Or to put it simply, Amdahl's law was wrong, proved by example many times over.
Perhaps you are merely trying to be outré but what exactly is wrong about:
"The speedup of a program using multiple processors in parallel computing is limited by the time needed for the sequential fraction of the program."
The flaw is the presumption that the "sequential fraction" of a program will always dominate as processor count increases. Here is a pretty good discussion. Basically, an embarrassingly large number of algorithms that Amdahl would have regarded as having limited scalability, turned out to have alternative formulations that scale out the yinyang. Ambahl would have been embarrassed by all the embarrassingly parallel parallelism going on, contrary to his dire and arguably self serving prediction.
They could have left ATI the fuck alone and concentrated on doing that which they were really good at. Chip design.
They improved ATI GPUs a lot by bringing in better process technology and throwing more engineers at it than ATI could afford. And they opened up the register specs, which is more than fine by me.
The only workstation class machine with which I have been completely happy is powered by an AMD 4 way Phenom II. Quiet, powerful, cheap, pick all three. And looking around, I would say that its successor is highly likely to be an AMD 6 way, 45 nm process chip. Best value by far for my money.
Today I can choose slightly less latency with Intel or significantly more value with AMD. Call me cheap, but I will take the value, thank you.
Wow, I am totally unimpressed with Intel tonight... the pattern of downmods for favorable comments about AMD strongly suggests hired astroturfers, right down there on a moral par with Apple.
It can't be made parallel. Each pass depends on the previous one. It's dumb that people think everything can be made parallel. Nine women can't have a baby in a month. That's been known for quite a while.
Perhaps you overstate the difficulty. There are many methods of making things parallel. While serializing constraints may in fact exist, it is rare that a problem cannot be factored in such a way as to make most of an algorithm parallelizable in spite of them. Or to put it simply, Amdahl's law was wrong, proved by example many times over. Or to put it equally simply: if you don't try, you can be sure of not succeeding.
It's no secret that RIMs QNX car application platform *actually* powers over 20 million vehicles on the road.
Wow, I didn't know RIM owned QNX until you pointed it out. It's too bad RIM just squats on it like a dog in a manger. QNX really needs to get free. If QNX got free under a copyleft license there would be gazillions of Linux devs in there working on it instantly. Just speaking for myself of course, but ask around if you doubt.
Maybe there are CNC algorithms that aren't easily parallelizable.
I doubt that, being somewhat familiar with the problem space.
It doesn't really matter which as far as the consumer is concerned -- in either case, they will want a chip that maximizes single-threaded performance.
Speak for yourself. I prefer to keep the money in my pocket, and spend it on more frequent full-box upgrades. This keeps me ahead of the curve on average. Example: in a past gig where money was no object I started life with a Core2 class desktop which was state of the art at the time, but no, even when money is no object the beancounters will reject the idea of a new box every six months. In short order my onetime shiny Intel box was being smoked by your bog standard mail order AMD box.
Or (more likely) they can be paralellized, but the CNC development teams haven't got around to doing that yet.
If you add up the user-affecting latency over the lifetime of the processor, and multiply by your hourly rate, you would have to be a fairly lowly serf to not pay for it several times over.
Oh indeed, they pay me peanuts. I should send somebody an email.
Or maybe I should just be thankful that today I have a supercomputer[1] class computer under my desk for which I paid peanuts.
Whatever makes a better processor is a good thing, but I find it ironic AMD promoting higher clock speeds after renaming their processors due to the clock speed wars.
It is not ironic, rather it is because returns from superscalar design are diminishing while feature size keeps shrinking and other incremental technology improvements keep delivering higher practical clock rates.
Agree. The multi-core trend was more to address inefficiency in CPU design, as well as technological limitations in clock speed.
More precisely, it is about seeking the best tradeoff in the Latency*Heat*Cost equation.
In short, GHz is important, as long as it's efficient.
Interesting proposition. I think its a little more complex than that. The main use of GHz today is to paper over the inefficiencies of current-generation single threaded software.
Single core performance is all that matters when processing a toolpath for CNC machining.
Rubbish. There is no way your CNC machining app will even get close to the minimum latency that a single AMD core is capable of. What you are really saying is that your vendor is slow to get a clue about parallel programming.
The only workstation class machine with which I have been completely happy is powered by an AMD 4 way Phenom II. Quiet, powerful, cheap, pick all three. And looking around, I would say that its successor is highly likely to be an AMD 6 way, 45 nm process chip. Best value by far for my money.
Today I can choose slightly less latency with Intel or significantly more value with AMD. Call me cheap, but I will take the value, thank you.
As for Ogg Vorbis, I suspect the patent FUD spread by Fraunhofer pretty much sealed its fate as far as commercial vendor adoption was concerned, which in turn has limited its uptake by the general public.
The fate of ogg is far from sealed. It has grabbed a dominant position in video game assets and is the tech of choice in many other contexts. Wisely, Frauenhofer has not made a peep about its troll patent portfolio. Anybody who uses ogg instead of mp3 today when they do not have to is an idiot, but no denying there is a good supply of such idiots.
Skype currently uses the SILK codec, which should give similar quality.
It doesn't matter what Skype uses, it's sunset technology of minor relevance to the larger landscape now that it's owned by Microsoft. Can you spell "Hotmail"?
Its painfully obvious from your brain vomit in the comment that you have little to no understanding of modern C++.
Dear cowardly wanker,
I do not hesitate to presume that I have written more and better C and C++11 than you. Come back when you are in a correct frame of mind to debate intelligently.
Dear Mr Coward, I am perfectly happy programming in a statically typed language for the applications I currently develop. However, I am against wanking in general, including the sort of wanking you just stooped to.
Slashdot Mirror
Perhaps you are merely trying to be outré but what exactly is wrong about:
"The speedup of a program using multiple processors in parallel computing is limited by the time needed for the sequential fraction of the program."
The flaw is the presumption that the "sequential fraction" of a program will always dominate as processor count increases. Here is a pretty good discussion. Basically, an embarrassingly large number of algorithms that Amdahl would have regarded as having limited scalability, turned out to have alternative formulations that scale out the yinyang. Ambahl would have been embarrassed by all the embarrassingly parallel parallelism going on, contrary to his dire and arguably self serving prediction.
They could have left ATI the fuck alone and concentrated on doing that which they were really good at. Chip design.
They improved ATI GPUs a lot by bringing in better process technology and throwing more engineers at it than ATI could afford. And they opened up the register specs, which is more than fine by me.
The only workstation class machine with which I have been completely happy is powered by an AMD 4 way Phenom II. Quiet, powerful, cheap, pick all three. And looking around, I would say that its successor is highly likely to be an AMD 6 way, 45 nm process chip. Best value by far for my money.
Today I can choose slightly less latency with Intel or significantly more value with AMD. Call me cheap, but I will take the value, thank you.
Wow, I am totally unimpressed with Intel tonight... the pattern of downmods for favorable comments about AMD strongly suggests hired astroturfers, right down there on a moral par with Apple.
It can't be made parallel. Each pass depends on the previous one. It's dumb that people think everything can be made parallel. Nine women can't have a baby in a month. That's been known for quite a while.
Perhaps you overstate the difficulty. There are many methods of making things parallel. While serializing constraints may in fact exist, it is rare that a problem cannot be factored in such a way as to make most of an algorithm parallelizable in spite of them. Or to put it simply, Amdahl's law was wrong, proved by example many times over. Or to put it equally simply: if you don't try, you can be sure of not succeeding.
You first say you picked all three "quiet, "powerful", "cheap". Then you say you dropped the powerful to get the cheap. I'm confused.
Do not be confused, gentle reader. You may understand that as "powerful enough; very powerful indeed". And please do not put words in my mouth.
What was not accurate about that?
It's no secret that RIMs QNX car application platform *actually* powers over 20 million vehicles on the road.
Wow, I didn't know RIM owned QNX until you pointed it out. It's too bad RIM just squats on it like a dog in a manger. QNX really needs to get free. If QNX got free under a copyleft license there would be gazillions of Linux devs in there working on it instantly. Just speaking for myself of course, but ask around if you doubt.
Excellent, this reinforces my decision to go with BMW.
Most CAM software are very CPU intensive for toolpath generation.
All the more reason to parallelize it, cutting latency drastically in the process.
Maybe there are CNC algorithms that aren't easily parallelizable.
I doubt that, being somewhat familiar with the problem space.
It doesn't really matter which as far as the consumer is concerned -- in either case, they will want a chip that maximizes single-threaded performance.
Speak for yourself. I prefer to keep the money in my pocket, and spend it on more frequent full-box upgrades. This keeps me ahead of the curve on average. Example: in a past gig where money was no object I started life with a Core2 class desktop which was state of the art at the time, but no, even when money is no object the beancounters will reject the idea of a new box every six months. In short order my onetime shiny Intel box was being smoked by your bog standard mail order AMD box.
Or (more likely) they can be paralellized, but the CNC development teams haven't got around to doing that yet.
Yes: slow to get a clue.
These days, it's a better value for me to spend the big bucks on Intel workstations and ride them out for an extra year.
Your strategy confuses me. In the "extra" year you will lose big.
If you add up the user-affecting latency over the lifetime of the processor, and multiply by your hourly rate, you would have to be a fairly lowly serf to not pay for it several times over.
Oh indeed, they pay me peanuts. I should send somebody an email.
Or maybe I should just be thankful that today I have a supercomputer[1] class computer under my desk for which I paid peanuts.
[1] As of not very long ago at all.
You first say you picked all three "quiet, "powerful", "cheap". Then you say you dropped the powerful to get the cheap. I'm confused.
Do not be confused, gentle reader. You may understand that as "powerful enough; very powerful indeed". And please do not put words in my mouth.
Whatever makes a better processor is a good thing, but I find it ironic AMD promoting higher clock speeds after renaming their processors due to the clock speed wars.
It is not ironic, rather it is because returns from superscalar design are diminishing while feature size keeps shrinking and other incremental technology improvements keep delivering higher practical clock rates.
Agree. The multi-core trend was more to address inefficiency in CPU design, as well as technological limitations in clock speed.
More precisely, it is about seeking the best tradeoff in the Latency*Heat*Cost equation.
In short, GHz is important, as long as it's efficient.
Interesting proposition. I think its a little more complex than that. The main use of GHz today is to paper over the inefficiencies of current-generation single threaded software.
Single core performance is all that matters when processing a toolpath for CNC machining.
Rubbish. There is no way your CNC machining app will even get close to the minimum latency that a single AMD core is capable of. What you are really saying is that your vendor is slow to get a clue about parallel programming.
The only workstation class machine with which I have been completely happy is powered by an AMD 4 way Phenom II. Quiet, powerful, cheap, pick all three. And looking around, I would say that its successor is highly likely to be an AMD 6 way, 45 nm process chip. Best value by far for my money.
Today I can choose slightly less latency with Intel or significantly more value with AMD. Call me cheap, but I will take the value, thank you.
Anybody who uses [mp3] instead of [ogg] today when they do not have to is an idiot, but no denying there is a good supply of such idiots.
I'm guessing by your context that you meant that the other way round?
Correct of couse. I could add "and anyboldy who posts the exact opposite of what they mean is an idiot".
Ogg may be better than the ancient mp3 codec but it falls flat compared to mp4.
Haha, very funny Mr clueless, you are a real card.
As for Ogg Vorbis, I suspect the patent FUD spread by Fraunhofer pretty much sealed its fate as far as commercial vendor adoption was concerned, which in turn has limited its uptake by the general public.
The fate of ogg is far from sealed. It has grabbed a dominant position in video game assets and is the tech of choice in many other contexts. Wisely, Frauenhofer has not made a peep about its troll patent portfolio. Anybody who uses ogg instead of mp3 today when they do not have to is an idiot, but no denying there is a good supply of such idiots.
Skype currently uses the SILK codec, which should give similar quality.
It doesn't matter what Skype uses, it's sunset technology of minor relevance to the larger landscape now that it's owned by Microsoft. Can you spell "Hotmail"?
Its painfully obvious from your brain vomit in the comment that you have little to no understanding of modern C++.
Dear cowardly wanker,
I do not hesitate to presume that I have written more and better C and C++11 than you. Come back when you are in a correct frame of mind to debate intelligently.
Now Google and Apple are oppressing us. What a joke.
No joke. In the end, the customer is king.
Google probably will applaud it publicly but this is absolutely a big threat to Android.
It's mainly a big threat to hubris, arrogance and certain nascent elements of evil at Google. Hard to complain about that.
Dear Mr Coward, I am perfectly happy programming in a statically typed language for the applications I currently develop. However, I am against wanking in general, including the sort of wanking you just stooped to.
chances that oracle will see the light?
Roughly the same as the chance that Larry Ellison will fly his MiG 25 straight into the side of a mountain.