General > General Technical Chat
Why do companies try to take patents out on standard schematics?
VK3DRB:
--- Quote from: Zero999 on November 13, 2021, 08:48:59 am ---
--- Quote from: VK3DRB on November 13, 2021, 06:39:48 am ---IBM made a monumental miscalculation by publishing the entire schematic, the BIOS listing and other technical documentation to the personal computer, thinking manufacturers would use this to make peripheral cards to support the PC. Instead, a number of companies used this information to make PC clones.
The rest, as they say, is history.
--- End quote ---
Maybe, but would everyone have a PC now, if they hadn't done that?
IBM clones only took off because they were cheap. At the time and for a good 10 or so years, they were well behind most other platforms and were generally poor value for money.
--- End quote ---
The clones had a lots of issues. One being the ISA bus timing specs which they could not really meet because the limits were not properly defined. All users could hope for was "100% IBM Compatible" in the clone advertisements, and even then there were sometimes problems.
Most of the early clone motherboards Hong Kong and Taiwan lacked signal quality testing in design, ICT, safe ESD practices and used poor quality components. So IBM did very well for quality until around the mid 90's when Taiwanese motherboard manufacturers in particular had improved greatly to the point where IBM started sourcing their Aptiva motherboards from these vendors. Also, Dell, HP, Compaq and Gateway were looming as strong competitors.
I actually developed and implemented the electronics testing hardware and software for the first PC plant in China, as part of an IBM-China joint venture in Tianjin in 1989 through to 1991. At this time, I felt it was the beginning of the end for PC's being manufactured in the West.
FOOTNOTE: I had to give the US ambassador James Lilley a tour of the Tianjin production line but the GPIB driven hipot tester had failed about two hours before his visit. I quickly put a jump in my code written in 8086 assembly language to ignore the error and act as if it were passing. As he did the tour, I told him the machine is designed to put 1760 VDC between active/neutral tied together and earth, to test for earth leakage. As I was "testing" it, he asked me "Is that what this is doing now?" Sweating like a pig, I answered, "That is what it is designed to do." As I said this, he was looking at the needle on the hipot tester meter that had not moved. I suspect he knew. :phew:
Zero999:
--- Quote from: VK3DRB on November 13, 2021, 12:37:43 pm ---
--- Quote from: Zero999 on November 13, 2021, 08:48:59 am ---
--- Quote from: VK3DRB on November 13, 2021, 06:39:48 am ---IBM made a monumental miscalculation by publishing the entire schematic, the BIOS listing and other technical documentation to the personal computer, thinking manufacturers would use this to make peripheral cards to support the PC. Instead, a number of companies used this information to make PC clones.
The rest, as they say, is history.
--- End quote ---
Maybe, but would everyone have a PC now, if they hadn't done that?
IBM clones only took off because they were cheap. At the time and for a good 10 or so years, they were well behind most other platforms and were generally poor value for money.
--- End quote ---
The clones had a lots of issues. One being the ISA bus timing specs which they could not really meet because the limits were not properly defined. All users could hope for was "100% IBM Compatible" in the clone advertisements, and even then there were sometimes problems.
Most of the early clone motherboards Hong Kong and Taiwan lacked signal quality testing in design, ICT, safe ESD practices and used poor quality components. So IBM did very well for quality until around the mid 90's when Taiwanese motherboard manufacturers in particular had improved greatly to the point where IBM started sourcing their Aptiva motherboards from these vendors. Also, Dell, HP, Compaq and Gateway were looming as strong competitors.
I actually developed and implemented the electronics testing hardware and software for the first PC plant in China, as part of an IBM-China joint venture in Tianjin in 1989 through to 1991. At this time, I felt it was the beginning of the end for PC's being manufactured in the West.
FOOTNOTE: I had to give the US ambassador James Lilley a tour of the Tianjin production line but the GPIB driven hipot tester had failed about two hours before his visit. I quickly put a jump in my code written in 8086 assembly language to ignore the error and act as if it were passing. As he did the tour, I told him the machine is designed to put 1760 VDC between active/neutral tied together and earth, to test for earth leakage. As I was "testing" it, he asked me "Is that what this is doing now?" Sweating like a pig, I answered, "That is what it is designed to do." As I said this, he was looking at the needle on the hipot tester meter that had not moved. I suspect he knew. :phew:
--- End quote ---
There there were far superior computers to the IBM, in the late 80s and early 90s. Amiga, Atari, Acorn and Mac all had better graphics and sound. If the average user had the money to buy a real IBM, they would normally something superior, for that price. If it wasn't for cheap clones, it wouldn't have caught on.
I didn't understand why IBM insisted on crappy graphics. They didn't even use all of the 256kiB on the VGA card properly. The highest resolution mode of 640x480 4-bit only used 150kiB. Fair enough, non-standard mode X could go up to 360x480 8-bit, but it didn't match the 4:3 aspect ratio of screens back then. It wouldn't have taken much to make a 600x400 8-bit mode, or use up some of the remaining 100kiB in the 640x480 mode for some sprites, to speed certain applications up. It would have made implementing a mouse cursor much easier. The lack of simple sprite support, meant it all had to be done in software, so the mouse cursor would often flicker, when placed over an area of the screen which was changing, or go jerky, when the processor was busy. It wasn't until the mid-90s when IBM clones overtook the other platforms in terms of bang for buck, with regards to graphics and sound.
coppercone2:
why? + ibm = normal
TimFox:
From the men's room wall at my university's computation center, ca. 1975:
IBM
UBM
We all BM
4 IBM
coppercone2:
--- Quote ---I agree that a simple circuits shouldn't require peer review, but is it worth trying to keep basic designs secret?
And even reasonably experienced people do occasionally make silly mistakes. The classic is using the wrong footprint, because the CAD library or data sheet contained an error, they missed.
I disagree with you about diluting the blame. Ultimately the person who signs off a design bares the ultimate responsibility, but playing the blame game all the time, is counterproductive.
--- End quote ---
The mistakes that are the worst and most overlooked are the simple ones. Double checking someones power dissipation, loads, etc.. often finds problems when something 'easy' was pushed out the door that is running a part hot or bad. Those are the best bugs to find because the explanations are concrete, the correction is simpler, and the re-cert process is easier/cheaper... and such a change is usually regarded as a solid win, even from nontechnical managers since its easy to understand and is the engineering equivalent of leaving your fly unzipped. So they just say 'thanks for noticing that', make a board revision and you are on your way without debating voodoo.
The simple mistakes are made by advanced people because either
1) too many rules of thumb
2) rush job
3) too boring/simple to verify and you are sure of yourself
4) overconfidence (LED are easy!!)
and they often have the most catastrophic consequences because thats how you take out a power rail to a control system, have a 'dead' light, explode MOV and so forth.
Would you rather
1) explain to some board why you want a better power factor correction circuit for the 2nd stage boost circuit because "its a good idea based on industry trends" despite "expectations to substantially increase cost"
2) tell the board "we need to replace a 0.04$ resistor with a 0.04$ resistor of a different size so we can get the temperature down from deep fryer to toasty, and it will slow down the telephone calls"
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version