Can you blame me for being inaccurate? I was playing with legos and playing kick-ball, not playing with computers (well, the Apple IIe in school back then, but I didn't really know what it was until later). I'm more focused on current and future technology, not ancient history (which is what Windows 1.0 is to modern computer tech).
Re:End of intel as likely as end of IBM
on
IBM Opts for AMD
·
· Score: 1
Who wins? So far, the customers.
Now that I think everyone can agree on! We (should)have processor cycles to spare (if it wasn't for the modern bloat today's programers write).
Well you are obviously wrong or they would have gone out of business by now! $5/month is not much of an income, per person, so they would need to have a fairly large user base just to keep their offices heated and the lights turned on.
Besides, Windows started out as just a "work" OS (as all computer were for "work" back in those days). People got used to Windows and when computers became more common in homes they went with what was familiar (or the only thing available), which was Windows. Developers started making Windows game, MS made APIs to help with that (DirectX anyone?), etc. With Linux getting more popular, not just for servers but for many different levels including the desktop, it's only natural that people would want some nice games to play when they don't "have" to be working.
Re:End of intel as likely as end of IBM
on
IBM Opts for AMD
·
· Score: 2, Insightful
Time to build a new fab, from the ground up, is around 4 years (and 4billion dollors). Last time I looked, AMD wasn't expanding very quickly or had the assets to build many fabs in parallel. Intel's current generation sucks, even Intel people will admit that. But the current direction is much more efficent and has great performance, so I definatly think we'll see Intel regain market share.
Re:End of intel as likely as end of IBM
on
IBM Opts for AMD
·
· Score: 1
I agree, all these "the end of Intel" comments are so off base it's comical. First, Intel is coming out with new chips that are already kicking the butt of AMD's best chips. Second, Intel still has over 50% of the market, no matter how you look at it. And third, AMD does not have the production capacity to provide chips for 100% of the market share even if they had it. That means that even if the market demanded more AMD, they would have to either turn them away or put them on a very very long waiting list (think year+). Why wait when you can get Intel right now? Intel isn't going anywhere, but AMD has always worked on very thin profit margins. If they can't respond with a very good line of new chips in the next 6 months, I can easily see their stock falling and profits going south very fast.
I'll give you that the data sets programs are using today are getting gigantic, which can easily lead to constant memory block swaping between the main memory and the caches. But when it comes to instruction caches, you obviously haven't heard of the 90/10 Locality rule of thumb: a program executes about 90% of its instructions in 10% of its code. That's because of branches, loops, the fact that there are large sections of code that are run only once, during initialization, and never run again, etc. So while the Java run time engine is larger than the L2 cache in all but the most expensive workstation processors, the majority of the instruction that are executed are only a small subset of the actual code, which can fit easily in typical L2 caches.
If you look at Intel's Core 2 Duo, the cache space is not "divided" as the number of cores increase. Each core, if running at full load, will have 2MB of cache (extreme edition anyway). That is a very respectivable cache size and would be a respectable single core processor. When one core is not running (like when running only Word), one core sleeps while the other core is given all of the cache.
Past marking ploys (GHz) were definately wrong, and trying to directly replace those metrics with the number of cores is also a bad choice. But don't you see that that is exactly what Intel is trying to prevent? The interviewee in the article is saying that more cores != more performance. Hence why desktop users will have no need for 8 cores or more. Most of the posts on this topic are along the lines of "ya right, more cores FTW!", which is a very uninformed mentality.
I don't know if you can qualify that is "good" or "bad", on either AMD's or Intel's part. Internal testing numbers show that using the FSB for communications with the RAM does not hurt performance. So you could postulate that AMD's solution is a bit of overkill for current needs. Of course, in 3-5 years we may all be using a system to what AMD is using right now due to an increased need for memory bandwidth, or we could still be where we are right now (I doubt it). But for right now, with current systems, neither is absolutely better than the other. Both meet the needs of the system and are not a bottleneck to performance.
I'm not too familiar with the specifics of the AMD implementation, but from what I have picked up they have a dedicated connection from the memory to the CPU instead of sharing a bus with other data (I/O, DMA traffic, etc). That's a nice setup and I don't think there is anything wrong with that. I have no idea if the test was with 2, 4, or 8 cores, but it's pretty safe to assume is was done with 2 cores. Therefore, when Core 2 Duo comes out the connection to the main memory should not be a performance bottleneck. The available bandwidth is plenty and even has some headroom to grow.
Ah, but will those applications truely take advantage of all those cores? Just because a program is multithreaded does not mean you can simply throw more cores at it and get ever increasing performance. That would denote a perfectly linear performance increase, which almost nothing does (a raytracer comes close on complex scenes).
The only real programs that will take advantage of large numbers of cores will be scientific applications. Games? You offload the AI to one core, the sound to one (or the nice sound card), network handling to another, I guess user input to one... The most computationaly intensive tasks are the AI work and the graphs, which is already done in the GPU. So unless you are going to run ultra realistic, human-like AI work, there really isn't much parallel work that can be done from games.
Multimedia can get a nice boost from those cores, but then you run into other issues such as hard drive speed and memory space (8 cores will quickly fill 2GB of memory with data if given the opportunity). Even then, does the increase in complexity both in the motherboard and huge power consumption justify shaving off a few minutes from transcoding some movies?
Workstation level work will definately use 8+ cores, there is no question. The issue is will your average or even hardcore desktop user see any benifit or have the benifit worth the cost? I think the person in the interview is right for the forseable future. Maybe in a decade or so when the CompSci people figure out new ways to do parallel work, 8 cors will make sense at home.
It depends on which memory bottleneck you are talking about. There is a memory hierarchy in computers, with the fastest also being the closest to the processor, the level 1 or L1 cache (usually split into separate data and instruction caches). This is then tied into a much larger, but slower, L2 cache (combined instruction and data lines). Some processors use an L3 cache, but not many these days. Current processors have L1 and L2 directly on the chip. If you see those die pictures they show off to the press, the largest areas of the chip are the caches. Finally, the chip can go across the front side bus and access the main system memory, which is very large compared to the L2 and L1 caches, but much slower in terms of number of cycles to access.
So which bottleneck are you refering to? The new Core 2 Duo chips of Intel's share the L2 cache and, as far as I can tell from the reviews I have read, this setup works very well. Both chips can share data very quickly or when executing a single sequential program one of the cores can use all of the L2 cache (which in the Extreme Edition verion is up to 4MB!). Or are you refering to the main memory? It is possible for both cores to need to access the main memory at the same time, but modern pre-fetching and aggress speculation techniques reduce how often that occurs and the timing penalties when they do occur. And of course, the larger the L2 cache the more memory can be stored on the chip at once, reducing the need to access the main memory very often. According to Intel's own internal testing, they had a very hard time using all of the bandwidth the current front side bus and memory offers, which means the main memory shouldn't be a bottleneck.
There was never "numerous intelligence agencies" agree with WMD's in Iraq. It was just a few field agents and even they said they think their intel is circumstantial and weak. But that was enough of an excuse to get thousands of American soldiers and thousands of Iraqi citizens killed.
At this point there is enough evidence to at least start an impeachment, and this time have to go all the way. Bush has committed actual crimes here. There might even be a few in there that can be considered treasonous! But will Congress even whisper the word 'impeach'? Nope, because the Republicans have the majority. Even if, as you say, there is a big fracture in the Republican party right now, they won't lift a finger to actualy do anything to stop Bush or punish him.
This is Bush's last term, he knows this, and he knows he is pretty much untouchable, so he does as he pleases. This was actually my greatest fear in giving him another term. His behavior during his first term was horrible and put the safty of the American people in more danger than any terrorist with bombs strapped to their bodies.
And yet, the Republican party cheered him! They called him our savior and a good leader and a good person. And they came out of the woodworks to vote for him again. And now you are saying that a big chunk of them regrette their decision? You'll excuse me if I don't find this reasuring or think very much of Republicans right now. They helped get a criminal elected to office. If they really hate what he is doing, they have the power to remove him, legally. But I double we will ever see a Republican Congress convict a Republican President.
I love all faith in the Democratic party to lead anything. We have too many different groups, each with their own agenda. The best analogy would trying to herd a bunch of cats.
The Republicans on the other hand are much more guided and on the same page with each other. You have two main groups, the social conservatives (right-wing religious types) and the fiscal conservatives. It's much easier to get the Republican base out in force to vote the same issues up and down than it is to get the Dems out to do the same.
I doubt we will see a Democratic majority for a while. The Republican may have made some major mistakes (and many illegal ones at that), but the general public still sees them as more able to lead than the Democrats. We will just have to wait for them to commit such a horrible crime that even the most staunch conservative can no longer stand to see them in power. We are almost there, but not yet.
That is definatly true, but places like those are the minority. The majority of newly trained programers and engineers will be coming from the State colleges and Universities. Just because your university doesn't have the same name recognition as MIT does not mean that you are incapable of meeting the performance requirements of a given job. Heck, I would say that MIT has made sure they the people they graduate are perceieved as being overly qualified. But is the person that graduated bottom of their class at MIT really better than the person who graduted within the top 10 at a State University? That's a question that no one can really answer without interviewing both students, which is a long and expensive process.
About the time AMD does, Intel will be moving towards their 45nm or 40nm or whatever their next smallest size is. Intel has the best fabs for a reason: they invested a LOT of money into fab R&D just after the bubble burst. They called is their One Generation Ahead strategy. While everyone else was trying to cope with the loss of capital and drop in stock prices, Intel want to make sure that they came out one generation of silicon manufacturing ahead and stayed that way. While they have in some respects (physical size of transisters) they have missed some advances in some other areas (SOI). In the end, Intel will probably stay ahead of AMD until they hit the physical wall on the size of transistors (can't go smaller than an atom!).
"Did you miss the part where a local Linux kernel exploit was used in the attack?"
True, it was part of the exploit, but first the attacker(s) had to get onto the system using a developer's account. Now I don't pretend to know what kind of rights the developer accounts have, but they would definately have more power than a "guest" account or the account used by apache if that had been hacked. This means that if the account hadn't been hacked, then the server wouldn't have been hacked (exploit requires user access to the server). The kernel exploit is bad, but at least they are upfront about it and don't try to hide it behind PR/marketing speak or not tell anyone about it.
You understand that the MPAA ratings boards is not a federal rating, right? If you don't agree with a movie's rating, inform the board and sight specific examples in the movie.
After all, who is it that has problems with movies rated R?
Children
Mentally challenged people (see children)
Bible-thumping ultra-religious types
As far as the children are concerned, the parents should be parents and screen what their young children watch (you know, like a parent). When you are talking about the different between PG13 and R, is it really that bad that your 16 year old saw a boobie for 0.75seconds? Is it going to warp him? If it does, see the mentally challenged people. As for the ultra-religious types, something tells me that they won't be interested in the type of movies that border on PG13-R. They would rather watch a movie that is condoned by their church instead of looking at evil, sinful stuff like the rest of us.
When it comes down to it, if you are offended by contents that 99% of adults don't even smirk at, you should be researching your film choices very very very very carefully. If you see something bad you just might go insane. As for the children, parents should be parents and take control of the information that their children consume. If you can't, you shouldn't be a parent.
Or, you know, they could... NOT WATCH MOVIES ABOVE THE RATING THEY ARE COMFORTABLE WITH. That is the simplest solution to this whole mess. Don't like the fact that there is sex and/or nudity in a movie? DON'T WATCH IT! See how easy that is? And no one is forced to bend to another's will.
If you look at TV, sooner or later an 'edited' version will be played on TV that has been 'sanatized' for those of weaker constitutions. The copyright holders can choose to provide an edited version, but they should never be forced into providing one.
But the rules in place are what currently allow competition! It's obvious you don't understand what laws are currently in place, such as the Common Carrier laws. Without regular, the communication lines, whether they are cable, telephone, or fiber optics, will be owned and USED by only one company. Right now if AT&T starts doing things that I don't like, I can go to Verizon or Qwest etc. But if the regulation are removed, AT&T can simply refuse to let me use another carrier since they own the lines in the ground up to my residence. And they can refuse to take any calls from Verizon customers. This is why we have regulations. It prevents the companies from being so territorial that in their pursuit of more money they make life worse for their customers. And since we only have a few competitors out there (how many nation wide telcos can you name?), we need these rules in place.
Going to the government isn't wanting a single person to fix the problems. Remember, our government represents the PEOPLE and is made up of representatives that the PEOPLE elect. We have placed power in both the Federal government, which deals with national issues, and in the State government, which focus on issues in the state. In issues like this, the Federal government, which is made of the PEOPLE, has the power to standup for the PEOPLE. It's not collectivist government, it's representative government.
In some cases the less the government meddles the better. But in some cases, such as this, the government needs to step in a fill the primary reason for having a government: to protect those that cannot protect themselves. In the end, it is the consumers vs the telcos. The consumers do not have the political or economic power to prevent the telcos from forcing us to take it up the ass. The only way we, as consumers, could force the telcos to change their behavior is either cut off all access to the rest of the world (stop watching TV, making phone calls, using the internet, etc) or get the government to stand up for us. Since this is not an even playing field we as consumers must rely on the government to fight this battle for us.
In this case, the less regulation the more likely the telcos are to screw us over and make our lives worse, all to make their bottem line get just a little bit bigger.
That only works when the market is playing fair. The GP used gas which is a good example of an unfair commodity. Just because businesses don't like the high price of gas doesn't mean they can just switch to another brand of gas. They all pretty much cost the same since the supply is controlled by a limited number of companies. The Internet is the same, with the majority of the network making up the backbone being owned by only a handful of companies. These companies are also the telcos! So even if the market doesn't want the BS pricing scheme from the telcos we don't have a choice. Not unless everyone wants to get together and build a new Internet. In the end, all the arguments about the market and supply and damand are just crap. Those ideas only work when the market is fair and there is competition. The gas supply and the Internet backbone are not fair.
Slashdot Mirror
Can you blame me for being inaccurate? I was playing with legos and playing kick-ball, not playing with computers (well, the Apple IIe in school back then, but I didn't really know what it was until later). I'm more focused on current and future technology, not ancient history (which is what Windows 1.0 is to modern computer tech).
Now that I think everyone can agree on! We (should)have processor cycles to spare (if it wasn't for the modern bloat today's programers write).
Besides, Windows started out as just a "work" OS (as all computer were for "work" back in those days). People got used to Windows and when computers became more common in homes they went with what was familiar (or the only thing available), which was Windows. Developers started making Windows game, MS made APIs to help with that (DirectX anyone?), etc. With Linux getting more popular, not just for servers but for many different levels including the desktop, it's only natural that people would want some nice games to play when they don't "have" to be working.
Time to build a new fab, from the ground up, is around 4 years (and 4billion dollors). Last time I looked, AMD wasn't expanding very quickly or had the assets to build many fabs in parallel. Intel's current generation sucks, even Intel people will admit that. But the current direction is much more efficent and has great performance, so I definatly think we'll see Intel regain market share.
I agree, all these "the end of Intel" comments are so off base it's comical. First, Intel is coming out with new chips that are already kicking the butt of AMD's best chips. Second, Intel still has over 50% of the market, no matter how you look at it. And third, AMD does not have the production capacity to provide chips for 100% of the market share even if they had it. That means that even if the market demanded more AMD, they would have to either turn them away or put them on a very very long waiting list (think year+). Why wait when you can get Intel right now? Intel isn't going anywhere, but AMD has always worked on very thin profit margins. If they can't respond with a very good line of new chips in the next 6 months, I can easily see their stock falling and profits going south very fast.
Speak up! I can't hear you too well these days. My bones creak so loud I can't hear you!
If you look at Intel's Core 2 Duo, the cache space is not "divided" as the number of cores increase. Each core, if running at full load, will have 2MB of cache (extreme edition anyway). That is a very respectivable cache size and would be a respectable single core processor. When one core is not running (like when running only Word), one core sleeps while the other core is given all of the cache.
Past marking ploys (GHz) were definately wrong, and trying to directly replace those metrics with the number of cores is also a bad choice. But don't you see that that is exactly what Intel is trying to prevent? The interviewee in the article is saying that more cores != more performance. Hence why desktop users will have no need for 8 cores or more. Most of the posts on this topic are along the lines of "ya right, more cores FTW!", which is a very uninformed mentality.
Then you are not a desktop user :) You are doing work that is more traditionaly done on a workstation.
I don't know if you can qualify that is "good" or "bad", on either AMD's or Intel's part. Internal testing numbers show that using the FSB for communications with the RAM does not hurt performance. So you could postulate that AMD's solution is a bit of overkill for current needs. Of course, in 3-5 years we may all be using a system to what AMD is using right now due to an increased need for memory bandwidth, or we could still be where we are right now (I doubt it). But for right now, with current systems, neither is absolutely better than the other. Both meet the needs of the system and are not a bottleneck to performance.
I'm not too familiar with the specifics of the AMD implementation, but from what I have picked up they have a dedicated connection from the memory to the CPU instead of sharing a bus with other data (I/O, DMA traffic, etc). That's a nice setup and I don't think there is anything wrong with that. I have no idea if the test was with 2, 4, or 8 cores, but it's pretty safe to assume is was done with 2 cores. Therefore, when Core 2 Duo comes out the connection to the main memory should not be a performance bottleneck. The available bandwidth is plenty and even has some headroom to grow.
The only real programs that will take advantage of large numbers of cores will be scientific applications. Games? You offload the AI to one core, the sound to one (or the nice sound card), network handling to another, I guess user input to one... The most computationaly intensive tasks are the AI work and the graphs, which is already done in the GPU. So unless you are going to run ultra realistic, human-like AI work, there really isn't much parallel work that can be done from games.
Multimedia can get a nice boost from those cores, but then you run into other issues such as hard drive speed and memory space (8 cores will quickly fill 2GB of memory with data if given the opportunity). Even then, does the increase in complexity both in the motherboard and huge power consumption justify shaving off a few minutes from transcoding some movies?
Workstation level work will definately use 8+ cores, there is no question. The issue is will your average or even hardcore desktop user see any benifit or have the benifit worth the cost? I think the person in the interview is right for the forseable future. Maybe in a decade or so when the CompSci people figure out new ways to do parallel work, 8 cors will make sense at home.
So which bottleneck are you refering to? The new Core 2 Duo chips of Intel's share the L2 cache and, as far as I can tell from the reviews I have read, this setup works very well. Both chips can share data very quickly or when executing a single sequential program one of the cores can use all of the L2 cache (which in the Extreme Edition verion is up to 4MB!). Or are you refering to the main memory? It is possible for both cores to need to access the main memory at the same time, but modern pre-fetching and aggress speculation techniques reduce how often that occurs and the timing penalties when they do occur. And of course, the larger the L2 cache the more memory can be stored on the chip at once, reducing the need to access the main memory very often. According to Intel's own internal testing, they had a very hard time using all of the bandwidth the current front side bus and memory offers, which means the main memory shouldn't be a bottleneck.
So what is the bottleneck you are refering to?
If they pass the savings on to me, sure. Food ain't free after all.
At this point there is enough evidence to at least start an impeachment, and this time have to go all the way. Bush has committed actual crimes here. There might even be a few in there that can be considered treasonous! But will Congress even whisper the word 'impeach'? Nope, because the Republicans have the majority. Even if, as you say, there is a big fracture in the Republican party right now, they won't lift a finger to actualy do anything to stop Bush or punish him.
This is Bush's last term, he knows this, and he knows he is pretty much untouchable, so he does as he pleases. This was actually my greatest fear in giving him another term. His behavior during his first term was horrible and put the safty of the American people in more danger than any terrorist with bombs strapped to their bodies.
And yet, the Republican party cheered him! They called him our savior and a good leader and a good person. And they came out of the woodworks to vote for him again. And now you are saying that a big chunk of them regrette their decision? You'll excuse me if I don't find this reasuring or think very much of Republicans right now. They helped get a criminal elected to office. If they really hate what he is doing, they have the power to remove him, legally. But I double we will ever see a Republican Congress convict a Republican President.
The Republicans on the other hand are much more guided and on the same page with each other. You have two main groups, the social conservatives (right-wing religious types) and the fiscal conservatives. It's much easier to get the Republican base out in force to vote the same issues up and down than it is to get the Dems out to do the same.
I doubt we will see a Democratic majority for a while. The Republican may have made some major mistakes (and many illegal ones at that), but the general public still sees them as more able to lead than the Democrats. We will just have to wait for them to commit such a horrible crime that even the most staunch conservative can no longer stand to see them in power. We are almost there, but not yet.
Sorry, Carly already beat you to that.
That is definatly true, but places like those are the minority. The majority of newly trained programers and engineers will be coming from the State colleges and Universities. Just because your university doesn't have the same name recognition as MIT does not mean that you are incapable of meeting the performance requirements of a given job. Heck, I would say that MIT has made sure they the people they graduate are perceieved as being overly qualified. But is the person that graduated bottom of their class at MIT really better than the person who graduted within the top 10 at a State University? That's a question that no one can really answer without interviewing both students, which is a long and expensive process.
About the time AMD does, Intel will be moving towards their 45nm or 40nm or whatever their next smallest size is. Intel has the best fabs for a reason: they invested a LOT of money into fab R&D just after the bubble burst. They called is their One Generation Ahead strategy. While everyone else was trying to cope with the loss of capital and drop in stock prices, Intel want to make sure that they came out one generation of silicon manufacturing ahead and stayed that way. While they have in some respects (physical size of transisters) they have missed some advances in some other areas (SOI). In the end, Intel will probably stay ahead of AMD until they hit the physical wall on the size of transistors (can't go smaller than an atom!).
"Did you miss the part where a local Linux kernel exploit was used in the attack?" True, it was part of the exploit, but first the attacker(s) had to get onto the system using a developer's account. Now I don't pretend to know what kind of rights the developer accounts have, but they would definately have more power than a "guest" account or the account used by apache if that had been hacked. This means that if the account hadn't been hacked, then the server wouldn't have been hacked (exploit requires user access to the server). The kernel exploit is bad, but at least they are upfront about it and don't try to hide it behind PR/marketing speak or not tell anyone about it.
After all, who is it that has problems with movies rated R?
- Children
- Mentally challenged people (see children)
- Bible-thumping ultra-religious types
As far as the children are concerned, the parents should be parents and screen what their young children watch (you know, like a parent). When you are talking about the different between PG13 and R, is it really that bad that your 16 year old saw a boobie for 0.75seconds? Is it going to warp him? If it does, see the mentally challenged people. As for the ultra-religious types, something tells me that they won't be interested in the type of movies that border on PG13-R. They would rather watch a movie that is condoned by their church instead of looking at evil, sinful stuff like the rest of us.When it comes down to it, if you are offended by contents that 99% of adults don't even smirk at, you should be researching your film choices very very very very carefully. If you see something bad you just might go insane. As for the children, parents should be parents and take control of the information that their children consume. If you can't, you shouldn't be a parent.
If you look at TV, sooner or later an 'edited' version will be played on TV that has been 'sanatized' for those of weaker constitutions. The copyright holders can choose to provide an edited version, but they should never be forced into providing one.
Going to the government isn't wanting a single person to fix the problems. Remember, our government represents the PEOPLE and is made up of representatives that the PEOPLE elect. We have placed power in both the Federal government, which deals with national issues, and in the State government, which focus on issues in the state. In issues like this, the Federal government, which is made of the PEOPLE, has the power to standup for the PEOPLE. It's not collectivist government, it's representative government.
In this case, the less regulation the more likely the telcos are to screw us over and make our lives worse, all to make their bottem line get just a little bit bigger.
That only works when the market is playing fair. The GP used gas which is a good example of an unfair commodity. Just because businesses don't like the high price of gas doesn't mean they can just switch to another brand of gas. They all pretty much cost the same since the supply is controlled by a limited number of companies. The Internet is the same, with the majority of the network making up the backbone being owned by only a handful of companies. These companies are also the telcos! So even if the market doesn't want the BS pricing scheme from the telcos we don't have a choice. Not unless everyone wants to get together and build a new Internet. In the end, all the arguments about the market and supply and damand are just crap. Those ideas only work when the market is fair and there is competition. The gas supply and the Internet backbone are not fair.
To being a kid again! /raises rootbeer float