Inventor Of USB Didn’t Make A Dime Off Of It

Megalith

Megalith

24-bit/48kHz
Staff member
Joined
Aug 20, 2006
Messages
13,000
He might not have made millions of dollars, but he did make computing easier for millions of people.

…the USB technology didn’t make money for anyone. It’s because Intel, who owns all patents to the technology as the first backer of Bhatt’s USB idea, decided to make it open and royalty free from the beginning.
 
USB is so widely spread pretty much because they don't require licensing of the technology.
 
Bhatt may have missed out on a chance to make hundreds of millions of dollars on this
Click to expand...

Riiiiiight, just like the people who made hundreds of millions of dollars on firewire...
 
If only we had gotten that "single connector" promise instead of the 8 or 10 we ended up with. Somehow I always have a ton of every type that I don't need and none of what I do need.
 
I am glad he didn't ...
USB is aweful. the spec is aweful and it really should not have become the defacto std..

usbC is sort of making up for it BUT damn... 10years too late
 
naib said:
I am glad he didn't ...
USB is aweful. the spec is aweful and it really should not have become the defacto std..

usbC is sort of making up for it BUT damn... 10years too late
Click to expand...

As compared to what? I'm not picking a fight, but I'm curious. Serial and ps/2 ports were horrid. USB solved that problem. FireWire never took off because of licensing and thunderbolt won't because it's too late to the game, requires too much from the hardware manufacturers in order to implement.
 
Koolthulu said:
If only we had gotten that "single connector" promise instead of the 8 or 10 we ended up with. Somehow I always have a ton of every type that I don't need and none of what I do need.
Click to expand...

Not to mention the fact that they needed to be flipped 3-4 times before finally connecting.
 
naib said:
I am glad he didn't ...
USB is aweful. the spec is aweful and it really should not have become the defacto std..

usbC is sort of making up for it BUT damn... 10years too late
Click to expand...

Haha, didn't use computers much before 1995? Because then you would know what aweful is. Probably think Windows 95 was not that big of deal too?

I can only guess you are trolling.....
 
naib said:
I am glad he didn't ...
USB is aweful. the spec is aweful and it really should not have become the defacto std..

e
Click to expand...


Yeah. The parallel port was just fine. The only reason they changed it was because Big Printer lobbied to force a new standard in order to force us to upgrade.
 
westrock2000 said:
Haha, didn't use computers much before 1995? Because then you would know what aweful is. Probably think Windows 95 was not that big of deal too?

I can only guess you are trolling.....
Click to expand...

My guess is trolling unless he missed having a billion serial connectors and adding them one by one until you ran out of IRQ addressed.
 
Seems we always wind up with the lowest performing of two options. VHS instead of Betamax. I vaguely remember something about VLB being faster than PCI too. Wasn't SCSI always faster than IDE/ATA? DRDOS was better than MSDOS. OS/2 usually worked better than Win3.1.
Win95? Don't get me started on the ultimate nightmare which is the registry, where software companies can hide so much crap that end users have absolutely no hope of ever finding every line inserted into it, so once you install a program, you can pretty much forget about ever getting rid of all of it if you do decide to remove it, you have to purchase special 'uninstall' software and even then don't know if you got everything, so when the machine starts acting up, there's no way to know for sure exactly what's wrong ("an error has occurred at 00h3f2995c", supposedly missing .dll's which are obviously still there, etc.), and then you have to format and reinstall. Arrrrgggghhhh. .
 
TheSoldier said:
As compared to what? I'm not picking a fight, but I'm curious. Serial and ps/2 ports were horrid. USB solved that problem. FireWire never took off because of licensing and thunderbolt won't because it's too late to the game, requires too much from the hardware manufacturers in order to implement.
Click to expand...
As an evolution away from ps2/DB15/... sure but thats not to say it is fault-free...
As an electrical engineer one of my initial problems with the spec is the blurring of the physical,hardware,protocol...
If you wanted to used the physical layer that was USB compliant you couldn't (firewire you can, CAN you can etc...)
The dimensions of the USB-A connector are exactly the same width as an RJ45 reciprocal. I have a piece of test equipment (and associated software) that I designed and there has been two separate incidents where I was dragged into the lab with disgruntled engineers or PhD's about "my software not working" only to see they have plugged a USB-A into the RJ45 as they reached behind to find a suitable "hole" FACEPALM

There is the entire orientation issue as well that everyone is fully aware of. As I said USB-C has solved a lot of the issues ( I like being able to goto bed in the dark and just plug power into my nexus 5x )


westrock2000 said:
Haha, didn't use computers much before 1995? Because then you would know what aweful is. Probably think Windows 95 was not that big of deal too?

I can only guess you are trolling.....
Click to expand...
hahaha... non sequitur much.. I still use GPIB IEEE-488 interconnections daily, RS422,RS232...
 
A while back, Maximum PC did an article of the 100 greatest PC innovations of all time. Solitaire was 100. USB was 1
 
He didn't make money directly from it, but I'd say he's doing pretty well for himself. From the sounds of it he was able to trade his idea for a job at Intel. (Or at least get them to pony up cash so he could follow through)

USB is definitely still the best connector out there. Name one other that is backwards compatible back to 1.0. I can't think of one. I can literally take a device made today, plug it into a computer from 20 years ago and it will work. VGA is the only thing that has outlasted USB, but it hasn't changed specifications either.

People complaining about the mini - micro versions need to realize smartphones didn't exist when they created the device so it would have been impossible know they needed to accommodate such small connectors. Hold a parallel cable (Or DB9 or even DB25) next to a usb cable and you'll quickly see that USB was still one of the smallest connectors out at the time. What's funny about the fact that it's keyed is that things used to not be keyed in computing so it was possible to put cables on backwards. Except the technology to detect a backwards cable wasn't around yet. Put a floppy cable on backwards and you can wipe a disk in the drive.

Comparing USB to PS/2, it was a ton easier to orientate a rectangle than a circle. I can plug a usb into the right port without having to get behind the tower but it was a lot harder doing that with PS/2. I could go on about the fact that PS/2 had dedicated, non plug and play ports for keyboard and mouse so you couldn't put the keyboard in the mouse port and vise versa, AND if you didn't have it plugged in before the pc booted it wouldn't detect it, so you had to shutdown and power it back on.

USB type C is the answer to the issue and should hopefully bring back a single standardized cable that can be used on every device. It has the technology now to allow for either end to be the host and can switch the signaling on the fly if it's plugged in upside down.
 
There are tons of people that have contributed huge strides to personal computing over the decades, really most were never compensated anything above normal pay.
 
xorbe said:
There are tons of people that have contributed huge strides to personal computing over the decades, really most were never compensated anything above normal pay.
Click to expand...

If you are contributing lots of high quality patents you usually receive a fair amount of extra compensation at many companies ... when I was at Intel in the late 90's they gave you $1000 if they decided to pursue a patent you developed and another $4000-5000 when it was granted (that was for each patent and some engineers did dozens of patents a year) ... also, the prolific designers with lots of patents usually had a fair number of stock options granted ... I suspect that most companies are similar ... you won't become a millionaire (depending on stock performance) but you are definitely compensated
 
xorbe said:
There are tons of people that have contributed huge strides to personal computing over the decades, really most were never compensated anything above normal pay.
Click to expand...


Read about the guy that invented LEDs for a prime example..... That guy got screwed.
 
Before usb a computer had AT, PS2, Serial, SCSI, Printer, Parallel and all sorts of other ports. It was a nightmare of cables and incompatible connectors.

YSe8SKd.gif
 
I liked FireWire better, but they were charging way too much in royalties per port and the cables. Plus even though it was better, it wasn't that much better.
 
Koolthulu said:
If only we had gotten that "single connector" promise instead of the 8 or 10 we ended up with. Somehow I always have a ton of every type that I don't need and none of what I do need.
Click to expand...

A 100x this. I just ordered a USB 3.0 A to B off of monoprice over the weekend, then 2 hours later found one in the garage... Called as soon as their customer service opened up earlier today to cancel and was told that it can't be cancelled anymore due to it being boxed already.:mad:
 
This is how all standards need to be formed. Patent it, but immediately make it 100% free for anyone to use. I hate how most companies rather keep stuff proprietary then you end up with these closed "standards". Like how devices have "apple docks" but it ONLY works for apple devices. It's ridiculous. Stuff that is widely used like that should be open.
 
So basically as time goes on you have to dig up old books written at the time to get the correct perspective.

Basically today Scsi is basically SAS or Serial Attached Storage or a flavor of SATA or serial Advanced Technology Adapter. Which is funny since all this tech happened because of the USB interface was so much faster than parallel and serial devices, and could be daisy chained like scsi.

All the drives started with seagates ST-506/412 controller until the speed of data was running into issues as that encoding of the data from the hard drive was done on the controller card not the hard drive unit.

What is interesting is that there was a ESDI bus that was faster than the CPU could do the encoding, which limited the speed of the data transfers to as fast as the math co-processor could encode the analog signal to ones and zeros.

The western digital and Compaq (before being bought by HP), built the controller that was the defacto standard for IDE until the SATA replaced the PATA interface. Parallel ATA was a twenty wires in plastic coatings that could send up stream or downsteam but not both at the same time. Serial bus was up stream or downstream, and then SATA took the best of both worlds by making a dedicated upstream and downstream set of wires.

The fun thing is the IDE stands for integrated drive electronics because of the encoder buitl into the drive to turn the magnetic resonance signal into ones and zeros before it left the drive unit instead of doing the encoding on the controller chip or worse the CPU. The various other names technically were written ahead of IDE, like enhanced, Intelligent, and so forth to describe subtypes of the IDE interface.

Crazy stuff but scsi used IDE drives with a different controller board most of the time. I remember a friend of mine taking the drive apart in a home made vaccum box, it was fish tank that had holes cut in the side with a glass cutter. then grommeted and dish washing ribber gloves supper glued to the grommets and then filled with helium so when the drive was taken apart no dust got on the platters.

crazy stuff the drive sorta worked when he ran a bunch of chips off the bread board to the drive but we did something and the thing suddenly started spinning faster and faster until the motor burned out. That was week's pay of life guard duty and his beer money so that never got repeated.

fun stuff but the thing is there have been dozens of bus that were faster in concept but the data was tossed at the system faster than it could use it so the actual speeds were far slower.

The thing to think about is that while apple kept taking the basic idea and trying to one up the PC clone market the apple computers usually were too bogged down with pork to actually run faster no matter how much money they threw at them. Take the adobe hard coded instructions. When you did their benchmark it was extremely fast but if you changed one pixel's color it had to do it in software and all the circuit or registers set aside for that benchmark test were wasted space on a small die. Which is why all the work was done on SGI machines that you used a dumb terminal to log into from be it apple's beos, IBM's Unix, Windows DOS or OS/2, windows... etc...

so the USB bus really was necessary to push the tech forward and while the guy might not make royalties on it, I;m sure being the guy who came up with the tech that they used the concepts for to take the best of the other previous bus to build tech for the next twenty years can not have hurt his job security. Then again most places you are an employee which means you are supposed to be take care of as all your work you do for the company is supposed to be works for hire unless otherwise specified... unless you are an I-9 then it get confusing.

At the end of the day just think the USB is a bi-directional serial bus, like the SATA standard that we use for all storage devices right now. Of course SATA is two channels like the old IDE but with one dedicated to upstream and one to downstream. This takes the idea of parallel bus which was twenty wires, plus electrical stuff to take up the rest of the pins, grounds neutrals basically necessary to complete an electrical circuit, of wires that could be upstream or downstream, verse serial which was only one circuit up or down and making two circuits, one up and one down. USB one was one circuit up or down with cross talk possible when the up or down needed to happen at the same time. USB three or the funny ones with the extra tall connector essentially send more amps and volts or total watts of current over the wires to basically send more data.

Fire wire was basically a dedicated parallel bus with a different form factor of connector that could be hot swapped. which is why it never really took off the extra components in the devices cost four to five times the cost and most of those costs were paying apple for the privilege of using a different connector. snicker. and people still said it was better and faster when it had to go through the cpu on the apple computer to do a software decode which could be done faster on a general purpose Intel or AMD processor as they did not need the IBM power PC Intel co-processor to do the compiling and encoding and decoding of the data from the fire wire device much like the old ESDI oh wait it was an ESDI bus that they did not play royalties on because it was an ANSI / IEEE standard. Still never figured out they could use the ESDI with a different connector and charge royalties for it but no one paid seagate for the ST-516/412 for the connectors in the days before I played with tech. Snicker I guess it does not matter they ended up making money reselling music than anything else. Still think it is funny they fight Apple music over the name because they were tech only then end up being a retailer of music more than a tech company.

So yes the USB may not have paid the guy royalties but it never qualified for anything other than a process patent since the tech was all on how to build it not how to connect it since it used existing standards to connect the devices. Funny I bet Intel laughed when this was written since if he tried suing they would have got fun in court when they showed that the USB was public domain as all tech that was submitted to IEEE and ANSI so that mkg would use it instead of coming up with there own solutions requiring a dozne extra chips on the mother boards making supporting the tech ten times harder for Intel and the ISVs. grin. systems vendors that resell computers in a box.
 
sorry I was typing the concept is simply patent it copyright left and require all updates to be contributed back to the forum that maintains the tech. Free to use but if you think of a way to make it better you have to submit that concept back to the forum. That is how ANSI and IEEE work right now. Some bodies like the blu ray forum refuse to have IEEE standards for that reason.
 
Firewire was great for it's time. I was an early adopter, but it is known to burn up some peripheral devices if you actually plug it in hot.. It has happened to my old Cannon video camera twice with different controllers. Fortunately my firewire audio interface is still good, but there are numerous reports of them dying because of Firewire issues..
 
kbrickley said:
If you are contributing lots of high quality patents you usually receive a fair amount of extra compensation at many companies ... when I was at Intel in the late 90's they gave you $1000 if they decided to pursue a patent you developed and another $4000-5000 when it was granted (that was for each patent and some engineers did dozens of patents a year) ... also, the prolific designers with lots of patents usually had a fair number of stock options granted ... I suspect that most companies are similar ... you won't become a millionaire (depending on stock performance) but you are definitely compensated
Click to expand...

Like I said, $4-5K over normal pay is ... basically normal pay. Yay small jelly club bonus.
 
xorbe said:
Like I said, $4-5K over normal pay is ... basically normal pay. Yay small jelly club bonus.
Click to expand...

Although stock options aren't as valuable today with some companies I was usually making 3-4 times my salary off of options when I was there and most of the patent crowd had more options than I did plus the $20-50K in patent bonuses they were getting

Nothing prevents these designers from leaving the company and trying to invent stuff and patent it on their own dime (plenty of people do) ... but if you want the safety of a paycheck and the benefits of all the lab software and company support to develop your patent-able ideas then you will make less for the big patents (but you might make more from the little ones) and you get to do it with a fraction of the risk
 
bman212121 said:
He didn't make money directly from it, but I'd say he's doing pretty well for himself. From the sounds of it he was able to trade his idea for a job at Intel. (Or at least get them to pony up cash so he could follow through)
Click to expand...

he worked at Intel since 1990. He's also one of the engineers that brought AGP, PCI Express and other changes in chip power management. Engineers largely don't get any credit for what they do, especially in technology. We probably wouldn't even know his name if he wasn't mentioned in a Intel commercial a few years ago, and even then I bet the people who remember think the actor in the commercial is actually the guy in real life.
 
This reminds me of the inventor of the computer mouse, Douglas Engelbart, who never made any royalties for his patent on the "X-Y position indicator for a display system".
 
NoNRG said:
This reminds me of the inventor of the computer mouse, Douglas Engelbart, who never made any royalties for his patent on the "X-Y position indicator for a display system".
Click to expand...

But weren't these people being paid by a company to work on projects? If you want to own or make money on a patent you have to front your own time and money. If someone else fronts the money then they typically take a piece or all of the money and credit.
 
You must log in or register to reply here.
Back
Top