One thing I've been really keen to know is what the utilisation is like on those supercomputers. We know they can do LINPACK really fast and more efficiently than the CPUs do, that's what you get for having a high ALU density, a few threads per core and wide SIMD structures. The question is: out of the algorithms that people intended to run on those supercomputers, then what level of efficiency are they hitting.
Are they still a net gain over a standard opteron-based machine? They may be, but I don't know the answer. What I heard about Roadrunner with its Cell chips was not so good.
And python compiles to the same x86 assembly as C++, so clearly turning C++ into python won't change any of the code that exists. A scala compiler doesn't have to compile the Java code that the entire industry expects to still compile perfectly 10 years after they first wrote it.
And who didn't have to deal with backward compatibility. Designing a language from scratch is a completely different problem from evolving one that's heavily used.
The City of London police covers 1.1 square miles with 813 police officers. That's one policeman for every 10 residents, interestingly, though obviously number of residents in the city is a fairly meaningless number.
The NYPD covers 468.9 with 35,284 police officers, or 75 per square mile.
If it's done through withholding then it's not really taken at all, you just don't get it to begin with. Your income is your post tax income. Also if you don't think of it as your money in the first place then it's more of a short term loan that you have to repay. It's all about the way you look at it and can be spun in all sorts of ways.
Organised crime isn't providing you the framework in which to earn the money. If they are providing something worth paying for, then it's no long really extortion. If you can't survive without paying then you could possibly try to avoid it, but it would be counterproductive over all. The more important difference, though, is that governments are (largely) democratic.
It is quite true that not everyone believes that compulsory taxation is necessary. Unfortunately evidence from around the world doesn't offer much hope that they're right.
But it's not slavery. If you weren't paying the taxes you simply wouldn't have the income to not pay from. The entire system would collapse. You'd have no roads to go to work on, no police to stop what you do have being stolen, no laws to prevent the company from exploiting you, no economic stabilisation to stop inflation running out of control and making your income worthless. You simply can't look at the gross income and claim that tax is theft from it, that's ridiculously naive.
Why not just be honest and stop pretending that the money that was "taken" in tax was ever yours to begin with? Without the tax the system wouldn't work and you wouldn't have been able to earn the money. Such is the way of the world and it might as well be accepted.
I'd happily pay slightly higher prices to have the tax included in the quoted price, too. If they also (as in the UK) display the charged tax on the receipt so much the better. The same can be true of service, if they like. Quote all service charges in the prices of the food and stop using "tips" as an artificial way to have higher food prices. If necessary, say on the receipt that a proportion of the food charge was specifically for carrying it to you as opposed to cooking it, which for whatever reason is already included.
The 0.99 gimmick should absolutely die, though, irrespective of whether tax is going to be added on or not.
Unfortunately from the physics simulations I've been working with I'm pretty sure that what they've done there is simply removed physics activity from the non-accelerated version rather than adding it to the accelerated one. A sneaky way of making GPU-accelerated PhysX look better. I'd be shocked if those book effects wouldn't be just as easy at the same framerate on the CPU unless there are truly ridiculous numbers of books.
Why do you need a free *implementation*? The whole point is that OpenCL is a free standard, which is at the heart of AMD's stream SDK (with AMD GPU and CPU backends) and also included in nvidia's SDK. Given that OpenCL has an ICD interface so that you can use any of the backends from any OpenCL-supporting software, I'm not sure what isn't present that you're after other, arguably, than current market penetration.
Not only that, but I just downloaded a Dickens book from google in ePub format. I loaded it into Stanza (made by Lexcycle, owned by Amazon) and exported it as mobipocket format. I then copied it onto my Kindle. It's perfectly readable. Converting between these formats without DRM isn't particularly difficult. Given that the DRM on all of the ebooks seems to be trivial to break, based on comments elsewhere on this thread and what I've seen online about Amazon's DRM, conversion from one ebook to another doesn't seem to be a problem. I was talking to a friend earlier about precisely what you're saying, that the music stores learned the DRM lesson and are changing, the e-book stores will too with time.
Pay £10 for a PAYG SIM, save £7/trip to the cinema on Orange Wednesdays scheme. Get two tickets a year and you're up already. It'd take a lot more trips than that to win on an annual pass, even if they are available.
What facts do you choose to teach them? You want to teach children techniques, using which they can discover the facts they need in the future. You don't want to guess what those facts might be, and try to teach them directly. History lessons, for example, are not really about memorising the dates of battles, but about learning how to compare historical sources and extract information. What facts can you think of that are both important, and will not be learnt by the children on their own?
Or in the UK where the terms "coke" and "cola" are synonymous. If anything, cola is simply not used. You go in and say "A coke please" and the staff have to say "Is Pepsi ok?" and then you just look at them as if they're crazy. Why would Pepsi not be ok?
Possibly, but very few of the people I know who use iPods use iTunes for anything other than copying the music onto it. There's clearly much more to it than that.
I just marked a code generation coursework at a UK university... second year level, admittedly, but later compiler courses are more theoretical on optimisations and so on, so that's not unreasonable.
> 14 789,56 is so much prettier than "14,789.56", isn't it ?
No, certainly not. Which says to me that it's entirely a matter of opinion, and hence you're right about spaces being the only acceptable form to stop confusion. Those horrible French quotes you mention are certainly an abomination too... do you think maybe that one's also based on what you experience when you're young?
People think they have evidence, and maybe they do. More importantly, the whole point of these experiments is to find out if it does work. That is to generate the evidence.
Quickbooks has a different focus from Quicken or Money though, and a much higher pricetag too as a result. Not totally sure what extra it does, but someone must feel it's only a little to spend. In reality of course it is if you're buying that for a company of even one employee, people don't buy Quickbooks for personal use on a regular basis.
Slashdot Mirror
One thing I've been really keen to know is what the utilisation is like on those supercomputers. We know they can do LINPACK really fast and more efficiently than the CPUs do, that's what you get for having a high ALU density, a few threads per core and wide SIMD structures. The question is: out of the algorithms that people intended to run on those supercomputers, then what level of efficiency are they hitting.
Are they still a net gain over a standard opteron-based machine? They may be, but I don't know the answer. What I heard about Roadrunner with its Cell chips was not so good.
No they don't, let's stop such silliness.
The GTX580 has 16 cores. The GTX280 has 32. The AMD 6970 has 24. The AMD Magny-Cours CPUs can have up to 16 (ish, if you don't mind that it's an MCM).
292 indeed. NVIDIA does an even better job of marketing than they do of building chips.
And python compiles to the same x86 assembly as C++, so clearly turning C++ into python won't change any of the code that exists. A scala compiler doesn't have to compile the Java code that the entire industry expects to still compile perfectly 10 years after they first wrote it.
And who didn't have to deal with backward compatibility. Designing a language from scratch is a completely different problem from evolving one that's heavily used.
Although that's such a bad dialog you have to copy it into a text editor and copy it back to be able to edit it properly.
It's never occurred to me to look where I would set that in a config file (short of checking if the old DOS batch files still work).
The City of London police covers 1.1 square miles with 813 police officers. That's one policeman for every 10 residents, interestingly, though obviously number of residents in the city is a fairly meaningless number.
The NYPD covers 468.9 with 35,284 police officers, or 75 per square mile.
If it's done through withholding then it's not really taken at all, you just don't get it to begin with. Your income is your post tax income. Also if you don't think of it as your money in the first place then it's more of a short term loan that you have to repay. It's all about the way you look at it and can be spun in all sorts of ways.
Organised crime isn't providing you the framework in which to earn the money. If they are providing something worth paying for, then it's no long really extortion. If you can't survive without paying then you could possibly try to avoid it, but it would be counterproductive over all. The more important difference, though, is that governments are (largely) democratic.
It is quite true that not everyone believes that compulsory taxation is necessary. Unfortunately evidence from around the world doesn't offer much hope that they're right.
But it's not slavery. If you weren't paying the taxes you simply wouldn't have the income to not pay from. The entire system would collapse. You'd have no roads to go to work on, no police to stop what you do have being stolen, no laws to prevent the company from exploiting you, no economic stabilisation to stop inflation running out of control and making your income worthless. You simply can't look at the gross income and claim that tax is theft from it, that's ridiculously naive.
Why not just be honest and stop pretending that the money that was "taken" in tax was ever yours to begin with? Without the tax the system wouldn't work and you wouldn't have been able to earn the money. Such is the way of the world and it might as well be accepted.
I'd happily pay slightly higher prices to have the tax included in the quoted price, too. If they also (as in the UK) display the charged tax on the receipt so much the better. The same can be true of service, if they like. Quote all service charges in the prices of the food and stop using "tips" as an artificial way to have higher food prices. If necessary, say on the receipt that a proportion of the food charge was specifically for carrying it to you as opposed to cooking it, which for whatever reason is already included.
The 0.99 gimmick should absolutely die, though, irrespective of whether tax is going to be added on or not.
Unfortunately from the physics simulations I've been working with I'm pretty sure that what they've done there is simply removed physics activity from the non-accelerated version rather than adding it to the accelerated one. A sneaky way of making GPU-accelerated PhysX look better. I'd be shocked if those book effects wouldn't be just as easy at the same framerate on the CPU unless there are truly ridiculous numbers of books.
Why do you need a free *implementation*? The whole point is that OpenCL is a free standard, which is at the heart of AMD's stream SDK (with AMD GPU and CPU backends) and also included in nvidia's SDK. Given that OpenCL has an ICD interface so that you can use any of the backends from any OpenCL-supporting software, I'm not sure what isn't present that you're after other, arguably, than current market penetration.
Not only that, but I just downloaded a Dickens book from google in ePub format. I loaded it into Stanza (made by Lexcycle, owned by Amazon) and exported it as mobipocket format. I then copied it onto my Kindle. It's perfectly readable. Converting between these formats without DRM isn't particularly difficult. Given that the DRM on all of the ebooks seems to be trivial to break, based on comments elsewhere on this thread and what I've seen online about Amazon's DRM, conversion from one ebook to another doesn't seem to be a problem. I was talking to a friend earlier about precisely what you're saying, that the music stores learned the DRM lesson and are changing, the e-book stores will too with time.
Pay £10 for a PAYG SIM, save £7/trip to the cinema on Orange Wednesdays scheme. Get two tickets a year and you're up already. It'd take a lot more trips than that to win on an annual pass, even if they are available.
What facts do you choose to teach them? You want to teach children techniques, using which they can discover the facts they need in the future. You don't want to guess what those facts might be, and try to teach them directly. History lessons, for example, are not really about memorising the dates of battles, but about learning how to compare historical sources and extract information. What facts can you think of that are both important, and will not be learnt by the children on their own?
Or in the UK where the terms "coke" and "cola" are synonymous. If anything, cola is simply not used. You go in and say "A coke please" and the staff have to say "Is Pepsi ok?" and then you just look at them as if they're crazy. Why would Pepsi not be ok?
Possibly, but very few of the people I know who use iPods use iTunes for anything other than copying the music onto it. There's clearly much more to it than that.
Easy, you look through the list of people on the conference programme committee, and see if any of them can be usefully cited.
Compilers, originally enough :) Unless that's not what you meant by your question.
Clearly... and having been told so I still am.
Maybe the even splittyness of it depends on the part of the country.
I just marked a code generation coursework at a UK university... second year level, admittedly, but later compiler courses are more theoretical on optimisations and so on, so that's not unreasonable.
Why? Most people in Britain say "14th of March".
> 14 789,56 is so much prettier than "14,789.56", isn't it ?
No, certainly not. Which says to me that it's entirely a matter of opinion, and hence you're right about spaces being the only acceptable form to stop confusion. Those horrible French quotes you mention are certainly an abomination too... do you think maybe that one's also based on what you experience when you're young?
Now, on to date formats...
People think they have evidence, and maybe they do. More importantly, the whole point of these experiments is to find out if it does work. That is to generate the evidence.
Quickbooks has a different focus from Quicken or Money though, and a much higher pricetag too as a result. Not totally sure what extra it does, but someone must feel it's only a little to spend. In reality of course it is if you're buying that for a company of even one employee, people don't buy Quickbooks for personal use on a regular basis.