Just wanted to let all the AMD fanbois out there who always trash Intel whenever they show the slightest move in the right direction (for their company and consumers) that if Intel disappeared tomorrow, the world of computing as you know it would crash to the ground like a bowling ball.
AMD may have had the upper hand on processor designs for the last few years, and even Intel is practically admitting that by switching their mainline core architecture to the Pentium M derivative (which is a PIII derivative). But Intel has had them beat in one area and probably will for a very long time: production capacity. Intel's fabs can crank out processors in a day that matches AMD's production in a year (exaggeration, but run with it). There is NO way that AMD could meet the demands of the business world in sheer scale of orders. If Intel closed up shop, computer CPUs would triple or quadrupal in price overnight due to scarcity. Fabs take YEARS to build, and can take even more years to full ramp up to full production while shaking out the process bugs. Intel has a LOT of experience with this due to the shear number of fabs that they own and operate.
Competition is good, for everyone. Give props where they are due though. Intel is turning around after making a very bad roadmap choice many years ago, and I think it will only benifit consumers in the long run. AMD had very good designs and will probably have some more good ones in the future. But don't pick on or the other exclusively and wish for the other to disappear. That would lead to a VERY bad situation for everyone, even you.
That's only if use assume that the current processor design used by Intel is used on the new 45nm. Intel has learned from it's mistake with NetBurst (P4) design descisions and are finally heading back in the right direction. So, by the time the 45nm comes out, there will also be a new architecture to place on the new chips. We'll see how things are then.
Oh I did! I knew VHS was going the way of the doodoo as soon as DVDs started hitting the shelves. How could anything like VHS survive DVDs? Smaller form factor, better quality (both in sound and video) (that doesn't degrade each time you play it), more extra features, etc, etc, etc. Everything they are holding up to "replace" DVDs are nothing more than increased storage/better video quality, but that is only benificial to people who do have HD TVs (which isn't many). Oh, and different/better/more draconian DRM features, which we all just LOVE! Nope, DVDs will be around longer than the industry wants them to be, but just as long as consumers want them.
Which sounds like AI to me:) There are seveal goals in AI, only one of which is to think like a human. Other goals include acting like a human (so it doesn't matter how the internals work, just how it acts), think logically (which would be related this kind of work and other intelligent data miners) and acting logically (very different than acting like a human:))
How was I MS bashing? All I said was this isn't some new, revolutionary work done exclusively at MS, but a combination of changes from RFCs, published papers, and probably some internal research. That isn't bashing, that's just bringing the fanboys down from their egotisitical high.
Yup, the original TCP stack sucked. Newer implementation do away with a LOT of memory copying for one, which alone can speed up the stack performance several times. Throw in a few more new tweaks and you can get even faster. All MS has done is take all those changes and used them.
It really varies from store to store, especially if you have already used your system. Some stores may not be willing to take returns for hardware after 7 days, which from what I hear about this bug is easy to pass through without finding the source of the problem or knowing you need to get a replacement.
Returning a broken product should always be the first action a consumer takes. But in this case, the XBox 360 was first produced in a limited quantity and then released in an even smaller amount to retailers. That means that MS artifically made them scarse in order to drive up apparent demand and value of the produce.
Now imagine waiting in line for a produce that you shouldn't have to wait in line for (think of McDonalds only making 10 Big Macs at each location every day, first come first served) and then think about that product not working. Not only can you not return it to the store for replacement (they are all sold out and will be for the next month), but the manufacturer cannot even replace the devices (don't know this for sure, but I wouldn't be suprised if MS doesn't have enough units in stock to replace all the faulty ones). So now you are stuck with a broken product and no way of fixing the situation because the manufacturer intentionally screwed you over (limit roll out, artifical inflation of value, etc). You would be pretty damn mad, just like everyone else in your position.
I haven't RTFA, but this guy is probably suing out of anger (I would be angry, I can tell you that), and probably from the promptings of a lawyer who can see a quick cash settlement or a possible class action lawsuit with an even bigger payout for him/his firm.
These are a few people who are true "alpha" males and are very successful in their respective fields. Of course, they aren't always the most loved people, but they do a hell of a job at leading.
Well, they could always hold off on their new feature patches for two weeks, wait for the stable version to be finished (it is only two weeks afterall) and then start submitting new feature patches. Duh.
There is a BIG difference between bug fixes and merges, and adding new features. I haven't RTFA, but based on the summary it seems like Linus wants to keep those two weeks dedicated to fixing bugs, merging changes, etc, not adding in new features that were not even coded until that two week window. That is what all the weeks leading up to the merge window is for. Once those two weeks hit, the focus should be 100% on making sure everything works, works together, and is stable. I say, good for Linus! Sometimes the only way to make people listen is to be tough with them.
This is a interesting point. There are definatly some people in the FOSS crowd that do believe that all software patents are evil and all source code should be free. But I'm not one of those. I think that something like PageRank is definately worthy of a patent. It's very none-obvious, very specifically defined and implemented. This is what patents where created to protect. The real problem is the people who patent ideas like "using a computer to buy goods online for delivery", which are so obvious as to be laughable to try and patent. These kind of patents hurt the industry and consumers, with only the lawyers coming out ahead. But PageRank is a very smart piece of coding and a very good concept, not the typical general/obvious software patents that you hear about on/. all the time.
LOL, my comment ment to be exactly as it was rated: funny. It was a small jab at the mentality here on/., not what I believe is an iron clad rule in the real world.
Supporting Open Source is not throwing away your money or doing something purely for philosophical reasons. Look at it this way: Google uses Linux/OSS heavily, so promoting future developers to improve current OSS projects and create new ones could potentially lead to yet more OSS that Google can use in their business. Google is probably banking on the idea that this investment will be much lower than whatever future OSS that is developed from this program ends up saving.
So you can support OSS with an eye on the bottome line. Maybe not the bottom line this quarter or next quarter, but sometime in the future. Look at IBM. They won't make a dime directly off of Linux, but they can use Linux to help them sell their servers. So it is in their interest to keep Linux improving and moving forward in performance. With IBM it's easy to see why they would want to support Linux. With Google it's a bit harder because they benifit from Linux in their daily operations of their company and services, which isn't directly seen by regular users.
You know he would use the treatment, but he would find some way to twist his reluctance of allowing stem cell research into being a champion of the field and how it would have died without him blah blah blah.
(Unfortunately, this is how all politicians act, doesn't matter the party affiliation)
It's pretty much just conjecture until Vista goes gold, but from what I have read in MaximumPC on their Vista coverage, the GDI is indead being removed. If a legacy app calls the GDI, it will pass through a special wrapper that will behave like GDI in some respects, but it will ultimatly call the new 3D vector graphics calls. Yes, there would be a slight performance hit, but not much of one.
Check out the latest issue of MaximumPC. They had a very quick overview of Vista in it, and one of the issues it covered was the Vector graphics and the end of the GDI. It's on page 28, under the section on "Longhorn Display Driver Model" and is a very interesting read.
Windows 95(mostly) and NT were written from scratch. To a point so was 2000 (based off of NT). But XP was not written from scratch, it was an upgrade from 2000. Vista is not based on XP, it was a total rewrite. Why else would it be taking over 4 years of development work? If it was just a new UI and a few changes under the hood, it should have been out years ago. The only reason it could be taking so long is either a) they are idiots and can't program or b) they did a drastic rewrite of the whole OS from ground up.
Except that they are removing the GDI functionality. All GDI calls will just act as special wrapers back to the vector display calls. MS is not making this OS to be a simple upgrade from XP. They started from scratch and they are compartmentalizing or outright removing a lot of legacy stuff (which is good, it leads to better design overall). The GDI is one such module that has been removed.
It takes months to fab a wafer of CPUs, so even if the design 'magically' appeared (all architecture design, HDL, schematics, AND layout data), it would still physically take time to build the layer masks, start the new design in the fab, and wait for it to finish. And THEN you would have to do silicon debug, because not every new chip design boots the first time. Make any changes needed, repeat fabbing process. That is why you can't turn around over night or even in a few weeks.
My bad, CPI is a better measurement for single pipelined architectures that doesn't use out of order execution (which our modern CPUs do). The IPC measurement is more applicable here. Playing with the architecture to increase the IPC, even if it sacrifices some clock speed, can pay off big time if the increase in IPC is greater than going with the older design at faster clock speed.
There is no 'right' design that everyone is shooting for. There are many, many, many variables in a modern CPU design that have to all be juggled. Like how big is the re-ordering buffer? If you increase it, you can have more instructions 'in-flight' at once. But if you increase that, you might have to increase the number of reservation stations on the functional units, or even increase the number of functional units to support more in-flight instrunction. And once you start talking about actually building the architecture in silicon, some things are practical while others aren't.
Intel engineers came out years ago and stated that they will be hitting the physical wall by 2010, if not sooner. And this isn't the 'we don't know how to get light any smaller' wall, it's the 'the gate is an atom thick' wall. Once you get that small, that's it, you can't get smaller using atoms. You'd have to goto subatomic particles to get smaller, which is a completely different ballgame.
And if anything, the battle between AMD and Intel should have taught everyone here on Slashdot that faster speed does not mean faster performance. There are MANY factors in architecture design that will improve or decrease overall performance. Sure, you can have a 4GHz CPU, but if it's cycles per instruction (CPI) is 100 while a 2GHz CPU has a CPI of 20, the 2GHz CPU will actually be FASTER than the 4GHz chip! Intel knows this, AMD knows this, and everyone who does serious computer design work knows this. Intel chose the wrong path with Netburst and they have known it for years. But you can't turn around one day, snap your fingers, and switch to another architecture company wide. It takes time, hard work, and a lot of people, which is why we are only seeing this change now and not back in 2002 like they would have wanted.
I'm happy with this change and I think playing with the architecture to get better CPI and instructions per cycle (IPC) is a better way to go than just cranking up the clock speed.
I'm currently in my seventh(7) internship at Intel. I'm lucky and got in on a very special program that takes local high school grads and gives them internships in design/manufacturing positions during the summers while in college. You have to be in a short list of majors (all technical of course) and maintain a high GPA.
Overall, they were great! Each year I was somewhere else, but mainly in design since that is where my interest lies. I got to work on Prescott, Cedermill a bit, etc. Great exposer to what it's like being an engineer.
My project last summer though was the best. My supervisor didn't treat me as just a lowly intern or throw a project that he was just kind of interested in. He treated me like a coworker and had me working on his person pet project (which succeeded beyond even his wildest dreams!) and it was very challenging and rewarding.
Intel treats its interns very well and rewards hard work. As they say around here, you get out what you put in. If you sit around not doing much and never leaving your cube (hehe, like me this summer), you don't end up doing much. But if you get out and talk to engineers and ask for work and take the initiative you get interesting work and a lot of respect.
Slashdot Mirror
AMD may have had the upper hand on processor designs for the last few years, and even Intel is practically admitting that by switching their mainline core architecture to the Pentium M derivative (which is a PIII derivative). But Intel has had them beat in one area and probably will for a very long time: production capacity. Intel's fabs can crank out processors in a day that matches AMD's production in a year (exaggeration, but run with it). There is NO way that AMD could meet the demands of the business world in sheer scale of orders. If Intel closed up shop, computer CPUs would triple or quadrupal in price overnight due to scarcity. Fabs take YEARS to build, and can take even more years to full ramp up to full production while shaking out the process bugs. Intel has a LOT of experience with this due to the shear number of fabs that they own and operate.
Competition is good, for everyone. Give props where they are due though. Intel is turning around after making a very bad roadmap choice many years ago, and I think it will only benifit consumers in the long run. AMD had very good designs and will probably have some more good ones in the future. But don't pick on or the other exclusively and wish for the other to disappear. That would lead to a VERY bad situation for everyone, even you.
That's only if use assume that the current processor design used by Intel is used on the new 45nm. Intel has learned from it's mistake with NetBurst (P4) design descisions and are finally heading back in the right direction. So, by the time the 45nm comes out, there will also be a new architecture to place on the new chips. We'll see how things are then.
Oh I did! I knew VHS was going the way of the doodoo as soon as DVDs started hitting the shelves. How could anything like VHS survive DVDs? Smaller form factor, better quality (both in sound and video) (that doesn't degrade each time you play it), more extra features, etc, etc, etc. Everything they are holding up to "replace" DVDs are nothing more than increased storage/better video quality, but that is only benificial to people who do have HD TVs (which isn't many). Oh, and different/better/more draconian DRM features, which we all just LOVE! Nope, DVDs will be around longer than the industry wants them to be, but just as long as consumers want them.
Which sounds like AI to me :) There are seveal goals in AI, only one of which is to think like a human. Other goals include acting like a human (so it doesn't matter how the internals work, just how it acts), think logically (which would be related this kind of work and other intelligent data miners) and acting logically (very different than acting like a human :))
How was I MS bashing? All I said was this isn't some new, revolutionary work done exclusively at MS, but a combination of changes from RFCs, published papers, and probably some internal research. That isn't bashing, that's just bringing the fanboys down from their egotisitical high.
Yup, the original TCP stack sucked. Newer implementation do away with a LOT of memory copying for one, which alone can speed up the stack performance several times. Throw in a few more new tweaks and you can get even faster. All MS has done is take all those changes and used them.
It really varies from store to store, especially if you have already used your system. Some stores may not be willing to take returns for hardware after 7 days, which from what I hear about this bug is easy to pass through without finding the source of the problem or knowing you need to get a replacement.
Now imagine waiting in line for a produce that you shouldn't have to wait in line for (think of McDonalds only making 10 Big Macs at each location every day, first come first served) and then think about that product not working. Not only can you not return it to the store for replacement (they are all sold out and will be for the next month), but the manufacturer cannot even replace the devices (don't know this for sure, but I wouldn't be suprised if MS doesn't have enough units in stock to replace all the faulty ones). So now you are stuck with a broken product and no way of fixing the situation because the manufacturer intentionally screwed you over (limit roll out, artifical inflation of value, etc). You would be pretty damn mad, just like everyone else in your position.
I haven't RTFA, but this guy is probably suing out of anger (I would be angry, I can tell you that), and probably from the promptings of a lawyer who can see a quick cash settlement or a possible class action lawsuit with an even bigger payout for him/his firm.
I can!
George W. Bush
Donald Trump
Bill Gates
Steve Jobs
These are a few people who are true "alpha" males and are very successful in their respective fields. Of course, they aren't always the most loved people, but they do a hell of a job at leading.
Well, they could always hold off on their new feature patches for two weeks, wait for the stable version to be finished (it is only two weeks afterall) and then start submitting new feature patches. Duh.
There is a BIG difference between bug fixes and merges, and adding new features. I haven't RTFA, but based on the summary it seems like Linus wants to keep those two weeks dedicated to fixing bugs, merging changes, etc, not adding in new features that were not even coded until that two week window. That is what all the weeks leading up to the merge window is for. Once those two weeks hit, the focus should be 100% on making sure everything works, works together, and is stable. I say, good for Linus! Sometimes the only way to make people listen is to be tough with them.
This is a interesting point. There are definatly some people in the FOSS crowd that do believe that all software patents are evil and all source code should be free. But I'm not one of those. I think that something like PageRank is definately worthy of a patent. It's very none-obvious, very specifically defined and implemented. This is what patents where created to protect. The real problem is the people who patent ideas like "using a computer to buy goods online for delivery", which are so obvious as to be laughable to try and patent. These kind of patents hurt the industry and consumers, with only the lawyers coming out ahead. But PageRank is a very smart piece of coding and a very good concept, not the typical general/obvious software patents that you hear about on /. all the time.
LOL, my comment ment to be exactly as it was rated: funny. It was a small jab at the mentality here on /., not what I believe is an iron clad rule in the real world.
Supporting Open Source is not throwing away your money or doing something purely for philosophical reasons. Look at it this way: Google uses Linux/OSS heavily, so promoting future developers to improve current OSS projects and create new ones could potentially lead to yet more OSS that Google can use in their business. Google is probably banking on the idea that this investment will be much lower than whatever future OSS that is developed from this program ends up saving.
So you can support OSS with an eye on the bottome line. Maybe not the bottom line this quarter or next quarter, but sometime in the future. Look at IBM. They won't make a dime directly off of Linux, but they can use Linux to help them sell their servers. So it is in their interest to keep Linux improving and moving forward in performance. With IBM it's easy to see why they would want to support Linux. With Google it's a bit harder because they benifit from Linux in their daily operations of their company and services, which isn't directly seen by regular users.
You must be new here
Open Source = Good
Closed Source = The Devil
You know he would use the treatment, but he would find some way to twist his reluctance of allowing stem cell research into being a champion of the field and how it would have died without him blah blah blah.
(Unfortunately, this is how all politicians act, doesn't matter the party affiliation)
It's pretty much just conjecture until Vista goes gold, but from what I have read in MaximumPC on their Vista coverage, the GDI is indead being removed. If a legacy app calls the GDI, it will pass through a special wrapper that will behave like GDI in some respects, but it will ultimatly call the new 3D vector graphics calls. Yes, there would be a slight performance hit, but not much of one.
Windows 95(mostly) and NT were written from scratch. To a point so was 2000 (based off of NT). But XP was not written from scratch, it was an upgrade from 2000. Vista is not based on XP, it was a total rewrite. Why else would it be taking over 4 years of development work? If it was just a new UI and a few changes under the hood, it should have been out years ago. The only reason it could be taking so long is either a) they are idiots and can't program or b) they did a drastic rewrite of the whole OS from ground up.
Except that they are removing the GDI functionality. All GDI calls will just act as special wrapers back to the vector display calls. MS is not making this OS to be a simple upgrade from XP. They started from scratch and they are compartmentalizing or outright removing a lot of legacy stuff (which is good, it leads to better design overall). The GDI is one such module that has been removed.
It takes months to fab a wafer of CPUs, so even if the design 'magically' appeared (all architecture design, HDL, schematics, AND layout data), it would still physically take time to build the layer masks, start the new design in the fab, and wait for it to finish. And THEN you would have to do silicon debug, because not every new chip design boots the first time. Make any changes needed, repeat fabbing process. That is why you can't turn around over night or even in a few weeks.
There is no 'right' design that everyone is shooting for. There are many, many, many variables in a modern CPU design that have to all be juggled. Like how big is the re-ordering buffer? If you increase it, you can have more instructions 'in-flight' at once. But if you increase that, you might have to increase the number of reservation stations on the functional units, or even increase the number of functional units to support more in-flight instrunction. And once you start talking about actually building the architecture in silicon, some things are practical while others aren't.
And if anything, the battle between AMD and Intel should have taught everyone here on Slashdot that faster speed does not mean faster performance. There are MANY factors in architecture design that will improve or decrease overall performance. Sure, you can have a 4GHz CPU, but if it's cycles per instruction (CPI) is 100 while a 2GHz CPU has a CPI of 20, the 2GHz CPU will actually be FASTER than the 4GHz chip! Intel knows this, AMD knows this, and everyone who does serious computer design work knows this. Intel chose the wrong path with Netburst and they have known it for years. But you can't turn around one day, snap your fingers, and switch to another architecture company wide. It takes time, hard work, and a lot of people, which is why we are only seeing this change now and not back in 2002 like they would have wanted.
I'm happy with this change and I think playing with the architecture to get better CPI and instructions per cycle (IPC) is a better way to go than just cranking up the clock speed.
I'll be right back, I have to go fill up my laptop's battery....
You think that if you want. I'm just relating my experience interning at a large technology company.
Overall, they were great! Each year I was somewhere else, but mainly in design since that is where my interest lies. I got to work on Prescott, Cedermill a bit, etc. Great exposer to what it's like being an engineer.
My project last summer though was the best. My supervisor didn't treat me as just a lowly intern or throw a project that he was just kind of interested in. He treated me like a coworker and had me working on his person pet project (which succeeded beyond even his wildest dreams!) and it was very challenging and rewarding.
Intel treats its interns very well and rewards hard work. As they say around here, you get out what you put in. If you sit around not doing much and never leaving your cube (hehe, like me this summer), you don't end up doing much. But if you get out and talk to engineers and ask for work and take the initiative you get interesting work and a lot of respect.