as a computer with nothing but USB for connectivity, the Cotton Candy from Fxitech comes pretty close--one USB male on one end, a female USB on the other end and an HDMI connector in the middle: http://www.fxitech.com/products/ --Chuck

While I think that the choice od conenctor here is a very bad one (USB connectors should only be used for USB), I think the external USBV-serial converter is a Good Thing. It is relatively easy to link an asynchronous serial device to a SUB port (one FTDI chip), it is very difficutl to do the reverse.By having an external converter (hidden in the cabel) you have a device that can be linked to a USB interfce or t o an synchronous serial port (possibly with a level shifter). -tony

On Sat, May 5, 2012 at 4:21 PM, Alexander Schreiber <als at thangorodrim.de> wrote: IIRC, that is actually mandated by law in some countries now, notably China. -- Joachim Thiemann :: http://jthiem.bitbucket.org

Yes, and I still haven't received an answer. That is NOT WHAT I AM ASKING ABOUT. I sam specifically considering only keyboards and mice. I happen to think that USB is a useful interface in some situations, for example for external memory devices. That doens't mean I have to think it's the best interface for all applications no matter what marketroids may claim. FWIW, I don;' consdier RS232 to be a universal interface either. It's one I use somethmes when I feel it's the appropriate choice. -tony

Tweo reaosns spring to mind : 1) It is over-complicated, considering the amount of data that has to be transfered, etc. 2) There is not 'interrupt' facility./ The host machine (OK, the USB inteface chip) has to keep on poollign the keyboard to see if a key has changed state, and keep on pollign the mouse to see if it has moved. That seesm to be totally the wrong thing to do.. Heck, even the HP9830 has an intettupt on keypress. -tony

On Fri, May 4, 2012 at 9:38 PM, Tony Duell <ard at p850ug1.demon.co.uk> wrote: My keyboard/mouse are connected using a _different_ universal interface: bluetooth. This frees up two USB ports for other things. In fact, It's quite possible to have a PC these days with nothing but USB connectors (and power, I guess). Video is typically not done for speed reasons, but you can buy USB video adapters. Joe. -- Joachim Thiemann :: http://jthiem.bitbucket.org

On 04/05/12 3:28 PM, Tony Duell wrote: They have the SAME connector, so from a usability pov it invites mistakes. Sometimes, even non-technical people use computers. I've noticed that they often don't identify sockets before attempting plugs. And proliferating physically incompatible connectors seems to make less sense than simply standardising a simple interface. Even if the computer doesn't blow up, it means a lot of pointless calls to help desk or "my son who knows about computers". And that problem gets bigger with increasing permutations of devices. --T

So does mine. Actually, it's a Type M which came off a PS/2 machine. I cut off the mini-DIN and soldered on a full-sized DIN plug. TRS-80 Mod1l I, I guess. Which of my machines has the same connector (even same gneder) for the serial and (TTL level) parallel ports? Hint, the conenctor is not a DB25 (and nor is it a card edge or a header plug). I hate to say this, but such people deserve all they get, including the magic smoke leaking out.But plugigng PS/2 devies int othe wrong sockets does no damage. I've seen a DEC field servoid force one of those '3 coaxes in a DA shell' conenctor (whatever it's called) on upside-down, thus swapping the red and blue signals... 10 seconds at most :-). Those are horrible header plugs... -tony

On an LCD TV we bought a few yewars ago there are 8 RCA phono sockets (along with lots of otehr conencotrs). These RCA phono sockets are : 3 for 'component video' -- luminance nad 2 chromanance signals. Coloured red, gereen amd blue 3 for an AV inoutm, Composite video (yellow) left (white) and red (right) audio 2 for audio output to an external amplifier (if to so wish), white and red again. So there are 3 red-coloured phono sockets with differnet functions. It makes having to plug colour-coded moni-DINs into keyboard and mouse conencotrs very simple. USB ius anything but simple. Have you read the standard? Anyway, next you'll be claiming that the 3 sockets on the wall here -- a 230V 13A mains socket, a coaxial TV aerial socket and a UK telephone sockey should all teally br the same type of conenctor and interchangeable. -tony

tony's not stupid; I think he sees why USB exists. But I think his point is that, while (like any flag day) it's in the interest of computer makers to impose it, it isn't in the interest of the end users. I agree with him. USB is great for hardware makers, since it enforces obsolescence, thus ensuring a stream of new sales - well, actually, it's not having USB that's good, but eliminating non-USB (and even tony doesn't seem to mind having USB for the things it is good at, just its imposition for other stuff it's unnecesary-to-negative for). It's not great for end users - for most of them it's a minor lose (and for a few it's a major lose and for a few it's a win). /~\ 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

But the discussion isn't how long USB has had a spec, or how long it's existed on the lab bench. It's about its appearance in mass-market machines, and, more specifically, its being imposed (ie, no alternative provided) for the connection of low-speed HCI devices such as keyboards and mice. USB is fine, in ts place. But I'm with tony in that I think that that place does not, possibly excepting some ridiculously space-constrained designs, include keyboards and mice. /~\ 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 May 5, 2012, at 11:02 PM, Lance Lyon wrote: I agree. All keyboards and mice (from original PC) had a micro controller in them (original was 8048). So arguing that USB *requires* more intelligence (as in the AT/PS2 keyboard/mouse doesn't) is wrong. OK, the stack is a little more involved but hard coding a stack for a simple device isn't all that hard. What USB does do, is (and what Apple and other MFR's have followed suit albeit more slowly) that there isn't a proliferation of connectors on the PC. Sourcing, stocking, managing multiple parts is expensive. Reducing the number of parts reduces costs. The other thing that USB allows, is that (especially in the case of Apple) if the keyboard has a hub, you can plug the mouse (or what ever) into the keyboard. I then only need 1 cable winding its way to the machine. USB memory sticks are convenient to plug into the keyboard. Don't have to reach down and find a spot for it on the back of the machine. And as many have mentioned previously, you don't have to explain to folks that the mouse and keyboard connectors while they look the same have to be plugged into different sockets. The crap about "well users should know better" is just that?crap. Good design/engineering doesn't require the user to make up for the shortfalls/short sightedness of the design. TTFN - Guy

On 6 May 2012 at 14:44, Toby Thain wrote: Don't almost all USB keyboards have 6-key rollover, where PS/2 keyboards generally have n-key rollover? Does heavy traffic on the host controller (say you have a hard disk plugged into the same port) interfere with USB keyboard operation? Just trying to ask some relevant questions... --Chuck

On 6 May 2012 at 13:06, Eric Smith wrote: Two simple demonstrations of good keyboard encoding that I'm aware of are these (the Model M fails both) 1. Press the ASDW keys and hold down. See if any other keys register. 2. Press right and left shift and hold down. Type "The quick brown fox jumps over the lazy dog". Since I don't have a USB keyboard (preferring model Ms for the feel), how do USB keyboards fare? --Chuck

On 06/05/12 4:47 PM, Chuck Guzis wrote: Apple Pro keyboard fails both. --Toby

No, you're not. I don't care for the weird resistance profile that's acquired the name "tactile feedback". I want to feel just steadily increasing back-pressure (ie, a simple spring) until the key strikes the hard stop at the end of its travel. In particular, no bump partway down and no mushy squish of substantially increased resistance for the last millimeter or so of travel. My favourite keyboard - though admittedly perhaps coloured rosy by nostalgic memory - was a particular model of Symbolics Lisp Machine keyboard I used in the late '80s. My second favourite, my favourite that is actually obtanium, is the Sun type-3. (When it's in good shape. A type-3 in bad shape borders on unusable, and I don't mean just distasteful to use; I mean it can actually be difficult to get keys to depress enough to register.) Mind you, I'll take a Model M over today's typical cheap crap keyboard, but that's not because I like it but because I dislike it less. /~\ 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

Not at all. I'm pretty much the opposite. Despite the fact that I also learned on manual typewriters and also play the piano, I find the long travel of the Model M uncomfortable. As for the sound, I used to like the click on the VT100, but these days, a loud keyboard is very distracting. There's exactly one Model M in our open plan office and the moment it goes into use, I have to reach for noise- cancelling headphones *and* run music through them to drown it out. I know many people complain about laptop keyboards, but I've come to quite like them. BLS

On 6 May 2012 at 21:39, blstuart at bellsouth.net wrote: I also play piano also, but it's a completely different skill, depending on the ability to moderate stroke force and speed, something that's completely irrelevant to computer keyboarding, not to mention the chording aspect. I've sometimes wondered if I could take the output of the MIDI keyboard here and work out some sort system of music-to-letters, not as abstract as Solresol. However, I'm convinced that I can't think that fast. The noise of a model M doesn't bother me, as I also have a Selectric III sitting by my desk here (now that's noisy!). If you've ever spent much time in a keypunch room, you're familiar with the much louder noise of a bunch of 026s or 029s chunking away. I prefer my NCR Cherry keyboard. Electrically, it's better and it's hard to beat the feel of a good Cherry keyswitch. I doubt that it understands USB. --Chuck

At 03:47 PM 5/6/2012, you wrote: asd asd asd asd I can't get a W when pressing asdw holding it down along with each preceding key. This is a PS/2 keyboard. I have lots of machines on this 8-port KVM and it doesn't speak USB. HE QUCI BROWN OX UMP OVER HE Z OG. Mine fails that test. 276 . [Science] It is very common to give natural language words technical meanings. --John McCarthy, NEW: a50mhzham at gmail.com ? N9QQB (amateur radio) "HEY YOU" (loud shouting) ? Second Tops (Set Dancing) ? FIND ME ON FACEBOOK 43? 7' 17.2" N by 88? 6' 28.9" W ? Elevation 815' ? Grid Square EN53wc LAN/Telecom Analyst ? Open-source Dude ? Musician ? Registered Linux User 385531

On 5/6/2012 1:47 PM, Chuck Guzis wrote: Northgate Omnikey Ultra (PS/2) passes. (For test 1 I can hold down 9 other keys and the 10th still registers properly. I need more fingers.) I should try this again at work on Monday -- there I have an Ultra hooked up via a PS/2->USB adapter... - Josh

Tony Duell wrote: As I said, cost. It's not that they don't do it because it is difficult; they don't do it because more difficulty means higher cost. If customers weren't willing to buy it without n-key rollover, they'd manage to include the diodes, by surface-mounting them on the back side of the PCB. However, the vast majority of customers are willing to buy the cheapest crap they can get.

It really saddens me when I compare the quality of HP stuff 20 or more years ago with stuff with that name on it today :-(. -tony

On Tue, May 8, 2012 at 7:39 PM, Tony Duell <ard at p850ug1.demon.co.uk> wrote: The lack of modern drivers for some of their printers, which are still going strong here, is a serious issue. I'm sorry for them that they made this stuff so well, it still works, I really am, but they should do the decent thing and support it. -- Yoyodyne Propulsion Systems: ?"The Future Begins Tomorrow" Visit us at: http://www.yoyodyne-propulsion.net -------- Doctor: You know when grownups tell you, "Everything's going to be fine" and you think they're probably lying to make you feel better? Amelia: Yeah. Doctor: Everything's going to be fine.

I'd love to have new HP gear. Unfortunately, it is all crap. No HP notebooks on my home.

At least here in Brazil, too expensive to be considered. --- Enviado do meu Motorola PT550 Meu site: http://www.tabalabs.com.br ----- Original Message ----- From: "Chuck Guzis" <cclist at sydex.com> To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Tuesday, May 08, 2012 6:51 PM Subject: Re: n-key rollover (was Re: What's the point of USB keyboards+mice...)

On 05/08/2012 08:11 PM, Toby Thain wrote: HP does (or did?) have a factory in Brazil. HP products with serial numbers nnnnBnnnnn were made in Brazil. (That's the old HP S/N scheme; not sure about the new one.)

I thought they did. I have seen HP calcualtors with 'made in Brazil' on them,. and a 'B' in the serial number. Old HP serial numbers have a letter in them that gives the coutry of manufacture. A = America (USA), B = Brazil, F = France, G = (West) Germany, 'Q' or 'U' =s England/United Kingdom, S = Singapore, etc. Mind you, this doesn't imply that HP had a factory in the speiifeid country. Ecan in the classic era, some products were built by other companies. I have seen 'J' (Japan) in the serial number of an 'HP' dot matrix printer which was clearly an Epson with different firmwre and an HPIL interface board added. And 'K' (Korea) on a monitor which was clearly a Samsung chassis built to an HP sprcification (It had connectors for an HP-HIL touchscreen option, etc). It was not built like a 'real' HP, and was downright painful to repair... But I think they did have a factory in Brazil -tony

Despite being a former Digital/Compaq worker who as sacked by HP I have a "new" (it's a factory re-furb) HP consumer style tower. The space saver keyboard was crap but the rest of the system is fine. I actually bought a new keyboard from our local supermarket. Its also labelled HP and that seems fine too. The keys have a nice length of trvael and I can type at a good speed. I have rang the HP support desk and they sorted out a minor problemI had. I think I would still buy a Lenovo laptop though as opposed to the HP... Dave

OK, I think we agree. The point I was making is that the cost is not just the cost of 100 diodes and soldering them in. It is difficult (not impossible, but difficult) to add components to a membrane. And a lot of keyboards, including the Type M and some horrible plastic modern thing I have here depend on the membranes being flat with no bumps (e.g. diodes). Yes, it's possible to get roudn both of those problems, but it adds to the complexity and cost. It is clearly possible to make a keyobard with full n key rollover But it would probalby be done most easilty (read cheaply) by not modifying the design of a normaal chap PC keybaord but by going to soemting like seaprate swithces on a PCB, or contacts that short traces on a PCB (like some old Commodore keyboards). One keyboard that certainyl could do full rollover (I ahve no diea if the interface microcontrolelr handled it properly) was the one for the HP9816. This is the only keyboard I have seen where hte switches are indicidually wired to IC inputs (in this case inputs on 4051 multiplexers) rather than being wired in a matrix. Unfortuantely true... Something has to be wrong when I can buy a complete PC keyboard (101 keys + encoder electronics) for less than the cost of half a dozen (reasoanble quality) keyswitches. -tony

USB ports are intrinsically much better than any other kind of port because the others are on the back of the machine, and USB ports are always on the front (except on some machines).

On Sun, 6 May 2012, Chuck Guzis wrote: WHY is the keyboard (and mouse) connector on the BACK? WHY was the power switch at the back? (I loved the PS/2 bellcrank linkage to put a power switch at the front with a long rod to the back!) When I set up the computer so that the connectors are where they belong, I can't get at the floppy drives. (I'm also not sure why the hard drive(s) needs to be at the front) When thin (LCD?) TVs came out, howcome all of the ads showed them hanging on the wall, but not having any of the requisite cables and accessory items (such as VCR)? OB_CC: likewise the hemipherical Apple. It's not such a "masterpiece of elegant design beauty" when you connect what is needed. Likewise, all of the iPOD ads always showed it being held at arms length overhead. Won't it work in other positions? If PADs are intended to be carried, howcome they don't have a handle, and a cover for the delicate screen. 'Course, with a folding screen cover, they now have the same form-factor as an ordinary clamshell.

Most mains wall socekt outlets over here have internal double-pole swithces which are sueful if the device doesn't have a real mains swtich :-) For countries where that's not the case (and indeed over here if you are plugging the machine into a multi-way mains adapter), it's quite useful to make up a mains plug to IEC socket lead with a double spole swithc somewhere along it. Add a neon indicator too ifyou like. I have one that I use when I am repariign something and don't want to use the internal mains swich, perpahs because it weoule be difficult to handle it without accidentally tocuhign the terminals. -tony

Is it really beyond people here (and others who often swap the devices plugged into their computer) to make little extension cables that bring the port conenctors from the back of the machine to somewhere more accessable? I have a such a cable ending in a DE9 plug conencted to oen fo the serial ports on this machien to plug in the cables from my HP handhelds, interface devices I am designing, etc. -tony

On 5/6/12 1:02 AM, Lance Lyon wrote: Strictly regarding keyboards and mice, one HUGE win for USB over PS/2 is genuine hot-plug capacity. I've never personally blown a PS/2 port, but I have worked on computers with blown ports, and invariably the user/owner was prone to swapping keyboards with the machine running. I don't kow if that's an odds thing or cumulative damage, but in my experience the PS/2 port is just not hot-pluggable in the long run. Doc

On May 7, 2012, at 14:49, "Chuck Guzis" <cclist at sydex.com> wrote: In the TI app note I mentioned, it often comes down to the ESD protection diodes. If the data pins are energized before Vcc or ground, you can get some pretty significant current through those diodes. This isn't the app note I was looking for (I have it filed away somewhere, maybe at work), but it does cover the basics: http://ics.nxp.com/support/documents/logic/pdf/hot.swap.pdf - Dave

On May 7, 2012, at 14:36, Doc Shipley <doc at vaxen.net> wrote: TI actually has a fascinating app note on the various potential failure modes of ICs in hotplug situations. I'll have to see if I can find it when I'm back at my computer. It was a big eye-opener for me. - Dave

Easy question: USB is a high-speed interface. It is already there, so why NOT use it for keyboard and mouse? Once developed the controller, it just works, you don't need to do it again, and it sells at millions. Why not? :)

On 5/4/2012 12:19 PM, Tony Duell wrote: I have a device on my desk here that's the size of a small paperback book; it's a subnotebook computer. I can connect keyboards, mice, scanners, printers, serial ports, parallel & scsi ports, cameras, calculators, phones, GPIB interfaces, external sound, video capture, hard drives, usb memory sticks, card readers and dozens of other things that I'm not able to bring to mind in the 20 seconds it took me to type that -- all to one of the pair of USB ports on the side. (No, not all at the same time, smartass :)). If this device had a specialized connector for each (or even most) of these it'd be completely covered in them and probably weigh 30 pounds :). This is the advantage of USB for modern computing. No, it's not suitable for a "classic" computer, but that's not the point and it's not why it was designed. (And at the same time, this entire, drawn-out thread isn't really suitable for "classic" computing either.) Is it a perfect standard? Nope. Does it do everything perfectly? Nope. Does it have a name that doesn't technically describe what it is/does that was probably chosen by a marketing department somewhere at Intel? Sure does. Is it well established, and does it work decently well for the purposes for which it was designed? Generally, yes. Does Tony have to like it? No :). - Josh

But y'all left out the humping dog, pole dancer, fan, book light, and coffee warmer. USB is faster, and slightly better cable standardization than RS232.

On 4 May 2012 at 18:29, ben wrote: That's easy--go shopping for one at Office Depot. --Chuck

To add all those at once you must have external hubs (since USB is not a bus), which will increase the volume somewhat. Actually, i have a similar-sized machine here. It fits in my pocket (normal size pocket) runs off interal primary cells, and can be programmed in 2 high level languages and assembly languate. It can be connected to external disk drives, tape drives, RS232 ports, parallel ports, HPIB devices, printers, pen plotters, video display units measuring instruments, other similar machines, and so. It can easily conenct ot 31 devies at a time, ther are ways to make it do rather more. Oh, and did I mention it doesn't need external hubs to do this. No, it's not USB. The thing is getting on for 28 years old... FIrstly, what would be the problem if it did waigh 30 pounds nad lived in a 19" rack module? Secondly, I assume you normally conenct a mouse and keyboard to it. That takes up 2 USBN ports,wheter built into the machine or on an external huyb. The volume of a SUB connector is comparable to that for a mini-DIN connecotr. SO it souldn't take up any more space to have PS/2 interfaces for the keyboard and mouse. You misunderstand me. Iam not quesitoning (yet) whether USB is a good idea for external memory drvies, or anything like that. I am specidicalyl askign why it is a good thing for keybaords and mice. So far nobody has managed to justivy that. -tony

On Sat, 5 May 2012, Josh Dersch wrote: Thank you The ability to add a keyboard and mouse on a machine, where adding a keyboard and mouse is truly OPTIONAL, is the first rational argument that I've heard for USB connection of keyboard and mouse being better than separate unique keyboard and mouse connectors. One other thing that USB has been good for is to standardize hardware interdacing across platforms. Otherwise, how do you add an EPROM programmer to a Mac? IFF you survive your journey through driver hell, at least the hardware can now interchange! There used to be a time when some Mac fanatics would claim that the Mac was a "better machine" because EVERY OEM peripheral available for it (both of them) would "just work", whereas the thousands of third party peripherals available for PC would require cabling and drivers. Now, finally, EVERYBODY can experience the joys of peripheral vendor "support"!

On May 5, 2012, at 11:21 PM, Fred Cisin wrote: Having been a Mac user for a long, long time, I'll admit that that's because the selection of devices available for the Mac was rather constrained. Yeah, anything that looked like a SCSI disk would "just work", just like anything that looks like USB Mass Storage will "just work" on any platform that even pretends to support USB. SCSI scanners, on the other hand, were always subject to the whims of device manufacturers; USB has made it neither better nor worse. I'm glad that digital cameras settled rather quickly on the sane idea of presenting themselves as USB Mass Storage, at least (since, at the base level, they're just acting as memory card storage adaptors). That would have been possible with SCSI, but who wants to hook up SCSI to a digital camera? Instead, you ended up with weird proprietary solutions over serial (e.g. Apple's QuickTake) or oddball devices like the Sony Mavica, which had the floppy drive built in. And, of course, with SCSI you needed some sort of intelligent device on both ends for managing the transactions. Obviously, this has nothing to do with keyboards or mice, which could be left well enough alone (the fact that the PS/2 port persists on most PC motherboards is a testament to the fact that it's not a bad idea), but it is nice to at least have keyboards and mice that can plug into the same port that every other device plugs into. Mandating it, of course, seems silly. I'll cop to that. I'm still using an ADB keyboard through a USB adaptor (ADB, of course, does require some level of intelligence in the device, though really not a whole lot more than a PS/2 keyboard). - Dave

On Tue, 1 May 2012, Tony Duell wrote: You would have them wire printers as DCE? My first printer was a Daisy-wheel (DTC-300 HytypeI) printing terminal. If it were wired DCE, then how would I connect it to a modem when I used it as a terminal instead of as a printer? GASP! I would need to make a cable with the "transmit" pin connected to the "receive" pin at the other end! I freely admit that some further standardization for serial printers would have helped. I had that printer for months before I finally got ink onto paper. And, in the USA, I'm aware of a documented homicide over the frustrations of interfacing serial printers to microcomputers. OTOH, I know somebody whose first experience with connecting a serial printer involved him reaching into a large bin of random DB25 cables, and selecting the first one that had the genders that he wanted. It turned out, by chance that that one had 2 crossed with 3 and some other crosses and ties-together between 4,5,6,8, and 20

On Tue, 1 May 2012, Doc Shipley wrote: ~35 years ago, a guy took a computer and a printer into a Computerland store. After 6 weeks without results, the user shot and killed the Computerland guy. Sorry, that's all that I remember. I mentioned the incident to Joe Campbell while he was writing his "RS232 Solution" book. He tracked it down and mentioned it in the book - there might be more detail there.

Terry Stewart wrote: Actually IBM got that one right. DTE (e.g., computers, printers) are supposed to use the male connector, and DCE (e.g., modems) are supposed to use the female. Using a DB25 of either gender for a parallel printer connector was brain-damaged.

Why? I have many things here that use DB25 conenctors for other than RS232 interfaces, so I don't see the problem with using it for a parallel printer port. My complaint about the IBM PC in this context is that none of the conenctors are labelled. Yes, the connector plates do look differnet if you know what to look for, so you can tell MDa from CGA from EGA monitor sockets (all DE9 sockets), but most users woulnd't know waht to look for. They should have been labelled.. The really brain-dead use of DB25s on a machine of that period is the HP150. That thing has a couple of DTE RS232 ports on DB25 _socket_. Yes, the wrong gender. That's bad enough, but there's an optional parallel+HPIL (not HP-HIL, I do know the differnce) board for that machine. The parallel port is on a DB25 _plug_, I guess to make sure you can't plug an HP serial cable in by accident. And would you believe the pins are a mirror image of the IBM parallel port. In other words, if they'd fitted a socket to the board, the IBM printer cable would plug straight in and work... I would bet they originally intended that and changed the conenctor gender at the last minute to prvvent lusers plugging serial cables into parallel ports or vice versa -tony

Hi I never even try to do serial without a breakout box and lights. Even that isn't always enough. I recall that Heathkit inverted one of the handshake signals from what everyone else used. Dwight

An awflul lot of RS232-interfced printers put the ready signal on pin 11. Why I do not know, but I've made many cables that brign that to something more standard (like CTS). Nothing to do with RS232, but I once raparied a Model II or model 12 (I forget which) in a Tandy store using a _Model 1_ printer cable. The owner of the machine had broken off a pin on the header plug on the back that connected to the printer cable, and the salesdroid said nothing could be done, it would ahve to go to the service centre which would take a few weeks. Said owner urgently needed to print his accounts. I grabbed a screwdrier (litterally) and took the machine apart. I noticed that the ribbon cable from said header went to a 34 pin card edge on one of the PCBs, and that it looked to have the normal pinout. I therefore asked for a Model 1 printer cable and conencred it straight to the PCB. Fed it out through a gap i nthe case and the owner went away happy with a non-original machine that printed fine. I have no idea if it was ever repaired... -tony

Well, that was pretty much the idea, as I understand it. But, as you say, And that's what a null modem is for. Conceptually, it's two back-to-back modems, implemented by non-straight-through wiring in the cable (or connector or whatever). Twisted-pair Ethernet appears to have learned nothing from the serial-line confusion; it went with two different wirings (`host' and `hub') and thus repeated the straight-vs-crossover cable issue in a new field. It would have made so much more sense, even more than for RS-232, for everything to have the same wiring, so there's only one kind of cable involved. Instead, we had two kinds of cables for a long time, and even now we have special hardware capabilities (called "auto-X" or "auto-MDI") to compensate for the whole mess. RS-232 can, to a degree, be excused by its vintage; when it was developed, the zeitgeist was much more heavily into the DTE/DCE dichotomy than now. Twisted-pair Ethernet had much less excuse. Agreed! Totally agreed. /~\ 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 Mon, Apr 30, 2012 at 11:14 AM, David Riley <fraveydank at gmail.com> wrote: If you have a USB 2.0 host controller which supports the USB debug interface (all Intel USB 2.0 host controllers do), and the USB 2.0 debug port is externally connectable on the debug target system, and you have the USB 2.0 debug connection device to connect between the debug host and the debug target, then you can kernel debug a Windows system over a USB 2.0 connection. It can be a pain to setup, but is useful when none of the other options (serial, 1394, Ethernet) are available on the debug target. -Glen