IA-64 is in absolutely no way related to the AMD64 (aka x86-64) instruction set. In fact, it has absolutely zero relation to the IA-32 instruction set, depsite the similarity in names. IA-64 is a complete new beast. There currently are and only ever have been two 64-bit x86 CPUs, teh Opteron and the Athlon64 (and you can easily count those as just one chip, especially considering they come from identical dies).
The support for AMD64 was only first added to the NetBSD kernel in 2001, much like it was first added to the Linux kernel at about the same time. This work was done on simulators and it was in no way fully functional. In fact, the port is still not officially supported, though apparently it works reasonably well, even with 32-bit x86 code, which is the real tricky part about AMD64 operating systems. Simply porting to AMD64 is just like porting to any other architecture, but setting up some form of dual-architecture system is what has complicated development, both in Linux and in *BSD (a lot of the kinks are still being worked out). On the upside, once this work is done and working for AMD64 it should be much easier to get it working on other operating systems.
Uhh, he lives in Thunder Bay, if the place is only 300km away that it's WAY closer than anything else! Can you name ANY other town that is within 300km of Thunder Bay? I'll give you a hint, you won't find any on most maps!
In any case, the levy on a 20GB iPod would be something like $110 if this goes through. Even with Canadian gas prices you can easily get do a 620km round trip on much less than $110 worth of gas (as long as you aren't driving something like a Suburban or a Ford Excretion). Heck, even with the CD-R levy ($0.59/CD, slight mistake in the article, $0.49/disk is the levy for CD-RWs) you would only need to buy 50 or 100 of them before you've covered the cost of gas (though it's hardly worth the 6+ hour trip for CD-Rs alone!).
Actually it's a $0.59 tariff per CD-R. The Toronto Star seems to be a bit confused, the $0.49 tariff is for each CD-RW (either that or there was a very recent change that hasn't been documented anywhere else).
Your laptop hard drive would not be covered under this tariff, as the tarriff specifically covers portable digital music players rather than the media used to store the data (at least they were smart in this sense, the original proposal tried to differentiate based on the type of media used in MP3 players, but that was removed for the current proposal).
On the upside, at least a lot of the money does seem to be going to the recording artists and songwriters of Canada. The record companies do get a cut, but at least half is supposed to be going straight to the artists. Now, mind you, it is only going to the Canadian artists, and the method used to determine how much these artists get is based entirely on sales (if the artists are selling a lot, doesn't that mean that people are buying their music and not stealing it?:> ).
The store deserves to be blamed if they are charging it at the cash, since the levy is paid by the CD manufacturer/importer! By the time London Drugs (or anyone else) gets the CDs the price has already been worked into it at least one level up the food chain, and probably two levels (the distributer bought from the importer who paid the levy).
You are correct though that most people don't know they've paid the levy. It currently sits at $0.21 a CD, or roughly half of the price we pay for blank CDs. Under the proposed tarriff that amount will jump up to $0.59 a CD. At that point in time the tarriff will make up about three quarters of the cost of blank CDs and it should push the price of CDs up dramatically.
Good to know that the next time I need to buy a CD for a new Linux ISO I'll be paying twice as much, but my hard earned money will go mostly to help Celine Dion pay for her next castle in France.
In my mind at least, there is only one elegant way to do address more than 4GB of memory, and that is to provide a plain old flat address space. If you're doing anything else, it's an ugly hack. We had these ugly hacks back in the bad old days of DOS. They sucked then and they still suck.
As for using very large ints for research, they are pretty rare for the most part. Research and scientific computing uses almost exlcusively floating point calculations. Yeah, there are some rare cases where large ints do come into play (encryption/decryption is a fairly common use of long longs), and if you hit one of those rare cases a 64-bit chip will just trounce a 32-bit chip. However, for the most part 64-bit ints are rare. If only one or two calculations out of a few thousand are long long ints, then the performance penalty for using a 32-bit chip is rather negligible.
The real advantage of a 64-bit chip has always been that it provides a flat address space that is much greater than that of a 32-bit processor. Neither OS X for the G5 or Windows for the Pentium/Xeon do this. Both allow more than 4GB of memory, but they need software tricks to accomplish it.
I don't think that you're problem is that your clueless about Microsoft products, but simply that you aren't bothering to read. Let's try it again, this is what I wrote:
"It's [WinXP 64-bit edition] available now for the Itanium and scheduled for Q3 of 2004 for AMD64"
I think that is pretty clear, no? Since the topic of discussion was operating systems for the AMD64 instruction, I didn't bother elaborating much on WinXP for Itanium. Yes, WinXP 64-bit for Itanium is available now, and it is a true 64-bit operating system now. WinXP 64-bit for AMD64 is not available now, it is 6-9 months away, but when it arrives it will be a true 64-bit operating system as well.
Sure you can write 64-bit applications for OS X, just like you could write 32-bit applications for Win3.1, but that hardly qualifies it as a 64-bit operating system!
All of the core operating system code is still 32-bit. OS X does NOT give you a flat 64-bit memory space, which is the most obvious sign that it's not a true 64-bit operating system. The kernel, the drivers and just about all of the core operating system is still a plain old 32-bit setup. They did include a few hacks to access more than 4GB of physical memory, but Intel proved that you can do that on a 32-bit processor years ago. The PPro and all follow-up x86 chips can support up to 64GB of physical memory. Like OS X, it does so through the use of ugly hacks.
Apple does also provide some 64-bit math libraries, which make use of the 64-bit integer registers. Nice, but not particularly important. It's fairly rare for most applications to need integers with more range than the 4 billion provided by 32-bit ints. However, when they are needed, being able to use native 64-bit integers is a big bonus. You can hack 64-bit integers together using two 32-bit integers, but that takes at least 3 times as long as with a proper 64-bit int.
As for Longhorn, I don't really know why you're bringing that up, it's years off and has absolutely nothing to do with the current discussion. WinXP 64-bit edition is just that, the 64-bit version of WinXP, no connection to Longhorn at all except that it comes from the same company. It's available now for the Itanium and scheduled for Q3 of 2004 for AMD64. This will be a true 64-bit operating system, providing a flat 64-bit address space with no ugly hacks or tricks. The kernel will be compiled in 64-bit mode, the memory manager will deal out address space using 64-bit pointers, and if your application wants to allocate 20GB of memory, you'll get it (assuming you've got sufficient physical + virtual memory).
Tiemtime has already changed. Current ETA has WinXP 64-bit for AMD64 arriving in Q3 of '04.
Of course, if you want 64-bit now, there are several Linux distributions that fully support the AMD64 instruction set (SuSE, Red Hat, Gentoo and Mandrake all have AMD64 distributions and there may be others).
Yes, XP Home will run on any 64-bit processor that uses either the AMD64 instruction set (Athlon64 or Opteron) or the IA-64 instruction set (Itanium or Itanium2), though it will run very poorly on the latter.
The AMD64 (aka x86-64) instruction set is a straight extension of the IA-32 (aka x86) instruction set to 32-bits, just like how IA-32 was a straight extension of the old 16-bit x86 to 32-bits. And just like how you could run old 16-bit operating systems on a 386 or 486 (and right up to todays Pentium 4 and even these Athlon64 chips), you can run a 32-bit operating system on these new 64-bit chips.
Not only does WinXP Home run on these Athlon64 chips, it actually runs quite well. They make for some of the fastest machines out there at running 32-bit code (the Athlon64 3200+ and the P4 3.2GHz are pretty comperable, depending on which benchmarks you look at).
Err, maybe because they AREN'T the first company to produce a lower-end computer with a 64-bit processor? HP is also selling Athlon64 desktops, and a number of the smaller tier 2 and tier 3 OEMs are as well.
Then of course there is Apple, but they don't really count:>
The mainstream media does tend to figure that consumer = Microsoft Windows. I don't know about you, but I'd sure as hell consider Mandrake Linux to be a "consumer" operating system. It's clearly not targeting business users the way that SuSE and Red Hat are, nor is it really a hobbyist OS like Gentoo (not that Gentoo is strictly useful for a hobby, just that the target market is for people who like to tweak their system a lot rather than the "turn it on and go" crowd). I don't know what that leaves other than it being a consumer operating system.
As for OS X, it definitely isn't a 64-bit operating system. Even the new 10.3 "Panther" version is no more a 64-bit operating system than Win3.1 was a 32-bit operating system. There are a few 64-bit elements (and probably sufficient for most Mac users for the time being), but it's still almost exclusively a 32-bit operating system. WinXP 64-bit for AMD64 will be a full-fledged 64-bit operating system from the ground up. This, unfortunately, means that it needs new drivers, which might be a bit of a problem early on.
Only if the law is enforced. Just look at all the fraudulant spam, the unlicensed pharmacies selling prescription drugs, the scams, the bestiality porn, etc. etc. that gets spammed to us now. Most spam is already blatently illegal, but as long as the law is not being enforced, it isn't going to help anything.
That being said, I really doubt that this spamhole project is going to help much either. Open relays are so 1990's for spammers! These days it almost all comes through open proxies. A similar project that used open proxies might be more effective, though it would need to do a little bit of trickery (ie pretending that the receiving mail server is accepting the message while not even trying to connect to said server).
That being said, I'm not sure that more open proxies on the internet is what we need, regardless of any good intentions behind them.
Much as I like Intel's Pentium M processor, AMD actually doesn't do that bad here. The Pentium M running at 1.6GHz consumes somewhere around 25W of power. AMD's AthlonXP-M chips for the "thin and light" market consume a maximum of about 25W at ~1.4 or 1.5GHz as well (unfortunately AMD does a piss-poor job of documeting their mobile processors, so a bit of guesswork is required). The Pentium M is a slightly faster processor, but the difference shouldn't be huge.
The AthlonXP-M "Desktop replacement" chips consume more power with no improvement in performance (except being available at higher clock speeds), so they don't do all that well, but then agian, their compeititon is the P4-M. Here AMD actually has the low-powered solution, as the P4-M chips quite the power hogs.
The highest power consuming Athlons run at about 70W. The highest power consuming P4's run at about 80W.
So why does the Athlon "run hot" while the P4 runs cooler? Because people stick tiny little heatsinks with insanely high RPM fans on Athlons but they stick big honking heatsinks with slow-spinning quiet fans on their P4s.
Of course, there's no reason why you can't put a bit heatsink and low-rpm fan on an Athlon. I did just that and it's VERY quiet.
Hmm, you're experience with Intel chipsets (or, more to the point, the drivers for said chipsets) must have been better then mine. Sure, Intel's drivers have almost always been better than those for VIA, ALi or SiS, but I'd hardly call them top-notch. Especially early revisions of certain chips were particulaly bad. For example, the first drivers for the PIIX4 southbridge (used with the 430TX and 440LX northbridges) were a disaster. Intel ended up getting them right, and by the time the 440BX came around (still using the PIIX4 southbridge) they were great. Similarly the early i8xx series of drivers were rather troublesome, but by the time the i845 chipset came out they were rock-solid.
When it comes to the competition though, nVidia has Intel beat solid. Their drivers simply install and work on any one of their chipsets. Intel has been working to improve their drivers, but they still have a little ways to go to match nVidia if you ask me. And if you want an nVidia chipset, you need an AMD processor (or an XBox:> ).
They certainly shouldn't be thrown in with the Cyrix line (any of them) since they have absolutely nothing in common with Cyrix except the name, and they got rid of that years ago.
The chips were designed by the Centaur/IDT team, the people that made the old Winchip series of processors. VIA bought both Cyrix line from from National Semi and Centaur Winchip from IDT. The kept the Cyrix name for a short time, but ditched everything else.
I know that most Slashdot readers rarely care for ACCURATE info, instead preferring to read incredibly biased editorials like the one linked above...
But, just in case anyone really does want the real info about TCG, here is a link to v.1.1b of the TCG specification, straight from the horses mouth so to speak.
Warning: this is a 300+ page technical document, might take a while to get through it!
No, the limit on the modern telco has always been known to be 4000 "baud" and 64Kbit/s, and no, we have not managed to reach this level yet (hint: there ain't no such thing as a "9600 baud" modem, only 9600 bits/second modems), nor are we ever going to, it's simple physics, just like what this article is talking about.
The question is always how close to the theoretical limit you can push things in the real world in a reliable and *economical* fashion. For modems, we managed to get them fairly close, albeit only in certain specific conditions (one end is digital). DSL isn't a work-around or a gimmick, it's a whole new system. Same deal here. We might manage to create a whole concept of IC design that changes all the rules, but that's got nothing to do with this article. All the paper said was that current transistor designs stop working at a certain point, regardless of what tricks you try and any material you use.
As you can see, memory bandwidth has only increased half as quickly as your processor speed and memory size (actually it's not quite that bad since the P4 reaches a higher percentage of it's theoretical peak than the old Pentium does). But it gets worse.
100MHz Pentium had ~ 300ns memory latency (rough guess here, I can't find any exact numbers, but it's in this range). 3.0GHz P4 has about 75ns memory latency.
Now THAT is a real killer, and the main reason why things like cache and memory prefetching have become such a big deal. Heck, even cache latency has become a big deal since you could easily wait for as many clock cycles to get data from cache as you used to wait to get data from memory. At 100MHz, you clock cycles are 10ns long, so you only need to wait for 30 clock cycles to get data from memory. For comparison, the L3 cache of the P4EE/Xeon has a latency of about 30-40 clock cycles.
You also get some similar numbers if you look at hard drive bandwidth and latency. Our hard drives our quite a bit faster now than they used to be, but as a fraction of the processor clock speed they are MUCH slower, particularly when you're talking about latency (roughly equal to seek time in hard drive speak).
A lot of it depends on just how much we continue to modify Moore's "Law". In his original paper, Moore stated that transistor density/cost would double every 12 months. Well, guess what, that hasn't been the case for ages, so Moore's law has been obsolete for a LONG time. We fairly quickly switched to doubling every 18 months, but even that is no longer holding true. Now people are claiming that Moore's Law states that transistor density will double every 24 months.
So, all we have to do to keep Moore's Law going is to keep changing the law. After all, if we decide that transistor density will double every 36 months than the "law" will last for twice as long as if we say it will double every 18 months.
I don't know who you're "inside connection" is, but whoever it is, they were smoking the crack pipe when they told you this. Either that or you were smoking the crack pipe when you posted this message. Either way, somewhere along the line, a lot of crack was smoked.
but we are already switching the fuel technology backbone to Hydrogen
Hehe, I always get a kick out of it when people start talking about our new "hydrogen based society" or some other garbage like that. It's incredible how many people seem to believe that you can generate power from hydrogen! Of course, anyone with an once of scientific knowledge can tell you, unless you're talking about nuclear fussion, than hydrogen is simply an energy carrier and not an energy source. You don't pick hydrogen off the magic hydrogen tree, you don't mine hydrogen from the ground and it definitely doesn't just materialize. You put energy into water, you get hydrogen and oxygen. You combine the two back together again at a later date and you get most (but not all) of the energy back. Long story short, you've basically made a cell (aka a "battery" in commonspeak). It's no coincedence that we call these things "fuel cells".
There may be ways to break down hydrocarbons cleanly, efficiently and *cheaply*, thus providing another source of hydrogen where you can get more energy out than you have to put in, but guess where those hydrocarbons come from? If you said, oil, you win the prize!
In any case, in a vain attempt to bring this back on-topic, nanotubes and the like do provide some interesting new long-term possibilities for producing ICs, but they are definitely not without their own set of constraints. No matter how you slice it, sooner or later you run into a minimum size. At some point in time you just don't have enough atoms left to keep your electrons where you would expect them to be. There's lots that can be done in new and different ways to help push these problems further back, but no matter what technology you chose you eventually hit the same sorts of limitations.
Long story short, don't hold your breath for nanotechnology to revolutionize ICs, and definitely don't hold your breath for a society "powered by hydrogen"!
To be a tiny bit pedantic, Moore's original paper talked about the number of transistors per integrated circuit at any given price point. You can always stick more transistors on the chip if you're willing to throw sufficient amounts of money at the problem, but to get those transistors for a reasonable price is another matter.
FWIW, Moore's original hypothesis was that the transistors/$ would double every 12 months, so his "law" hasn't been correct for quite some time. We had been seeing a doubling of transistors about every 18 months for a while, but now it's more like every 24 months. With the current troubles that Intel, AMD and IBM all seem to be having at implementing their new 90nm manufacturing process, it seems likely that the pace will continue to slow.
First and foremost, "Centrino" is a 100% pure marketing term, nothing at all new or technical behind it. All Centrino means is that you havea laptop that has an Intel Pentium-M processor with an Intel motherboard chipset and an Intel wireless chip. If you use a Pentium-M processor and Intel motherboard chipset but a non-Intel wireless chipset (eg one of the Broadcom chips that include 802.11g support which is lacking in the Intel chip), you can't call your laptop a "Centrino" laptop and more importantly, you don't get the millions of dollars in advertising money from Intel.
Intel is just planning on integrating the motherboard chipset and the wireless chip onto a single piece of silicon. This has been the plan all along for their notebook chipsets, but now it looks like they're also planning on integrating the wireless chip into their motherboard chipsets for desktops as well.
Initially it will probably be targeted at OEMs of the Dell and HPaq variety, so compatiability with older boards (or future boards for that matter) will be a non-issue. For these guys, saving a few dollars on parts is a fairly big deal, even if it means that users have a hard time getting replacement parts on their own (in fact, this may be seen as a benefit to the OEMs).
As for the future, I'd imagine that Intel will try to put as much of a software radio on board as they possibly can. Anything that can bring more of the total system dollars into Intel's pocket is a good thing in their books. It used to be that Intel ONLY made the CPU in most computers. Now, Intel is responsible for the CPU, the memory controller, hard drive controller, extra I/O controller (USB, serial, keyboard, etc.), sound, modem, NIC, etc. etc. Heck, Intel is even competing with nVidia to be the #1 supplier of graphics processors in the world now. The more things that they can integrate, the larger chunk of the total cost of a computer goes into Intel's pockets. And if the integration is of the software variety and needs a little extra processor umph, that's an extra bonus for Intel since it gives users more reason to upgrade to the latest and greatest processor.
Slashdot Mirror
IA-64 is in absolutely no way related to the AMD64 (aka x86-64) instruction set. In fact, it has absolutely zero relation to the IA-32 instruction set, depsite the similarity in names. IA-64 is a complete new beast. There currently are and only ever have been two 64-bit x86 CPUs, teh Opteron and the Athlon64 (and you can easily count those as just one chip, especially considering they come from identical dies).
The support for AMD64 was only first added to the NetBSD kernel in 2001, much like it was first added to the Linux kernel at about the same time. This work was done on simulators and it was in no way fully functional. In fact, the port is still not officially supported, though apparently it works reasonably well, even with 32-bit x86 code, which is the real tricky part about AMD64 operating systems. Simply porting to AMD64 is just like porting to any other architecture, but setting up some form of dual-architecture system is what has complicated development, both in Linux and in *BSD (a lot of the kinks are still being worked out). On the upside, once this work is done and working for AMD64 it should be much easier to get it working on other operating systems.
Uhh, he lives in Thunder Bay, if the place is only 300km away that it's WAY closer than anything else! Can you name ANY other town that is within 300km of Thunder Bay? I'll give you a hint, you won't find any on most maps!
In any case, the levy on a 20GB iPod would be something like $110 if this goes through. Even with Canadian gas prices you can easily get do a 620km round trip on much less than $110 worth of gas (as long as you aren't driving something like a Suburban or a Ford Excretion). Heck, even with the CD-R levy ($0.59/CD, slight mistake in the article, $0.49/disk is the levy for CD-RWs) you would only need to buy 50 or 100 of them before you've covered the cost of gas (though it's hardly worth the 6+ hour trip for CD-Rs alone!).
Actually it's a $0.59 tariff per CD-R. The Toronto Star seems to be a bit confused, the $0.49 tariff is for each CD-RW (either that or there was a very recent change that hasn't been documented anywhere else).
:> ).
Your laptop hard drive would not be covered under this tariff, as the tarriff specifically covers portable digital music players rather than the media used to store the data (at least they were smart in this sense, the original proposal tried to differentiate based on the type of media used in MP3 players, but that was removed for the current proposal).
On the upside, at least a lot of the money does seem to be going to the recording artists and songwriters of Canada. The record companies do get a cut, but at least half is supposed to be going straight to the artists. Now, mind you, it is only going to the Canadian artists, and the method used to determine how much these artists get is based entirely on sales (if the artists are selling a lot, doesn't that mean that people are buying their music and not stealing it?
The store deserves to be blamed if they are charging it at the cash, since the levy is paid by the CD manufacturer/importer! By the time London Drugs (or anyone else) gets the CDs the price has already been worked into it at least one level up the food chain, and probably two levels (the distributer bought from the importer who paid the levy).
You are correct though that most people don't know they've paid the levy. It currently sits at $0.21 a CD, or roughly half of the price we pay for blank CDs. Under the proposed tarriff that amount will jump up to $0.59 a CD. At that point in time the tarriff will make up about three quarters of the cost of blank CDs and it should push the price of CDs up dramatically.
Good to know that the next time I need to buy a CD for a new Linux ISO I'll be paying twice as much, but my hard earned money will go mostly to help Celine Dion pay for her next castle in France.
In my mind at least, there is only one elegant way to do address more than 4GB of memory, and that is to provide a plain old flat address space. If you're doing anything else, it's an ugly hack. We had these ugly hacks back in the bad old days of DOS. They sucked then and they still suck.
As for using very large ints for research, they are pretty rare for the most part. Research and scientific computing uses almost exlcusively floating point calculations. Yeah, there are some rare cases where large ints do come into play (encryption/decryption is a fairly common use of long longs), and if you hit one of those rare cases a 64-bit chip will just trounce a 32-bit chip. However, for the most part 64-bit ints are rare. If only one or two calculations out of a few thousand are long long ints, then the performance penalty for using a 32-bit chip is rather negligible.
The real advantage of a 64-bit chip has always been that it provides a flat address space that is much greater than that of a 32-bit processor. Neither OS X for the G5 or Windows for the Pentium/Xeon do this. Both allow more than 4GB of memory, but they need software tricks to accomplish it.
I don't think that you're problem is that your clueless about Microsoft products, but simply that you aren't bothering to read. Let's try it again, this is what I wrote:
"It's [WinXP 64-bit edition] available now for the Itanium and scheduled for Q3 of 2004 for AMD64"
I think that is pretty clear, no? Since the topic of discussion was operating systems for the AMD64 instruction, I didn't bother elaborating much on WinXP for Itanium. Yes, WinXP 64-bit for Itanium is available now, and it is a true 64-bit operating system now. WinXP 64-bit for AMD64 is not available now, it is 6-9 months away, but when it arrives it will be a true 64-bit operating system as well.
Sure you can write 64-bit applications for OS X, just like you could write 32-bit applications for Win3.1, but that hardly qualifies it as a 64-bit operating system!
All of the core operating system code is still 32-bit. OS X does NOT give you a flat 64-bit memory space, which is the most obvious sign that it's not a true 64-bit operating system. The kernel, the drivers and just about all of the core operating system is still a plain old 32-bit setup. They did include a few hacks to access more than 4GB of physical memory, but Intel proved that you can do that on a 32-bit processor years ago. The PPro and all follow-up x86 chips can support up to 64GB of physical memory. Like OS X, it does so through the use of ugly hacks.
Apple does also provide some 64-bit math libraries, which make use of the 64-bit integer registers. Nice, but not particularly important. It's fairly rare for most applications to need integers with more range than the 4 billion provided by 32-bit ints. However, when they are needed, being able to use native 64-bit integers is a big bonus. You can hack 64-bit integers together using two 32-bit integers, but that takes at least 3 times as long as with a proper 64-bit int.
As for Longhorn, I don't really know why you're bringing that up, it's years off and has absolutely nothing to do with the current discussion. WinXP 64-bit edition is just that, the 64-bit version of WinXP, no connection to Longhorn at all except that it comes from the same company. It's available now for the Itanium and scheduled for Q3 of 2004 for AMD64. This will be a true 64-bit operating system, providing a flat 64-bit address space with no ugly hacks or tricks. The kernel will be compiled in 64-bit mode, the memory manager will deal out address space using 64-bit pointers, and if your application wants to allocate 20GB of memory, you'll get it (assuming you've got sufficient physical + virtual memory).
Tiemtime has already changed. Current ETA has WinXP 64-bit for AMD64 arriving in Q3 of '04.
Of course, if you want 64-bit now, there are several Linux distributions that fully support the AMD64 instruction set (SuSE, Red Hat, Gentoo and Mandrake all have AMD64 distributions and there may be others).
Yes, XP Home will run on any 64-bit processor that uses either the AMD64 instruction set (Athlon64 or Opteron) or the IA-64 instruction set (Itanium or Itanium2), though it will run very poorly on the latter.
The AMD64 (aka x86-64) instruction set is a straight extension of the IA-32 (aka x86) instruction set to 32-bits, just like how IA-32 was a straight extension of the old 16-bit x86 to 32-bits. And just like how you could run old 16-bit operating systems on a 386 or 486 (and right up to todays Pentium 4 and even these Athlon64 chips), you can run a 32-bit operating system on these new 64-bit chips.
Not only does WinXP Home run on these Athlon64 chips, it actually runs quite well. They make for some of the fastest machines out there at running 32-bit code (the Athlon64 3200+ and the P4 3.2GHz are pretty comperable, depending on which benchmarks you look at).
Err, maybe because they AREN'T the first company to produce a lower-end computer with a 64-bit processor? HP is also selling Athlon64 desktops, and a number of the smaller tier 2 and tier 3 OEMs are as well.
:>
Then of course there is Apple, but they don't really count
The mainstream media does tend to figure that consumer = Microsoft Windows. I don't know about you, but I'd sure as hell consider Mandrake Linux to be a "consumer" operating system. It's clearly not targeting business users the way that SuSE and Red Hat are, nor is it really a hobbyist OS like Gentoo (not that Gentoo is strictly useful for a hobby, just that the target market is for people who like to tweak their system a lot rather than the "turn it on and go" crowd). I don't know what that leaves other than it being a consumer operating system.
As for OS X, it definitely isn't a 64-bit operating system. Even the new 10.3 "Panther" version is no more a 64-bit operating system than Win3.1 was a 32-bit operating system. There are a few 64-bit elements (and probably sufficient for most Mac users for the time being), but it's still almost exclusively a 32-bit operating system. WinXP 64-bit for AMD64 will be a full-fledged 64-bit operating system from the ground up. This, unfortunately, means that it needs new drivers, which might be a bit of a problem early on.
Only if the law is enforced. Just look at all the fraudulant spam, the unlicensed pharmacies selling prescription drugs, the scams, the bestiality porn, etc. etc. that gets spammed to us now. Most spam is already blatently illegal, but as long as the law is not being enforced, it isn't going to help anything.
That being said, I really doubt that this spamhole project is going to help much either. Open relays are so 1990's for spammers! These days it almost all comes through open proxies. A similar project that used open proxies might be more effective, though it would need to do a little bit of trickery (ie pretending that the receiving mail server is accepting the message while not even trying to connect to said server).
That being said, I'm not sure that more open proxies on the internet is what we need, regardless of any good intentions behind them.
Much as I like Intel's Pentium M processor, AMD actually doesn't do that bad here. The Pentium M running at 1.6GHz consumes somewhere around 25W of power. AMD's AthlonXP-M chips for the "thin and light" market consume a maximum of about 25W at ~1.4 or 1.5GHz as well (unfortunately AMD does a piss-poor job of documeting their mobile processors, so a bit of guesswork is required). The Pentium M is a slightly faster processor, but the difference shouldn't be huge.
The AthlonXP-M "Desktop replacement" chips consume more power with no improvement in performance (except being available at higher clock speeds), so they don't do all that well, but then agian, their compeititon is the P4-M. Here AMD actually has the low-powered solution, as the P4-M chips quite the power hogs.
FWIW Tom's Hardware did a comparison of two nearly identical notebooks, one with a Pentium-M and the other with an AthlonXP-M.
The highest power consuming Athlons run at about 70W. The highest power consuming P4's run at about 80W.
So why does the Athlon "run hot" while the P4 runs cooler? Because people stick tiny little heatsinks with insanely high RPM fans on Athlons but they stick big honking heatsinks with slow-spinning quiet fans on their P4s.
Of course, there's no reason why you can't put a bit heatsink and low-rpm fan on an Athlon. I did just that and it's VERY quiet.
Hmm, you're experience with Intel chipsets (or, more to the point, the drivers for said chipsets) must have been better then mine. Sure, Intel's drivers have almost always been better than those for VIA, ALi or SiS, but I'd hardly call them top-notch. Especially early revisions of certain chips were particulaly bad. For example, the first drivers for the PIIX4 southbridge (used with the 430TX and 440LX northbridges) were a disaster. Intel ended up getting them right, and by the time the 440BX came around (still using the PIIX4 southbridge) they were great. Similarly the early i8xx series of drivers were rather troublesome, but by the time the i845 chipset came out they were rock-solid.
:> ).
When it comes to the competition though, nVidia has Intel beat solid. Their drivers simply install and work on any one of their chipsets. Intel has been working to improve their drivers, but they still have a little ways to go to match nVidia if you ask me. And if you want an nVidia chipset, you need an AMD processor (or an XBox
They certainly shouldn't be thrown in with the Cyrix line (any of them) since they have absolutely nothing in common with Cyrix except the name, and they got rid of that years ago.
The chips were designed by the Centaur/IDT team, the people that made the old Winchip series of processors. VIA bought both Cyrix line from from National Semi and Centaur Winchip from IDT. The kept the Cyrix name for a short time, but ditched everything else.
I know that most Slashdot readers rarely care for ACCURATE info, instead preferring to read incredibly biased editorials like the one linked above...
But, just in case anyone really does want the real info about TCG, here is a link to v.1.1b of the TCG specification, straight from the horses mouth so to speak.
Warning: this is a 300+ page technical document, might take a while to get through it!
No, the limit on the modern telco has always been known to be 4000 "baud" and 64Kbit/s, and no, we have not managed to reach this level yet (hint: there ain't no such thing as a "9600 baud" modem, only 9600 bits/second modems), nor are we ever going to, it's simple physics, just like what this article is talking about.
The question is always how close to the theoretical limit you can push things in the real world in a reliable and *economical* fashion. For modems, we managed to get them fairly close, albeit only in certain specific conditions (one end is digital). DSL isn't a work-around or a gimmick, it's a whole new system. Same deal here. We might manage to create a whole concept of IC design that changes all the rules, but that's got nothing to do with this article. All the paper said was that current transistor designs stop working at a certain point, regardless of what tricks you try and any material you use.
Interesting, but there are a lot of other issues at work here. Take, for example, memory bandwidth.
100MHz Pentium had ~ 533MB/s of memory bandwidth
3.0C P4 has 6400MB/s of memory bandwidth
533MB/s / 100MHz = 5.33B
6400MB/s / 3000MHz = 2.13B
As you can see, memory bandwidth has only increased half as quickly as your processor speed and memory size (actually it's not quite that bad since the P4 reaches a higher percentage of it's theoretical peak than the old Pentium does). But it gets worse.
100MHz Pentium had ~ 300ns memory latency (rough guess here, I can't find any exact numbers, but it's in this range). 3.0GHz P4 has about 75ns memory latency.
300ns / 100MHz = 0.003 s^2
75ns / 3000MHz = 0.000025 s^2
Now THAT is a real killer, and the main reason why things like cache and memory prefetching have become such a big deal. Heck, even cache latency has become a big deal since you could easily wait for as many clock cycles to get data from cache as you used to wait to get data from memory. At 100MHz, you clock cycles are 10ns long, so you only need to wait for 30 clock cycles to get data from memory. For comparison, the L3 cache of the P4EE/Xeon has a latency of about 30-40 clock cycles.
You also get some similar numbers if you look at hard drive bandwidth and latency. Our hard drives our quite a bit faster now than they used to be, but as a fraction of the processor clock speed they are MUCH slower, particularly when you're talking about latency (roughly equal to seek time in hard drive speak).
A lot of it depends on just how much we continue to modify Moore's "Law". In his original paper, Moore stated that transistor density/cost would double every 12 months. Well, guess what, that hasn't been the case for ages, so Moore's law has been obsolete for a LONG time. We fairly quickly switched to doubling every 18 months, but even that is no longer holding true. Now people are claiming that Moore's Law states that transistor density will double every 24 months.
So, all we have to do to keep Moore's Law going is to keep changing the law. After all, if we decide that transistor density will double every 36 months than the "law" will last for twice as long as if we say it will double every 18 months.
I don't know who you're "inside connection" is, but whoever it is, they were smoking the crack pipe when they told you this. Either that or you were smoking the crack pipe when you posted this message. Either way, somewhere along the line, a lot of crack was smoked.
but we are already switching the fuel technology backbone to Hydrogen
Hehe, I always get a kick out of it when people start talking about our new "hydrogen based society" or some other garbage like that. It's incredible how many people seem to believe that you can generate power from hydrogen! Of course, anyone with an once of scientific knowledge can tell you, unless you're talking about nuclear fussion, than hydrogen is simply an energy carrier and not an energy source. You don't pick hydrogen off the magic hydrogen tree, you don't mine hydrogen from the ground and it definitely doesn't just materialize. You put energy into water, you get hydrogen and oxygen. You combine the two back together again at a later date and you get most (but not all) of the energy back. Long story short, you've basically made a cell (aka a "battery" in commonspeak). It's no coincedence that we call these things "fuel cells".
There may be ways to break down hydrocarbons cleanly, efficiently and *cheaply*, thus providing another source of hydrogen where you can get more energy out than you have to put in, but guess where those hydrocarbons come from? If you said, oil, you win the prize!
In any case, in a vain attempt to bring this back on-topic, nanotubes and the like do provide some interesting new long-term possibilities for producing ICs, but they are definitely not without their own set of constraints. No matter how you slice it, sooner or later you run into a minimum size. At some point in time you just don't have enough atoms left to keep your electrons where you would expect them to be. There's lots that can be done in new and different ways to help push these problems further back, but no matter what technology you chose you eventually hit the same sorts of limitations.
Long story short, don't hold your breath for nanotechnology to revolutionize ICs, and definitely don't hold your breath for a society "powered by hydrogen"!
To be a tiny bit pedantic, Moore's original paper talked about the number of transistors per integrated circuit at any given price point. You can always stick more transistors on the chip if you're willing to throw sufficient amounts of money at the problem, but to get those transistors for a reasonable price is another matter.
FWIW, Moore's original hypothesis was that the transistors/$ would double every 12 months, so his "law" hasn't been correct for quite some time. We had been seeing a doubling of transistors about every 18 months for a while, but now it's more like every 24 months. With the current troubles that Intel, AMD and IBM all seem to be having at implementing their new 90nm manufacturing process, it seems likely that the pace will continue to slow.
First and foremost, "Centrino" is a 100% pure marketing term, nothing at all new or technical behind it. All Centrino means is that you havea laptop that has an Intel Pentium-M processor with an Intel motherboard chipset and an Intel wireless chip. If you use a Pentium-M processor and Intel motherboard chipset but a non-Intel wireless chipset (eg one of the Broadcom chips that include 802.11g support which is lacking in the Intel chip), you can't call your laptop a "Centrino" laptop and more importantly, you don't get the millions of dollars in advertising money from Intel.
Intel is just planning on integrating the motherboard chipset and the wireless chip onto a single piece of silicon. This has been the plan all along for their notebook chipsets, but now it looks like they're also planning on integrating the wireless chip into their motherboard chipsets for desktops as well.
Initially it will probably be targeted at OEMs of the Dell and HPaq variety, so compatiability with older boards (or future boards for that matter) will be a non-issue. For these guys, saving a few dollars on parts is a fairly big deal, even if it means that users have a hard time getting replacement parts on their own (in fact, this may be seen as a benefit to the OEMs).
As for the future, I'd imagine that Intel will try to put as much of a software radio on board as they possibly can. Anything that can bring more of the total system dollars into Intel's pocket is a good thing in their books. It used to be that Intel ONLY made the CPU in most computers. Now, Intel is responsible for the CPU, the memory controller, hard drive controller, extra I/O controller (USB, serial, keyboard, etc.), sound, modem, NIC, etc. etc. Heck, Intel is even competing with nVidia to be the #1 supplier of graphics processors in the world now. The more things that they can integrate, the larger chunk of the total cost of a computer goes into Intel's pockets. And if the integration is of the software variety and needs a little extra processor umph, that's an extra bonus for Intel since it gives users more reason to upgrade to the latest and greatest processor.