|
|
I have translated the following text from Russian. I would appreciate knowing if I have used any incorrect terminology, and also knowing what this is all about. (No hurry, I have sent in the translation but would still like to learn something). Development of a specialized processor for image processing based on the K1804 sectioned system, including development of microcode. (What is a sectioned system?) Integration of software into a finished system for a television with digital signal processing, with improved user qualities. (How does digital signal processing differ from what is used now?) Software implementation of a bus for data exchange between 12C microcircuits. (How does one implement a bus, or does one use another word?) Development of software for Teletext 1.5 and 2.X functions and the use of Teletext to create an on-screen menu (OSD). (What is an OSD?) Creation of a database of parameters for controlling a set of ITT Digit 3000 microcircuits. Development and implementation of software for automatic detection of components installed in a television chassis and and configuration of the related control code. (Rusty at Kiwanis was telling us about this - the new TVs have a computer built into them that tells you what is in the TV and how to fix it). Software implementation of reception of infrared remote control signals according to RC-4 and RC-5 protocols. (What are these protocols?) Writing software for self-diagnosing a television chassis and indicating errors using a light or audio signal. Writing control software for work with various television tuners. Implementation of functions for adjusting a television receiver using digital signal processing during servicing. When writing software, he used primarily Aztech 6502 assembler language and Aztech 6502 C compiler language. The code was debugged using EMU 3000 and TPU 2040 intracircuit emulators and TAMINO debugging software. (Is this commonly used software?)
15 responses total.
Re #0: Some of these terms are familiar, some I can guess at, some are probably language-specfic colloquialisms which don't resemble the English equivalents enough to suggest a meaning. "sectioned system": no idea. Maybe it means "modular", except that the Russian word for "module" is "module" IIRC. It might also mean something like a Harvard architecture processor, where code and data are in separate memory spaces. There are too many possibilities. A 6502 is not "sectioned" in this sense, many DSP chips are. >(How does digital signal processing differ from what is used now?) Digital signal processing is radically different from historical television practice, which is to use analog components. DSP can remove ghosting, stabilize sync, and do all kinds of magic. >Software implementation of a bus for data exchange between 12C >microcircuits. >(How does one implement a bus, or does one use another word?) Is that "12C", or "I2C"? I2C is a convention for relatively low-speed data transmission between chips; it is cost-effective because it only requires 2 wires and can inter-connect many devices on them. A bus is just a system for routing data. 16 wires in parallel with some control signals thrown in is a common scheme for a bus between a CPU and its memory and other devices. An OSD in this context is almost certainly an on-screen display. In this context, RC-4 and RC-5 are probably infrared remote-control signalling formats. The same acronyms mean other things in e.g. cryptography. >When writing software, he used primarily Aztech 6502 assembler >language and Aztech 6502 C compiler language. The code was >debugged using EMU 3000 and TPU 2040 intracircuit emulators >and TAMINO debugging software. >(Is this commonly used software?) Aztec (no "h") made some pretty good assemblers when I was doing stuff a while back. Wouldn't surprise me if they were still in use in Russia. I can't tell you anything about the emulators and debugger. An emulator is a device that plugs in in place of the CPU, and both executes the same program and produces the same electrical signals. It also has high-speed recording and trapping circuits which allow the software engineer to analyze the behavior of the program as it interacts with the rest of the system in real time and excruciating detail. Sometimes that's the only thing that will get the job done.
Thanks for the info. Russian typewriters make no distinction between 1 and I (they don't have a 1) so I could not tell the difference and guessed wrong. The Russian was spelled 'Aztech' (in the Roman alphabet) - must be they are used to 'tech'. One of our Kiwanis volunteers has been learning to fix TVs with computer chips in them, that will self-diagnose - I think he said these were the high definition TVs. Are all HDTVs digital? Any idea when there will be a switch from broadcasting both analog and digital (I presume that is being done now) to just digital? Originally it was supposed to be 20006 (at which point Kiwanis would be flooded with analog-only TVs). What other devices besides TVs and cars now contain 'computers'?
It might be easier to ask what devices *don't* contain computers, nowadays. Microcontrollers have gotten inexpensive enough that they're often used where logic circuits would have been used in the past. Cable boxes, TVs, VCRs, microwave ovens, stereo equipment, thermostats, washing machines, dryers...any number of devices. If it doesn't have a computer in it yet, it will probably have one eventually. Microcontrollers have also revolutionized things like elevators, traffic light controllers, and railroad crossing signals, in many cases taking over tasks that were once performed by banks of relays.
My spoon? Nope. My fork? Nope. My Leatherman? Nope. My glasses? Nope. My pants? Nope. My boots? Nope. This oil can on my desk? Nope. My desk? Nope. My backpacking tent? Nope. My sleeping bag? Nope. My table lamp? Nope. This stapler? Nope. I'm glad that there are still a lot of things in my life that don't have computers in them. I don't have a fondness for batteries and wall-warts all over, to say nothing of the complex technology involved, the likely bug in any software burnt into its feeble ROM or dealing with one button that has fifteen functions.
My printer (a 24-pin) probably has some microcontrollers in it. It is so far from intuitive how to program it that I no longer even try but just use the default font. Too many buttons to push in just the right sequence. I presume this solution is now much cheaper than the older methods, which had more moving parts. Jim has a smooth-top stove that no longer works at all, probably because of a dead diode. My stove does not have diodes. Are there cases where the newer circuitry is more reliable than the older mechanical switches? I would think the digital answering machines would last longer, since there are no belts to wear out. And they can also do things the older machines could not, such as erase, or listen to, only one message out of six.
I agree about the nuisance of batteries and wall-warts (we need a new type of power unit based on a quantum singularity), but I find that it is microcontrollers that are among the most reliable components of my home, compared to plumbing, windows and door, roofs, lawns, furniture, and paint.
I can think of some things that have become more reliable because of microcontrollers. Car engines, for example. They used to need a tuneup every 6,000 miles or so to keep running well. They require a lot less attention, now, because the computer can adjust for a certain amount of wear and tear. Also, most cars nowadays have no mechanical speedometer cable, since the speedometer is directly controlled by the car's computer. That means no noisy or broken speedometer cables, and no low-speed bobble. A lot of home appliances that seem less reliable nowadays are less reliable not because of microcontrollers, but because of shoddy manufacturing. A good example are all the TVs RCA has put out in the last ten years with bad solder joints. Companies have figured out that most customers will pay extra for added features, but won't pay extra for quality. The microcontrollers themselves are generally exceedingly reliable, being low-power devices based on proven designs.
Our TV repairman, any time he gets a TV which loses its picture after it heats up, goes straight to the solder joints. In each brand it is different ones that tend to go bad, and it is pretty predictable. (By 'our' I mean Kiwanis'). We met him because a large TV was donated that was not working but seemed worth fixing so we called around to various shops to see if someone wanted it for parts. He said to drop it off, and next day said he could drop it off, fixed, for free. (At which point we 'hired' him as a volunteer). Bad solder joint. Jim and John had spent hours testing all the components. Perhaps most customers don't expect to keep their purchases past the warranty period? What are typical warranty periods on electronic goods?
If the warranty period is 30 day, it might last a year. If the warranty period is 90 days, it might last three years.
Re #2: If the "2" in "I2C" was a superscript, or if it was ever written "IIC", then it's the 2-wire perpheral bus I was talking about. If it is mentioned in the context of connecting the control unit to the frequency synthesizer (channel selector), volume and tone controls, and picture controls, then it's almost certain it's I(squared)C. There are several different HDTV "standards" worldwide. The one for the USA uses a digital format for the over-the-air signal, so any TV which can handle it is going to handle things as numbers after the conversion from the intermediate frequency and probably all the way to selecting the brightness of pixels on the display. If it's electronic and has any kind of button-driven multi-function capability, it's almost certainly got some kind of digital smarts. Many of the really simple chips are dedicated-purpose (ASICs, or Application-Specific ICs) but even some car-alarm key fobs are driven by a programmed microcontroller. Your printer could not exist without one. Re #6: I think if we subjected microcontrollers to the weather directly, or flushed crap down them, that we'd find they're not as reliable as some people's experience with more-sheltered applications might indicate. It's the sensors and actuators where the rubber meets the road, and they have the dirty jobs. Re #7: The inherent reliability is something I'd question; a bad connection to a sensor can kill the engine in a modern car. What they don't do is degrade gracefully, so your pollution control gear will still be working as long as the ignition fires and the oxygen sensor reports good data. That's a big plus for the EPA. Especially for cars, microcontrollers are essential to meeting pollution and efficiency specs. There's no way that you could compensate spark timing for RPM, manifold pressure, intake temperature, fuel octane and exhaust-gas recirculation without digital controls; a mechanical distributor only manages the first two. You'd never be able to run the fuel mixture in the rich-lean-rich-lean sequence which gives the best emissions reduction without them either. There are even more little tweaks here and there, like retarding the spark briefly to reduce power output while the transmission is shifting which lets the clutches lock up faster and with less wear and tear. The nitty-gritty stuff of running an engine gets extremely detailed and arcane! You'd find them almost impossible to handle with mechanical or analog controls. The point about sloppiness is a good one, and it applies to cars (one of my particular areas of expertise) also. One of the great features of the sensors is that they can offer feedback about how upstream components are working, which allows the computer to adjust them. This in turn allows far-out-of-spec components to function just fine... or lets the manufacturer use cheaper, less-consistent components and smooth out the variations in software. Manufacturers have no incentive to avoid failures after the warranty period, unless they want repeat sales. Case in point: I have an Aiwa bookshelf stereo with a 3 CD changer and dual cassette. The CD changer now refuses to play most of my CDs, but they play fine in my little Magnavox portable and my JVC boom box. I am going to take the thing apart sometime to see if it just needs cleaning, but if it's more than that I'm going to junk it... and never buy another Aiwa again.
Actually, my impression is that car systems degrade a lot more gracefully than they used to. I had, for a time, a 1975 fuel-injected VW. it had no microcontroller, just an analog computer. It didn't degrade gracefully at all. If one sensor quit, it usually died entirely. Modern engine computers have a lot of "limp-home" modes to let them keep running if something goes wrong. The extreme example is the Cadillac Northstar engine, which can be run for a considerable distance without any coolant, without suffering any damage. It does this trick by firing on four cylinders until those four get too hot, then firing on the other four while the first set pumps air. It switches back and forth like that to keep the internal temperatures down.
The first CD players (more than 10 years old) did not play poor quality CDs well, no matter how expensively they were built. More gadgets have since been incorporated to deal with scratches. So don't blame AIWA in particular. Instead of junking the AIWA, could we have it to try to fix? Today we went through most of 20 broken tape decks and got about half working (the others mostly need belts that we don't have in our collection) - cleaning rollers or reindexing or turning off PAUSE or replacing belts fixed them all. One had an electronic problem and got recycled - the chip would cost much more than a belt if available. Mechanical problems are usually fairly simple to diagnose and cheap to fix. Several of the tape decks had something called AUtomatic Music Search System, which never seemed to work. (You hit FF while it is playing and it is supposed to go the next track). This is controlled electronically.
Re #12: This unit is less than 3 years old, and I'm going to try to fix it myself. (It played the occasional CD perfectly, but refused to play most.) Without CD function it is essentially an AM/FM/cassette. If the CD is dead I am going to look at making some other use of its speakers, which would make the receiver unit rather useless. If I have no use for the speakers, Kiwanis can have the whole kit and kaboodle including original packing. The remote works fine.
The public library has one good book on CD repair. If you have not ever
cleaned the lens, it may overheat and burn out the optics. I don't know if
optics that are going bad would play good CDs but not bad ones, but bad optics
make it start to skip, and then simply not recognize that there is a CD.
Our VCR repairman at Kiwanis might be willing to give you some idea of whether
it is fixable if you describe the exact symptom. He usually pokes around in
them and says 'bad optics' so we recycle them.
Units that have CD and tape combined are often built cheaply, to a
price.
That's why I prefer component systems.
Response not possible - You must register and login before posting.
|
|
- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss