First off, none of the low-power states C0->C4 stash to a system management RAM (yet). Second, the lower Cx states flush the cache, but they don't flush in response to heat, in that case they perform a Geyserville transaction which lowers frequency and voltage. Only if you exceed the thermal diode does it go tits-up. Now there's word it may save state in future Cx states, but I sincerely doubt anyone would be able to get inside the on-die ram, since it will sit behind control register busses. You might as well hack the ucode -- which hasn't been done in over 13 years of existence.
Getting back to the low-power C4 states, this is the same thing that occurs when HALT is executed. If someone has the technology to invade motherboard ram, or to invade the page table, then of course they can break this, but it is by NO MEANS TIED JUST TO INTEL CHIPS.
This idiot spouts off two paragraphs of something he barely understands, and then fills the rest with panicked-fluff.
But AMD went off and did their own implementation of Intel's Vanderpol (VT) secure virtualization.
So either they want a fight, or they want to save face.
But you are 100% correct: anything Intel invent's AMD can use, and vice versa. They only caveat is that they are not required to supply implementation details, just patented methods.
I'd just let AMD be part of the venture and laugh. It's a double whammy: use fab capacity to ship near-zero profit SKUs of its latest-and-greatest embedded processor instead of CPUs it can earn a profit from. I can't believe AMD went along with it to begin with. At least Intel is demonstrating business acumen by trying to earn a profit.
Sooner or later you have to branch, or speculate, or go to main memory (or *gasp* I/O). So exploiting parallelism in arbitrary code is quite a challenge. Having worked on optimizing the linpack benchmark for a small "supercomputer" (64 CPUs) during college, it is quite a bit of work to decompose a problem this way.
However, the trend is not "well written" parallel apps, but rather parallel process scheduling: like antivirus + network stack + device drivers + MP3 player + excel + VNC + compiler... all getting their own processor core.
Dual-core will be the new commodity in 2006. Manufacturing costs for a die size with dual core is nil, so it is a bonus (die size of Core Duo is less than P4).
Um, OK... let's now apply your half-assed knowledge:
a) modern pop music already massively overcompressed due to the studio trend of squeezing something into every frequency range (there is very little dynamic range in modern music)
b) the problem is due to compensating for high ambient dB by increasing the player's volume
b) compressing 120 dB of your favorite pop music is still 120 dB of volume because if you compress it so that there are no "dangerous" peaks, you have a DC signal. duuuhhhhh.
The primary critics of video games are the people that do not play them.
Yeah, and how many propenents of video games don't have kids? Exactly.
I know plenty of gamers who think GTA goes way over the top for something targeted at kids (ratings aside, they know their primary audience). They also think parents aren't educated enough, or are too fucking lazy, so we all end up suffering for the sake of the fuckup parents.
Gamers tend to become elitist snobs to anyone who brings up regulation of video games. That's the wrong way to affect change. Maybe this political party will smack some sense into more than a few people and realize some of these games are violence porn (How many opponents of videogame regulation would buy their kid a Hustler? Raise your hands high! Thought so.)
the nastiest, sleaziest marketers in the business,
Do you know any, or are you just making shit up? 'cause boy oh boy, those bunny people are so sleazy.
I think maybe you are confusing "marketing" with "volume pricing" and "price fixing for preferred customers".
Intel's marketing is painfully benign compared to just about every other industry, like the Marlboro (cigarrets are cool), or Kellogs (sugary cereal for kids is part of a balanced breakfast), or McDonalds, or British Petroleum (we care about the environment), or Ford (our new suv is "green"). Now THOSE are some sleazy marketing examples, lying to kids qualifies.
Intel has ditched the p4 architecture and are focusing specifically on low power.
If you actually read technical reviews (anandtech and tomshardware), you will see that the dual core Yonah hovers around 13 W when the USB2.0 is not active. With 4 core on the next process, it wouldn't surprise me if it had an average power of ~10W.
So, the frying eggs joke should be focused at AMD, and not Intel.
Easily solution if I was Diebold: I'd only hire Republican lawyers, analysts, fincial advisors, interns, janitors... top to bottom, all GOP loyals. There are no laws for discriminating against political party. Even if there were, just need a few background checks into voter registration.
Thanks for trying, but it is absolutely NOT perfectly clear what that means. You missed the point of my entire post. Stallman's wording is an example of "intellectual eliteness" getting in the way of progress. He thinks the concept is clear, but gives a poor analogy.
But hey, at least you weren't condescending!
Some people get this concept immediately, the vast majority do not (try asking your mother if she understands it).
Can I sell free software? Why or why not? What are the circumstances? People give away beer all the time (parties, sporting events...) so why isn't beer free? Speech costs money too, like speechwriters, or airtime on television, or mic time at a podium... and it can cost legal fees if I slander or yell "FIRE" in a cinema.
Free software is a matter of liberty?
So I have the liberty to give it away or sell it? Then it isn't always free if I can sell it...
...the propeller-heads that coined the term "free as in beer, not as in speech" thought it was so cute and witty that it didn't matter of the average Joe still can't understand the concept.
The overloaded naming convention is one of the largest obstacles to this concept. And it seemed like (prior to finally "getting it") whenever I asked an expert to explain it, I was spoken down to as an inferior for not grasping such and obvious concept.
Maybe a better communicator can rename the entire movement and save some confusion.
I recall a debate that adventure games were stagnantly based on Eurocentric Medieval world views. While many, many RPGs of the time were exactly that, I enjoyed Zelda because it was Japan's take odd take on this theme. Here's a guy dressed like Robin Hood with a sword, traipsing through dungeons and forests doing some of the things a questing hobbit might do, but that's where the similarities end. Zelda, like Pac Man, was an early indication of the fountainhead of cyber-culutre that would issue from Japan over the next two decades.
Since you've demonstrated a lack of reading comprehension I'll refer you to page 5 of your article. The AMD processors are clocked at 2 and 2.2 ghz. What's your point?
My point is, AMD is saying, look our 2.2 is so darned badass we'll call it 4800 to compete with intel's high GHz numbers. So on one hand AMD wants credit for running 2.2 GHz and still competing with Intel's 4.0 GHz CPU. Now here's the rub...
Today (on the the hand), AMD compares against Yonah 2.2GHz, and suddenly AMD wants to ignore the previous statement and say, well, we're really competitive because we're not 4800 MHz, we're really just 2.2 GHz, so it's fair. The fact that a mobile product is competeing with the flagship turns years of arguments on this board on their heads.
So I see this as: AMD proponents wants their cake and wants to eat it, too.
That's what I'm calling shenanegans on. (Well, I called "bloody stool" on it, but only because it sounded more edgy.)
CPUs have achieved a level of performance that makes them all essentially interchangable. Performance deltas in benchmarks typically run less than 5% and bottlenecks can be traced to other system components (notably graphics cards) depending on what's being exercised.
Hey hey this is supposed to be a flamewar, no rational comments please!
However, if anything has kicked bloody it'd be the previous generation Pentium parts and therein lies my disagreement with you.
No argument there. Haven't we been reading Tom/Anand's websites for like three years seeing intel consistently trying it's hardest to play the 1% performance game on the desktop? It gets tedious. I understand Intel can't just nuke a flagship product, or suddenly come out and say, "Uh, gee, we suck." No 30+billion dollar corporation in their right mind would do that. But it's squeezing every last drop out of a overdesigned architecture... It's the marketing that kills me. AMD made a smart decision to put a memcontroller on die, and that one simple step bought them 12~14+ months of performance domination. Underneath, both AMD and Intel archs are based on ISAs from 20 years ago, and the extent to which the ISA reflects the microarchitecture, there's really nothing new under the sun from either of them w.r.t. microarch (well, IA64 was a revolution, but look how much performance boost we got outta that!), and now it's all process process process => frequency. I'm really excited to see the race to the bottom-Watt, at least it's new. Where am I going with this... Oh, right. Yes, this originally just started out as a "case of the mondays troll", and now it's a brain dead zombied wednesday...
Itty, bitty mobile processor Yonah, at 2GHz, with no 64-bit extensions, kicks the bloody shit out of AMDs top of the line offering on almost all the benchmarks. Can't wait to see their 64bit mobile part in a few months.
It may be in recognition that AMD64 chips will eventually be able to outperform even SSE3 optimized Intel code, if they cannot already.
Wait, SSE3 is FP, which is 80 to 128bit, or 64bit integer packed using the FP stack. How can AMD64 matter when *64 in AMD marketing refers to integer? AMD only has FP performance advantage.
Unless it Skype is pure integer, I don't see how your argument is valid on the bases of 64 bits.
Slashdot Mirror
...technical articles.
Where to begin.
First off, none of the low-power states C0->C4 stash to a system management RAM (yet). Second, the lower Cx states flush the cache, but they don't flush in response to heat, in that case they perform a Geyserville transaction which lowers frequency and voltage. Only if you exceed the thermal diode does it go tits-up. Now there's word it may save state in future Cx states, but I sincerely doubt anyone would be able to get inside the on-die ram, since it will sit behind control register busses. You might as well hack the ucode -- which hasn't been done in over 13 years of existence.
Getting back to the low-power C4 states, this is the same thing that occurs when HALT is executed. If someone has the technology to invade motherboard ram, or to invade the page table, then of course they can break this, but it is by NO MEANS TIED JUST TO INTEL CHIPS.
This idiot spouts off two paragraphs of something he barely understands, and then fills the rest with panicked-fluff.
Let's all move along. Nothing to see here.
But AMD went off and did their own implementation of Intel's Vanderpol (VT) secure virtualization.
So either they want a fight, or they want to save face.
But you are 100% correct: anything Intel invent's AMD can use, and vice versa. They only caveat is that they are not required to supply implementation details, just patented methods.
Very strange bedfellows.
Hmm, you certainly do sound like quite the authority, and insightful post has changed my mind. I thank you.
Retard.
I'd just let AMD be part of the venture and laugh. It's a double whammy: use fab capacity to ship near-zero profit SKUs of its latest-and-greatest embedded processor instead of CPUs it can earn a profit from. I can't believe AMD went along with it to begin with. At least Intel is demonstrating business acumen by trying to earn a profit.
These ARE companies after all folks, not martyrs.
properly written programs
Almost.
Sooner or later you have to branch, or speculate, or go to main memory (or *gasp* I/O). So exploiting parallelism in arbitrary code is quite a challenge. Having worked on optimizing the linpack benchmark for a small "supercomputer" (64 CPUs) during college, it is quite a bit of work to decompose a problem this way.
However, the trend is not "well written" parallel apps, but rather parallel process scheduling: like antivirus + network stack + device drivers + MP3 player + excel + VNC + compiler... all getting their own processor core.
That's the future, baby.
That's a very good point: we all took it on identical T43p IBM laptops, with the same corporate configuration. We'll try again on various desktops...
Four of my friends took it and we all got 27!!!
Something smells rotten in Denmark...
Dual-core will be the new commodity in 2006. Manufacturing costs for a die size with dual core is nil, so it is a bonus (die size of Core Duo is less than P4).
Um, OK... let's now apply your half-assed knowledge:
a) modern pop music already massively overcompressed due to the studio trend of squeezing something into every frequency range (there is very little dynamic range in modern music)
b) the problem is due to compensating for high ambient dB by increasing the player's volume
b) compressing 120 dB of your favorite pop music is still 120 dB of volume because if you compress it so that there are no "dangerous" peaks, you have a DC signal. duuuhhhhh.
...involved with serious patent law. WTF? Next we'll have an actor as president of the free world... oh, wait...
The primary critics of video games are the people that do not play them.
Yeah, and how many propenents of video games don't have kids? Exactly.
I know plenty of gamers who think GTA goes way over the top for something targeted at kids (ratings aside, they know their primary audience). They also think parents aren't educated enough, or are too fucking lazy, so we all end up suffering for the sake of the fuckup parents.
Gamers tend to become elitist snobs to anyone who brings up regulation of video games. That's the wrong way to affect change. Maybe this political party will smack some sense into more than a few people and realize some of these games are violence porn (How many opponents of videogame regulation would buy their kid a Hustler? Raise your hands high! Thought so.)
I'm surprised at the cost alone. $16 mil for paper systems, $90 mil for electronic? Something is missing in that article...
the nastiest, sleaziest marketers in the business,
Do you know any, or are you just making shit up? 'cause boy oh boy, those bunny people are so sleazy.
I think maybe you are confusing "marketing" with "volume pricing" and "price fixing for preferred customers".
Intel's marketing is painfully benign compared to just about every other industry, like the Marlboro (cigarrets are cool), or Kellogs (sugary cereal for kids is part of a balanced breakfast), or McDonalds, or British Petroleum (we care about the environment), or Ford (our new suv is "green"). Now THOSE are some sleazy marketing examples, lying to kids qualifies.
Intel has ditched the p4 architecture and are focusing specifically on low power.
If you actually read technical reviews (anandtech and tomshardware), you will see that the dual core Yonah hovers around 13 W when the USB2.0 is not active. With 4 core on the next process, it wouldn't surprise me if it had an average power of ~10W.
So, the frying eggs joke should be focused at AMD, and not Intel.
Easily solution if I was Diebold: I'd only hire Republican lawyers, analysts, fincial advisors, interns, janitors... top to bottom, all GOP loyals. There are no laws for discriminating against political party. Even if there were, just need a few background checks into voter registration.
Except when there is no balance of power, and the authorities are in on it.
What secret gov't agency could this guy have gone to that wouldn't have covered it up, since they are all staffed by Bush cronies?
Thanks for trying, but it is absolutely NOT perfectly clear what that means. You missed the point of my entire post. Stallman's wording is an example of "intellectual eliteness" getting in the way of progress. He thinks the concept is clear, but gives a poor analogy.
But hey, at least you weren't condescending!
Some people get this concept immediately, the vast majority do not (try asking your mother if she understands it).
Can I sell free software? Why or why not? What are the circumstances? People give away beer all the time (parties, sporting events...) so why isn't beer free? Speech costs money too, like speechwriters, or airtime on television, or mic time at a podium... and it can cost legal fees if I slander or yell "FIRE" in a cinema.
Free software is a matter of liberty?
So I have the liberty to give it away or sell it? Then it isn't always free if I can sell it...
See? A very poor analogy.
...the propeller-heads that coined the term "free as in beer, not as in speech" thought it was so cute and witty that it didn't matter of the average Joe still can't understand the concept.
The overloaded naming convention is one of the largest obstacles to this concept. And it seemed like (prior to finally "getting it") whenever I asked an expert to explain it, I was spoken down to as an inferior for not grasping such and obvious concept.
Maybe a better communicator can rename the entire movement and save some confusion.
Eh, in the words of the ZIL interpreter:
Not bloody likely.
Cool! Now Linux desktops can be as annoying as Windows XP.
...how am I supposed to feel about this again???
No, wait.
Cool! Now Linux desktops can compete with Windows XP.
No, wait...
I recall a debate that adventure games were stagnantly based on Eurocentric Medieval world views. While many, many RPGs of the time were exactly that, I enjoyed Zelda because it was Japan's take odd take on this theme. Here's a guy dressed like Robin Hood with a sword, traipsing through dungeons and forests doing some of the things a questing hobbit might do, but that's where the similarities end. Zelda, like Pac Man, was an early indication of the fountainhead of cyber-culutre that would issue from Japan over the next two decades.
Since you've demonstrated a lack of reading comprehension I'll refer you to page 5 of your article. The AMD processors are clocked at 2 and 2.2 ghz. What's your point?
My point is, AMD is saying, look our 2.2 is so darned badass we'll call it 4800 to compete with intel's high GHz numbers. So on one hand AMD wants credit for running 2.2 GHz and still competing with Intel's 4.0 GHz CPU. Now here's the rub...
Today (on the the hand), AMD compares against Yonah 2.2GHz, and suddenly AMD wants to ignore the previous statement and say, well, we're really competitive because we're not 4800 MHz, we're really just 2.2 GHz, so it's fair. The fact that a mobile product is competeing with the flagship turns years of arguments on this board on their heads.
So I see this as: AMD proponents wants their cake and wants to eat it, too.
That's what I'm calling shenanegans on. (Well, I called "bloody stool" on it, but only because it sounded more edgy.)
CPUs have achieved a level of performance that makes them all essentially interchangable. Performance deltas in benchmarks typically run less than 5% and bottlenecks can be traced to other system components (notably graphics cards) depending on what's being exercised.
Hey hey this is supposed to be a flamewar, no rational comments please!
However, if anything has kicked bloody it'd be the previous generation Pentium parts and therein lies my disagreement with you.
No argument there. Haven't we been reading Tom/Anand's websites for like three years seeing intel consistently trying it's hardest to play the 1% performance game on the desktop? It gets tedious. I understand Intel can't just nuke a flagship product, or suddenly come out and say, "Uh, gee, we suck." No 30+billion dollar corporation in their right mind would do that. But it's squeezing every last drop out of a overdesigned architecture... It's the marketing that kills me. AMD made a smart decision to put a memcontroller on die, and that one simple step bought them 12~14+ months of performance domination. Underneath, both AMD and Intel archs are based on ISAs from 20 years ago, and the extent to which the ISA reflects the microarchitecture, there's really nothing new under the sun from either of them w.r.t. microarch (well, IA64 was a revolution, but look how much performance boost we got outta that!), and now it's all process process process => frequency. I'm really excited to see the race to the bottom-Watt, at least it's new. Where am I going with this... Oh, right. Yes, this originally just started out as a "case of the mondays troll", and now it's a brain dead zombied wednesday...
Yessir, that'd be kicking the bloody shit. Yup. Oh yes.
Yup, 2.0GHz vs. AMD marketing machine's 4800+. Thanks for proving my point.
You AMD fanboys have the shortest memories every.
Well, there is something called Anti-Trust to keep capitalism fair. I think that's the issue here.
But check this out:
http://www.anandtech.com/cpuchipsets/showdoc.aspx
Itty, bitty mobile processor Yonah, at 2GHz, with no 64-bit extensions, kicks the bloody shit out of AMDs top of the line offering on almost all the benchmarks. Can't wait to see their 64bit mobile part in a few months.
It may be in recognition that AMD64 chips will eventually be able to outperform even SSE3 optimized Intel code, if they cannot already.
Wait, SSE3 is FP, which is 80 to 128bit, or 64bit integer packed using the FP stack. How can AMD64 matter when *64 in AMD marketing refers to integer? AMD only has FP performance advantage.
Unless it Skype is pure integer, I don't see how your argument is valid on the bases of 64 bits.
You should read this post:
7 11403
http://slashdot.org/comments.pl?sid=177277&cid=14