Disney == Bad. I'm hoping AOL tries to buy them out soon like some rumors indicate.
However, that's not the real reason why I won't be buying that movie. Am I the only one here that thinks Tron sucked? I saw it again a few years ago and no, the geek factor is not 8, it's off the charts. In fact, it's so high it ruined the whole movie. I couldn't stand to finish the movie.
In typical Disney fashion, it wasn't a movie written by geeks for geeks, but rather by geeks for geeks' Moms and children who need a personified analogy for all the parts of the computer in order to better understand computers. I could go on about how it could easily be the opening scene in the "Computers for Dummies" puppet show, but this is beginning to bore me...
That point may be true for ESR. I think ESR is often an arraogant jackass who will likely leave the cause when it no longer suits his pocketbook anymore. He is definitely quite pleased with how his goals line up with the fanatical open source movement, and he never seems to truly go out on a limb to advance open source. His motives clearly are to promote himself, and when he starts fading from peoples' memories, he never fails to embarrass himself yet again. Granted, he has apparently developed some very key pieces of software and has generously given them back to the community. However, he leaves no question that he is in it for the money and recognition.
RMS, on the other hand knows that his movement will help people (ultimately) by providing them freedoms he is willing to fight for at the expense of his reputation and pocketbook. He is a true libertarian (check out his home page at stallman.org if you have any doubts whatsoever). GNU is one of his many strong passions and is possibly the only one with with which I completely relate and sympathize -- I will add that he is exceptionally wise to maintain a strong distance between his GNU agenda and his other personal agendas so as to not dissociate himself from his GNU supporters. His goals are to free software ("free" as in the verb to free from restraints and regulations). Contrary to what many people believe, the goal is NOT to provide free software ("free" as in the adjective for cost).
The fact that you and many others misunderstand his agenda is precisely why he continues to push it with all his energy. I have an enormous amount of respect for RMS and what he has selflessly created. He is often criticized for being a relentless control-freak. Although this seems a bit exaggerated, I don't doubt that his agenda would fail were he not of this nature.
You are completely wrong and a very poor judge of the Gnu and Linux communities if you think Linux will die when Microsoft becomes easier to use or when Linux becomes unnecessary to support. GNU is all about freedom -- the freedom to program, modify programs to suit you, modify programs for others, and redistribute modified programs. The philosophy has nothing to do with ease of use, and quite frankly cares little whether common man ever uses the software except that common man has power over the government which has the power to advance the cause or seriously impede or abolish the struggle for free software.
BUT -- to earn popular support from the masses, Gnu/Linux products must be free in cost and must be easy to use. The end product, not the philosophy, has to appeal to the masses. Note this has nothing to do with the philosophy of programming freedom. This is merely a means to an end. I agree there is indeed a shame in prematurely trying to sell a product to people not ready for it. Unfortunately, some GNU followers misinterpret the goals of gaining acceptance with trying to obtain marketshare. It is still too early to obtain marketshare, but it is not too early to obtain popular support for the ideals. Note that RMS does not encourage forcing the product on those not yet ready -- I challenge you to provide a counterexample. ESR on the other hand seems to be an unfortunate troll, unknowingly bent on destroying the open source movement.
Your marxist comments are completely irrelevant to the discussion. Just because one can equate RMS' fervor with that of a priest doesn't mean the GNU philosophy is meant to delude or control as does many religions of the spirit. GNU is a governmental policy, grounded in the physical world. There are few, if any, parallels between GNU and spirituality.
I know this could put a slant on the postings that are approved, but assuming/. actually generates traffic for other sites such as wired and space.com and pbs.org and nasa.gov, etc, this should, in turn, increase the add revenue of those sites. Why can't/. get a cut of that (probably laughably small) sum of money somehow?
I know this would mean that/. would post more links to stories on the sites that comply, but at the same time, it would be drivin not by number of links posted, but by actual click-thrus. This means that the dumb stories don't actually earn revenue for/.
So what this effectively reduces/. to is a discussion wrapper for select stories on other sites. This would kind of out-source the discussion layer from those sites.
One flaw is that the other sites might want revenue from the ad viewership attributed to the discussion about that story, since without that other site's story, there would be no traffic at/.
First off, support costs spread across all customers. I know most computer geeks would like to think everyone else calls support all the time. I don't know any companies that provide such good support that it can add more than 50% to the total cost of selling OEM windoze OSs (assuming more recent M$ OSs which is precisely what the discussion is about). Remember that support calls are not limited to OS-only issues, but also drivers and hardware troubleshooting, therefore the costs are not all attributed to the OS. You cannot make the argument that drivers are part of the OS any more than you can say the hardware is part of the OS.
Secondly, you only further drive my point that the OS is absolutely not the cheapest part of the system. In particular, I address your quote of my previous post. You only make that quote more correct.
I think what you are trying to say is that the OS is more of the cost of a system than seems obvious. While that may be true, it is becoming much less the case in the XP and 2K versions of windoze. My neighbor is almost 70. I installed Xp on his machine and he has not once called me for help in the past 2.5 months since then.
M$ is removing the support burden from the OEM reseller. Ideally, the OEM version of the software should be getting more expensive due to this factor, but it doesn't seem to be the case since users also want to upgrade to the retail product to take advantage of the "less hassle" factor.
What I think is that you don't really have a point and you like to argue. You started off beefing that the open-source movement is not consumer-oriented and finished up by rambling micromanaging the irrelevant portion of the discussion.
The whole issue of RMS and ESR pushing this open-source movement almost at the level of a religion is where it started. You want this movement to be consumer-oriented, but we are still in the evangelism stage and not ready for consumers. We will be in the evangelism stage for a long time, hence the "religious" nature of their communications. They continue such an extreme doctrine because their philosophy is still spreading, and it is spreading at a very rapid rate. Since this is a critical time, they don't want their views to be diluted while they spread.
No, what the $0 means is that if you want to remove the OS to save money, you will save a whopping $0, so it's not worth your effort.
I'm sure Dell has to pay a significant amount of money for their licenses.
The $35 OS is not by any means the cheapest part of the system. The keyboard and mouse combine to cost approximately $10. The integrated audio costs less than $10. The integrated video costs around $10. The north/southbridge chipsets cost about $30. The fans are less than $5 each. The 3.5" floppy is less than $10. The CDROM drive is less than $35. The case is probably $30. The power supply is less than $20. Non-M$ bundled software is nearly free.
The ONLY things that likely exceed the cost of M$ OS are the monitor, the CPU, and maybe the DVDROM or CDRW drive.
At profit margins that are less than $50 for the low-end PCs, $35 is a large piece of the pie.
For palmtops, the margin is very similar. The battery and the high-end embedded processors and memory are still pretty expensive. In the future, manufacturers will likely take a loss on the initial sale of their hardware and try to recoup their costs on the sale of services. Here, M$ is killing them. But, M$ is trying to remove the service source of revenue as well, so hardware manufacturers will be even worse off.
You are comparing two completely different business models.
The first is the essence of capitalism, where consumers are purchasing their commodities from manufacturers. The consumers typically have the clear agenda to buy a high-performance platform for a cheap price, with a high degree of interoperability and a wide variety of software. The PC offers all this for a fantastic price because of the enormous amount of competition out there.
ESR and RMS are not trying to sell free software to you, obviously, because you can't sell free software. They are not participating in the capitalistic marketplace. They are not at the point where they give a damn about you, especially considering you are a gimme, gimme, gimme type of guy. They don't even have a suite of products to push. They many very well never have a product line to push, but that's ok.
What they do have is a platform, an OS, a compiler, and a huge slew of products that are in a state of constant development. Despite what some may think, these products are not appropriate for you or any other typical user.
What RMS (and to a significantly lesser extent, ESR) are trying to sell is the concept of free software. They are trying to sell their ideology to what I consider two significant bodies of people. One is the companies or governments that have the potential or revenue to fund free software. The other is the large body of software developers who will individually create ideas or write source code to contribute to the base that already exists.
This is an extremely tough sell, since companies want to see an immediate source of revenue as a result of their effort. The sell involves convincing the large entities that free software is better for them in the long run either financially or simply for the freedom from software patents and undue restrictions. Of course individuals may be an easier sell, but even if you convince someone that free software is good, that is good but you still need individual contributors to advance free software.
This is an ongoing effort for both of RMS and ESR. They don't care about you (yet). You are a consumer, and they are not after consumers yet. They need stakeholders and contributors who are ready to make a commitment to their cause.
Don't be so quick to shoot down their cause. The cause has not advanced to the point that you, the consumer, are in their target. If everyone were to dismiss their cause, they might never be able to deliver to you the tools/software/OS/platform you need to see.
I think you have to use your imagination here to figure out how the DMCA could cover WC3.
I'm probably a bigger Bliz fan than most here, and I'm sure that's saying a lot. But I'm also a proponent of individual rights and identify the need for fewer, not more, copyright protections.
With that out of the way, it would be easy to interpret the "multiplayer version" of WC3 as your copy of the game plus Battle.net. If you consider them together as a system, then you are essentially replacing a part of the system with your own part of the system and thereby circumventing the copy protection.
If you assume that multiplayer WC3 does NOT work without Battle.net, then you can assume that the multiplayer version of the game was not intended to work without copy protection, since that was built into Battle.net. Replacing Battle.net removes copy protection.
I hate it, and I too am not a lawyer, but I think this argument can be sold to any judge.
What I could discern from your message is that you are confusing Hammer with a desktop PC processor. You probably didn't understand my original post, so I'll elaborate.
In order for AMD to compete to any extent with IA64, AMD is targetting Hammer to the workstation and server market. Very few people are going to "upgrade for the speed benefits". Even though it's AMD, the price will be way too high for the home PC market. AMD is looking to increase its margins by entering the workstation and server market with Hammer, so it would defeat their purpose to target the home PC market with this platform.
With that in mind, I would offer that M$ is desparately trying to displace Unix in heavy computing environments and even moreso displace Linux and Apache in the server space. To do this, they need to attack Linux on all fronts. And if you recognize that AMD will consume more than a negligible portion of the marketplace (which is definitely arguable at this point since nobody can predict what will turn out) and you assume that M$ will not yield any of their marketshare to Linux, then you could conclude that AMDs active support of Linux forces M$ to support win64 on the Hammer platform.
The reason you know about the 64-bit Linux and IA64 Linux projects is because linux is open source. You don't know what Microsoft is cooking up because they are not open source. Just because you don't know what they are working on doesn't mean they are in trouble here.
I don't think Linux' cross-platform portability has anything to do with their potential in this marketplace. The market is in somewhat of a fluctuation point. That means there is a new set of platforms out there, a new market, and a whole load of applications that need to be created. If there were no degree of backward compatibility, all products would need to be created from the ground up. That would put all software developers on level turf.
However, since there is definitely a degree of backward compatibility (i.e. Hammer will run 32-bit apps, IA64 will run 32-bit apps in a lesser mode) the potential for Linux to wedge itself into the marketplace is not so great.
Linux does have several things going for it however. First, workstations were historically Unix-based, and Linux will be accepted nicely. Workstations historically run X, so again, Linux is a natural. It is multi-platform, so users may be more likely to have used it before, thus more inclined to use it again (if they liked it of course).
Linux has several things working against it though. First, NT-based OSs have a significant market share. As long as M$ has a product available, it will have no problem maintaining market share unless a competitor (i.e. IBM, Sun, etc) seriously markets Linux and does NOT offer an NT product as an alternative. Read: AMD is doing very little to push Linux by simply demonstrating it. Granted, they make silicon, not products, so either way they will have little influence over the acceptance of Linux over a competing OS.
What Linux does for AMD is simply allow Hammer to be an alternative processor to Intel IA64. It forces M$ to directly support Hammer or resign to allowing Linux ownership of at least the market share that AMD has. It also forces companies who are trying to sell Linux solutions (again IBM, Sun, etc) to consider providing Hammer in their product line. In effect, AMD is using Linux as a tool. Linux is NOT using AMD as a tool.
Linux will not be a beneficiary of this effort without a serious benefactor and some serious marketing funding.
Then you can't run the server in a closed network behind a firewall that won't open this port. That's one of the biggest reasons to have a bnet emulator.
If Blizzard did this, and the CD key checks were part of the server software they distributed, then people would be able to run pirated CDs. This violates their fundamental goal of preventing piracy.
For example, consider I got a copy of this server. Then I would get one of likely hundreds of valid keys off the internet, and my friends would all use their pirated keys. We would all use a copy of the original disc image to install our software, but use the pirated key when installing. We could all play on the same server, but hundreds of similar servers could run all using the same pirated keys since there is no way to determine whether the servers are authenticating common keys.
Thanks! Where would we be without clarifications?
on
Intel's Big Chip
·
· Score: 3, Informative
Thanks for your "clarifications". You have saved us all from a life of ignorance.
What you meant to say (and what the article said), is that 464mm^2 is size of the actual die size of the processor This includes the CPU and the caches. The CPU is a relatively small portion of the processor die, and noting there is 3MB of L3, the total cache may amount to 2/3 of the die size. The square on top of the athlon is also the entire processor die: cpu, caches and all.
Also, L3 cache can never perform "equivalently" to L2 or L1 cache unless it runs at core speed. And I can tell you now, it doesn't -- or they wouldn't need L1 and L2. The L3 cache probably runs at something like 10 access cycles or more. It's not difficult to engineer 10 access cycles into any pipeline -- it's impossible. Which is precisely why it's not L1.
I'm quite sure the engineers at Intel have done their modeling homework and determined that however fast the L4 memory may be, the L3 will improve performance by that much more.
Remember, this processor is not meant to go on you or any other Joe Sixpack's desktop. It is meant to sit inside the workstations on the desks of engineers and in the racks of high-bandwidth servers. These platforms are specifically designed to run hundreds of tasks simultaneously and handle staggeringly high memory bandwidths. It has nothing to do with "complicated instructions." The L3 exists for swapping out large pages of memory in large bursts from a significantly larger sized L4 memory (think on the order of 100's of GB) from L5 memory (local drives and SANs) that has an incomprehendable virtual memory space.
This has absolutely nothing to do with mainstream. I'm quite certain an OS already exists that will run on the platform. An IA-64 Linux is well under way (try http://www.linuxia64.org) and you can bet that Compaq, HP, Dell, and Intel have put a total of more than 100x your lifetime earnings into developing software for that platform.
Intel could not care less whether you or 99.9% of the/. readers out there ever buy an IA-64. They don't give a crap about your market segment, but I'm sure if you want to drop $10K+ on a IA64 workstation, be my guest. Your choices are limited. Either choose IA64 or UltraSparc. Or maybe if AMD ever gets a design win, you might get a chance to buy a Hammer box.
What a great movie. A bit demented, and largely resting on the shoulders of a young Brad Pitt as "Early" and the slow (yet so beautiful) Juliette Lewis as Adele.
Sadly, though, this was prior to his celebrity status resulting from X-Files. That was back in the day before casting directors would recruit Duchovny to save an otherwise hopelessly pathetic sci-fi movie from utter disaster.
He showed adequate talent in Kalifornia, hinting that he can cover roles other than the paranoia expert. Although, he could easily finish out his career in sci-fi movies. He really just needs a more talented agent that can help him screen out the bad ones.
Given the data on the CDs is still PCM but with a minor tweak on the samples to induce confusion in computer CDROM equipment, why couldn't Phillips just argue that they are merely trying to fix a problem that the RIAA has caused?
From what little I understand of it, some simple analog filtering present in most CD players essentially defeats the scheme. And if this is the case, I would hardly think their scheme could possibly be considered digital copy protection.
Re:Clustering Athlon XP machines questions
on
AthlonXP Released
·
· Score: 1
"High-powered", "reliable", and "Athlon" all in the same sentence?
Jokes aside, I don't think anyone will be developing said workstations because they will simply not be able to sell the volume they need to make it affordable enough to sell. Catch22.
That's a very good point. Unfortunately, I think if AMD went for a proprietary, integrated rambus-style memory architecture, the royalties and bad press would kill them.
Re:Customers prefer numbers to letters
on
AthlonXP Released
·
· Score: 1
On the other hand, if I had Win7 and Win8 were out, I would think that it wasn't that bad to be only one major revision behind the current product.
If I had WinME and WinXP came out, then nobody ever spoke another word about ME except to say that it sucked, then I would probably upgrade.
My bet is that MS goes back to some numbering scheme after this, but this XP thing is to accomplish the market "flush" that their shareholders are craving. I.e. WinXP 2.0, WinXP 3.0, etc.
Honestly, I don't think MS will give AMD the time of day. I'm sure AMD is crossing their fingers and hoping to get sued so they can get that extra "free" publicity. Since, as we know, no publicity is bad publicity when you are not a marketshare leader.
Yes, we all know the theory behind a deeply-pipelined sequencer, and that if you were truly able to divide each operation into an infinite number of infinitely-small steps, then your processor could run infinitely fast.
Of course this does not take into account changes of flow, but you are incorrect to say that shorter pipelines are faster during heavy branched code. In theory, shorter and longer pipelines perform identically with heavily-branched code. This is because the amount of time required to compute the next instruction address remains constant for a given process. So although the more deeply-pipelined machine must abort more stages due to each mispredicted branch, the faster clock speed makes it a wash.
In actuality, longer pipelines perform far better for several reasons. The first you mentioned already is branch prediction. A correctly-predicted branch suffers no penalty and therefore the higher instruction committal rate can be maintained. And as has long been understood in the academic world, branches can be predicted with a very high rate of accuracy. Of course the rate of mispredicted branches for short and deeply-piplined machines are the same for a given prediction method.
Then, you can analyze the vast majority of multimedia code today. It is of course the killer-app for most home users and is one of the primary driving forces for today's pc market. Multimedia code is very static. Data are processed in packets or bursts of fixed length. Modern processors provide instruction buffers that eliminate the branch penalty for loops of static loop count. It follows that deeply-pipelined processors are a natural fit for multimedia code. If you look at the latest offerings of DSPs in the industry, you will see the industry has become aware of this in the past few years.
However creating a deeply-pipelined machine is not as simple as you let on. If it were, everyone would be doing it, and of course AMD would not need to concede the GHz race so early. Stage overhead can become cumbersome, and operations may be atomic, requiring parallel units instead of further pipelining. Memory systems must be redesigned. Control registers become very complex. Event handling and context switching even moreso.
Finally, you missed my original point. It is not simply rearchitecting the pentium that is getting Intel to the many GHz clock rate. Intel will be relying on a much faster process with much smaller geometries to get there. This is one of the main reasons why AMD won't be able to keep up. Simply shrinking a design to a smaller process does not yield the great speeds one would think. The main reason for this is that portions of the pipeline must be rearchitected to remove the critical paths to truly take advantage of the smaller process. This requires many novel improvements in the memory architecture so that instructions can be fetched in a more pipelined manner and so L2 accesses can be minimized, since as I've mentioned before, speeds do not linearly scale with transistor size.
Well...let's take a look at this as it compares to Moore's Law, which says, essentially, that the top speed of microprocessors will double every 18 months.
Actually, what Moore's Law essentially says is that the number of transistors on a chip will double every 18 months. The speed somewhat follows, but we have seen that simple scaling of transistor size is not sufficient to increase the speed linearly.
Take AMD for example. AMD stays with basically the same microarchitecture as when they first crossed the 1 GHz boundary, over 18 months ago. What are they at, 1400 MHz? That's a 40% increase in the past ~18 mos. Hmm...
Then you look at Intel. Intel practically abandoned the P3 to work on the P4, knowing the P3 was a dead end due to critical paths when scaling up the speed. The reason being that there are some parts of the microarchitecture that simply don't scale linearly with the rest of the process, primarily the memory system. Intel realized that the GHz race will guarantee market share, and has effectively succeeded in maintaining "Moore's Law" in the speed realm by scaling from 1GHz to 2Ghz in the same 18 mos. Sure, but it requires a reimplementation to do it.
If you scale these rates over 6 years, Intel has, yes the 2^4=16x increase you are predicting. AMD on the other hand has but a 1.4^4=3.8x improvement over the next 6 years. End result, Intel would have the 32GHz machine, and AMD would have the 1.4GHz*3.8 = 5.32 GHz Athlon that they call the Athlon 30K which actually performs as well as a 7 GHz P4, (yet still heats the small city.)
This really sounds bad for AMD, not to mention their incredibly-shrinking market share.
Fortunately it's not marketed to the average user
on
2.2 GHz Xeon
·
· Score: 1
You think Intel will waste money marketing this thing to the average user? Hell no. You think the average user would buy it at a few $K per chip even if Intel were stupid enough to try to market it to them?
Just because this news bit makes it past some/. editors doesn't mean it's intended for the/. mainstream. Hell, even us/. (l)users don't even know what to do with a 2.2GHz machine, so you know your mom and dad won't have a clue.
You can rest assured that the average corporation that buys servers will appreciate the hell out of these. Or what -- did you think they'd be satisfied with an old 450MHz Athlon system?
If you want low power, then buy a mobile processor
on
2.2 GHz Xeon
·
· Score: 1
Each chip has a niche in the market. It's good that you noted that this is a server chip because that's its intended market. A server chip is a high-speed, high-bandwidth, heavy IO, efficient multithreaded and multi-processing platform. If you want a low-power (i.e. low-heat) solution, you might think about buying a laptop. You should only pay extra money for things that are important to you, but should you be critical if other people have other priorities?
All this speed is encouraging programmers to turn out applications much more quickly than they used to. I don't think lazy is in their vocabulary. If you don't believe me, why don't you apply to your nearest software firm for that nice cushy job you think you'll get?
Or would you rather wait until 2010 while they hand-code WinXP in assembly so you can run it on your 1994 laptop? I can guarantee you almost nobody would contribute to open-source projects if they required only hand-coded submissions.
Slashdot Mirror
From the article:
The gun is accurate to within seven metres.
Disney == Bad. I'm hoping AOL tries to buy them out soon like some rumors indicate.
However, that's not the real reason why I won't be buying that movie. Am I the only one here that thinks Tron sucked? I saw it again a few years ago and no, the geek factor is not 8, it's off the charts. In fact, it's so high it ruined the whole movie. I couldn't stand to finish the movie.
In typical Disney fashion, it wasn't a movie written by geeks for geeks, but rather by geeks for geeks' Moms and children who need a personified analogy for all the parts of the computer in order to better understand computers. I could go on about how it could easily be the opening scene in the "Computers for Dummies" puppet show, but this is beginning to bore me...
That point may be true for ESR. I think ESR is often an arraogant jackass who will likely leave the cause when it no longer suits his pocketbook anymore. He is definitely quite pleased with how his goals line up with the fanatical open source movement, and he never seems to truly go out on a limb to advance open source. His motives clearly are to promote himself, and when he starts fading from peoples' memories, he never fails to embarrass himself yet again. Granted, he has apparently developed some very key pieces of software and has generously given them back to the community. However, he leaves no question that he is in it for the money and recognition.
RMS, on the other hand knows that his movement will help people (ultimately) by providing them freedoms he is willing to fight for at the expense of his reputation and pocketbook. He is a true libertarian (check out his home page at stallman.org if you have any doubts whatsoever). GNU is one of his many strong passions and is possibly the only one with with which I completely relate and sympathize -- I will add that he is exceptionally wise to maintain a strong distance between his GNU agenda and his other personal agendas so as to not dissociate himself from his GNU supporters. His goals are to free software ("free" as in the verb to free from restraints and regulations). Contrary to what many people believe, the goal is NOT to provide free software ("free" as in the adjective for cost).
The fact that you and many others misunderstand his agenda is precisely why he continues to push it with all his energy. I have an enormous amount of respect for RMS and what he has selflessly created. He is often criticized for being a relentless control-freak. Although this seems a bit exaggerated, I don't doubt that his agenda would fail were he not of this nature.
You are completely wrong and a very poor judge of the Gnu and Linux communities if you think Linux will die when Microsoft becomes easier to use or when Linux becomes unnecessary to support. GNU is all about freedom -- the freedom to program, modify programs to suit you, modify programs for others, and redistribute modified programs. The philosophy has nothing to do with ease of use, and quite frankly cares little whether common man ever uses the software except that common man has power over the government which has the power to advance the cause or seriously impede or abolish the struggle for free software.
BUT -- to earn popular support from the masses, Gnu/Linux products must be free in cost and must be easy to use. The end product, not the philosophy, has to appeal to the masses. Note this has nothing to do with the philosophy of programming freedom. This is merely a means to an end. I agree there is indeed a shame in prematurely trying to sell a product to people not ready for it. Unfortunately, some GNU followers misinterpret the goals of gaining acceptance with trying to obtain marketshare. It is still too early to obtain marketshare, but it is not too early to obtain popular support for the ideals. Note that RMS does not encourage forcing the product on those not yet ready -- I challenge you to provide a counterexample. ESR on the other hand seems to be an unfortunate troll, unknowingly bent on destroying the open source movement.
Your marxist comments are completely irrelevant to the discussion. Just because one can equate RMS' fervor with that of a priest doesn't mean the GNU philosophy is meant to delude or control as does many religions of the spirit. GNU is a governmental policy, grounded in the physical world. There are few, if any, parallels between GNU and spirituality.
No kidding. That was a good pun. Mod it UP.
I know this could put a slant on the postings that are approved, but assuming /. actually generates traffic for other sites such as wired and space.com and pbs.org and nasa.gov, etc, this should, in turn, increase the add revenue of those sites. Why can't /. get a cut of that (probably laughably small) sum of money somehow?
/. would post more links to stories on the sites that comply, but at the same time, it would be drivin not by number of links posted, but by actual click-thrus. This means that the dumb stories don't actually earn revenue for /.
/. to is a discussion wrapper for select stories on other sites. This would kind of out-source the discussion layer from those sites.
/.
I know this would mean that
So what this effectively reduces
One flaw is that the other sites might want revenue from the ad viewership attributed to the discussion about that story, since without that other site's story, there would be no traffic at
Oh well. Back to the drawing board.
First off, support costs spread across all customers. I know most computer geeks would like to think everyone else calls support all the time. I don't know any companies that provide such good support that it can add more than 50% to the total cost of selling OEM windoze OSs (assuming more recent M$ OSs which is precisely what the discussion is about). Remember that support calls are not limited to OS-only issues, but also drivers and hardware troubleshooting, therefore the costs are not all attributed to the OS. You cannot make the argument that drivers are part of the OS any more than you can say the hardware is part of the OS.
Secondly, you only further drive my point that the OS is absolutely not the cheapest part of the system. In particular, I address your quote of my previous post. You only make that quote more correct.
I think what you are trying to say is that the OS is more of the cost of a system than seems obvious. While that may be true, it is becoming much less the case in the XP and 2K versions of windoze. My neighbor is almost 70. I installed Xp on his machine and he has not once called me for help in the past 2.5 months since then.
M$ is removing the support burden from the OEM reseller. Ideally, the OEM version of the software should be getting more expensive due to this factor, but it doesn't seem to be the case since users also want to upgrade to the retail product to take advantage of the "less hassle" factor.
What I think is that you don't really have a point and you like to argue. You started off beefing that the open-source movement is not consumer-oriented and finished up by rambling micromanaging the irrelevant portion of the discussion.
The whole issue of RMS and ESR pushing this open-source movement almost at the level of a religion is where it started. You want this movement to be consumer-oriented, but we are still in the evangelism stage and not ready for consumers. We will be in the evangelism stage for a long time, hence the "religious" nature of their communications. They continue such an extreme doctrine because their philosophy is still spreading, and it is spreading at a very rapid rate. Since this is a critical time, they don't want their views to be diluted while they spread.
No, what the $0 means is that if you want to remove the OS to save money, you will save a whopping $0, so it's not worth your effort.
I'm sure Dell has to pay a significant amount of money for their licenses.
The $35 OS is not by any means the cheapest part of the system. The keyboard and mouse combine to cost approximately $10. The integrated audio costs less than $10. The integrated video costs around $10. The north/southbridge chipsets cost about $30. The fans are less than $5 each. The 3.5" floppy is less than $10. The CDROM drive is less than $35. The case is probably $30. The power supply is less than $20. Non-M$ bundled software is nearly free.
The ONLY things that likely exceed the cost of M$ OS are the monitor, the CPU, and maybe the DVDROM or CDRW drive.
At profit margins that are less than $50 for the low-end PCs, $35 is a large piece of the pie.
For palmtops, the margin is very similar. The battery and the high-end embedded processors and memory are still pretty expensive. In the future, manufacturers will likely take a loss on the initial sale of their hardware and try to recoup their costs on the sale of services. Here, M$ is killing them. But, M$ is trying to remove the service source of revenue as well, so hardware manufacturers will be even worse off.
You are comparing two completely different business models.
The first is the essence of capitalism, where consumers are purchasing their commodities from manufacturers. The consumers typically have the clear agenda to buy a high-performance platform for a cheap price, with a high degree of interoperability and a wide variety of software. The PC offers all this for a fantastic price because of the enormous amount of competition out there.
ESR and RMS are not trying to sell free software to you, obviously, because you can't sell free software. They are not participating in the capitalistic marketplace. They are not at the point where they give a damn about you, especially considering you are a gimme, gimme, gimme type of guy. They don't even have a suite of products to push. They many very well never have a product line to push, but that's ok.
What they do have is a platform, an OS, a compiler, and a huge slew of products that are in a state of constant development. Despite what some may think, these products are not appropriate for you or any other typical user.
What RMS (and to a significantly lesser extent, ESR) are trying to sell is the concept of free software. They are trying to sell their ideology to what I consider two significant bodies of people. One is the companies or governments that have the potential or revenue to fund free software. The other is the large body of software developers who will individually create ideas or write source code to contribute to the base that already exists.
This is an extremely tough sell, since companies want to see an immediate source of revenue as a result of their effort. The sell involves convincing the large entities that free software is better for them in the long run either financially or simply for the freedom from software patents and undue restrictions. Of course individuals may be an easier sell, but even if you convince someone that free software is good, that is good but you still need individual contributors to advance free software.
This is an ongoing effort for both of RMS and ESR. They don't care about you (yet). You are a consumer, and they are not after consumers yet. They need stakeholders and contributors who are ready to make a commitment to their cause.
Don't be so quick to shoot down their cause. The cause has not advanced to the point that you, the consumer, are in their target. If everyone were to dismiss their cause, they might never be able to deliver to you the tools/software/OS/platform you need to see.
I think you have to use your imagination here to figure out how the DMCA could cover WC3.
I'm probably a bigger Bliz fan than most here, and I'm sure that's saying a lot. But I'm also a proponent of individual rights and identify the need for fewer, not more, copyright protections.
With that out of the way, it would be easy to interpret the "multiplayer version" of WC3 as your copy of the game plus Battle.net. If you consider them together as a system, then you are essentially replacing a part of the system with your own part of the system and thereby circumventing the copy protection.
If you assume that multiplayer WC3 does NOT work without Battle.net, then you can assume that the multiplayer version of the game was not intended to work without copy protection, since that was built into Battle.net. Replacing Battle.net removes copy protection.
I hate it, and I too am not a lawyer, but I think this argument can be sold to any judge.
mega
What I could discern from your message is that you are confusing Hammer with a desktop PC processor. You probably didn't understand my original post, so I'll elaborate.
In order for AMD to compete to any extent with IA64, AMD is targetting Hammer to the workstation and server market. Very few people are going to "upgrade for the speed benefits". Even though it's AMD, the price will be way too high for the home PC market. AMD is looking to increase its margins by entering the workstation and server market with Hammer, so it would defeat their purpose to target the home PC market with this platform.
With that in mind, I would offer that M$ is desparately trying to displace Unix in heavy computing environments and even moreso displace Linux and Apache in the server space. To do this, they need to attack Linux on all fronts. And if you recognize that AMD will consume more than a negligible portion of the marketplace (which is definitely arguable at this point since nobody can predict what will turn out) and you assume that M$ will not yield any of their marketshare to Linux, then you could conclude that AMDs active support of Linux forces M$ to support win64 on the Hammer platform.
The reason you know about the 64-bit Linux and IA64 Linux projects is because linux is open source. You don't know what Microsoft is cooking up because they are not open source. Just because you don't know what they are working on doesn't mean they are in trouble here.
I don't think Linux' cross-platform portability has anything to do with their potential in this marketplace. The market is in somewhat of a fluctuation point. That means there is a new set of platforms out there, a new market, and a whole load of applications that need to be created. If there were no degree of backward compatibility, all products would need to be created from the ground up. That would put all software developers on level turf.
However, since there is definitely a degree of backward compatibility (i.e. Hammer will run 32-bit apps, IA64 will run 32-bit apps in a lesser mode) the potential for Linux to wedge itself into the marketplace is not so great.
Linux does have several things going for it however. First, workstations were historically Unix-based, and Linux will be accepted nicely. Workstations historically run X, so again, Linux is a natural. It is multi-platform, so users may be more likely to have used it before, thus more inclined to use it again (if they liked it of course).
Linux has several things working against it though. First, NT-based OSs have a significant market share. As long as M$ has a product available, it will have no problem maintaining market share unless a competitor (i.e. IBM, Sun, etc) seriously markets Linux and does NOT offer an NT product as an alternative. Read: AMD is doing very little to push Linux by simply demonstrating it. Granted, they make silicon, not products, so either way they will have little influence over the acceptance of Linux over a competing OS.
What Linux does for AMD is simply allow Hammer to be an alternative processor to Intel IA64. It forces M$ to directly support Hammer or resign to allowing Linux ownership of at least the market share that AMD has. It also forces companies who are trying to sell Linux solutions (again IBM, Sun, etc) to consider providing Hammer in their product line. In effect, AMD is using Linux as a tool. Linux is NOT using AMD as a tool.
Linux will not be a beneficiary of this effort without a serious benefactor and some serious marketing funding.
Then you can't run the server in a closed network behind a firewall that won't open this port. That's one of the biggest reasons to have a bnet emulator.
For example, consider I got a copy of this server. Then I would get one of likely hundreds of valid keys off the internet, and my friends would all use their pirated keys. We would all use a copy of the original disc image to install our software, but use the pirated key when installing. We could all play on the same server, but hundreds of similar servers could run all using the same pirated keys since there is no way to determine whether the servers are authenticating common keys.
What you meant to say (and what the article said), is that 464mm^2 is size of the actual die size of the processor This includes the CPU and the caches. The CPU is a relatively small portion of the processor die, and noting there is 3MB of L3, the total cache may amount to 2/3 of the die size. The square on top of the athlon is also the entire processor die: cpu, caches and all.
Also, L3 cache can never perform "equivalently" to L2 or L1 cache unless it runs at core speed. And I can tell you now, it doesn't -- or they wouldn't need L1 and L2. The L3 cache probably runs at something like 10 access cycles or more. It's not difficult to engineer 10 access cycles into any pipeline -- it's impossible. Which is precisely why it's not L1.
I'm quite sure the engineers at Intel have done their modeling homework and determined that however fast the L4 memory may be, the L3 will improve performance by that much more.
Remember, this processor is not meant to go on you or any other Joe Sixpack's desktop. It is meant to sit inside the workstations on the desks of engineers and in the racks of high-bandwidth servers. These platforms are specifically designed to run hundreds of tasks simultaneously and handle staggeringly high memory bandwidths. It has nothing to do with "complicated instructions." The L3 exists for swapping out large pages of memory in large bursts from a significantly larger sized L4 memory (think on the order of 100's of GB) from L5 memory (local drives and SANs) that has an incomprehendable virtual memory space.
This has absolutely nothing to do with mainstream. I'm quite certain an OS already exists that will run on the platform. An IA-64 Linux is well under way (try http://www.linuxia64.org) and you can bet that Compaq, HP, Dell, and Intel have put a total of more than 100x your lifetime earnings into developing software for that platform.
Intel could not care less whether you or 99.9% of the /. readers out there ever buy an IA-64. They don't give a crap about your market segment, but I'm sure if you want to drop $10K+ on a IA64 workstation, be my guest. Your choices are limited. Either choose IA64 or UltraSparc. Or maybe if AMD ever gets a design win, you might get a chance to buy a Hammer box.
What a great movie. A bit demented, and largely resting on the shoulders of a young Brad Pitt as "Early" and the slow (yet so beautiful) Juliette Lewis as Adele.
Sadly, though, this was prior to his celebrity status resulting from X-Files. That was back in the day before casting directors would recruit Duchovny to save an otherwise hopelessly pathetic sci-fi movie from utter disaster.
He showed adequate talent in Kalifornia, hinting that he can cover roles other than the paranoia expert. Although, he could easily finish out his career in sci-fi movies. He really just needs a more talented agent that can help him screen out the bad ones.
Given the data on the CDs is still PCM but with a minor tweak on the samples to induce confusion in computer CDROM equipment, why couldn't Phillips just argue that they are merely trying to fix a problem that the RIAA has caused?
From what little I understand of it, some simple analog filtering present in most CD players essentially defeats the scheme. And if this is the case, I would hardly think their scheme could possibly be considered digital copy protection.
Jokes aside, I don't think anyone will be developing said workstations because they will simply not be able to sell the volume they need to make it affordable enough to sell. Catch22.
That's a very good point. Unfortunately, I think if AMD went for a proprietary, integrated rambus-style memory architecture, the royalties and bad press would kill them.
If I had WinME and WinXP came out, then nobody ever spoke another word about ME except to say that it sucked, then I would probably upgrade.
My bet is that MS goes back to some numbering scheme after this, but this XP thing is to accomplish the market "flush" that their shareholders are craving. I.e. WinXP 2.0, WinXP 3.0, etc.
Honestly, I don't think MS will give AMD the time of day. I'm sure AMD is crossing their fingers and hoping to get sued so they can get that extra "free" publicity. Since, as we know, no publicity is bad publicity when you are not a marketshare leader.
Yes, we all know the theory behind a deeply-pipelined sequencer, and that if you were truly able to divide each operation into an infinite number of infinitely-small steps, then your processor could run infinitely fast.
Of course this does not take into account changes of flow, but you are incorrect to say that shorter pipelines are faster during heavy branched code. In theory, shorter and longer pipelines perform identically with heavily-branched code. This is because the amount of time required to compute the next instruction address remains constant for a given process. So although the more deeply-pipelined machine must abort more stages due to each mispredicted branch, the faster clock speed makes it a wash.
In actuality, longer pipelines perform far better for several reasons. The first you mentioned already is branch prediction. A correctly-predicted branch suffers no penalty and therefore the higher instruction committal rate can be maintained. And as has long been understood in the academic world, branches can be predicted with a very high rate of accuracy. Of course the rate of mispredicted branches for short and deeply-piplined machines are the same for a given prediction method.
Then, you can analyze the vast majority of multimedia code today. It is of course the killer-app for most home users and is one of the primary driving forces for today's pc market. Multimedia code is very static. Data are processed in packets or bursts of fixed length. Modern processors provide instruction buffers that eliminate the branch penalty for loops of static loop count. It follows that deeply-pipelined processors are a natural fit for multimedia code. If you look at the latest offerings of DSPs in the industry, you will see the industry has become aware of this in the past few years.
However creating a deeply-pipelined machine is not as simple as you let on. If it were, everyone would be doing it, and of course AMD would not need to concede the GHz race so early. Stage overhead can become cumbersome, and operations may be atomic, requiring parallel units instead of further pipelining. Memory systems must be redesigned. Control registers become very complex. Event handling and context switching even moreso.
Finally, you missed my original point. It is not simply rearchitecting the pentium that is getting Intel to the many GHz clock rate. Intel will be relying on a much faster process with much smaller geometries to get there. This is one of the main reasons why AMD won't be able to keep up. Simply shrinking a design to a smaller process does not yield the great speeds one would think. The main reason for this is that portions of the pipeline must be rearchitected to remove the critical paths to truly take advantage of the smaller process. This requires many novel improvements in the memory architecture so that instructions can be fetched in a more pipelined manner and so L2 accesses can be minimized, since as I've mentioned before, speeds do not linearly scale with transistor size.
Actually, what Moore's Law essentially says is that the number of transistors on a chip will double every 18 months. The speed somewhat follows, but we have seen that simple scaling of transistor size is not sufficient to increase the speed linearly.
Take AMD for example. AMD stays with basically the same microarchitecture as when they first crossed the 1 GHz boundary, over 18 months ago. What are they at, 1400 MHz? That's a 40% increase in the past ~18 mos. Hmm...
Then you look at Intel. Intel practically abandoned the P3 to work on the P4, knowing the P3 was a dead end due to critical paths when scaling up the speed. The reason being that there are some parts of the microarchitecture that simply don't scale linearly with the rest of the process, primarily the memory system. Intel realized that the GHz race will guarantee market share, and has effectively succeeded in maintaining "Moore's Law" in the speed realm by scaling from 1GHz to 2Ghz in the same 18 mos. Sure, but it requires a reimplementation to do it.
If you scale these rates over 6 years, Intel has, yes the 2^4=16x increase you are predicting. AMD on the other hand has but a 1.4^4=3.8x improvement over the next 6 years. End result, Intel would have the 32GHz machine, and AMD would have the 1.4GHz*3.8 = 5.32 GHz Athlon that they call the Athlon 30K which actually performs as well as a 7 GHz P4, (yet still heats the small city.)
This really sounds bad for AMD, not to mention their incredibly-shrinking market share.
Just because this news bit makes it past some /. editors doesn't mean it's intended for the /. mainstream. Hell, even us /. (l)users don't even know what to do with a 2.2GHz machine, so you know your mom and dad won't have a clue.
You can rest assured that the average corporation that buys servers will appreciate the hell out of these. Or what -- did you think they'd be satisfied with an old 450MHz Athlon system?
All this speed is encouraging programmers to turn out applications much more quickly than they used to. I don't think lazy is in their vocabulary. If you don't believe me, why don't you apply to your nearest software firm for that nice cushy job you think you'll get?
Or would you rather wait until 2010 while they hand-code WinXP in assembly so you can run it on your 1994 laptop? I can guarantee you almost nobody would contribute to open-source projects if they required only hand-coded submissions.