But my point is that a 64-bit machine won't have a larger bus than a 32-bit machine.
The 32-bit machine is already transfering at 64-bit or 128-bit.
The original poster was arguing that you would gain speed by the increased bus size. I'm saying, no, because the word size of the computer hasn't been strongly connected to the data-bus from the CPU to memory since the first days of cache.
I'll agree with you (and never did argue) that the memory accesses aren't happening in a critical-word-first order, and that the bus size is not the same size as the cache-line. (I may have incorrectly stated that to begin with, but I didn't argue the point, because I was wrong.)
But the data-bus size and the word size of the computer are most definitly not linked in modern computers, and there's no reason to think that there would be a speed increase for a wider data bus.
Adjacent Cache Line Prefetch (Enabled/Disabled) - to enable/disable the adjacent cache line prefetch mode. When disabled, only one 64 byte line from the 128 byte sector is prefetched (which contains the requested data). When enabled - both lines are prefetched no matter whether they have or have not the requested data.
Wow, it's almost like they're saying that it grabs 64 bytes at a time. And on most production hardware, in fact, the entire 128 byte sector at a time.
Well, not all at one time, they have to break it up to go over the data bus...
Except that every access to memory hits an enitre cache-line. I don't doubt that the external bus is not the entire cache-line wide.
But every access to memory does extract a whole cache-line (or at least more than a single data-bus size connection) at a time. Otherwise you eliminate the relevance of proximity, which is one of the two reasons that cache is useful.
Actually, I did deal with program in the 16-bit segement 16-bit address space day and age, also with the interaction with EMS, copying data in and out of it like a secondary store.
Unlike most kids I had fun as a child. If by fun you mean, drilling holes in my head by dealing with braindead architectures.
But now-a-days the data bus line is not related to the size of the computing word.
CPUs request data in cache-line sizes, and not in word-sizes. Since the CPU is requesting in cache-line sizes, then having a bus wide enough to transport the cache-line will make things faster, anything else expects waste.
This likely has to do with a number of nifty factors. For instance with GMP, where you're performing extremely large math. Normally you're stuck with doing something like this:
MOV32 reg, [src1]
ADD32 reg, [src2]
MOV32 [dst], reg
MOV32 reg, [src1+1]
ADC32 reg, [src2+1]
MOV32 [dst+1], reg
Which works nice and all, but it's a ripple carry adder. Ripple carry is SLOW, because you have to wait on definitive resolution of the carry out from one add before you can start with the next add.
On the 64-bit machine, you can take advantage of numerous Carry Look Ahead schemes. Fundamentally, and theoretically, you can break down the carry flag for each bit of the 64-bit add into two levels of logic. No ripple, and both 32-bit adds effectively go through in parallel.
The "coolest" thing that you can actually do in x86 is called memory value forwarding. (Or something like that).
Basically, you assign an internal register to cache the value of the memory access in an unarchitected register. This means that you can write some code like:
This has been true since the 486. (because that's where I learned that, sometimes, it's just not worth putting it into a register on the x86.)
"Limited registers" and "more load/store" are partially informed reasons why x86-64 is faster than x86-32. But in actual truth, it's in "smaller bottleneck for describing dependencies amoung register values."
The Pentium4 has a ton of GPRs. You just can only access 8 of them at a time. Now, you can access 16 of them at a time. This means better dependency description. Not to "you can keep more in registers than in memory".
I think that it is safe to hold off on 64 bit for your personal desktop until a larger share of applications are compiled with 64 bit optimizations, but unlike the 16 -> 32 bit shift, I suspect the results will be underwhelming except for extremely memory consuming applications.
Very telling, in fact, I think you'll find that the 16-32bit shift only helped in applications that were extremely memory consuming for their time.
Odd thing is, it's about 2^16 times easier to exhaust a 16-bit memory address space than it is a 32-bit memory address space.
Oh posh! Deciding their needs on numerical values of supported RAM, users and CPUs.
Windows Server 2003 minimum requirements say a Pentium 133. Are you going to take them at face value on that? No one cares about the requirements when picking their operating system.
The average person *if they are deciding to upgrade* is going to ask their closest geek advisor (if you have a computer, you know one. Or you know how to get in contact with one) and they'll be able to tell them which windows they should get.
That or, they're going to go down to the local Best Buy or what-have-you, and ask one of the representatives there. And they will also tell them which windows they should get... with a slightly higher margin of error than the local geek advisor.
Either way, do you think the average consumer has trouble picking out which car to buy? I mean, Civic, Corolla, all those makes and models just make my head spin!
What version will I get on my new computer? Mac OSX 10.whatever-they-get-up-next-time. I don't run PCs... more accurately, I suppose now, would be to say, that I don't run Windows unless I have to. Which is at work, and occationally at home. (you know, some of those darned games, just won't get ported)
If you weren't refering to me specifically, then I suggest using the pronoun "one", as it's much clearer that you're not talking to me. "Well, I was wondering what version one would get on a new PC." (Note the subjunctive; makes it clear that this is a hypothetical generic subposition, and not an actual rhretorical question.) I won't bother with comma usage (I hate it, also), but question marks are appreciated when asking questions.
Dell will not sell a crippled Operating System, of this you can be assured. While they do sell Windows XP Home Edition, you can't really argue that it's entirely crippled, as the most notable feature the SMP support is simply a lack of a SMP-enabled kernel. Honestly, for Joe Sixpack, there's no difference between XP Pro, and XP Per.
Now, if you want to argue that Windows Vista Starter edition won't allow you to make a home network, it's possible. But if Dell is selling the computer with a network card, then I'd place money on being able to make a home network on it.
Plus, everyone's running a home network now. And in order to access the internet, you need 90% of the infrastructure to build a home network (100% if it's broadband through a cable/DSL modem that links with an ethernet cord).
Joe Sixpack will be fine with his copy of Windows. Whichever one Dell put on it.
No, they will get whatever OS they had pre-installed on their brand new PC.
You might want to look at new OS adoption rates outside of geeks. It's very, very low.
Once it actually does start to increase (ma' and pa' installing new operating systems on older hardware) the various flavors will be well established, and they will have their closest geek advisor tell them exactly what they need to be running.
Ah yes, it'll be so much easier for the support people when you have to ask:
Tech: "What OS are you running?" Clueless: "What do you mean OS?" Tech: "It's the thing running your computer." Clueless: "Oh, Dell." Tech: "Dell is the manufacturer... who makes the software on it?" Clueless: "OH!!! I'm sorry. OpenOffice." Tech: "*sigh* Ok, open up a terminal." Clueless: "How do I do that?" Tech: "Well, if you're using Linux GNOME, find the little icon that is located somewhere on some panel that's a little screen like thing, and click that. If you're using Linux KDE, then mostly the same procedure, except everything is going to look different. If you're running MacOSX then open Finder, go to the Applications directory, then go into Utilities then double click on the one that says "Terminal" if you're running ZetaOS then........"
Gimme a break. Tech support people dealing with customers will always suck until people learn how to understand computers, and then they won't need tech support for anything but those damned esoteric problems that really aren't their fault.
No, it can't be that they're segmenting the product so that they can target their product to better suit the needs of their users.
I mean, RHEL, vs RH personal.
Or SUSE Linux with SUSE Enterprise Linux, SUSE Personal Edition, SUSE Trial Edition, etc etc..
Everyone does this shit. Not to sound like a Microsoft apologist, but what says product segmentation is going for money. Are you REALLY going to be buying more than one copy?
I mean, if you *are* going to buy more than one copy for the same computer, then it is milking... but these are essentially mutually exclusive purchases.
So, wtf? Be mad at Microsoft for something that really is their fault, not what everyone else is doing to, and extrapolating stupid things to goad the whole irrational anti-Microsoft mentality.
Crap, I'm so getting modded down for this post *shakes his head*
Windows XP Personal Edition Windows XP Professional Edition Windows XP Personal Edition N Windows XP Professional Edition N Windows XP... Windows XP...
I don't remember the other two off the top of my head, but I know they're there.
Windows Server 2003 Standard Edition Windows Server 2003 Enterprise Edition Windows Server 2003 Datacenter Edition Windows Server 2003 High-Performance Computing Windows Small Business Server 2003
(I left out the x64, and ia64 flavors, as I don't believe those are actually really all that much of "flavors" but they are sold seperately, so I guess you could make it 7.)
So, wow. Despite actually having the names for the flavors of Windows Vista. What's the news here? That Microsoft has targetted its products since XP? That's is old news.
If you were actually me, you'd know that I did *not* turn down Microsoft.
Did I "sell-out"? Yes, but that's the only way I was going to make money. All the other companies that are interested in hiring F/OSS people care in general too much about GPA, of which, mine is not awesome.
Literal reason from the mouth of the recruiter for IBM as to why I would not be hired: "Well, you only have a 3.0 GPA, and we don't hire anyone with less than a 3.5" (could be higher)
Unfortunately, in our world as it is right now, one needs to make money in order to survive. With a huge student loan debt over my head, it was way too frustrating to be working in a career field that wasn't related to the degree I obtained, but worse: did not even require the degree.
Yes, I'm a sell-out, but sometimes you have to do somethings that you're not proud of in order just to get ahead in life.
I might be better known in the linux circles, just because of the publicity that CherryOS gave the PearPC community.
I was sick of working for $8/hour with the skills that I had, or worse working piecemeal at a job that was dirty and annoying for at one point $500 a month because there wasn't any other work that I could have done. So when Microsoft called, I was willing to go work for them.
While they say they're working towards Open Source, they're not. Many parts of the company are working strongly to work with F/OSS, and improve MS's image to the F/OSS community, there are other groups that are expressly devoted towards skewing data, and spinning F/OSS as worse than MS's products.
He did not say on his page that he was actually offered any sort of job. He says that he was contacted by a Microsoft representative about a job. And that Micrsoft attempted to recruit him.
Correct me if I'm wrong, but at some point your job offer at Microsoft started as a telephone interview.
"Microsoft tries to recruit me" title from TFA. Now, agreed that doesn't say "job offer" at all.
But considering that I, also as an open source developer, with a lot of Computing experience was practically guarenteed a position somewhere at Microsoft when they called me to try and recruit me. Somehow, I don't think it would be that far of a stretch to say, that if ESR had expressed any interest in getting hired by Microsoft that they would have picked him up right away.
And just for the record: How do you "buyout" a community that makes Open Source software to ensure that they can't continue working? Hire them into your fold... for ANYTHING, and then tell them that they can't work on FOSS as a matter of company policy.
I personally don't see any reason why the Guards should be under *any* federal branch unless they're federalised.
If they're federalised, then they're usually just as much a combat military organization force as the Army or Air Force, so it makes sense for them to be under the DoD.
But unless they're being assigned to combat, then yeah, DHS definitely. For instance, if they're federalised to handle a national crisis like this hurricane. (although, I don't imagine the Guards in this instance *are* even federalised, and if they are, then why?)
The National Guard should be functioning as a self-sufficient element of the state government to which they belong. If the DoD wants a group trained to fill a special role that they'd like a backup for, then show the state the money (which is possibly how they do it now, I don't know.)
But the Governor of that state (or whatever Commander in Chief they have of their Guard) should be responsible for their Guards, independent of the Federal government in any way.
Unfortunately for having one state's Guard help another state; I can see how you can file that under "interstate traffic", and thus it would be federally controlled. Who knows... but there shouldn't be a reason why the Louisiana Guard couldn't have been on call, and ready to go the second the Hurrican passed.
Slashdot Mirror
But my point is that a 64-bit machine won't have a larger bus than a 32-bit machine.
The 32-bit machine is already transfering at 64-bit or 128-bit.
The original poster was arguing that you would gain speed by the increased bus size. I'm saying, no, because the word size of the computer hasn't been strongly connected to the data-bus from the CPU to memory since the first days of cache.
I'll agree with you (and never did argue) that the memory accesses aren't happening in a critical-word-first order, and that the bus size is not the same size as the cache-line. (I may have incorrectly stated that to begin with, but I didn't argue the point, because I was wrong.)
But the data-bus size and the word size of the computer are most definitly not linked in modern computers, and there's no reason to think that there would be a speed increase for a wider data bus.
Wow, it's almost like they're saying that it grabs 64 bytes at a time. And on most production hardware, in fact, the entire 128 byte sector at a time.
Well, not all at one time, they have to break it up to go over the data bus...
Except that every access to memory hits an enitre cache-line. I don't doubt that the external bus is not the entire cache-line wide.
But every access to memory does extract a whole cache-line (or at least more than a single data-bus size connection) at a time. Otherwise you eliminate the relevance of proximity, which is one of the two reasons that cache is useful.
You are definitely someone with too much time on your hands. ;)
But then, I suppose, so am I...
Actually, I did deal with program in the 16-bit segement 16-bit address space day and age, also with the interaction with EMS, copying data in and out of it like a secondary store.
Unlike most kids I had fun as a child. If by fun you mean, drilling holes in my head by dealing with braindead architectures.
But now-a-days the data bus line is not related to the size of the computing word.
CPUs request data in cache-line sizes, and not in word-sizes. Since the CPU is requesting in cache-line sizes, then having a bus wide enough to transport the cache-line will make things faster, anything else expects waste.
This likely has to do with a number of nifty factors. For instance with GMP, where you're performing extremely large math. Normally you're stuck with doing something like this:
MOV32 reg, [src1]
ADD32 reg, [src2]
MOV32 [dst], reg
MOV32 reg, [src1+1]
ADC32 reg, [src2+1]
MOV32 [dst+1], reg
Which works nice and all, but it's a ripple carry adder. Ripple carry is SLOW, because you have to wait on definitive resolution of the carry out from one add before you can start with the next add.
On the 64-bit machine, you can take advantage of numerous Carry Look Ahead schemes. Fundamentally, and theoretically, you can break down the carry flag for each bit of the 64-bit add into two levels of logic. No ripple, and both 32-bit adds effectively go through in parallel.
Speed++
The problem is that it's architected registers.
The "coolest" thing that you can actually do in x86 is called memory value forwarding. (Or something like that).
Basically, you assign an internal register to cache the value of the memory access in an unarchitected register. This means that you can write some code like:
ROR [mem], 1
ADD [mem], 2
ROL [mem], 1
And it will go faster than:
MOV reg, [mem]
ROR reg, 1
ADD reg, 2
ROL reg, 1
MOV [mem], reg
This has been true since the 486. (because that's where I learned that, sometimes, it's just not worth putting it into a register on the x86.)
"Limited registers" and "more load/store" are partially informed reasons why x86-64 is faster than x86-32. But in actual truth, it's in "smaller bottleneck for describing dependencies amoung register values."
The Pentium4 has a ton of GPRs. You just can only access 8 of them at a time. Now, you can access 16 of them at a time. This means better dependency description. Not to "you can keep more in registers than in memory".
Why do you think that people using 64-bit linux are running a 32-bit userland?
They have the source code, you got back, you recompile, you get at 64-bit binary.
Linux is 64-bit kernel and userland.
WINDOWS is 64-bit kernel and device drivers, with 64/32-bit libraries (often time both at the same time) and 32-bit binaries.
I think that it is safe to hold off on 64 bit for your personal desktop until a larger share of applications are compiled with 64 bit optimizations, but unlike the 16 -> 32 bit shift, I suspect the results will be underwhelming except for extremely memory consuming applications.
Very telling, in fact, I think you'll find that the 16-32bit shift only helped in applications that were extremely memory consuming for their time.
Odd thing is, it's about 2^16 times easier to exhaust a 16-bit memory address space than it is a 32-bit memory address space.
Oh posh! Deciding their needs on numerical values of supported RAM, users and CPUs.
Windows Server 2003 minimum requirements say a Pentium 133. Are you going to take them at face value on that? No one cares about the requirements when picking their operating system.
The average person *if they are deciding to upgrade* is going to ask their closest geek advisor (if you have a computer, you know one. Or you know how to get in contact with one) and they'll be able to tell them which windows they should get.
That or, they're going to go down to the local Best Buy or what-have-you, and ask one of the representatives there. And they will also tell them which windows they should get... with a slightly higher margin of error than the local geek advisor.
Either way, do you think the average consumer has trouble picking out which car to buy? I mean, Civic, Corolla, all those makes and models just make my head spin!
No. They'll figure it out.
What version will I get on my new computer? Mac OSX 10.whatever-they-get-up-next-time. I don't run PCs... more accurately, I suppose now, would be to say, that I don't run Windows unless I have to. Which is at work, and occationally at home. (you know, some of those darned games, just won't get ported)
If you weren't refering to me specifically, then I suggest using the pronoun "one", as it's much clearer that you're not talking to me. "Well, I was wondering what version one would get on a new PC." (Note the subjunctive; makes it clear that this is a hypothetical generic subposition, and not an actual rhretorical question.) I won't bother with comma usage (I hate it, also), but question marks are appreciated when asking questions.
Dell will not sell a crippled Operating System, of this you can be assured. While they do sell Windows XP Home Edition, you can't really argue that it's entirely crippled, as the most notable feature the SMP support is simply a lack of a SMP-enabled kernel. Honestly, for Joe Sixpack, there's no difference between XP Pro, and XP Per.
Now, if you want to argue that Windows Vista Starter edition won't allow you to make a home network, it's possible. But if Dell is selling the computer with a network card, then I'd place money on being able to make a home network on it.
Plus, everyone's running a home network now. And in order to access the internet, you need 90% of the infrastructure to build a home network (100% if it's broadband through a cable/DSL modem that links with an ethernet cord).
Joe Sixpack will be fine with his copy of Windows. Whichever one Dell put on it.
No, they will get whatever OS they had pre-installed on their brand new PC.
You might want to look at new OS adoption rates outside of geeks. It's very, very low.
Once it actually does start to increase (ma' and pa' installing new operating systems on older hardware) the various flavors will be well established, and they will have their closest geek advisor tell them exactly what they need to be running.
Microsoft didn't want to release Windows XP Starter edition, or Windows XP N.
GOVERNMENT made them release those.
So, now Microsoft figures that it will make up some Starter edition boxes, then the Governments can't complain.
Makes sense to me. Why don't you get it yet?
Ah yes, it'll be so much easier for the support people when you have to ask:
Tech: "What OS are you running?"
Clueless: "What do you mean OS?"
Tech: "It's the thing running your computer."
Clueless: "Oh, Dell."
Tech: "Dell is the manufacturer... who makes the software on it?"
Clueless: "OH!!! I'm sorry. OpenOffice."
Tech: "*sigh* Ok, open up a terminal."
Clueless: "How do I do that?"
Tech: "Well, if you're using Linux GNOME, find the little icon that is located somewhere on some panel that's a little screen like thing, and click that. If you're using Linux KDE, then mostly the same procedure, except everything is going to look different. If you're running MacOSX then open Finder, go to the Applications directory, then go into Utilities then double click on the one that says "Terminal" if you're running ZetaOS then........"
Gimme a break. Tech support people dealing with customers will always suck until people learn how to understand computers, and then they won't need tech support for anything but those damned esoteric problems that really aren't their fault.
My point is, that you're going to buy one or the other. You're not going to be buying all 7 flavors.
This isn't pokemon "gotta catch 'em all", this is targetted marketing. EVERYONE DOES IT.
No, it can't be that they're segmenting the product so that they can target their product to better suit the needs of their users.
I mean, RHEL, vs RH personal.
Or SUSE Linux with SUSE Enterprise Linux, SUSE Personal Edition, SUSE Trial Edition, etc etc..
Everyone does this shit. Not to sound like a Microsoft apologist, but what says product segmentation is going for money. Are you REALLY going to be buying more than one copy?
I mean, if you *are* going to buy more than one copy for the same computer, then it is milking... but these are essentially mutually exclusive purchases.
So, wtf? Be mad at Microsoft for something that really is their fault, not what everyone else is doing to, and extrapolating stupid things to goad the whole irrational anti-Microsoft mentality.
Crap, I'm so getting modded down for this post *shakes his head*
What like with Windows XP? XP has six flavors:
Windows XP Personal Edition
Windows XP Professional Edition
Windows XP Personal Edition N
Windows XP Professional Edition N
Windows XP...
Windows XP...
I don't remember the other two off the top of my head, but I know they're there.
Or, maybe it's like Server 2003
Which comes in 5 flavors:
Windows Server 2003 Standard Edition
Windows Server 2003 Enterprise Edition
Windows Server 2003 Datacenter Edition
Windows Server 2003 High-Performance Computing
Windows Small Business Server 2003
(I left out the x64, and ia64 flavors, as I don't believe those are actually really all that much of "flavors" but they are sold seperately, so I guess you could make it 7.)
So, wow. Despite actually having the names for the flavors of Windows Vista. What's the news here? That Microsoft has targetted its products since XP? That's is old news.
If you were actually me, you'd know that I did *not* turn down Microsoft.
Did I "sell-out"? Yes, but that's the only way I was going to make money. All the other companies that are interested in hiring F/OSS people care in general too much about GPA, of which, mine is not awesome.
Literal reason from the mouth of the recruiter for IBM as to why I would not be hired: "Well, you only have a 3.0 GPA, and we don't hire anyone with less than a 3.5" (could be higher)
Unfortunately, in our world as it is right now, one needs to make money in order to survive. With a huge student loan debt over my head, it was way too frustrating to be working in a career field that wasn't related to the degree I obtained, but worse: did not even require the degree.
Yes, I'm a sell-out, but sometimes you have to do somethings that you're not proud of in order just to get ahead in life.
I'll take the opposite side of the coin.
I might be better known in the linux circles, just because of the publicity that CherryOS gave the PearPC community.
I was sick of working for $8/hour with the skills that I had, or worse working piecemeal at a job that was dirty and annoying for at one point $500 a month because there wasn't any other work that I could have done. So when Microsoft called, I was willing to go work for them.
While they say they're working towards Open Source, they're not. Many parts of the company are working strongly to work with F/OSS, and improve MS's image to the F/OSS community, there are other groups that are expressly devoted towards skewing data, and spinning F/OSS as worse than MS's products.
To ESR's defense (yeah, I know, I'm nuts)
He did not say on his page that he was actually offered any sort of job. He says that he was contacted by a Microsoft representative about a job. And that Micrsoft attempted to recruit him.
Correct me if I'm wrong, but at some point your job offer at Microsoft started as a telephone interview.
Because mine did.
you came off like a professional and stable person.
Really... um... did you EXPECT ESR to react in a rational and polite manner?
It's like George Bush offering a job to Michael Moore. What do you think Michael Moore would do? Politely decline and keep it private?
Hell no!
That's the thing with vocal individuals... they're really loud.
"Microsoft tries to recruit me" title from TFA. Now, agreed that doesn't say "job offer" at all.
But considering that I, also as an open source developer, with a lot of Computing experience was practically guarenteed a position somewhere at Microsoft when they called me to try and recruit me. Somehow, I don't think it would be that far of a stretch to say, that if ESR had expressed any interest in getting hired by Microsoft that they would have picked him up right away.
And just for the record: How do you "buyout" a community that makes Open Source software to ensure that they can't continue working? Hire them into your fold... for ANYTHING, and then tell them that they can't work on FOSS as a matter of company policy.
THAT is how you buyout open source software.
Makes sense.
Still doesn't mean that the National Guard of a particular state can't act as an individual autonomous unit.
I personally don't see any reason why the Guards should be under *any* federal branch unless they're federalised.
If they're federalised, then they're usually just as much a combat military organization force as the Army or Air Force, so it makes sense for them to be under the DoD.
But unless they're being assigned to combat, then yeah, DHS definitely. For instance, if they're federalised to handle a national crisis like this hurricane. (although, I don't imagine the Guards in this instance *are* even federalised, and if they are, then why?)
The National Guard should be functioning as a self-sufficient element of the state government to which they belong. If the DoD wants a group trained to fill a special role that they'd like a backup for, then show the state the money (which is possibly how they do it now, I don't know.)
But the Governor of that state (or whatever Commander in Chief they have of their Guard) should be responsible for their Guards, independent of the Federal government in any way.
Unfortunately for having one state's Guard help another state; I can see how you can file that under "interstate traffic", and thus it would be federally controlled. Who knows... but there shouldn't be a reason why the Louisiana Guard couldn't have been on call, and ready to go the second the Hurrican passed.