This is no flamebait. The single-precision floating point numbers are on the order of 250 GFLOPS/s. However double-precision is less than 10.
Cell will be good for graphics and gaming. It won't be useful for most scientific apps because of the double-precision issue. It won't be used on the desktop because of backwards compatibility and programmability issues.
The Cell will be heavily used in game consoles, thats about it. Don't buy into the IBM marketing machine.
Sorry but having 8 SPE "coprocessors" or whatever you call them is not for general-purpose computing. Programmers can hardly deal with multhreading on a symmetric multiprocessor. What makes you think that programming a heterogeneous machine will be easier?
No kidding! I use a Dell in my living area just for that very reason. They employ an army of mechanical engineers for this very reason. My wife, with her HVAC and ME degree, was actually recruited by Dell to work on quiet enclosures and cooling.
On the other hand, the workstation in my basement is a homebuilt Athlon system that sounds like a jet engine.
Dude, if you had 3 Western Digital drives fail in a row and had 3 of 4 Maxtor drives fail, I would seriously consider that something else is your problem.
Where do you buy your drives? Some shady operation that resells returned drives? Maybe your power supply or IDE controllers are shoddy.
In the last 8 years, of the 15 or so hard drives I've bought and used, one has failed. Most were Western Digital, a few Maxtors, and a couple IBM drives. Do I have good luck? Maybe, but my anecdotal experience pretty much matches the MTTF data from manufacturers.
Is there anything else out there that compares to Microsoft PowerPoint? It is a phenomenal program and I can't find an equivalent competitor. The features in the "Custom Animation" are amazing.
Earth calling Motorola: your phones are crap. I got burnt on a T720 and went back to nokia. I'd have taken almost any nokia in preference to any Motorola. My advice to Motorola: Build a usable, responsive interface.
When I worked at Motorola in 2000-2002, Motorola phones had superior RF and mechanics (except for the terrible antenna on the V120 and early V60 phones). Yes, the software is not so great. A StarTAC has been known to be dropped from a motorcyle, going 70 mph, ran over by a truck, and it still turned on and made a phone call. Dipped in a swimming pool? Let her dry out for a couple weeks and give it a try.
Back when I used to work at Motorola in Schaumburg, the CEO sent out a company-wide email saying how he was displeased at the number of employees seen with Nokia and other non-Motorola phones. So he offered free Motorola phones to the first 1,000 employees that responded and urged the rest to buy a Motorola.
He was especially pissed at the salesmen, trying to sign the big carriers to promote Motorola phones, who had Nokia's hanging from their belt! Makes sense for the visible people I guess.
The only people who thinks it was Microsoft who brought cheap PC to the market is those who have no knowledge of computers whatsoever. The PC became cheap because the open standard that let anyone do a clone without expensive patents and copyrights standing in the way except for the bios. The competition that enabled was what drove the PC price from 8000$ down to todays prices.
Gee, do you think the masses have "knowledge of computers". The OP says that everybody hates Microsoft. If somebody did a nation-wide poll, the percentage of Microsoft "haters" would be less than 10%.
And it is arguable whether Microsoft had nothing to do with cheap computing. If IBM OS/2 won, things might be different today. Microsoft is not a hardware company, and because of that, hardware is a commodity. If the hardware company that developed the PC wins the OS monopoly, things might not be the same today.
Who hates Microsoft? Only techno-geeks that have an open-source agenda.
On the other hand, talk to older folks and they see Microsoft as the company that brought usable and affordable personal computing to the masses. They see Microsoft as one of the companies that fattened their 401k or mutual funds.
Its the number of number of transistors per square inch on integrated circuits that doubles every 18 months not the speed.
That would be Moore's Law which I never mentioned in my previous post. If you look at the 90s, performance of microprocessors did mostly double every 18 months. Granted that this hasn't been exact in the last 4 years, but not that far off.
3Ghz PCs made an appearance over 18 months ago - how many 6GHz ones have you seen?
You silly fool. Of course clock rate isn't the only factor in performance. Extra transistors lead to bigger caches, better branch predictors, a larger instruction window, etc. With a 3GHz P4, which can retire up to 3 instructions per cycle, a single cache miss (that takes say 100ns to fill) equates to up to 900 lost opportunities to retire instructions. Now tell me, is going to a 6GHz clock better then saving a few cache misses here and there?
Really? Do they really "dominate" in the way of performance? I haven't looked at the latest benchmarks. How much faster is AMD's best compared to Intel's? 20%? 30%? This is really not that significant when processors double in speed every 18 months.
Geee...lets see: Intel has 85% market share and has a market cap of over 10 times that of AMD. Intel operates at a profit margin of 22.68% whereas AMD is at 2.89%. And yet the fanboys are declaring "AMD's dominance". How pathetic. Wake me up in 2015 if AMD still "dominates".
Maybe the fanboys should compare some basic financial statistics of Intel and AMD. This stuff doesn't change overnight.
Look at the size of these companies compared to Intel. It is must easier for little guys to displace each other than for a little guy to displace the king.
The ultimate limit on the number of cores you can put on a single chip is the available pin bandwidth. At a point, there simply isn't enough bandwidth available to supply instructions and data.
I expect AMD to become bigger than Intel in the next 10 years
That is quite naive. AMD has close to 15,000 employees and is a $9 billion company. Intel has 85,000 employees and is a $140 billion company. This doesn't change overnight, and yes, 10 years is overnight when you consider companies of this size. Intel will have to make several more mistakes like Itanium. Plus AMD has a long ways to go to match the manufacturing capabilities of Intel.
You can mess up security policies and implementations with Java, but it is much harder to shoot yourself in the foot. The JVM may have bugs, but because it is used for all Java applications, it is likely well-debugged and secure
Language features eliminate security problems. For example, the Java JVM does something incredibly advanced: bounds checking!
Yes, a mini base station would be required on the plane for cellphones to work. This would then be patched into the rest of the system via satellite.
Even if you are flying at 1000 feet and your signal wasn't distorted by the aluminum shell, handoffs woudln't work flying that fast. There is a small window of opportunity for cell-to-cell handoffs. It differs between CDMA, AMPS, GSM, and etc. This is also the reason the old Japanese PDC system wouldn't work in cars. Handoffs were too slow to work beyond 20 mph or so.
Photographers using traditional film have argued exactly the same for almost five years now and digital photography took off anyway.
Yes, digital has mostly replaced 35mm (for color anyways) and is encroaching on medium format. However I am skeptical that Moore's law applies to CCD and CMOS photo sensors. The low-hanging fruit has been grabbed and the higher megapixel sensors, coming to the prosumer market today, are quite noisy in order to get 7 or 8 megapixels. Plus I think you will see a point when sub $1000 cameras simply start to level off on megapixels, because they aren't needed at a point.
A digital sensor that can do with 4x5 film does is a long ways off. Plus you are ignoring the aspects of a view camera, such as tilts/swings/rise/etc, that cannot be reproduced in photoshop.
But I really don't care what happens am an not clinging to old technology. My 40-year old view camera lenses will be usable with that new digital 4x5 sensor invented 20 years from now anyways. I am a hobbyist and I use what I enjoy. If my living was photography, I might feel differently about digital.
Slashdot Mirror
This is no flamebait. The single-precision floating point numbers are on the order of 250 GFLOPS/s. However double-precision is less than 10.
Cell will be good for graphics and gaming. It won't be useful for most scientific apps because of the double-precision issue. It won't be used on the desktop because of backwards compatibility and programmability issues.
The Cell will be heavily used in game consoles, thats about it. Don't buy into the IBM marketing machine.
Sorry but having 8 SPE "coprocessors" or whatever you call them is not for general-purpose computing. Programmers can hardly deal with multhreading on a symmetric multiprocessor. What makes you think that programming a heterogeneous machine will be easier?
No kidding! I use a Dell in my living area just for that very reason. They employ an army of mechanical engineers for this very reason. My wife, with her HVAC and ME degree, was actually recruited by Dell to work on quiet enclosures and cooling.
On the other hand, the workstation in my basement is a homebuilt Athlon system that sounds like a jet engine.
Dude, if you had 3 Western Digital drives fail in a row and had 3 of 4 Maxtor drives fail, I would seriously consider that something else is your problem.
Where do you buy your drives? Some shady operation that resells returned drives? Maybe your power supply or IDE controllers are shoddy.
In the last 8 years, of the 15 or so hard drives I've bought and used, one has failed. Most were Western Digital, a few Maxtors, and a couple IBM drives. Do I have good luck? Maybe, but my anecdotal experience pretty much matches the MTTF data from manufacturers.
Is there anything else out there that compares to Microsoft PowerPoint? It is a phenomenal program and I can't find an equivalent competitor. The features in the "Custom Animation" are amazing.
I agree. If you haven't already read "Guns, Germs, and Steel: The Fates of Human Societies", I highly recommend it. Everything happens for a reason and because the chips were in place...
Earth calling Motorola: your phones are crap. I got burnt on a T720 and went back to nokia. I'd have taken almost any nokia in preference to any Motorola. My advice to Motorola: Build a usable, responsive interface.
When I worked at Motorola in 2000-2002, Motorola phones had superior RF and mechanics (except for the terrible antenna on the V120 and early V60 phones). Yes, the software is not so great. A StarTAC has been known to be dropped from a motorcyle, going 70 mph, ran over by a truck, and it still turned on and made a phone call. Dipped in a swimming pool? Let her dry out for a couple weeks and give it a try.
You don't need to tell me that. After 2 years, I was ready to croak and decided to leave my good job behind and bail.
Back when I used to work at Motorola in Schaumburg, the CEO sent out a company-wide email saying how he was displeased at the number of employees seen with Nokia and other non-Motorola phones. So he offered free Motorola phones to the first 1,000 employees that responded and urged the rest to buy a Motorola.
He was especially pissed at the salesmen, trying to sign the big carriers to promote Motorola phones, who had Nokia's hanging from their belt! Makes sense for the visible people I guess.
Talk to me when you are 45 with kids in college.
The only people who thinks it was Microsoft who brought cheap PC to the market is those who have no knowledge of computers whatsoever. The PC became cheap because the open standard that let anyone do a clone without expensive patents and copyrights standing in the way except for the bios. The competition that enabled was what drove the PC price from 8000$ down to todays prices.
Gee, do you think the masses have "knowledge of computers". The OP says that everybody hates Microsoft. If somebody did a nation-wide poll, the percentage of Microsoft "haters" would be less than 10%.
And it is arguable whether Microsoft had nothing to do with cheap computing. If IBM OS/2 won, things might be different today. Microsoft is not a hardware company, and because of that, hardware is a commodity. If the hardware company that developed the PC wins the OS monopoly, things might not be the same today.
Who hates Microsoft? Only techno-geeks that have an open-source agenda.
On the other hand, talk to older folks and they see Microsoft as the company that brought usable and affordable personal computing to the masses. They see Microsoft as one of the companies that fattened their 401k or mutual funds.
So you see not everybody hates Microsoft.
How about the consumer video recorder (VHS or Betamax...take you pick)? Never before could the consumer time shift television shows.
You silly fool
Its the number of number of transistors per square inch on integrated circuits that doubles every 18 months not the speed.
That would be Moore's Law which I never mentioned in my previous post. If you look at the 90s, performance of microprocessors did mostly double every 18 months. Granted that this hasn't been exact in the last 4 years, but not that far off.
3Ghz PCs made an appearance over 18 months ago - how many 6GHz ones have you seen?
You silly fool. Of course clock rate isn't the only factor in performance. Extra transistors lead to bigger caches, better branch predictors, a larger instruction window, etc. With a 3GHz P4, which can retire up to 3 instructions per cycle, a single cache miss (that takes say 100ns to fill) equates to up to 900 lost opportunities to retire instructions. Now tell me, is going to a 6GHz clock better then saving a few cache misses here and there?
dominance in the way of performance you dolt.
Really? Do they really "dominate" in the way of performance? I haven't looked at the latest benchmarks. How much faster is AMD's best compared to Intel's? 20%? 30%? This is really not that significant when processors double in speed every 18 months.
Geee...lets see: Intel has 85% market share and has a market cap of over 10 times that of AMD. Intel operates at a profit margin of 22.68% whereas AMD is at 2.89%. And yet the fanboys are declaring "AMD's dominance". How pathetic. Wake me up in 2015 if AMD still "dominates".
Maybe the fanboys should compare some basic financial statistics of Intel and AMD. This stuff doesn't change overnight.
Look at nVidia vs the rest... ATI, 3dFX, S3
Look at the size of these companies compared to Intel. It is must easier for little guys to displace each other than for a little guy to displace the king.
The ultimate limit on the number of cores you can put on a single chip is the available pin bandwidth. At a point, there simply isn't enough bandwidth available to supply instructions and data.
I expect AMD to become bigger than Intel in the next 10 years
That is quite naive. AMD has close to 15,000 employees and is a $9 billion company. Intel has 85,000 employees and is a $140 billion company. This doesn't change overnight, and yes, 10 years is overnight when you consider companies of this size. Intel will have to make several more mistakes like Itanium. Plus AMD has a long ways to go to match the manufacturing capabilities of Intel.
What would you recommend instead?
Java/J2EE/JSP
You can mess up security policies and implementations with Java, but it is much harder to shoot yourself in the foot. The JVM may have bugs, but because it is used for all Java applications, it is likely well-debugged and secure
Language features eliminate security problems. For example, the Java JVM does something incredibly advanced: bounds checking!
Yes, a mini base station would be required on the plane for cellphones to work. This would then be patched into the rest of the system via satellite.
Even if you are flying at 1000 feet and your signal wasn't distorted by the aluminum shell, handoffs woudln't work flying that fast. There is a small window of opportunity for cell-to-cell handoffs. It differs between CDMA, AMPS, GSM, and etc. This is also the reason the old Japanese PDC system wouldn't work in cars. Handoffs were too slow to work beyond 20 mph or so.
You are between the ages of 18-25 for sure.
It shows on how naive and smart you think you are.
Photographers using traditional film have argued exactly the same for almost five years now and digital photography took off anyway.
Yes, digital has mostly replaced 35mm (for color anyways) and is encroaching on medium format. However I am skeptical that Moore's law applies to CCD and CMOS photo sensors. The low-hanging fruit has been grabbed and the higher megapixel sensors, coming to the prosumer market today, are quite noisy in order to get 7 or 8 megapixels. Plus I think you will see a point when sub $1000 cameras simply start to level off on megapixels, because they aren't needed at a point.
A digital sensor that can do with 4x5 film does is a long ways off. Plus you are ignoring the aspects of a view camera, such as tilts/swings/rise/etc, that cannot be reproduced in photoshop.
But I really don't care what happens am an not clinging to old technology. My 40-year old view camera lenses will be usable with that new digital 4x5 sensor invented 20 years from now anyways. I am a hobbyist and I use what I enjoy. If my living was photography, I might feel differently about digital.
Great post! Have you considered scanning your 4x5 transparencies with an Epson 4870? That is what I am eying up...
Why? 8x10 cameras have existed for 100 years. Using modern film and a drum scanner will create a digital image with more than 1Gb of pixel data.
Even my 4x5 camera yields over 100 megapixels when scanning film with a $300 Epson flatbed.