When it comes to desktop, laptop and server chips, Intel's pretty much got a lock on the market but everyone can see the writing on the wall: mobile chips and architectures are the future of computing thanks to the popularity of smartphones, but Intel doesn't have anything to offer in that regard.
Server market is a different ball game. Xeon are only in the low end server. IBM has lead in middle and high end servers with their P and Z systems. P and Z chips are custom designed for IBM systems only.
If it is matter of core-war, IBM's latest Power7 also has 8 cores. It is actually based on 45nm technology compared to Intel's latest 32nm. What makes Power7 exciting is that it has on-die 32MB L3 cache. They achieved this by introducing eDRAM (embedded DRAM) in the technology. Both Nehalem-EX and Power7 are targeting low-end server market, so it should be interesting battle.
Note: Damage caused by overclocking AMDâ(TM)s GPUs above factory-set overclocking is not covered by AMDâ(TM)s product warranty, even when such overclocking is enabled via AMD software.
Agreed. And there are more reliable ways to do background search than randomly putting your name on Google. Because it is equally likely that if there is a Nobel Prize winner with same name as yours, they wouldn't be thrilled either.
AMD has multiple "nodes" per technology. So in 45nm itself, they have 7 to 9 nodes. Each node represents performance improvement over the previous one by using new technology innovations. It is still 45nm technology, but you may add, for example, higher stress liner to improve mobility, hence more current and hence performance. It doesn't change any of the basic groundrules. These nodes are typically in 3-6 months range (rather than 18 month as said by Moore's law). But then these nodes don't really improve performance by 2x either. The first node is the hardest - get the ground rules right, get a yielding process etc. Once the foundation is set, it is relatively easier to experiment with new process technologies.
Many people have wondered what the difference is between the Computer Science education given in the average public university versus one given in an Ivy League university.
Pardon my ignorance, but we had many supercomputers (more recent one Roadrunner) which use multiple CPUs (and accelerators). Can't we use the programming 'tricks' or 'models' or 'techniques' used there for efficiently using multicores ? I understand multicore has significantly less communication overhead, but overall philosophy of synchronizing, message passing, shared memory etc wouldn't be completely irrelevent ?
IBM has a much larger fab facilities then Intel had (maybe still).
No, they do not. There only 300mm line is in East Fishkill, NY. The aging 200mm line in Burlington will not be upgraded. IBM doesn't have capacity for low price and high volume manufacturing. IBM's main business is servers which always demand more than the best available in the market. Since there is huge profit margin on servers, IBM invests in getting that extra performance even if it is not yield friendly. You don't sell millions of big iron (Z series) machines. The development cost of that technology is becoming prohibitively expensive. That is why (among other reasons) decided to go with partnership model where the cost is amortized. The partners get access to latest and greatest technology in return.
) In China, there is no mandated 40-hour work week (limit)
Nor is it in US. We in IT/Electronics work 60-hour week regardless. Yes, if I log in from home after dinner to check progress on work, it is work. No, I don't do it out of fun, we have to make a final check before going to bed so that things can run smoothly (hopefully) till morning. So yes we get paid for 40 hours but we work more than that.
I am less productive when I get interrupted every 10-15 min from a pop window. Just because you can ask, people don't spend time thinking or looking for it themselves. One can argue for and against such a thing and it depends on your work. Analogy holds that online books are good for manuals (instant search), but when I want to read a novel, nothing beats a physical paper. Similarly if you working which requires more thinking and analysis, you are better off with less interruption.
I have also limited checking emails to 3 times a day. If there is an emergency, there is a phone and you can stop by my cube.
If one looks at http://www.top500.org/ list and compare the CPU frequencies of the top supercomputers - all BlueGene CPUs were running at less than a GHz. And it seemed those low power cores were key to HPC (high performance computing). Cell and opteron - both run at multiple GHz and (presumably consume more power). IBM still has next generation of BlueGene/Q in works and is also for +Petaflop computation.
I was thinking of putting good megapixel network cameras for surveillance but then I needed to buy another set of good cameras to monitor the surveillance cameras and then another set of cameras...
Slashdot Mirror
Sorry I don't have any sympathy for LulzSec or whatever their name (Geeze at least get a cooler name or they are bunch of high school drop outs).
Server market is a different ball game. Xeon are only in the low end server. IBM has lead in middle and high end servers with their P and Z systems. P and Z chips are custom designed for IBM systems only.
Agreed. Not exactly, but this comic says it as well. http://abstrusegoose.com/291
I got notification from them as well but it went to the SPAM folder. Sorry.
There are many patents on random number generator. http://www.google.com/search?hl=en&source=hp&q=random+number+generator+patent&btnG=Google+Search
http://arstechnica.com/hardware/news/2009/09/ibms-8-core-power7-twice-the-muscle-half-the-transistors.ars
Note: Damage caused by overclocking AMDâ(TM)s GPUs above factory-set overclocking is not covered by AMDâ(TM)s product warranty, even when such overclocking is enabled via AMD software.
http://www.nytimes.com/2009/02/10/arts/design/10fair.html
http://www.nokia.com/A4879144
In Soviet Russia, water spins electric field !
Now, if you'll excuse me, I've got to get back to stalking this one chick....
Sorry to disappoint you. That is a guy with female login-id.
Agreed. And there are more reliable ways to do background search than randomly putting your name on Google. Because it is equally likely that if there is a Nobel Prize winner with same name as yours, they wouldn't be thrilled either.
Why do you hate India ?
AMD has multiple "nodes" per technology. So in 45nm itself, they have 7 to 9 nodes. Each node represents performance improvement over the previous one by using new technology innovations. It is still 45nm technology, but you may add, for example, higher stress liner to improve mobility, hence more current and hence performance. It doesn't change any of the basic groundrules. These nodes are typically in 3-6 months range (rather than 18 month as said by Moore's law). But then these nodes don't really improve performance by 2x either. The first node is the hardest - get the ground rules right, get a yielding process etc. Once the foundation is set, it is relatively easier to experiment with new process technologies.
Or 1/3rd US population ?
Tuition for sure :)
Do you mean no more flying cars either ? :(
Pardon my ignorance, but we had many supercomputers (more recent one Roadrunner) which use multiple CPUs (and accelerators). Can't we use the programming 'tricks' or 'models' or 'techniques' used there for efficiently using multicores ? I understand multicore has significantly less communication overhead, but overall philosophy of synchronizing, message passing, shared memory etc wouldn't be completely irrelevent ?
No, they do not. There only 300mm line is in East Fishkill, NY. The aging 200mm line in Burlington will not be upgraded. IBM doesn't have capacity for low price and high volume manufacturing. IBM's main business is servers which always demand more than the best available in the market. Since there is huge profit margin on servers, IBM invests in getting that extra performance even if it is not yield friendly. You don't sell millions of big iron (Z series) machines. The development cost of that technology is becoming prohibitively expensive. That is why (among other reasons) decided to go with partnership model where the cost is amortized. The partners get access to latest and greatest technology in return.
Nor is it in US. We in IT/Electronics work 60-hour week regardless. Yes, if I log in from home after dinner to check progress on work, it is work. No, I don't do it out of fun, we have to make a final check before going to bed so that things can run smoothly (hopefully) till morning. So yes we get paid for 40 hours but we work more than that.
I have also limited checking emails to 3 times a day. If there is an emergency, there is a phone and you can stop by my cube.
If one looks at http://www.top500.org/ list and compare the CPU frequencies of the top supercomputers - all BlueGene CPUs were running at less than a GHz. And it seemed those low power cores were key to HPC (high performance computing). Cell and opteron - both run at multiple GHz and (presumably consume more power). IBM still has next generation of BlueGene/Q in works and is also for +Petaflop computation.
It is much superior in Windows. Linux has long way to go. Not everyone is compiling kernel out there.
I was thinking of putting good megapixel network cameras for surveillance but then I needed to buy another set of good cameras to monitor the surveillance cameras and then another set of cameras ...