[...]one of the '4k' resolutions once the necessary displayport and HDMI revisions to run them above 30Hz settle down
You should be OK with DisplayPort for 4K, it has been around for a while. HDMI is more recent and therefore more marginal.
And I totally agree about waiting for 60Hz, 30Hz feels very sluggish for interactive work. I just got a 120Hz monitor and that feels pretty slick for desktops (as well as games, of course!)
"C++, Objective-C, Perl, Python, Java, PHP, C#, D and Go all have block syntax that's derived from C"
And C got the block syntax from B which got it from BCPL which was a simplified version of CPL which was influnced by the first block structured language, ALGOL.
I was taugh ALGOL at University, though I had already been "mentally mutilated beyond hope of regeneration" by BASIC before that...
What it couldn't do (that ENIAC could) is store its program.
Nope, ENIAC couldn't do that either:
"The freeze on design in 1943 meant that the [ENIAC] computer design would lack some innovations that soon became well-developed, notably the ability to store a program."
Everyone who went to school before 1996 was taught that ENIAC was the world's first GP computer.
It depends where you went to school. I was taught that EDSAC was the first fully programmable computer. Earlier devices (including ENIAC) had to be physically re-configured to run each different program using cables and switches, rather than just loading a new program into the same memory that is used for data.
Even if we had known about Colossus at that time (and it is possible that some of my teachers did...) it would not have qualified as a stored-program computer.
On a slight tangent, I've been wondering about this "things are getting worse" meme as it relates to just about anything related to humanity that can be tracked over time.
This is a natural consequence of random changes to personal and social preferences over time. You grow up with the particular set of preferences that is accepted by the majority and they become your norm. Over time the preferences in society change and the majority opinion changes so you naturally move from being in the majority to the minority. What was once the accepted majority view becomes a minority view and your opinons become out of step with the rest of society. The older you get, the more this happens and the more you feel out of touch with the "modern" world.
The real problem comes when we consider the majority opinion to be morally superior to the minority opinion. Since these opinions change randomly over time it becomes inevitable that views and attitudes that we form when young become less morally acceptable when we get old, and the attitude of those around us becomes less morally acceptable to us. So we are left with the choice of conforming to the current morality or sticking with the morality that was prevalent when we first formed our moral framework.
The alternative is to follow an absolute moral framework that does not change over time. In this case you will always be out of step with the rest of the world, but at least it won't get worse as you get older!
There are major problems in the world (e.g. poverty, disease, and conflict). Our best hope for reducing these problems is factual observation and logical reasoning - i.e. science.
No, science is just a tool that can be used to predict the outcome of any changes that we make to the world. Science is amoral and has been been used to create the issues you describe as much as it has alleviated them.
Science does not say that poverty, disease or conflict are "problems" so it cannot give "hope" that they can be reduced.
Which makes me wonder, would it (in principle) be worth designing a chip with an ISA that is based explicitly on single-assignment-form, thereby avoiding both the need for transformations by the compiler and (more importantly) transformations by the CPU at run-time?
If the ISA is based on single assignment then the compiler will still have to transform the code into this form. In practice the compiler does this anyway when it can and super-scaler hardware does a good job of executing this kind of code. The problem (returning to the subject of the thread) is that not all code fits this pattern and in these cases a single-assignment ISA will perform particularly badly so it is not suitable for general purpose processors.
Researchers have be developing ISAs for decades (including the single-assignment approach on dataflow computers in the 1980s) and the current mainstream ISAs are a reflection of that research. There is no magic new way of executing general purpose code.
Using abstract classes in C++ (with multiple inheritance) [...] doesn't really guide the coding process to protect yourself the way compiler recognized interfaces do.
I'll bite.
The AC above makes it clear that we are talking about people using language features appropriately, so how do "compiler-recognized interfaces" provide such a significant benefit over appropriate use of C++11 abstract base classes?
Uhh, USB 3.0 has a battery charge specification that allows for 1.5A
Yes it does, but that is designed for charging batteries and doesn't allow data transfer at that current. So while you can technically power one of these chips using a cable with a USB connector on it, that isn't what I would call a USB-powered device.
Well, if the peak is 6w, that is easily in the power envelope of USB. (6w = 5v@1.1a)
The maximum current for USB 2.0 is 500mA and for USB 3.0 it is 900mA, so 6W is well outside the power budget for a USB-powered device. 4.5W is possible on USB 3.0 as long as you don't want power for anything else like a screen, memory or other peripherals. It is more manageable with the 3.5W option but at 600MHz you are way below the performance of equivalent ARM-based parts with the same power profile, so you have to really want x86 in your device.
Science is not a "political decision", but you're right that China and India are great contributors toward the problem.
Science is not a political decision, it just makes predictions. The issue becomes political as soon as you apply value judgements to those predictions by, for example, calling Golobal Warming a "problem". Science does not say that global warming is a problem, it just predicts different outcomes based on various possible courses of action. Science cannot make value judgements about those outcomes, it is inherently amoral.
Please note that I do believe that global warming is a problem and that action needs to be taked at a personal, national and international level to reduce the impact of climate change. But that is based on applying my moral viewpoint to the underlying science, it does not come from the science alone.
Intel is second comer to the party and thus the ecosystem is clearly stacked against them, but they still managed to get their components in the market.
It is amazing what you can acheive if you are prepared to lose (sorry, I mean "Contra Revenue") $1B/quarter by massively subsidsing development and silicon costs.
I'd like to see some evidence that the performance gain due to more registers outweighs the performance loss due to fewer pointers per cache in the majority of cases.
I'd like to see some evidence for your claim that pointer-intensive 64-bit programs run more slowly than pointer-intensive 32-bit programs.
For what workloads do the extra general purpose registers of x86-64 outweigh the cache hit from larger pointers?
Almost every workload, I would imagine, because those extra registers significantly reduce the need to go to the cache in the first place and can increase the IPC. If you are using so many pointers that the size in the cache is significant then your workload probably has so much code and data divergence that cache occupancy is not the limiting factor on performance.
But if you have some figures that contradict this I'd be interested in seeing them.
The cryosphere page at University of Illinois-Champagne shows that we are currently seeing more sea ice than the average, and the levels have been sharply rising the last few years.
It is the same effect: The ice on the land is melting and flowing into the sea where some of it re-freezes.
The area of ice is increasing, the mass of ice is decreasing.
"Go is the runner-up but still significantly slower with medium effect size: the average Go program is 18.7 times slower than the average C program. Programs in other languages are much slower than Go programs, with medium to large effect size (4.6–13.7 times slower than Go on average)."
My only objection is that they classify Go as "procedural" along with C, Ada, PL/1 and FORTRAN. It may not have inheritance (a good thing in my book!) but it has many OO features including support for abstraction and encapsulation.
You helpfully describe two different approaches to tackling ideas about reality, but I'm not sure it is a good idea to personify it in this way. It is better to look at how people behave in specific situations rather than apply one or other characteristic to everything an individual does. Any individual scientist will make some theories and attempt to disprove them but they will also accept other theories without proof.
I have a science degree so I probably count as a scientist but I don't apply the scientific method to everything I do. This is partly because I am too lazy to use it for everything, partly because I know the danger of "overthinking" things and partly because there are things in my life for which the scientific method simply does not apply because they are not measurable or repeatable.
People may not always use science deliberately, but using direct experience along with inductive and deductive reasoning is the bedrock of scientific discovery.
The bedrock of scientific discovery is reason applied to experimentation and measurement, not experience. Science requires repeatability and experience is not repeatable.
More importantly, the use of logic and reason is not restricted to science.
The scientific method is not the only way to gain knowledge, but it is also not the only method in which people perform scientific studies.
If you are not using the scientific method then you are not doing science. The clue is in the name.
[Science] does not, by any stretch of the imagination, represent the sole mechanism for understanding the world.
Yes it does. Without logical arguments there is no verifiable way to ensure you know anything. Under circumstances where too much is unknown you can use very weak arguments, but they must still be backed up by some form of logical reasoning.
Again, logic and reason are not science, they are tools that are used by science that can also be used outside of science to gain understanding of the world.
Science exists because it is the only process of understanding the world in a way that can provide useful results. Or at least the only way we have found so far.
Nonsense. People understood the world before long science was invented, and very little of our useful understanding of the world comes from science. Most of it comes from direct experience or the experience passed on to us by others.
Science exists because people invented it. It survives because it is a useful tool for making predications about some of the simpler aspects of the world. It does not, by any stretch of the imagination, represent the sole mechanism for understanding the world.
An independent Scotland would already be in the EU
Not really. The SNP White Paper was clear that this was a matter for negotiation: "Following a vote for independence, the Scottish Government will immediately seek discussions with the Westminster Government and with the member states and institutions of the EU to agree the process whereby a smooth transition to full EU membership can take place on the day Scotland becomes an independent country."
The President of the European Commission said "A new independent state would, by the fact of its independence, become a third country with respect to the EU and the Treaties would no longer apply on its territory" and when interviewed said it would be "extremely difficult, if not impossible" for an independent Scotland to secure membership.
What makes the crash of 2008 any different from the series of crashes that preceeded it? The UK has come out of it relatively stronger than most of the rest of Europe.
The UK has mostly go over losing the Empire and not being the world power that it used to be. It will be interesting to see how the US copes as the same thing happens to it over the next few decades. I fear it could go very badly...
Slashdot Mirror
[...]one of the '4k' resolutions once the necessary displayport and HDMI revisions to run them above 30Hz settle down
You should be OK with DisplayPort for 4K, it has been around for a while. HDMI is more recent and therefore more marginal.
And I totally agree about waiting for 60Hz, 30Hz feels very sluggish for interactive work. I just got a 120Hz monitor and that feels pretty slick for desktops (as well as games, of course!)
"C++, Objective-C, Perl, Python, Java, PHP, C#, D and Go all have block syntax that's derived from C"
And C got the block syntax from B which got it from BCPL which was a simplified version of CPL which was influnced by the first block structured language, ALGOL.
I was taugh ALGOL at University, though I had already been "mentally mutilated beyond hope of regeneration" by BASIC before that...
What it couldn't do (that ENIAC could) is store its program.
Nope, ENIAC couldn't do that either:
"The freeze on design in 1943 meant that the [ENIAC] computer design would lack some innovations that soon became well-developed, notably the ability to store a program."
http://en.wikipedia.org/wiki/E...
ENIAC was the first general purpose, stored program computer.
Nope, the program was not stored in the computer, it was configured using switches and cables.
But perhaps you meant to say "EDSAC" :)
Everyone who went to school before 1996 was taught that ENIAC was the world's first GP computer.
It depends where you went to school. I was taught that EDSAC was the first fully programmable computer. Earlier devices (including ENIAC) had to be physically re-configured to run each different program using cables and switches, rather than just loading a new program into the same memory that is used for data.
Even if we had known about Colossus at that time (and it is possible that some of my teachers did...) it would not have qualified as a stored-program computer.
[..] When you have a million versions of closed-source MS Office files floating around, this shit happens. Another reason to use open formats.
Is there an Internet Law that says "Whatever the real cause of the problem, there is always someone who will blame Microsoft"?
On a slight tangent, I've been wondering about this "things are getting worse" meme as it relates to just about anything related to humanity that can be tracked over time.
This is a natural consequence of random changes to personal and social preferences over time. You grow up with the particular set of preferences that is accepted by the majority and they become your norm. Over time the preferences in society change and the majority opinion changes so you naturally move from being in the majority to the minority. What was once the accepted majority view becomes a minority view and your opinons become out of step with the rest of society. The older you get, the more this happens and the more you feel out of touch with the "modern" world.
The real problem comes when we consider the majority opinion to be morally superior to the minority opinion. Since these opinions change randomly over time it becomes inevitable that views and attitudes that we form when young become less morally acceptable when we get old, and the attitude of those around us becomes less morally acceptable to us. So we are left with the choice of conforming to the current morality or sticking with the morality that was prevalent when we first formed our moral framework.
The alternative is to follow an absolute moral framework that does not change over time. In this case you will always be out of step with the rest of the world, but at least it won't get worse as you get older!
There are major problems in the world (e.g. poverty, disease, and conflict). Our best hope for reducing these problems is factual observation and logical reasoning - i.e. science.
No, science is just a tool that can be used to predict the outcome of any changes that we make to the world. Science is amoral and has been been used to create the issues you describe as much as it has alleviated them.
Science does not say that poverty, disease or conflict are "problems" so it cannot give "hope" that they can be reduced.
Which makes me wonder, would it (in principle) be worth designing a chip with an ISA that is based explicitly on single-assignment-form, thereby avoiding both the need for transformations by the compiler and (more importantly) transformations by the CPU at run-time?
If the ISA is based on single assignment then the compiler will still have to transform the code into this form. In practice the compiler does this anyway when it can and super-scaler hardware does a good job of executing this kind of code. The problem (returning to the subject of the thread) is that not all code fits this pattern and in these cases a single-assignment ISA will perform particularly badly so it is not suitable for general purpose processors.
Researchers have be developing ISAs for decades (including the single-assignment approach on dataflow computers in the 1980s) and the current mainstream ISAs are a reflection of that research. There is no magic new way of executing general purpose code.
Using abstract classes in C++ (with multiple inheritance) [...] doesn't really guide the coding process to protect yourself the way compiler recognized interfaces do.
I'll bite.
The AC above makes it clear that we are talking about people using language features appropriately, so how do "compiler-recognized interfaces" provide such a significant benefit over appropriate use of C++11 abstract base classes?
Uhh, USB 3.0 has a battery charge specification that allows for 1.5A
Yes it does, but that is designed for charging batteries and doesn't allow data transfer at that current. So while you can technically power one of these chips using a cable with a USB connector on it, that isn't what I would call a USB-powered device.
Well, if the peak is 6w, that is easily in the power envelope of USB. (6w = 5v@1.1a)
The maximum current for USB 2.0 is 500mA and for USB 3.0 it is 900mA, so 6W is well outside the power budget for a USB-powered device. 4.5W is possible on USB 3.0 as long as you don't want power for anything else like a screen, memory or other peripherals. It is more manageable with the 3.5W option but at 600MHz you are way below the performance of equivalent ARM-based parts with the same power profile, so you have to really want x86 in your device.
Science is not a "political decision", but you're right that China and India are great contributors toward the problem.
Science is not a political decision, it just makes predictions. The issue becomes political as soon as you apply value judgements to those predictions by, for example, calling Golobal Warming a "problem". Science does not say that global warming is a problem, it just predicts different outcomes based on various possible courses of action. Science cannot make value judgements about those outcomes, it is inherently amoral.
Please note that I do believe that global warming is a problem and that action needs to be taked at a personal, national and international level to reduce the impact of climate change. But that is based on applying my moral viewpoint to the underlying science, it does not come from the science alone.
Intel is second comer to the party and thus the ecosystem is clearly stacked against them, but they still managed to get their components in the market.
It is amazing what you can acheive if you are prepared to lose (sorry, I mean "Contra Revenue") $1B/quarter by massively subsidsing development and silicon costs.
That should be "The Doctor to teach kids to code".
The programme is "Doctor Who", the lead character is "The Doctor".
I wish I could mod "-1 Xenophobic"...
I'd like to see some evidence that the performance gain due to more registers outweighs the performance loss due to fewer pointers per cache in the majority of cases.
I'd like to see some evidence for your claim that pointer-intensive 64-bit programs run more slowly than pointer-intensive 32-bit programs.
For what workloads do the extra general purpose registers of x86-64 outweigh the cache hit from larger pointers?
Almost every workload, I would imagine, because those extra registers significantly reduce the need to go to the cache in the first place and can increase the IPC. If you are using so many pointers that the size in the cache is significant then your workload probably has so much code and data divergence that cache occupancy is not the limiting factor on performance.
But if you have some figures that contradict this I'd be interested in seeing them.
The cryosphere page at University of Illinois-Champagne shows that we are currently seeing more sea ice than the average, and the levels have been sharply rising the last few years.
It is the same effect: The ice on the land is melting and flowing into the sea where some of it re-freezes.
The area of ice is increasing, the mass of ice is decreasing.
The biggest suprise for me is how well Go does:
"Go is the runner-up but still significantly slower with medium effect size: the average Go program is 18.7 times slower than the average C program. Programs in other languages are much slower than Go programs, with medium to large effect size (4.6–13.7 times slower than Go on average)."
My only objection is that they classify Go as "procedural" along with C, Ada, PL/1 and FORTRAN. It may not have inheritance (a good thing in my book!) but it has many OO features including support for abstraction and encapsulation.
A scientist has an idea about reality [...]
A non-scientist has an idea about reality [...]
You helpfully describe two different approaches to tackling ideas about reality, but I'm not sure it is a good idea to personify it in this way. It is better to look at how people behave in specific situations rather than apply one or other characteristic to everything an individual does. Any individual scientist will make some theories and attempt to disprove them but they will also accept other theories without proof.
I have a science degree so I probably count as a scientist but I don't apply the scientific method to everything I do. This is partly because I am too lazy to use it for everything, partly because I know the danger of "overthinking" things and partly because there are things in my life for which the scientific method simply does not apply because they are not measurable or repeatable.
People may not always use science deliberately, but using direct experience along with inductive and deductive reasoning is the bedrock of scientific discovery.
The bedrock of scientific discovery is reason applied to experimentation and measurement, not experience. Science requires repeatability and experience is not repeatable.
More importantly, the use of logic and reason is not restricted to science.
The scientific method is not the only way to gain knowledge, but it is also not the only method in which people perform scientific studies.
If you are not using the scientific method then you are not doing science. The clue is in the name.
[Science] does not, by any stretch of the imagination, represent the sole mechanism for understanding the world.
Yes it does. Without logical arguments there is no verifiable way to ensure you know anything. Under circumstances where too much is unknown you can use very weak arguments, but they must still be backed up by some form of logical reasoning.
Again, logic and reason are not science, they are tools that are used by science that can also be used outside of science to gain understanding of the world.
Science exists because it is the only process of understanding the world in a way that can provide useful results. Or at least the only way we have found so far.
Nonsense. People understood the world before long science was invented, and very little of our useful understanding of the world comes from science. Most of it comes from direct experience or the experience passed on to us by others.
Science exists because people invented it. It survives because it is a useful tool for making predications about some of the simpler aspects of the world. It does not, by any stretch of the imagination, represent the sole mechanism for understanding the world.
An independent Scotland would already be in the EU
Not really. The SNP White Paper was clear that this was a matter for negotiation:
"Following a vote for independence, the Scottish Government will immediately seek discussions with the Westminster Government and with the member states and institutions of the EU to agree the process whereby a smooth transition to full EU membership can take place on the day Scotland becomes an independent country."
The President of the European Commission said "A new independent state would, by the fact of its independence, become a third country with respect to the EU and the Treaties would no longer apply on its territory" and when interviewed said it would be "extremely difficult, if not impossible" for an independent Scotland to secure membership.
That was before 2008.
What makes the crash of 2008 any different from the series of crashes that preceeded it? The UK has come out of it relatively stronger than most of the rest of Europe.
The UK has mostly go over losing the Empire and not being the world power that it used to be. It will be interesting to see how the US copes as the same thing happens to it over the next few decades. I fear it could go very badly...