OK, so all those snarky posts about how the U.S. is a worthless third world country because Foxconn was going to open factories here? GUESS WHAT: Now the U.S. is a worthless third world country because Foxconn isn't going to open factories here!
That's the beuatiful thing about already have a predetermined conclusion, the facts always fit.
Funny you should hate on Medfield when a Razer I with Medfield has better battery life than a krait Razer M with the exact same screen and battery. But it looks like you never let facts get in the way of your koolaid.
You know what I love? When the exact same people who say that an Intel workstation with a 6-core CPU being used for heavy compiling/CAD/etc. etc. is "wasted" and "overkill" but that a 256 core ARM chip on your cellphone will be insanely great... because... uh... Angry Birds is the most parellizable program in human history?
Haswell is a (probably) ~1.6 Billion transistor chip that obviously costs more than a SoC that is really designed for tablets. Interesting then that a ~1.6 Billion transistor chip that includes similar functionality to the SoC uses about the same amount of power as that tablet SoC while including vastly more performance.
If you want cheap, Atoms are already out now that are quite cost competitive with ARM chips, and 22nm Atoms will be out next year.
Oh and as for "release dates" the Exynos has just barely begun to reach the market and Haswell will be out and about at around the same time that most Cortex A-15s really come into the market as well. Considering I've had to listen to "A15 will kill Intel!!!!" for over 2 years as if they were already coming out of faucets like water, I'm not too worried about part availability.
So here we are in the ARM vs. Intel Evolution: 2008: ARM is superior, Intel can NEVER scale its power consumption down below 100 watts!!
2009-2010: ARM is still superior! Atom sucks at performance and uses 10 WHOLE WATTS, thats more than 10X ARM! The Cortex A9 will annihilate Intel!
2011: ARM performance dominance is just around the corner! Ignore those useless benchmarks of Cortex A9 vs. Atom! So what if Atom has higher performance, IT SUCKS DOWN MORE POWER AND POWER CONSUMPTION IS ALL THAT MATTERS!
2012: Medfield sucks! Who cares if it gets better battery life than a dual-core 28nm Krait when put into Motorala Razers with the exact same! See, we have benchmarks where the higher-clocked Krait gets 10% better performance (in some benchmarks while losing in others that we ignore)! WHO CARES THAT ATOM IS MORE POWER EFFICIENT THE ONLY THING THAT MATTERS IS MORE PERFORMANCE! INTEL IS STILL OVERPRICED EVEN THOUGH THE RAZER I AND RAZER M HAVE THE SAME PRICE!
2013: Uh... at least ARMs are cheap when you intentionally compare chips desiged for cellphones to Intel's desktop chips and pretend that Atom doesn't exist. ARM WILL DESTROY INTEL!
Really.. please show me the same demo running on... let's say... the A6 on the newest iPad. I'm sure it will get *much* better framerates.... (or not).
You completely missed the point of my post, but I can see that Intel is gradually working its way up Ghandi's list. It's getting to be between the "then they laugh at you" and the "then they fight you" stages....
Good thing then that Haswell's idle power draw is 20x better than Ivy Bridge's, meaning that it is probably about the same as the Cortex A-15 (or maybe even better).
I'm not saying that Haswell belongs in a smartphone.. I'm also saying that unless you downclock that Exynos you don't want it in a smartphone either. I *am* saying that the blind assumption that ARM == efficiency tends to disintegrate when confronted with facts. I'm also saying that if Haswell can run at 8 watts, the whole "x86 wastes powar!" line is going to sound pretty silly when next-generation 22nm Atoms show up.
No, I'm saying that on a chromebook with a SoC (that stands for "system on a chip" you know...) the total power consumption of the SoC running a web benchmark that likely requires little or no wireless network power due to caching is equivalent to the power consumption of a low-power Haswell part (that is similar to a SoC but with a separate south-bridge MCM).
Oh, and if the Kraken benchmark is anything remotely similar to any other web browser benchmark I've ever seen, the CPU/GPU on the SoC are not being taxed at 100% for the whole time the benchmark is running, so an *average* power consumption of 8 watts is even *worse* than it looks compared to the Heaven benchmark which is one of the most CPU & GPU intensive benchmarks available anywhere.
Thanks for making me think about the comparison a bit more, it looks like I was being a little too nice to ARM when I said the power consumption was "equal" when in fact it is likely skewed in ARM's favor due to the lighter workload on the ARM chip that would consume even more power if truly stressed.
" efficiency now matters more than gains in raw performance"
Sure, so why don't you start off by telling us why an Exynos Cortex A-15 chip running a web benchmark is using about 8 watts of power, with the display turned off so only SoC power is being measured, while Intel has already demoed a full-blown Haswell running Unigine Heaven at... 8 watts.
So when the miraculous Cortex A-15 uses the same amount of power as the supposedly "bloated" x86 Haswell, while Haswell is running a benchmark that is massively more intensive than a web-browser test, who is really making the most "efficient" platform?
On the Google+ thread there are some recommendations for Yakuake, which Linus might find useful since I'm sure he does quite a bit of work from the terminal.
And so now, scientists and engineers look at data and suggest that infrastructure may not be adequate to provide safety margins for possible weather conditions -- and you imply that is alarmist!
No you idiot: I simply said that in the 1970's it was known that NY would be hit by hurricanes. That means that when something that is known to happen actually happens it is not OMG GLOBAL WARMING MITT ROMNEY IS EVIL AND HATES BLACK PEOPLE!!! Instead it's: Yeah, we knew this would happen and it did happen and we took steps to prepare for it and "global warming" shouldn't be used as a substitute for logical thought.
You do have a smoke detector in your house right? Because you know a fire could break out right? So if a fire breaks out that's automatically proof that Global Warming is going to kill all of us? Not really.
Yeah, back in the 1970's the Citigroup Center in New York needed an emergency retrofit due to a design flaw in bolts used to hold the building together. Basically, wind-shear from.. wait for it... a hurricane could topple the building. (http://en.wikipedia.org/wiki/Citigroup_Center)
So in the 1970's it was common knowledge that New York could and would be hit by hurricanes and it was considered a real enough threat that the engineers went on an emergency retrofitting job to fix the problem once it was discovered. In 2012 a CAT 1 Hurricane actually hits New York, which was 100% expected, and frankly weaker than predicted hurricanes that could hit New York. Of course these inconvenient facts won't deter the alarmist conclusion: GLOBAL WARMING!!!
[Posted from Firefox installed on a non-European computer, which apparently violates the laws of physics since Firefox usage share is zero outside of Europe since we don't have a "browser ballot" to do our thinking for us here]
Look at the article in detail. Isn't it funny how the A-15 is the super-miracle chip that is going to stick it to Intel in the server world! Oh wait.. now that super-miracle chip is the 64-bit ARM Miracle Chip (TM) and the A-15 has been relegated to smartphones instead of taking over the server world.
Fortunately, Intel is completely incapable of making any improvements to its chips whatsoever, so ARM's victory in 2014 is assured.
Oh wait.. the first real A15s just launched literally this month and except for Samsung they won't even be on sale from other manufacturers until next year.
Now we're going to be hearing non-stop about how the 64-bit ARMs will be here next Tuesday and take over the world and put Big-Bad Intel out of business so that Apple & Samsung can sell us non-modifiable devices with locked-down hardware apparently this is supposed to make Linux take over... somehow...
I was promised a NUKULAR OPOCALIPSE yesterday, and here we are with satellite images that don't even show the entire East Coast as a glowing radioactive wasteland.
I think that this complete lack of NUKULAR meltdown is 100% absolute irrefutable proof of two key concepts: 1. Capitalism is an abject failure and we need U.N. control of everything and everyone right now. 2. NUKULAR power is obviously far too dangerous and should be banned right now before somebody doesn't get killed again JUST LIKE FUKUSHIMA.
No... 5 years ago the top of the line *mobile* CPUs had 2 cores. Today the top of the line mobile CPUs have 4 much more powerful cores. It turns out that the mobile CPUs are now used on most desktops too, but there are still more powerful desktop CPUs.
If, however, you want the top of the line *desktop* CPU, the 3930K has 6 cores and wipes the floor with any quad core CPU in existence... assuming you are running software that can actually take advantage of all those cores. Since a whole lot of even "multithreaded" software cannot do so, you are basically not having to pay for unneeded silicon and power by getting a quad-core CPU.
So ITER is an international version of the Space Shuttle which was an intentionally lousy design that "succeeded" by maximizing the number of contractors in different congressional districts that got government $$$. The difference is that there was still enough residual talent left at NASA for the Space Shuttle to at least take flight. Not so much for ITER.
If Bill Gates really wanted to help the world, he'd take $30 Billion and make it a prize for whoever can get an operation fusion reactor running. No awards go to maximizing sub-contractor payouts in this scenario. Instead, success would actually be the objective of the project instead of failure + guaranteed taxpayer funded payouts for the next 40+ years.
Just before the AMD board executed him, Dirk Meyer screamed out: "MOAR COARZ!" And today we have Bulldozer.
The moral of the story is, MOAR != more all of the time. Especially in a freakin' cellphone, where, despite what some Slashdotters think, the primary use case is *not* performing massively parallel scientific simulations.
That's the problem though... the K20 is definitely not "general" but highly specialized. Throw the right type of problem through optimized CUDA code and it'll run great. Throw it the wrong type of computational problem and it'll go nowhere fast. That's specialized instead of general.
You have it backwards... graphics have been around since the time of the caveman. "Graph theory" only came into existence in the late 19th century and took its original cues from hand-drawn graphs..which are a type of graphics. Plus, adding and multiplying numbers, which is basically what the K20 does on a huge scale, is by no means an operation that is limited to graph theory.
GPU means graphical processing unit. Now consumer GPUs are pressed into service for compute tasks like BOINC & folding, but they are still GPUs (they can still do graphics).
Does Nvidia even bother to put in the graphics-specific silicon and output hardware on the K20, or should these things really be called.. I dunno.. "compute accelerators" or something like that?
Intel's 22nm transistors certainly do. The overall chips don't because the price differential between even a top-line 8 Core Sandy Bridge Xeon chip/system and the Power7 chips/systems that actually have the high-end performance you are talking about is similar to the price differential between the chip in my cellphone and the high-end Xeon chip.
I know guys that do CPU design for IBM and they will flat out tell you that Intel has a better process. The difference is that IBM is making chips for million dollar+ servers with huge legacy needs in markets where even Itanium isn't trying to compete. At that point, you can afford to design CPUs with 200+ watt TDPs and exotic liquid cooling systems that are made in tiny quantities compared to what Intel & AMD churn out.
Slashdot Mirror
OK, so all those snarky posts about how the U.S. is a worthless third world country because Foxconn was going to open factories here?
GUESS WHAT: Now the U.S. is a worthless third world country because Foxconn isn't going to open factories here!
That's the beuatiful thing about already have a predetermined conclusion, the facts always fit.
Funny you should hate on Medfield when a Razer I with Medfield has better battery life than a krait Razer M with the exact same screen and battery. But it looks like you never let facts get in the way of your koolaid.
You know what I love? When the exact same people who say that an Intel workstation with a 6-core CPU being used for heavy compiling/CAD/etc. etc. is "wasted" and "overkill" but that a 256 core ARM chip on your cellphone will be insanely great... because... uh... Angry Birds is the most parellizable program in human history?
Haswell is a (probably) ~1.6 Billion transistor chip that obviously costs more than a SoC that is really designed for tablets. Interesting then that a ~1.6 Billion transistor chip that includes similar functionality to the SoC uses about the same amount of power as that tablet SoC while including vastly more performance.
If you want cheap, Atoms are already out now that are quite cost competitive with ARM chips, and 22nm Atoms will be out next year.
Oh and as for "release dates" the Exynos has just barely begun to reach the market and Haswell will be out and about at around the same time that most Cortex A-15s really come into the market as well. Considering I've had to listen to "A15 will kill Intel!!!!" for over 2 years as if they were already coming out of faucets like water, I'm not too worried about part availability.
So here we are in the ARM vs. Intel Evolution:
2008: ARM is superior, Intel can NEVER scale its power consumption down below 100 watts!!
2009-2010: ARM is still superior! Atom sucks at performance and uses 10 WHOLE WATTS, thats more than 10X ARM! The Cortex A9 will annihilate Intel!
2011: ARM performance dominance is just around the corner! Ignore those useless benchmarks of Cortex A9 vs. Atom! So what if Atom has higher performance, IT SUCKS DOWN MORE POWER AND POWER CONSUMPTION IS ALL THAT MATTERS!
2012: Medfield sucks! Who cares if it gets better battery life than a dual-core 28nm Krait when put into Motorala Razers with the exact same! See, we have benchmarks where the higher-clocked Krait gets 10% better performance (in some benchmarks while losing in others that we ignore)! WHO CARES THAT ATOM IS MORE POWER EFFICIENT THE ONLY THING THAT MATTERS IS MORE PERFORMANCE!
INTEL IS STILL OVERPRICED EVEN THOUGH THE RAZER I AND RAZER M HAVE THE SAME PRICE!
2013: Uh... at least ARMs are cheap when you intentionally compare chips desiged for cellphones to Intel's desktop chips and pretend that Atom doesn't exist. ARM WILL DESTROY INTEL!
Really.. please show me the same demo running on... let's say... the A6 on the newest iPad. I'm sure it will get *much* better framerates.... (or not).
You completely missed the point of my post, but I can see that Intel is gradually working its way up Ghandi's list. It's getting to be between the "then they laugh at you" and the "then they fight you" stages....
Good thing then that Haswell's idle power draw is 20x better than Ivy Bridge's, meaning that it is probably about the same as the Cortex A-15 (or maybe even better).
I'm not saying that Haswell belongs in a smartphone.. I'm also saying that unless you downclock that Exynos you don't want it in a smartphone either. I *am* saying that the blind assumption that ARM == efficiency tends to disintegrate when confronted with facts. I'm also saying that if Haswell can run at 8 watts, the whole "x86 wastes powar!" line is going to sound pretty silly when next-generation 22nm Atoms show up.
No, I'm saying that on a chromebook with a SoC (that stands for "system on a chip" you know...) the total power consumption of the SoC running a web benchmark that likely requires little or no wireless network power due to caching is equivalent to the power consumption of a low-power Haswell part (that is similar to a SoC but with a separate south-bridge MCM).
Oh, and if the Kraken benchmark is anything remotely similar to any other web browser benchmark I've ever seen, the CPU/GPU on the SoC are not being taxed at 100% for the whole time the benchmark is running, so an *average* power consumption of 8 watts is even *worse* than it looks compared to the Heaven benchmark which is one of the most CPU & GPU intensive benchmarks available anywhere.
Thanks for making me think about the comparison a bit more, it looks like I was being a little too nice to ARM when I said the power consumption was "equal" when in fact it is likely skewed in ARM's favor due to the lighter workload on the ARM chip that would consume even more power if truly stressed.
" efficiency now matters more than gains in raw performance"
Sure, so why don't you start off by telling us why an Exynos Cortex A-15 chip running a web benchmark is using about 8 watts of power, with the display turned off so only SoC power is being measured, while Intel has already demoed a full-blown Haswell running Unigine Heaven at... 8 watts.
So when the miraculous Cortex A-15 uses the same amount of power as the supposedly "bloated" x86 Haswell, while Haswell is running a benchmark that is massively more intensive than a web-browser test, who is really making the most "efficient" platform?
Exynos Source: http://www.anandtech.com/show/6422/samsung-chromebook-xe303-review-testing-arms-cortex-a15/7
Haswell Demo Video: http://www.youtube.com/watch?v=cKvVdhkgAxg
I only pay for applications with bad money instead of good money, so I'm fine with the ads.
"With this round of CPUs, AMD has split its clock speed Turbo range into 'Max' and 'Max All Cores.' "
Remember Episode 200 of Stargate and the "Set Weapons to Maximum!" line?
On the Google+ thread there are some recommendations for Yakuake, which Linus might find useful since I'm sure he does quite a bit of work from the terminal.
And so now, scientists and engineers look at data and suggest that infrastructure may not be adequate to provide safety margins for possible weather conditions -- and you imply that is alarmist!
No you idiot: I simply said that in the 1970's it was known that NY would be hit by hurricanes. That means that when something that is known to happen actually happens it is not OMG GLOBAL WARMING MITT ROMNEY IS EVIL AND HATES BLACK PEOPLE!!! Instead it's: Yeah, we knew this would happen and it did happen and we took steps to prepare for it and "global warming" shouldn't be used as a substitute for logical thought.
You do have a smoke detector in your house right? Because you know a fire could break out right? So if a fire breaks out that's automatically proof that Global Warming is going to kill all of us? Not really.
Yeah, back in the 1970's the Citigroup Center in New York needed an emergency retrofit due to a design flaw in bolts used to hold the building together. Basically, wind-shear from.. wait for it... a hurricane could topple the building. (http://en.wikipedia.org/wiki/Citigroup_Center)
So in the 1970's it was common knowledge that New York could and would be hit by hurricanes and it was considered a real enough threat that the engineers went on an emergency retrofitting job to fix the problem once it was discovered. In 2012 a CAT 1 Hurricane actually hits New York, which was 100% expected, and frankly weaker than predicted hurricanes that could hit New York. Of course these inconvenient facts won't deter the alarmist conclusion: GLOBAL WARMING!!!
And obnoxiously dumb the second time!
[Posted from Firefox installed on a non-European computer, which apparently violates the laws of physics since Firefox usage share is zero outside of Europe since we don't have a "browser ballot" to do our thinking for us here]
Look at the article in detail. Isn't it funny how the A-15 is the super-miracle chip that is going to stick it to Intel in the server world! Oh wait.. now that super-miracle chip is the 64-bit ARM Miracle Chip (TM) and the A-15 has been relegated to smartphones instead of taking over the server world.
Fortunately, Intel is completely incapable of making any improvements to its chips whatsoever, so ARM's victory in 2014 is assured.
Hey Guys,
Don't you just love how it's 2010 and the Cortex A15 already out on the market!
http://www.electronics-eetimes.com/en/arm-in-servers-push-describes-the-cortex-a15-cpu.html?cmp_id=7&news_id=222903607
Oh wait.. the first real A15s just launched literally this month and except for Samsung they won't even be on sale from other manufacturers until next year.
Now we're going to be hearing non-stop about how the 64-bit ARMs will be here next Tuesday and take over the world and put Big-Bad Intel out of business so that Apple & Samsung can sell us non-modifiable devices with locked-down hardware apparently this is supposed to make Linux take over... somehow...
I was promised a NUKULAR OPOCALIPSE yesterday, and here we are with satellite images that don't even show the entire East Coast as a glowing radioactive wasteland.
I think that this complete lack of NUKULAR meltdown is 100% absolute irrefutable proof of two key concepts:
1. Capitalism is an abject failure and we need U.N. control of everything and everyone right now.
2. NUKULAR power is obviously far too dangerous and should be banned right now before somebody doesn't get killed again JUST LIKE FUKUSHIMA.
No... 5 years ago the top of the line *mobile* CPUs had 2 cores.
Today the top of the line mobile CPUs have 4 much more powerful cores.
It turns out that the mobile CPUs are now used on most desktops too, but there are still more powerful desktop CPUs.
If, however, you want the top of the line *desktop* CPU, the 3930K has 6 cores and wipes the floor with any quad core CPU in existence... assuming you are running software that can actually take advantage of all those cores. Since a whole lot of even "multithreaded" software cannot do so, you are basically not having to pay for unneeded silicon and power by getting a quad-core CPU.
So ITER is an international version of the Space Shuttle which was an intentionally lousy design that "succeeded" by maximizing the number of contractors in different congressional districts that got government $$$. The difference is that there was still enough residual talent left at NASA for the Space Shuttle to at least take flight. Not so much for ITER.
If Bill Gates really wanted to help the world, he'd take $30 Billion and make it a prize for whoever can get an operation fusion reactor running. No awards go to maximizing sub-contractor payouts in this scenario. Instead, success would actually be the objective of the project instead of failure + guaranteed taxpayer funded payouts for the next 40+ years.
Just before the AMD board executed him, Dirk Meyer screamed out: "MOAR COARZ!" And today we have Bulldozer.
The moral of the story is, MOAR != more all of the time. Especially in a freakin' cellphone, where, despite what some Slashdotters think, the primary use case is *not* performing massively parallel scientific simulations.
That's the problem though... the K20 is definitely not "general" but highly specialized. Throw the right type of problem through optimized CUDA code and it'll run great. Throw it the wrong type of computational problem and it'll go nowhere fast. That's specialized instead of general.
You have it backwards... graphics have been around since the time of the caveman. "Graph theory" only came into existence in the late 19th century and took its original cues from hand-drawn graphs..which are a type of graphics. Plus, adding and multiplying numbers, which is basically what the K20 does on a huge scale, is by no means an operation that is limited to graph theory.
"Nobody has a product that lets you work and play that can be your tablet and your PC. Not at any price point,"
That's actually a true statement. Ballmer's problem is that it is still a true statement after Surface debuts.
GPU means graphical processing unit. Now consumer GPUs are pressed into service for compute tasks like BOINC & folding, but they are still GPUs (they can still do graphics).
Does Nvidia even bother to put in the graphics-specific silicon and output hardware on the K20, or should these things really be called.. I dunno.. "compute accelerators" or something like that?
Intel's 22nm transistors certainly do. The overall chips don't because the price differential between even a top-line 8 Core Sandy Bridge Xeon chip/system and the Power7 chips/systems that actually have the high-end performance you are talking about is similar to the price differential between the chip in my cellphone and the high-end Xeon chip.
I know guys that do CPU design for IBM and they will flat out tell you that Intel has a better process. The difference is that IBM is making chips for million dollar+ servers with huge legacy needs in markets where even Itanium isn't trying to compete. At that point, you can afford to design CPUs with 200+ watt TDPs and exotic liquid cooling systems that are made in tiny quantities compared to what Intel & AMD churn out.