But isn't the problem that the use of Java and like languages continues through the higher level courses, dumbing them down in order to avoid having to use the more complex and appropriate languages like C & C++?
You are spreading FUD. Just because a piece of software is written to use FOSS systems and libraries does NOT mean it must be given away for free. As others have mentioned, UT2K4 and Doom3 both work on Linux and both cost money. But hey, don't let reality get in the way of your FUDing.
Geeze, I don't know what you are smoking but cut back a bit. What Intel is releasing in the next few months has been years in the pipeline. Same with AMD. It just so happens that because of some release date slips from AMD and some good luck for Intel in overcoming some challenges sooner than thought, Intel can actually choose to release their 45nm product at a competitive time instead of playing catchup to AMD. Believe me, there are enough cool stuff in the pipelines over at Intel to keep you drooling through the end of the decade. They aren't holding it back to screw you over, but because this stuff is hard to make! It takes time to get it right and be able to produce it at on massive scales that can actually be sold at a profit. So chill, cut back a bit on the smokes, and believe me, you will not be disappointed.
Netflix did not pick a side in the HD DVD vs BluRay war. Since they don't have to worry about finite shelf space like BlockBuster does, they are able to provide both formats. I recently changed setting in my Netflix account to provide BluRay discs when they are available and just got one in the mail late last week. So don't go around spreading FUD.
Do you think that once we identify the 'engine' driving a hurricane we could throw a metaphorical wrench in it before causes another Katrina? Also, should we even try? While preventing a hurricane from making landfall and destroying cities would be good for us humans, are the effects of hurricanes an important part of ecosystems/global weather patterns? Basically, can we stop them and if so, would it be a good idea to even try?
So, if a local/state government agency overreacts and spend hundreds of thousands of dollars on something that you basic citizen wouldn't even look at twice (see Mooninites), they can sue the group/person/etc that they overreacted too for large sums of money? How is this not a gigantic loophole just asking for $$$$$ to be "recovered" from people/groups that disagree with said government agencies?
If a group posts fliers and holds rallies against some government official because he is corrupt, couldn't he simply call in the police/feds on the group as a "possible terrorist group", ransack their offices, etc, run up a huge bill and then sue the group out of existence under this new bill?
Based on the article, "TRIPS" is nothing more than a Out-Of-Order(OOO) SuperScalar based processor. So unless the article is grossly simplifying (possible), this is nothing but a PR stunt. And based on the quote from one of the Professors about building it on "nanoscale" technology (um, been doing that for years now), my vote is pure PR BS.
And as an aside, the reason modern CPUs are designed to "only" issue 4 instructions per cycle instead of 16 is because after years of careful research and testing real work applications, 4 instructions is almost always the maximum number of instructions any program can concurrently issue, due to issues like branches, cache-misses, data dependencies, etc. Makes me question just how much these "professors" really know.
The biggest reason to get these cards over other existing ones is for DirectX 10.
I disagree completely. DX10 is a non-issue right now because there are no games to take advantage of it. However, the new processing unit design (no pixel and vertex processors, but a unified processor) greatly increases performance of current games, sometimes significantly. That's a pretty damn good reason for getting these cards (at least from Nvidia right now).
I do agree with you that there is a definite disconnect between industry and academia, which is why I got my MS and then went straight into industry to do real work.
I think you are talking about a different level of multi-threading than I was referring to (and the kind that AMD is researching). The kind of threading you are talking about would be at the programming language level, such as a multi-threaded web server or database server. These already exist and do make great use of raw processing power. But if you are talking about hardware dynamically generating threads from a single threaded program, that is a much different type of multi-threading. Those types of threads are created from sources such as loops (multiple iterations executing at once), function calls (function in one thread and the continuation in another), error handlers, and so on. These suffer from data dependencies between threads. You have to track them and either stall on them or predict them and suffer a missprediction penalty. For programs that are inherently data independent this method works extremely well. But for integer based programs, which most desktop applications are, the interdependence between threads is so high that you are going to have much more stalls and/or misspredictions, completely offsetting the threading performance. And that hasn't even taking into account how you get these inter-thread dependency values between different cores without incurring further performance issues.
Basically, I believe threading should be done at the programming language level and not at the hardware level. Hardware is simply incapable of producing enough parallelism to increase performance enough to justify the investment in hardware, silicon, and effort. Stick to programming good multi-threaded applications (such as web and DB servers) if you really want to take advantage of multi-core systems.
Those are all good ideas that have already been explored. The bottom line in most of these designs is that you don't get much ROI in the form of decreased execution time. For highly specialized applications that have little data inter-dependence, there is a significant increase, but you could get the same anyway by making the program multi-threaded and told to use both cores. Parallel programming is not simple for most applications since they cannot be broken down into many if any parallel tasks to make the error worth the time and effort. The area that Intel/AMD/etc should be targeting is multi-tasking: play a video game while encoding a movie and running Folding@Home without a performance drop in any of the programs.
I did my MS thesis on a topic very similar to this. Trust me, it's not worth it. While some applications that have inherent parallelism (image manipulation, movie encoding/decoding, etc) can see between 2x to 4x improvements when dynamically threaded, the majority of your basic applications are too linear and have too many dependencies between instructions for dynamic threading to really be worth the investment in hardware.
I've used the Vista betas. My first though when I used them were "pretty", just like when I used Mac OSX for the first time. After that, I wasn't impressed by its functionality or features. Now if we had gone from Windows 95/98 to that then I would agree it was worth the wait and the price. But from XP to Vista, it really isn't much of an upgrade.
Also, what makes you think they don't already have a port for DX10 for XP ready or almost complete? They wouldn't release it right now because it would directly compete with one of the selling points on Vista, but they could see it as necessary if Vista doesn't sell well right out of the gate.
I just wanted to respond to this post by saying that is exactly how it should be! Peoples' lives do not cease to exist when they walk in their employer's front door. It is much better to allow people to keep their work and personal lives separate by allowing webmail systems for person emails and cell phones for personal calls. Kudos to your company for recognizing that employees are people and if you treat them as such they will have a much better perception of their work place and be happier about working for you.
It's not a question of MS wanting to make people who can't afford Vista happy. Most people can afford Vista, they just don't want it (for very good reasons). And since MS has been wooing game developers and card manufacturers with the promise of DX10 being the greatest advance in gaming since texture and lighting engines, its safe to bet that MS will be forced to port DX10 back to XP due to the low uptake of Vista. Vista, as a whole, is crap. It took over 5 years to complete and at most it's Windows with a prettier interface (Max OSX) with a very very annoying security system and no drivers (Linux). Why pay $250 - $500 for an operating system 5 years behind the times that doesn't give you anything over XP but DX10?
MS has said that DX10 will be Vista only, so if you are using XP you won't be able to use any of the DX10 features of the Source engine. Of course when MS realizes that almost nobody is buying/using Vista and DX10, they'll make a port for DX10 back to XP. They just won't do it for a year or two.
I agree. I read the summary and was simply dumb struck with how oblivious those Music Exec. are to the true cause of their suffering. Maybe instead of blaming everyone else, look in the mirror you idiots! You have fought digital online music since your first heard the term "MP3". Top that off with suing your own customers in mass, and its no wonder your revenue is falling! People don't want to buy from you!
Eh, everyone makes mistakes. Rambus is a great example of a something that sounded good in the begining, then quickly took a turn in the wrong direction. But as long as people (and companies) learn from their mistakes, they won't repeat them in the future.
While all the technical reasons you listed are valid when looking towards future products, it is naive to think that Intel hasn't seen the same things you have. Intel does not employ stupid people (or at least for very long), so Intel will most likely have very good solutions to those problems by the time they are needed.
If you are going to the make the chips smaller how hard is it to come out with a true quad-core?
Not very. It just takes time to go through all the design -> verification -> layout -> production steps that all new processors require. Yes, the two dual-core processors in a single package is a minor cheat, but it's not meant to be the only quad-core Intel produces. It simply let the consumer have a quad-core chip now which will help push software manufacturers produce programs that utilize the processing resources made available by quad-core. Its an easy bet that Intel already has a true quad-core design in the pipeline and it will be out to consumers in due time. So stop bashing on Intel because they couldn't snap their fingers and have a working quad-core appear on shelves. After all, Intel's solution is a hell of lot better and less expensive than AMD's quad-core solution (2 dual-core chips in the same system).
I've seen two types of "newbies". The first are "newbies" or people who are ignorant of the field/topic but are completely willing to learn from someone with experience. Newbies are considerate users that are a pleasure to help (and also a rarity these days).
The second type are "n00bies". These are people who know just as little as newbies, but think they are experts and everyone else that doesn't agree with them are idiots. N00bies are much more common and are the source of 99% of the headaches for IT professionals.
There is also the issue of "trade secrets" present in the drivers. While you wouldn't think there would be much secrets in a piece of software that acts as a go-between for the OS and hardware, there are more and more cases of drivers actually performing significant "magic" to speed up performance. Releasing that information could give competitors valuable information that could hurt Nvidia. All in all, there are many very good reasons for keeping the driver closed and only a few for opening it up.
But Nvidia could never legally release their drivers as open source. Many people fail to realize that Nvidia has to license many of the technologies in their drivers from other companies and patent holders (just like ATI, Intel, et al must as well). So, even if the CEO of Nvidia wants to release their driver code as open source, they cannot legally do so without the go-ahead from all the license holders, which will never happen.
Apple's iTunes DRM. I can burn as many copies of songs to CDs as I like with nothing getting in the way. I can use my iPod on 5 different computers with no issues. I can plug my iPod into a home or car stereo system and play the music with no issues.
About the only annoying features of their DRM is trying to play the songs under Linux. But even then, someone has already found a work around that isn't that difficult. Overall, I hold Apple's DRM up as a good example of DRM. People can use their songs easily while providing protection from simple sharing for the content producers.
Slashdot Mirror
But isn't the problem that the use of Java and like languages continues through the higher level courses, dumbing them down in order to avoid having to use the more complex and appropriate languages like C & C++?
You are spreading FUD. Just because a piece of software is written to use FOSS systems and libraries does NOT mean it must be given away for free. As others have mentioned, UT2K4 and Doom3 both work on Linux and both cost money. But hey, don't let reality get in the way of your FUDing.
Geeze, I don't know what you are smoking but cut back a bit. What Intel is releasing in the next few months has been years in the pipeline. Same with AMD. It just so happens that because of some release date slips from AMD and some good luck for Intel in overcoming some challenges sooner than thought, Intel can actually choose to release their 45nm product at a competitive time instead of playing catchup to AMD. Believe me, there are enough cool stuff in the pipelines over at Intel to keep you drooling through the end of the decade. They aren't holding it back to screw you over, but because this stuff is hard to make! It takes time to get it right and be able to produce it at on massive scales that can actually be sold at a profit. So chill, cut back a bit on the smokes, and believe me, you will not be disappointed.
Netflix did not pick a side in the HD DVD vs BluRay war. Since they don't have to worry about finite shelf space like BlockBuster does, they are able to provide both formats. I recently changed setting in my Netflix account to provide BluRay discs when they are available and just got one in the mail late last week. So don't go around spreading FUD.
Do you think that once we identify the 'engine' driving a hurricane we could throw a metaphorical wrench in it before causes another Katrina? Also, should we even try? While preventing a hurricane from making landfall and destroying cities would be good for us humans, are the effects of hurricanes an important part of ecosystems/global weather patterns? Basically, can we stop them and if so, would it be a good idea to even try?
If a group posts fliers and holds rallies against some government official because he is corrupt, couldn't he simply call in the police/feds on the group as a "possible terrorist group", ransack their offices, etc, run up a huge bill and then sue the group out of existence under this new bill?
And as an aside, the reason modern CPUs are designed to "only" issue 4 instructions per cycle instead of 16 is because after years of careful research and testing real work applications, 4 instructions is almost always the maximum number of instructions any program can concurrently issue, due to issues like branches, cache-misses, data dependencies, etc. Makes me question just how much these "professors" really know.
All three of them...
I disagree completely. DX10 is a non-issue right now because there are no games to take advantage of it. However, the new processing unit design (no pixel and vertex processors, but a unified processor) greatly increases performance of current games, sometimes significantly. That's a pretty damn good reason for getting these cards (at least from Nvidia right now).
I think you are talking about a different level of multi-threading than I was referring to (and the kind that AMD is researching). The kind of threading you are talking about would be at the programming language level, such as a multi-threaded web server or database server. These already exist and do make great use of raw processing power. But if you are talking about hardware dynamically generating threads from a single threaded program, that is a much different type of multi-threading. Those types of threads are created from sources such as loops (multiple iterations executing at once), function calls (function in one thread and the continuation in another), error handlers, and so on. These suffer from data dependencies between threads. You have to track them and either stall on them or predict them and suffer a missprediction penalty. For programs that are inherently data independent this method works extremely well. But for integer based programs, which most desktop applications are, the interdependence between threads is so high that you are going to have much more stalls and/or misspredictions, completely offsetting the threading performance. And that hasn't even taking into account how you get these inter-thread dependency values between different cores without incurring further performance issues.
Basically, I believe threading should be done at the programming language level and not at the hardware level. Hardware is simply incapable of producing enough parallelism to increase performance enough to justify the investment in hardware, silicon, and effort. Stick to programming good multi-threaded applications (such as web and DB servers) if you really want to take advantage of multi-core systems.
Those are all good ideas that have already been explored. The bottom line in most of these designs is that you don't get much ROI in the form of decreased execution time. For highly specialized applications that have little data inter-dependence, there is a significant increase, but you could get the same anyway by making the program multi-threaded and told to use both cores. Parallel programming is not simple for most applications since they cannot be broken down into many if any parallel tasks to make the error worth the time and effort. The area that Intel/AMD/etc should be targeting is multi-tasking: play a video game while encoding a movie and running Folding@Home without a performance drop in any of the programs.
I did my MS thesis on a topic very similar to this. Trust me, it's not worth it. While some applications that have inherent parallelism (image manipulation, movie encoding/decoding, etc) can see between 2x to 4x improvements when dynamically threaded, the majority of your basic applications are too linear and have too many dependencies between instructions for dynamic threading to really be worth the investment in hardware.
Also, what makes you think they don't already have a port for DX10 for XP ready or almost complete? They wouldn't release it right now because it would directly compete with one of the selling points on Vista, but they could see it as necessary if Vista doesn't sell well right out of the gate.
I just wanted to respond to this post by saying that is exactly how it should be! Peoples' lives do not cease to exist when they walk in their employer's front door. It is much better to allow people to keep their work and personal lives separate by allowing webmail systems for person emails and cell phones for personal calls. Kudos to your company for recognizing that employees are people and if you treat them as such they will have a much better perception of their work place and be happier about working for you.
It's not a question of MS wanting to make people who can't afford Vista happy. Most people can afford Vista, they just don't want it (for very good reasons). And since MS has been wooing game developers and card manufacturers with the promise of DX10 being the greatest advance in gaming since texture and lighting engines, its safe to bet that MS will be forced to port DX10 back to XP due to the low uptake of Vista. Vista, as a whole, is crap. It took over 5 years to complete and at most it's Windows with a prettier interface (Max OSX) with a very very annoying security system and no drivers (Linux). Why pay $250 - $500 for an operating system 5 years behind the times that doesn't give you anything over XP but DX10?
MS has said that DX10 will be Vista only, so if you are using XP you won't be able to use any of the DX10 features of the Source engine. Of course when MS realizes that almost nobody is buying/using Vista and DX10, they'll make a port for DX10 back to XP. They just won't do it for a year or two.
I agree. I read the summary and was simply dumb struck with how oblivious those Music Exec. are to the true cause of their suffering. Maybe instead of blaming everyone else, look in the mirror you idiots! You have fought digital online music since your first heard the term "MP3". Top that off with suing your own customers in mass, and its no wonder your revenue is falling! People don't want to buy from you!
Eh, everyone makes mistakes. Rambus is a great example of a something that sounded good in the begining, then quickly took a turn in the wrong direction. But as long as people (and companies) learn from their mistakes, they won't repeat them in the future.
While all the technical reasons you listed are valid when looking towards future products, it is naive to think that Intel hasn't seen the same things you have. Intel does not employ stupid people (or at least for very long), so Intel will most likely have very good solutions to those problems by the time they are needed.
And Intel has up coming signal chip quad-core CPUs. To think they don't have one in their pipeline would be ignorant.
Not very. It just takes time to go through all the design -> verification -> layout -> production steps that all new processors require. Yes, the two dual-core processors in a single package is a minor cheat, but it's not meant to be the only quad-core Intel produces. It simply let the consumer have a quad-core chip now which will help push software manufacturers produce programs that utilize the processing resources made available by quad-core. Its an easy bet that Intel already has a true quad-core design in the pipeline and it will be out to consumers in due time. So stop bashing on Intel because they couldn't snap their fingers and have a working quad-core appear on shelves. After all, Intel's solution is a hell of lot better and less expensive than AMD's quad-core solution (2 dual-core chips in the same system).
The second type are "n00bies". These are people who know just as little as newbies, but think they are experts and everyone else that doesn't agree with them are idiots. N00bies are much more common and are the source of 99% of the headaches for IT professionals.
I hope this cleared up the confusion :)
There is also the issue of "trade secrets" present in the drivers. While you wouldn't think there would be much secrets in a piece of software that acts as a go-between for the OS and hardware, there are more and more cases of drivers actually performing significant "magic" to speed up performance. Releasing that information could give competitors valuable information that could hurt Nvidia. All in all, there are many very good reasons for keeping the driver closed and only a few for opening it up.
But Nvidia could never legally release their drivers as open source. Many people fail to realize that Nvidia has to license many of the technologies in their drivers from other companies and patent holders (just like ATI, Intel, et al must as well). So, even if the CEO of Nvidia wants to release their driver code as open source, they cannot legally do so without the go-ahead from all the license holders, which will never happen.
About the only annoying features of their DRM is trying to play the songs under Linux. But even then, someone has already found a work around that isn't that difficult. Overall, I hold Apple's DRM up as a good example of DRM. People can use their songs easily while providing protection from simple sharing for the content producers.