Yep, this guy is reinventing the wheel, and has no clue how processors are designed. He only knows about assembly language, but the impact of his proposed changes to the micro-architecture [which is what determines the frequency and overall performance of you processor]
One more mode to artificially extend the register space will make the decoding of instructions more complex, the bypassing of data harder, and the dependencies computations for multiple instruction issue (a critical part of the processor) slower...
And you don't even gain much at all out of this! As you pointed out, any RISC architecture is better than that... but no-one has the fabs that Intel has, or the resources to put so many engineers to optimize the chip to such performance levels.
But don't get it wrong. If the Pentiums are the best processors around, it's in spite of the x86 instruction set.
What is your ultimate motivation in life? Why do you keep following the high-tech industry? Are you simply a geek or do you see yourself interested in completely different things as time evolves? Would you define yourself as a 21st century "humaniste"?
Don't get me wrong, I agree with you and place the 1st amendment on the top of my list but...
"which part of 'Congress shall make no law'" is not a correct argument. The 1st amendment is not the only sentence in the US constitution. Therefore, 'Congress shall make no law' applies as long as there is no other sentence somewhere else in the constitution that says Congress should do something.
For instance, if the constitution said that Congress can pass laws to protect decency, the 1st amendment would be de facto restricted.
In effect, the constitution is a bunch of principles we'd like to have, and let the courts balance them whenever they conflict. It is only in that context that the "shall make no law" shows its strength: it clearly was not written with exceptions in mind, therefore court tend to give it a high priority, which is great! But it's not an absolute.
Apple dumped Motorola with the clones fiasco. Since then, Motorola understood that their only friend was the embedded market, and they stopped trying to fight Intel on frequency.
Now you know why designers think the x86 architecture is messed-up: with roughly 50 times more engineering, it only gets at best 50% faster than a clean architecture designed by a small group.
You have no idea how the size of the design group at Intel and Motorola compare. They don't compare!
"not obvious to a person skilled in the art" is the key sentence that should decide whether a patent gets granted or not. Most software patents are obvious and should have been rejected, not because they are software, but because they are obvious. If we applied the law, we wouldn't be in such a mess.
Why could Ford patent "round headlights" and prevent other car manufacturers from using such headlights for 20 years more easily than a new software algorithm that noone had ever thought of before?
If you invent something that is truly new, you deserve to own it [for a while].
Good point! This actually reminds me of how it used to work on the Lisa (the Macintosh didn't keep that part of the design).
So let's say in the dock (task bar), there is this picture of an empty letter. It's pretty obvious that if you click on that, you will be able to start writing a new letter. The fact that Word is launched in the background is irrelevant.
Now hear this: once you start typing a little bit in this letter, a new icon is created in the dock, with a picture of your current letter. Therefore the dock now contains two "letters": the one you are currently typing, which should be easily distinguishable from the blank one that you can still use, would you want to work on a second letter.
How you decide whether to use Word or another competing program could all be in the appearance of the blank letter. Marketing could express itself nicely so that a Word "create a letter" icon would look more fancy than the simpletext equivalent, and so on... It's kind of like having an ad on the user's desktop!
running programs should look different from program launchers in my opinion. (That's a mistake I think OSX makes, kind of mixing the two)
Wrong! You have an "old" mindset: why does it matter if a program is in main memory (launched) or on disk? From my dumb user perspective, I couldn't care less. What matters is that when I click on the icon of the application, it comes to the front. Whether it has to be launched from disk or already happens to be loaded is irrelevent.
Makes sense?
Alain.
Slashdot Mirror
One more mode to artificially extend the register space will make the decoding of instructions more complex, the bypassing of data harder, and the dependencies computations for multiple instruction issue (a critical part of the processor) slower...
And you don't even gain much at all out of this! As you pointed out, any RISC architecture is better than that... but no-one has the fabs that Intel has, or the resources to put so many engineers to optimize the chip to such performance levels.
But don't get it wrong. If the Pentiums are the best processors around, it's in spite of the x86 instruction set.
Alain.
What is your ultimate motivation in life? Why do you keep following the high-tech industry? Are you simply a geek or do you see yourself interested in completely different things as time evolves? Would you define yourself as a 21st century "humaniste"?
A
B
"which part of 'Congress shall make no law'" is not a correct argument. The 1st amendment is not the only sentence in the US constitution. Therefore, 'Congress shall make no law' applies as long as there is no other sentence somewhere else in the constitution that says Congress should do something.
For instance, if the constitution said that Congress can pass laws to protect decency, the 1st amendment would be de facto restricted.
In effect, the constitution is a bunch of principles we'd like to have, and let the courts balance them whenever they conflict. It is only in that context that the "shall make no law" shows its strength: it clearly was not written with exceptions in mind, therefore court tend to give it a high priority, which is great! But it's not an absolute.
Alain.
Apple dumped Motorola with the clones fiasco. Since then, Motorola understood that their only friend was the embedded market, and they stopped trying to fight Intel on frequency.
Now you know why designers think the x86 architecture is messed-up: with roughly 50 times more engineering, it only gets at best 50% faster than a clean architecture designed by a small group.
You have no idea how the size of the design group at Intel and Motorola compare. They don't compare!
Alain.Software is not the problem: obviousness is!
"not obvious to a person skilled in the art" is the key sentence that should decide whether a patent gets granted or not. Most software patents are obvious and should have been rejected, not because they are software, but because they are obvious. If we applied the law, we wouldn't be in such a mess.
Why could Ford patent "round headlights" and prevent other car manufacturers from using such headlights for 20 years more easily than a new software algorithm that noone had ever thought of before?
If you invent something that is truly new, you deserve to own it [for a while].
Knuth's letter is very weak in that respect.
Alain.
So let's say in the dock (task bar), there is this picture of an empty letter. It's pretty obvious that if you click on that, you will be able to start writing a new letter. The fact that Word is launched in the background is irrelevant.
Now hear this: once you start typing a little bit in this letter, a new icon is created in the dock, with a picture of your current letter. Therefore the dock now contains two "letters": the one you are currently typing, which should be easily distinguishable from the blank one that you can still use, would you want to work on a second letter.
How you decide whether to use Word or another competing program could all be in the appearance of the blank letter. Marketing could express itself nicely so that a Word "create a letter" icon would look more fancy than the simpletext equivalent, and so on... It's kind of like having an ad on the user's desktop!
Alain.
Wrong! You have an "old" mindset: why does it matter if a program is in main memory (launched) or on disk? From my dumb user perspective, I couldn't care less. What matters is that when I click on the icon of the application, it comes to the front. Whether it has to be launched from disk or already happens to be loaded is irrelevent. Makes sense? Alain.