What rate would you set this Fair Tax at so that you would raise enough money to fund all the programmes you need to fund?
Would you have different rates for different classes of goods? E.g., gas/electricity, baby clothes, food, luxury goods, cars,...
What about the problem that rich people simply don't spend all their money, therefore a fair tax could actually just mean they pay even less money than they do now on their earnings (tax avoidance notwithstanding). There's only so many sports cars you can buy.
There is an argument to introduce VAT whilst reducing corporation tax - corporations are offshoring more and more of their money in order to avoid tax, so you should just get the tax at the point of sale instead.
In reality governments merely tweak the tax rules because doing a major change will be highly disruptive.
Not much is known about Intel's yields. There have been suggestions that 22nm did not start off very well, and Intel overcame that with mass production. They're not making 22nm Atoms right now either, and in addition you forget that Samsung runs its own fabs as well.
Intel likes a very large margin on its CPU sales, does Samsung have an internal market where they're paying a 60% margin to their fab department for the SoC? Regardless, most SoC cost estimates in product teardowns put the price of ARM-based SoCs around $20, give or take a few dollars. Intel isn't selling Celeron 847s for that price, certainly not with all the support chippery that it requires. Sure, Haswell will be more SoC-like. Atom is a better comparison here, but the SoC Atoms currently have very poor graphics and the A15 chip is generally outperforming it by quite a margin.
Intel can't afford to pricematch against a $30 SoC for long... however I'm surprised the Celeron 847 has such a high tray price. More likely it's available for around $60, and the quality of other components has been compromised to attain a $199 pricepoint.
The 28nm Exynos 5450 is coming out next year, with two more cores, twice the GPU and a faster clock speed.
Haswell is simply not going to compete in this area of the market, where price, power consumption and performance come together. It will compete at higher price points, maybe even at low power in single-core ULV (ultra-low-clock too) variants, but not all three.
Intel's 22nm next-generation Atoms are coming out in 2014.
The next generation of 22nm Atoms with the new core has been delayed until 2014.
A best of breed current generation Atom (dual-core, quad-thread) is thoroughly beaten in many benchmarks by this slightly slower clock-speed, first generation, 32nm dual-core Cortex A15 product. Next year brings a lot of 28nm quad-core A15s at 2GHz including the Exynos 5450.
And because it is cheap - no Intel tax - you can get a very decent computing device using it for $250, which is a darn sight better than the pricing of Atom netbooks, AMD netbooks and certainly anything called an Ultrabook.
Seems to me that the issue is low pay and poor working conditions.
You would really have to love flying to get into it as a job with the rewards you've outlined above.
Maybe if they had raised the pay to be commensurate with the responsibility of the role, the unsociable hours and so on, and paid them when they turned up at the airport rather than when they entered the cabin door, there would be more people becoming pilots.
TSMC's 28nm process ramped really slowly, and couldn't cope with the quantities that Apple wanted.
Maybe TSMC's 20nm will work nicely, and not be late, etc, but form doesn't suggest it.
The fact that Apple had to agree to the price rise shows that they have no other options right now.
In addition, who says this isn't merely a 5 year price review by Samsung on their fab contract with Apple, or that another fab would be able to offer significantly more competitive pricing?
I'm sure Samsung would be happy to continue manufacturing for Apple, as it is still income.
But maybe this is flexing some of their muscles to fire a warning shot over to Apple about Apple's recent lawsuit games. On the other hand, this is the first price rise for Apple in five years of manufacturing (where I imagine pricing is on a per-wafer or per-mm^2 basis - not clarified).
It's not hard to imagine that 32nm is more expensive, and that inflation over that time will have raised costs as well, that Samsung have merely invoked a five year price review clause in their contract with Apple to raise pricing to a reasonable level for the next five years.
But nobody thought to show a photo of the guy the police named to the victim during any of this?
The real question - why is the police say what they did to the guy - has been completely forgotten and ignored. That would have made a much better Newsnight story too.
Mistakes were made of course during this show - but that's probably because two of the existing long-term Newsnight editors were moved on because of the previous furore over not showing a show naming someone as a paedophile merely on the say-so of members of the public. The new editorial team (and new director general) are going to make mistakes because they are new to it all. The director general probably resigned because the job role of "public archery target" was not what he signed up for, and this provided a convenient out.
The Z2460's base clock remains 1.3GHz, it's only when the thermal conditions allow it that the CPU will boost up to 2GHz.
What does this mean in reality? Yes, short running benchmarks run at 2GHz before the chip gets hot, reporting high scores as a result. This is a flaw with the benchmarks, especially the javascript ones that are gravy for a CPU design that emphasises clock speed at the cost of multiple cores (and hyperthreading really doesn't add much).
And in this case, the little guy wants a 60% margin on its products, and is wondering why the big guy, with far smaller margins, is still so popular.
Hardly anybody is complaining about performance on today's Cortex A9 based mobile phones. Next year's dual-core A15 phones (and today's Apple A6 phones) will make it even less of an issue. What people do want is decent graphics (Intel is notorious for behind behind the curve here in all markets) and hardware acceleration for media, security, and so on. The ARM SoCs provide this already. Intel is not differentiating their products at all, except for x86 support - increasingly irrelevant today, especially in mobile which is all ARM already.
I agree that ARM is safe in mobile and tablets. That doesn't mean that Intel won't get a slice of the market, but I don't think that Intel will ever have the flexibility and price advantages of ARM.
Moving to Atom would give phones more CPU power
That's debatable as well, because Intel only offers a single-core, hyperthreaded Atom at 1.3GHz (with some turbo features), that only performs well in Javascript benchmarks. Quad-core A9s and dual-core A15s are more than competitive, and definitely include far better graphics these days - something that people will make use of (games, UI, etc).
ARM have a product for everyone to license. This also adds 64-bits to the mix, in time for 8GB phones and tablets in 2014/2015.
x86 compatibility in servers - only if your running Windows Server. I haven't seen Microsoft crowing about server sales for quite some time, Linux is really the leader here. And Linux runs on ARM. Java runs on ARM. Perhaps it is only a matter of time... Power consumption is already an issue in data centres...
Another post by someone who doesn't consider that ARM is targetting both performance/dollar and performance/watt.
NVIDIA actually adds an extra core because the cores are small enough to do that, and thus have a special low-power core for idling processes. It works very well, it's a shame that NVIDIA's mobile graphics are so lame in comparison.
I have seen no proof that ARM doesn't scale, and indeed this design is targeting on-chip clusters (4x4cores) using the new interconnects that ARM has just started making available.
It is incredibly difficult to scale down a power hungry design to low power - it needs to be done during the design. Haswell will include a lot of that in its design, but it still won't be a 1W core. ARM's designs are targeting ultra low power consumption as a standard mode of operation, hence the big.LITTLE designs (actively switch to/from ultra-low-power ISA compatable cores and high-performance cores).
Consumers are also not asking for more, more, more performance. ARM SoCs include dedicated hardware for the multimedia and security that you mention. Intel's new Atoms (not out yet) may perform better than their old ones, but nobody wants to do comparative tests except in Javascript benchmarks that are extremely well optimised for x86.
This type of early release of a core design is standard in the embedded and SoC world.
It takes some time after the core is available for it to be integrated into the SoCs that third parties are designing and building. That's why there is a gap between ARM "launching" a core design, and it being included within a device that you can buy on the high street.
Today's 2GB mobile phone will be 2014's 4GB mobile phone, and 2016's 8GB mobile phone (using TSV & WideIO for ultra-low-power connectivity to the SoC).
Of course today's home computer is 16GB, and probably 64GB in 2016... and I'm not going to claim that it won't be needed because somehow computers continue to make use of all the RAM you ram inside the box. 8K monitors and 4K video will probably use a load of that memory just for OS imagery...
And this is pretty much what ARM is doing with big.LITTLE - powerful ARM Cortex A15 cores coupled with very low power ARM Cortex A7 cores - both have exactly the same ISA. Depending on software need, things either run on the A7s (most of the time probably) and then the A15s kick in when something actually stresses the A7s.
The policy for doing so does have to take into account task power as well as instantaneous power. A slow-running job on the A7s could complete far faster on the A15s, and thus actually use less power overall. Something running constantly but not doing much should run on the A7s.
Given that even today's mobile phone SoCs have dedicated hardware "cores" for encryption, video encoding and decoding, etc, this is only Intel trying to generalise the functionality back into the CPU - which is pretty much all Intel know - rather than the more suitable, lower power, dedicated function blocks that are but pinheads even on today's SoCs, never mind the 10nm SoCs in years to come.
Today's quad-core mobile phone SoC on 32nm could be a 16-core cluster-on-a-chip on 16nm, and 64-cores on 10nm.
Never mind the compute-assist on the GPUs in these SoCs - Exynos 5 is over 70 GFLOPS for example, and supports OpenCL 1.1. Never mind the 4K video decode/encode support...
Would you say this is the way forward for other GPU vendors to offer open source drivers - by embedding the actual implementation within the chip itself? Does this also have a benefit in offloading the high-level processing from the system CPU by running that within the GPU as well?
Does the VideoCore IV also incorporate it's own CPU-function that the embedded firmware runs on (I believe the VideoCore lines are fully fledged CPUs in their own right, with large vector/dsp/etc engines? Therefore it's the engines that are totally hidden, with the GPU presenting an API that maps very closely to operations within OpenGL?).
Is there an OpenCL interface on the VideoCore IV? Can one be added by upgrading the firmware, or by implementing it in userland and RPC to a lower level interface on the VideoCore IV?
Other components, for example the GPU, are still very proprietary and closed.
From raspberry pi blog:
As of right now, all of the VideoCore driver code which runs on the ARM is available under a FOSS license (3-Clause BSD to be precise). The source is available from our new userland repository on GitHub. If you’re not familiar with the status of open source drivers on ARM SoCs this announcement may not seem like such a big deal, but it does actually mean that the BCM2835 used in the Raspberry Pi is the first ARM-based multimedia SoC with fully-functional, vendor-provided (as opposed to partial, reverse engineered) fully open-source drivers, and that Broadcom is the first vendor to open their mobile GPU drivers up in this way.
Yup, that's the GPU. Now the question is how much of the work is done by the firmware on the GPU rather than the driver running on the CPU?
You've also got to consider China with their MIPS64 based Longsoon SoCs. Might not be a threat anytime soon, but with government backing and a desire to do things themselves, it probably will grow and grow, and inevitably start growing outside China as well.
You are aware that different ARM designs have different pipeline lengths depending on their market position? Or is all your knowledge of ARM based upon an ancient university lecture about simple RISC Fetch/Decode/Execute pipelines that the first ARM employed?
Slashdot Mirror
What rate would you set this Fair Tax at so that you would raise enough money to fund all the programmes you need to fund?
Would you have different rates for different classes of goods? E.g., gas/electricity, baby clothes, food, luxury goods, cars, ...
What about the problem that rich people simply don't spend all their money, therefore a fair tax could actually just mean they pay even less money than they do now on their earnings (tax avoidance notwithstanding). There's only so many sports cars you can buy.
There is an argument to introduce VAT whilst reducing corporation tax - corporations are offshoring more and more of their money in order to avoid tax, so you should just get the tax at the point of sale instead.
In reality governments merely tweak the tax rules because doing a major change will be highly disruptive.
Not much is known about Intel's yields. There have been suggestions that 22nm did not start off very well, and Intel overcame that with mass production. They're not making 22nm Atoms right now either, and in addition you forget that Samsung runs its own fabs as well.
Intel likes a very large margin on its CPU sales, does Samsung have an internal market where they're paying a 60% margin to their fab department for the SoC? Regardless, most SoC cost estimates in product teardowns put the price of ARM-based SoCs around $20, give or take a few dollars. Intel isn't selling Celeron 847s for that price, certainly not with all the support chippery that it requires. Sure, Haswell will be more SoC-like. Atom is a better comparison here, but the SoC Atoms currently have very poor graphics and the A15 chip is generally outperforming it by quite a margin.
Intel can't afford to pricematch against a $30 SoC for long ... however I'm surprised the Celeron 847 has such a high tray price. More likely it's available for around $60, and the quality of other components has been compromised to attain a $199 pricepoint.
Let me know when an 8W Haswell is available for under $40, as it would have to cost for it to be viable for a $250 netbook.
The 8W Haswells will be used for Ultra-ultrabooks, or long-life-ultrabooks, costing at least $799.
This Exynos 5250 is a 32nm SoC.
The 28nm Exynos 5450 is coming out next year, with two more cores, twice the GPU and a faster clock speed.
Haswell is simply not going to compete in this area of the market, where price, power consumption and performance come together. It will compete at higher price points, maybe even at low power in single-core ULV (ultra-low-clock too) variants, but not all three.
Intel's 22nm next-generation Atoms are coming out in 2014.
The next generation of 22nm Atoms with the new core has been delayed until 2014.
A best of breed current generation Atom (dual-core, quad-thread) is thoroughly beaten in many benchmarks by this slightly slower clock-speed, first generation, 32nm dual-core Cortex A15 product. Next year brings a lot of 28nm quad-core A15s at 2GHz including the Exynos 5450.
And because it is cheap - no Intel tax - you can get a very decent computing device using it for $250, which is a darn sight better than the pricing of Atom netbooks, AMD netbooks and certainly anything called an Ultrabook.
Seems to me that the issue is low pay and poor working conditions.
You would really have to love flying to get into it as a job with the rewards you've outlined above.
Maybe if they had raised the pay to be commensurate with the responsibility of the role, the unsociable hours and so on, and paid them when they turned up at the airport rather than when they entered the cabin door, there would be more people becoming pilots.
TSMC doesn't have a 32nm process.
TSMC's 28nm process ramped really slowly, and couldn't cope with the quantities that Apple wanted.
Maybe TSMC's 20nm will work nicely, and not be late, etc, but form doesn't suggest it.
The fact that Apple had to agree to the price rise shows that they have no other options right now.
In addition, who says this isn't merely a 5 year price review by Samsung on their fab contract with Apple, or that another fab would be able to offer significantly more competitive pricing?
I'm sure Samsung would be happy to continue manufacturing for Apple, as it is still income.
But maybe this is flexing some of their muscles to fire a warning shot over to Apple about Apple's recent lawsuit games. On the other hand, this is the first price rise for Apple in five years of manufacturing (where I imagine pricing is on a per-wafer or per-mm^2 basis - not clarified).
It's not hard to imagine that 32nm is more expensive, and that inflation over that time will have raised costs as well, that Samsung have merely invoked a five year price review clause in their contract with Apple to raise pricing to a reasonable level for the next five years.
But nobody thought to show a photo of the guy the police named to the victim during any of this?
The real question - why is the police say what they did to the guy - has been completely forgotten and ignored. That would have made a much better Newsnight story too.
Mistakes were made of course during this show - but that's probably because two of the existing long-term Newsnight editors were moved on because of the previous furore over not showing a show naming someone as a paedophile merely on the say-so of members of the public. The new editorial team (and new director general) are going to make mistakes because they are new to it all. The director general probably resigned because the job role of "public archery target" was not what he signed up for, and this provided a convenient out.
Replying again with some more concrete information, that you may be interested in:
From http://hexus.net/tech/news/cpu/47517-samsungs-exynos-5-dual-faces-off-intels-atom-n570/ and http://www.anandtech.com/show/6422/samsung-chromebook-xe303-review-testing-arms-cortex-a15/6
Latest dual-core Exynos 5 thrashes a dual-core Atom with HyperThreading.
The Z2460's base clock remains 1.3GHz, it's only when the thermal conditions allow it that the CPU will boost up to 2GHz.
What does this mean in reality? Yes, short running benchmarks run at 2GHz before the chip gets hot, reporting high scores as a result. This is a flaw with the benchmarks, especially the javascript ones that are gravy for a CPU design that emphasises clock speed at the cost of multiple cores (and hyperthreading really doesn't add much).
And in this case, the little guy wants a 60% margin on its products, and is wondering why the big guy, with far smaller margins, is still so popular.
Hardly anybody is complaining about performance on today's Cortex A9 based mobile phones. Next year's dual-core A15 phones (and today's Apple A6 phones) will make it even less of an issue. What people do want is decent graphics (Intel is notorious for behind behind the curve here in all markets) and hardware acceleration for media, security, and so on. The ARM SoCs provide this already. Intel is not differentiating their products at all, except for x86 support - increasingly irrelevant today, especially in mobile which is all ARM already.
I agree that ARM is safe in mobile and tablets. That doesn't mean that Intel won't get a slice of the market, but I don't think that Intel will ever have the flexibility and price advantages of ARM.
Moving to Atom would give phones more CPU power
That's debatable as well, because Intel only offers a single-core, hyperthreaded Atom at 1.3GHz (with some turbo features), that only performs well in Javascript benchmarks. Quad-core A9s and dual-core A15s are more than competitive, and definitely include far better graphics these days - something that people will make use of (games, UI, etc).
ARM have a product for everyone to license. This also adds 64-bits to the mix, in time for 8GB phones and tablets in 2014/2015.
x86 compatibility in servers - only if your running Windows Server. I haven't seen Microsoft crowing about server sales for quite some time, Linux is really the leader here. And Linux runs on ARM. Java runs on ARM. Perhaps it is only a matter of time... Power consumption is already an issue in data centres...
Another post by someone who doesn't consider that ARM is targetting both performance/dollar and performance/watt.
NVIDIA actually adds an extra core because the cores are small enough to do that, and thus have a special low-power core for idling processes. It works very well, it's a shame that NVIDIA's mobile graphics are so lame in comparison.
I have seen no proof that ARM doesn't scale, and indeed this design is targeting on-chip clusters (4x4cores) using the new interconnects that ARM has just started making available.
It is incredibly difficult to scale down a power hungry design to low power - it needs to be done during the design. Haswell will include a lot of that in its design, but it still won't be a 1W core. ARM's designs are targeting ultra low power consumption as a standard mode of operation, hence the big.LITTLE designs (actively switch to/from ultra-low-power ISA compatable cores and high-performance cores).
Consumers are also not asking for more, more, more performance. ARM SoCs include dedicated hardware for the multimedia and security that you mention. Intel's new Atoms (not out yet) may perform better than their old ones, but nobody wants to do comparative tests except in Javascript benchmarks that are extremely well optimised for x86.
This type of early release of a core design is standard in the embedded and SoC world.
It takes some time after the core is available for it to be integrated into the SoCs that third parties are designing and building. That's why there is a gap between ARM "launching" a core design, and it being included within a device that you can buy on the high street.
Today's 2GB mobile phone will be 2014's 4GB mobile phone, and 2016's 8GB mobile phone (using TSV & WideIO for ultra-low-power connectivity to the SoC).
Of course today's home computer is 16GB, and probably 64GB in 2016... and I'm not going to claim that it won't be needed because somehow computers continue to make use of all the RAM you ram inside the box. 8K monitors and 4K video will probably use a load of that memory just for OS imagery...
And this is pretty much what ARM is doing with big.LITTLE - powerful ARM Cortex A15 cores coupled with very low power ARM Cortex A7 cores - both have exactly the same ISA. Depending on software need, things either run on the A7s (most of the time probably) and then the A15s kick in when something actually stresses the A7s.
The policy for doing so does have to take into account task power as well as instantaneous power. A slow-running job on the A7s could complete far faster on the A15s, and thus actually use less power overall. Something running constantly but not doing much should run on the A7s.
Given that even today's mobile phone SoCs have dedicated hardware "cores" for encryption, video encoding and decoding, etc, this is only Intel trying to generalise the functionality back into the CPU - which is pretty much all Intel know - rather than the more suitable, lower power, dedicated function blocks that are but pinheads even on today's SoCs, never mind the 10nm SoCs in years to come.
Today's quad-core mobile phone SoC on 32nm could be a 16-core cluster-on-a-chip on 16nm, and 64-cores on 10nm.
Never mind the compute-assist on the GPUs in these SoCs - Exynos 5 is over 70 GFLOPS for example, and supports OpenCL 1.1. Never mind the 4K video decode/encode support...
Would you say this is the way forward for other GPU vendors to offer open source drivers - by embedding the actual implementation within the chip itself? Does this also have a benefit in offloading the high-level processing from the system CPU by running that within the GPU as well?
Does the VideoCore IV also incorporate it's own CPU-function that the embedded firmware runs on (I believe the VideoCore lines are fully fledged CPUs in their own right, with large vector/dsp/etc engines? Therefore it's the engines that are totally hidden, with the GPU presenting an API that maps very closely to operations within OpenGL?).
Is there an OpenCL interface on the VideoCore IV? Can one be added by upgrading the firmware, or by implementing it in userland and RPC to a lower level interface on the VideoCore IV?
Other components, for example the GPU, are still very proprietary and closed.
From raspberry pi blog:
As of right now, all of the VideoCore driver code which runs on the ARM is available under a FOSS license (3-Clause BSD to be precise). The source is available from our new userland repository on GitHub. If you’re not familiar with the status of open source drivers on ARM SoCs this announcement may not seem like such a big deal, but it does actually mean that the BCM2835 used in the Raspberry Pi is the first ARM-based multimedia SoC with fully-functional, vendor-provided (as opposed to partial, reverse engineered) fully open-source drivers, and that Broadcom is the first vendor to open their mobile GPU drivers up in this way.
Yup, that's the GPU.
Now the question is how much of the work is done by the firmware on the GPU rather than the driver running on the CPU?
You've also got to consider China with their MIPS64 based Longsoon SoCs. Might not be a threat anytime soon, but with government backing and a desire to do things themselves, it probably will grow and grow, and inevitably start growing outside China as well.
You are aware that different ARM designs have different pipeline lengths depending on their market position? Or is all your knowledge of ARM based upon an ancient university lecture about simple RISC Fetch/Decode/Execute pipelines that the first ARM employed?
Haswell will be available for $20 in its underclocked ultra-low-power configuration?