Not loose. Loose is your girlfriend on a Friday night with the football team.
Lose goddamnit!
How many other so-called 'educated' idiots are going to keep making the same mistake?
Well thought out post. The world is so cowed by your mental and literary awesomeness that I think the heavens just opened up and showed the entire Eastern Seaboard the second coming. All because God was jealous of the display of pure distilled intellect.
Actually not. The PCI implementations are different now then they were when the case was filed. It is fairly improbable that currently used buffering and switching techniques, which are are far faster than the ones sued over, would still be covered. It would take another entirely new lawsuit to cover that topic if Apple say's they changed something in response to the original suit.
That's why I said "if". I don't keep up on these techniques anymore.
The patent system seems to be rigged for abuse, especially if you have to patent something which is later marketed as intuitive. In some cases, gestures are patented, which I really don't get, but it's backed up by legalese about how it's supposed to be good for the electronic device.
For anyone wondering with this idea of DMA, or Direct Memory Access.
Disk reads, in comparison to CPU speeds, are very very very slow. Consider a 2GHz CPU, and a disk with 10ms seek times (that is it takes 10ms to find what one is looking for on a disk). 2GHz is 2 billion Hz, which means that the CPU "loops" 2 billion times per second. 10ms is one 100th of a second.
When a program wants to read from a disk, if it were to just wait this means that the 10ms seek time would mean that roughly 20 million CPU cycles would be wasted. So instead what happens is the operating system tells the disk what it wants to read and where it wants it written in RAM, and then the disk does all of the work copying data from the disk to RAM and tells the CPU when it's done. This way, the CPU can use those 20 million cycles for something else. This is called DMA.
There is then the idea of predictive caching, where if you can guess what the user wants to read, you can save them time. For example, normally when a program reads a file, if it reads the first kilobyte of the file, there is a good chance that it will want to read the next kilobyte as well. The OS cannot know this for sure, so it says "Well, you're probably going to read the next kilobyte, so let's read it from the disk." If the OS guesses correctly, everything runs faster. It's a similar idea to how if I were to say "Good mor" you could guess with reasonable certainty that the next three letters would be "ning".
Of course, you can't do this perfectly, but there are very good algorithms for this. Now, what OPTi has patented seems to be somewhat more complicated than this, but this is sort of the basic idea behind it. And there's the whole issue of willfulness - if Apple knew about this but chose to do it anyway (for whatever reasons) they were quite wrong.
But $19 million is like what? Profits from a couple of days sales at best.
For anyone wondering with this idea of DMA, or Direct Memory Access.
Disk reads, in comparison to CPU speeds, are very very very slow. Consider a 2GHz CPU, and a disk with 10ms seek times (that is it takes 10ms to find what one is looking for on a disk). 2GHz is 2 billion Hz, which means that the CPU "loops" 2 billion times per second. 10ms is one 100th of a second.
When a program wants to read from a disk, if it were to just wait this means that the 10ms seek time would mean that roughly 20 million CPU cycles would be wasted. So instead what happens is the operating system tells the disk what it wants to read and where it wants it written in RAM, and then the disk does all of the work copying data from the disk to RAM and tells the CPU when it's done. This way, the CPU can use those 20 million cycles for something else. This is called DMA.
There is then the idea of predictive caching, where if you can guess what the user wants to read, you can save them time. For example, normally when a program reads a file, if it reads the first kilobyte of the file, there is a good chance that it will want to read the next kilobyte as well. The OS cannot know this for sure, so it says "Well, you're probably going to read the next kilobyte, so let's read it from the disk." If the OS guesses correctly, everything runs faster. It's a similar idea to how if I were to say "Good mor" you could guess with reasonable certainty that the next three letters would be "ning".
Of course, you can't do this perfectly, but there are very good algorithms for this. Now, what OPTi has patented seems to be somewhat more complicated than this, but this is sort of the basic idea behind it. And there's the whole issue of willfulness - if Apple knew about this but chose to do it anyway (for whatever reasons) they were quite wrong.
But $19 million is like what? Profits from a couple of days sales at best.
This case isn't about DMA, DMA has been around for about 40 years now. DMA is also a bit different than you describe here, but for the purposes of this thread DMA doesn't matter and I'm passing on the opportunity to pontificate more on that detail.
The case is about predictive cache snooping, just one of the ways to make a bus act faster than it is signal-wise. These techniques change over time. According to the original story "OPTi argued that virtually every Mac sold infringes on the patent".
Well if every Mac sold, over 5-7 years, actually willfully infringed the judgement wold have been in the 100s of millions at a minimum. It wasn't so the judge did make a rather significant narrowing of the original complaint showing there is more than one legally acceptable way to do cache snooping.
Since there was no court imposed ongoing royalties on top of the $19 million it is also reasonable to conclude that the judged infringement is in the past - technology did it's thing and moved on.
The legal system is done with the case unless Apple decides to appeal, and that isn't a legal decision as our little loser lover thinks it is. It is strictly a business decision at this point. A business decision dealing with small enough, actually trivial in this instance, worst case losses and large enough upside to warrant pursuing an appeal purely for principle. A business decision like that isn't driven by accountants dollars and cents analysis. It is driven by personality and moment-by-moment emotional state of the decision maker.
I guess NanoAkron just hasn't been around top-level decision makers on a bad day to see the pissant things they will do when they are in a bad mood and some unlucky or stupid moron crosses them at that moment. It can be quite fun to stand back and watch the ensuing rug-dances and policy-dagger twisting. Business is a cruel mistress and lawyers are just tools of the trade.
This case isn't about DMA, DMA has been around for about 40 years now. DMA is also a bit different than you describe here, but for the purposes of this thread DMA doesn't matter and I'm passing on the opportunity to pontificate more on that detail.
The case is about predictive cache snooping, just one of the ways to make a bus act faster than it is signal-wise. These techniques change over time. According to the original story "OPTi argued that virtually every Mac sold infringes on the patent".
Well if every Mac sold, over 5-7 years, actually willfully infringed the judgement wold have been in the 100s of millions at a minimum. It wasn't so the judge did make a rather significant narrowing of the original complaint showing there is more than one legally acceptable way to do cache snooping.
Since there was no court imposed ongoing royalties on top of the $19 million it is also reasonable to conclude that the judged infringement is in the past - technology did it's thing and moved on.
The legal system is done with the case unless Apple decides to appeal, and that isn't a legal decision as our little loser lover thinks it is. It is strictly a business decision at this point. A business decision dealing with small enough, actually trivial in this instance, worst case losses and large enough upside to warrant pursuing an appeal purely for principle. A business decision like that isn't driven by accountants dollars and cents analysis. It is driven by personality and moment-by-moment emotional state of the decision maker.
I guess NanoAkron just hasn't been around top-level decision makers on a bad day to see the pissant things they will do when they are in a bad mood and some unlucky or stupid moron crosses them at that moment. It can be quite fun to stand back and watch the ensuing rug-dances and policy-dagger twisting. Business is a cruel mistress and lawyers are just tools of the trade.
I agree with everything you've said except for one thing. The judge can't impose royalties. The companies must work that out for themselves. That becomes a separate civil contractual issue. If the two companies can't agree on that, then an arbitrator is usually agreed to. But as often as not, the two parties leave the table with nothing accomplished, while the infringer works out a different methodology to bypass the patent, which is almost always possible in software disputes.
Great win for the patent holders against a rich, willfully infringing company. They should have gone after more though -- Apple has almost $29 billion in cash.
I agree with everything you've said except for one thing. The judge can't impose royalties. The companies must work that out for themselves. That becomes a separate civil contractual issue. If the two companies can't agree on that, then an arbitrator is usually agreed to. But as often as not, the two parties leave the table with nothing accomplished, while the infringer works out a different methodology to bypass the patent, which is almost always possible in software disputes.
Wrong. A judge can impose royalties and a compulsory license. Also, an infringer can work out a different methodology, but the patent owner can issue patent continuances that cover anticipated, different methodologies.
Wrong. A judge can impose royalties and a compulsory license. Also, an infringer can work out a different methodology, but the patent owner can issue patent continuances that cover anticipated, different methodologies.
They can not. They can only order, at best, that both parties work out a solution.
This court case is an example of how stupid the residents of Texas really are.
Please don't put all Texans in one basket - I am embarrassed every time a patent case goes to the court in Marshall. It seems the court almost always sides on the patent holders side (which more and more seem to be patent warehousers).
We need patent reform but we also need judicial reform. Lets not let everybody that wants to file a case in Marshall, TX cause the idiots over there have such a high percentage win rate for these "ambulance chasing" patent attorneys.
jOhn
Quote:
Originally Posted by aka
Give Texas back to Mexico!
AGAIN - Those of us that live here (legally), and have family here, would not agree. Texas has a number of fine things to offer - it just so happens that a certain judge in Marshall is not one of them. That is if you appreciate fair handed adjudication when it comes to patent law.
Comments
Lose!
Lose lose lose lose lose lose lose lose lose.
Not loose. Loose is your girlfriend on a Friday night with the football team.
Lose goddamnit!
How many other so-called 'educated' idiots are going to keep making the same mistake?
Well thought out post. The world is so cowed by your mental and literary awesomeness that I think the heavens just opened up and showed the entire Eastern Seaboard the second coming. All because God was jealous of the display of pure distilled intellect.
Actually not. The PCI implementations are different now then they were when the case was filed. It is fairly improbable that currently used buffering and switching techniques, which are are far faster than the ones sued over, would still be covered. It would take another entirely new lawsuit to cover that topic if Apple say's they changed something in response to the original suit.
That's why I said "if". I don't keep up on these techniques anymore.
The patent system seems to be rigged for abuse, especially if you have to patent something which is later marketed as intuitive. In some cases, gestures are patented, which I really don't get, but it's backed up by legalese about how it's supposed to be good for the electronic device.
Not for the device, for the user.
Disk reads, in comparison to CPU speeds, are very very very slow. Consider a 2GHz CPU, and a disk with 10ms seek times (that is it takes 10ms to find what one is looking for on a disk). 2GHz is 2 billion Hz, which means that the CPU "loops" 2 billion times per second. 10ms is one 100th of a second.
When a program wants to read from a disk, if it were to just wait this means that the 10ms seek time would mean that roughly 20 million CPU cycles would be wasted. So instead what happens is the operating system tells the disk what it wants to read and where it wants it written in RAM, and then the disk does all of the work copying data from the disk to RAM and tells the CPU when it's done. This way, the CPU can use those 20 million cycles for something else. This is called DMA.
There is then the idea of predictive caching, where if you can guess what the user wants to read, you can save them time. For example, normally when a program reads a file, if it reads the first kilobyte of the file, there is a good chance that it will want to read the next kilobyte as well. The OS cannot know this for sure, so it says "Well, you're probably going to read the next kilobyte, so let's read it from the disk." If the OS guesses correctly, everything runs faster. It's a similar idea to how if I were to say "Good mor" you could guess with reasonable certainty that the next three letters would be "ning".
Of course, you can't do this perfectly, but there are very good algorithms for this. Now, what OPTi has patented seems to be somewhat more complicated than this, but this is sort of the basic idea behind it. And there's the whole issue of willfulness - if Apple knew about this but chose to do it anyway (for whatever reasons) they were quite wrong.
But $19 million is like what? Profits from a couple of days sales at best.
I didn't read the boring article. I just saw that Apple was slapped with another fine lawsuit.
ahhh. with that inquisitive mind, your education will advance by leaps and bounds.
but it's not too late for you to change. i'm pulling for you...
Lose!
Lose lose lose lose lose lose lose lose lose.
Not loose. Loose is your girlfriend on a Friday night with the football team.
Lose goddamnit!
How many other so-called 'educated' idiots are going to keep making the same mistake?
SEE! As has been suggested by others here, we CAN educate in the process. It's about time for the forum to use its power.
If Texas secedes, can we put them in a basket then? By "we" I mean "The United States of America."
Yes, but they won't be able to keep the name Texas anymore, the USA owns the patent for that.
For anyone wondering with this idea of DMA, or Direct Memory Access.
Disk reads, in comparison to CPU speeds, are very very very slow. Consider a 2GHz CPU, and a disk with 10ms seek times (that is it takes 10ms to find what one is looking for on a disk). 2GHz is 2 billion Hz, which means that the CPU "loops" 2 billion times per second. 10ms is one 100th of a second.
When a program wants to read from a disk, if it were to just wait this means that the 10ms seek time would mean that roughly 20 million CPU cycles would be wasted. So instead what happens is the operating system tells the disk what it wants to read and where it wants it written in RAM, and then the disk does all of the work copying data from the disk to RAM and tells the CPU when it's done. This way, the CPU can use those 20 million cycles for something else. This is called DMA.
There is then the idea of predictive caching, where if you can guess what the user wants to read, you can save them time. For example, normally when a program reads a file, if it reads the first kilobyte of the file, there is a good chance that it will want to read the next kilobyte as well. The OS cannot know this for sure, so it says "Well, you're probably going to read the next kilobyte, so let's read it from the disk." If the OS guesses correctly, everything runs faster. It's a similar idea to how if I were to say "Good mor" you could guess with reasonable certainty that the next three letters would be "ning".
Of course, you can't do this perfectly, but there are very good algorithms for this. Now, what OPTi has patented seems to be somewhat more complicated than this, but this is sort of the basic idea behind it. And there's the whole issue of willfulness - if Apple knew about this but chose to do it anyway (for whatever reasons) they were quite wrong.
But $19 million is like what? Profits from a couple of days sales at best.
This case isn't about DMA, DMA has been around for about 40 years now. DMA is also a bit different than you describe here, but for the purposes of this thread DMA doesn't matter and I'm passing on the opportunity to pontificate more on that detail.
The case is about predictive cache snooping, just one of the ways to make a bus act faster than it is signal-wise. These techniques change over time. According to the original story "OPTi argued that virtually every Mac sold infringes on the patent".
Well if every Mac sold, over 5-7 years, actually willfully infringed the judgement wold have been in the 100s of millions at a minimum. It wasn't so the judge did make a rather significant narrowing of the original complaint showing there is more than one legally acceptable way to do cache snooping.
Since there was no court imposed ongoing royalties on top of the $19 million it is also reasonable to conclude that the judged infringement is in the past - technology did it's thing and moved on.
The legal system is done with the case unless Apple decides to appeal, and that isn't a legal decision as our little loser lover thinks it is. It is strictly a business decision at this point. A business decision dealing with small enough, actually trivial in this instance, worst case losses and large enough upside to warrant pursuing an appeal purely for principle. A business decision like that isn't driven by accountants dollars and cents analysis. It is driven by personality and moment-by-moment emotional state of the decision maker.
I guess NanoAkron just hasn't been around top-level decision makers on a bad day to see the pissant things they will do when they are in a bad mood and some unlucky or stupid moron crosses them at that moment. It can be quite fun to stand back and watch the ensuing rug-dances and policy-dagger twisting. Business is a cruel mistress and lawyers are just tools of the trade.
This case isn't about DMA, DMA has been around for about 40 years now. DMA is also a bit different than you describe here, but for the purposes of this thread DMA doesn't matter and I'm passing on the opportunity to pontificate more on that detail.
The case is about predictive cache snooping, just one of the ways to make a bus act faster than it is signal-wise. These techniques change over time. According to the original story "OPTi argued that virtually every Mac sold infringes on the patent".
Well if every Mac sold, over 5-7 years, actually willfully infringed the judgement wold have been in the 100s of millions at a minimum. It wasn't so the judge did make a rather significant narrowing of the original complaint showing there is more than one legally acceptable way to do cache snooping.
Since there was no court imposed ongoing royalties on top of the $19 million it is also reasonable to conclude that the judged infringement is in the past - technology did it's thing and moved on.
The legal system is done with the case unless Apple decides to appeal, and that isn't a legal decision as our little loser lover thinks it is. It is strictly a business decision at this point. A business decision dealing with small enough, actually trivial in this instance, worst case losses and large enough upside to warrant pursuing an appeal purely for principle. A business decision like that isn't driven by accountants dollars and cents analysis. It is driven by personality and moment-by-moment emotional state of the decision maker.
I guess NanoAkron just hasn't been around top-level decision makers on a bad day to see the pissant things they will do when they are in a bad mood and some unlucky or stupid moron crosses them at that moment. It can be quite fun to stand back and watch the ensuing rug-dances and policy-dagger twisting. Business is a cruel mistress and lawyers are just tools of the trade.
I agree with everything you've said except for one thing. The judge can't impose royalties. The companies must work that out for themselves. That becomes a separate civil contractual issue. If the two companies can't agree on that, then an arbitrator is usually agreed to. But as often as not, the two parties leave the table with nothing accomplished, while the infringer works out a different methodology to bypass the patent, which is almost always possible in software disputes.
I agree with everything you've said except for one thing. The judge can't impose royalties. The companies must work that out for themselves. That becomes a separate civil contractual issue. If the two companies can't agree on that, then an arbitrator is usually agreed to. But as often as not, the two parties leave the table with nothing accomplished, while the infringer works out a different methodology to bypass the patent, which is almost always possible in software disputes.
Wrong. A judge can impose royalties and a compulsory license. Also, an infringer can work out a different methodology, but the patent owner can issue patent continuances that cover anticipated, different methodologies.
Wrong. A judge can impose royalties and a compulsory license. Also, an infringer can work out a different methodology, but the patent owner can issue patent continuances that cover anticipated, different methodologies.
They can not. They can only order, at best, that both parties work out a solution.
If you think otherwise, show some case law.
Apple loves to sue others and warn others of not to use their patents. Apple should listen to themselves. Glad they lost 19$ million.
Whatever Doosh, OPTi doesn't actually make anything they just sit on patents which they will never use nor are they capable of using.
Not all Texans are stupid!
Quote:
Originally Posted by hillstones
This court case is an example of how stupid the residents of Texas really are.
Please don't put all Texans in one basket - I am embarrassed every time a patent case goes to the court in Marshall. It seems the court almost always sides on the patent holders side (which more and more seem to be patent warehousers).
We need patent reform but we also need judicial reform. Lets not let everybody that wants to file a case in Marshall, TX cause the idiots over there have such a high percentage win rate for these "ambulance chasing" patent attorneys.
jOhn
Give Texas back to Mexico!
Maybe this guy's 10 gallon hat is on too tight!
jOhn