Umm, that's a really bad idea. The whole point of requiring a system call was to ensure that the code is evicted from the instruction cache. The call was required in the past, you could simply usually get away without calling it. Usually, that address range wouldn't be in the icache to begin with, if it hadn't held code before - but that's not guaranteed.
In other words, that system call had nothing to do with e.g. NX to begin with, although now they are related.
There's more to free software than free as in beer. Free as in speech allows disparate volunteer programmers to create a base of commonly-used code, to standardize the essential operating system and tools, so that nobody's unjustly at the mercy of a company that uses control of such tools as a weapon.
When it comes to free services, however, free as in beer is about the long & the short of it. Moreover, unlike Linux, SourceForge is not free to run, and each new "customer" costs OSDN $$.
So maybe it would have been better if he'd said, "you get what OSDN pays for." It's charity.
You don't have any humanities courses? Or is "management" humanities? I don't think anybody who pushes "well-roundedness" has this set of courses in mind. More likely they'd be in favor of not allowing you to fill requirements that should make you take Readin' & Ritin' or Some Country's History with Management and How to Be a CEO. You are responsible for the roundness of your courseload, and if you feel that, having filled requirements set by pro-liberal-arts ppl with shit, that you have wasted your time, it's because you picked shitty classes, not that the philosophy of well-roundedness sux.
Hehehe - I'm being so preachy to a complete stranger who's prolly 4 years older than me. [S]he was just trying to blow off some serious frustration, and I scream uselessly at him/her. Been a while since I've posted on Slashdot...
I couldn't write this before, because it warned me that changing my theme would cause my post to be lost.
There are two themes included with this install, Modern (the default) and Classic. Modern isn't too ugly, it's very round. Not Mac-y, but I can put up with it.
Classic is a knockoff of Apple's standard theme. It's supposed to be indiscriminable, but it's obviously a fake. Scroll bars have an arrow on each end and a proportional tab - an impossibility unless you've used a shareware tool to customize your theme. The accent color (used for scroll tabs etc.) is PURPLE. Yuck.
And, dashing my only real hope for it, it doesn't restore standard Mac behavior to the text boxes. Damn.
I forgot one thing in the previous post: the page source display doesn't format the HTML based on tags or anything. It's all plain black. Got some catchin' up with iCab to do there....
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I'm typing this using NS 6 Mac, and it seems I'm the only one not having major problems...
The first impression was from the installer, which doesn't use Apple's APIs to draw the controls on the screen. Easily one of the worst bastardizations of the Mac interface to be seen on the Mac. It seems that none of it uses Mac APIs for any control-drawing or standard interaction. Right now, if I double-click a word and drag, first the word is selected but when I drag it switches to selecting by character: nonstandard, crappier behavior. Expect that.
I haven't noticed any glaring visual glitches in my usual sites. However, when I clicked the Slashdot Login nutty little box (it forgot all my cookies, apparently), the entire table element that contained it jumped a little to the left. Unexpected, really incredibly strange (I can't imagine how a bug could cause elements to *jump around* without some kind of aliasing - it erased the space left behind when it moved), but I guess within tolerances. Which sums up my NS6 experience till now - still no crashes.
Page rendering times, btw, are incredible. It beats iCab (and iCab beats IE).
But compared to iCab in other ways, this is falling short. There's no standard Mac interface to be seen, and going back to the old Preferences dialog is pretty sad. Although there's no ad filtering, they were thoughtful enough to add an "image blocking" feature, which gets an entire Preferences tab to itself (and there aren't very many tabs), whose sole purpose is to stop loading of images. Shouldn't there be a menu item for that? Practically no cookie control either - it's either "load" or "don't load". You can tell it to accept all cookies from a certain site by clicking "remember this decision" when it asks you the first time (I think), but I haven't seen where to undo that operation.
Of course, it has the IE-style popup sidebar that iCab (semi-)lacks, but I'm pretty happy without one.
Other Netscape-standard behavior is also a pain. If I type "arstechnica" into the navigation bar, it takes me not to "http://www.arstechnica.com" but to a Netscape Search page that doesn't even list Ars' homepage in the first page of search results. Don't expect any overhauls of this kinda thing, it's mainly a facelift.
But it's an update, and it hasn't crashed on me yet. Although being (relatively) independent of the Mac APIs causes nonstandard behavior, which is a pain, that means that it'll probably be maintained better on our platform than before, also.
And it DOESN'T INSTALL ANY EXTENSIONS INTO THE SYSTEM FOLDER! This is the reason I'll be keeping it on my HD as an update to NS4.7 - IE is not for me. And iCab (although less frequently than ever) sometimes doesn't cut it.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Their mission is to seek out and protect their clients from the "bad parts" of the Internet.
What on the Internet is more evil than the alarming number (thousands!) of people hawking scams to technologically illiterate people? They should be protecting people from themselves!
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
This is ssssaaaaadddddd.... No offense, but this should be -1 offtopic -1 misinformative. The article compares several processors at various speeds using an Intel benchmark, and it's apparent that they perform differently. That's the point of this article. And the point of this post is? This post couldn't possibly be informative because the sole nugget of information is already prerequisite to understanding the article it's posted to in the first place.
Not only that, but once the author Informed us that MHz mean nothing (which it doesn't), he laments that "they" won't know the real performance of various chips until they start looking at the "real" simple quantitative properties of processors (at least as he seems to be seeing it), the "mips and flops". I have news for you, Anaxagoras. There is no way to place a general performance score on any processor. How many FLOPS a processor can do makes no difference if you're using a word processor. The maximum number of instructions per second it can do can't be compared across processors that don't run the same ISA. And what's the representative program that says what instructions will determine the MIPS of the processor? Hence the science of benchmarking... but the author of the article used a benchmark!
And what's more, this post is *not* in response to anything. I went and reloaded the thread at -1 and this remained, totally unsolicited.
I'm sorry, I don't normally beat up on people like this. But it irks me when I see a comment with the length and value of a pedagogical sig get more moderation points than so many other people (e.g. moi) get for hard research, typing, and thought.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I'm not sure why I'm responding to an offtopic AC with no reply e-mail address, but... The Cube is built on the same motherboard as every other Mac currently shipping. It uses the same video, sound, USB, firewire, etc. circutry, so Linux should work fine on it. It is true that the PowerBook w/ FireWire didn't boot Linux when it first came out, but that was (as far as I could tell by trying to install it on one) due to the trackpad not behaving like a generic mouse, and it was before the new motherboards were well supported.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
A few weeks ago, MacNN mentioned this. It appears that the PowerBook version is an orderable part, meaning that Mac support is there now. Mebbe they forgot to update their Windows info page with the news...
An abridged repost of something I wrote before, to someone who couldn't see the use of this product:
I have a compatible PowerBook which I use for programming and art. The computer supports driving 2 monitors to make an extended work area, running the LCD at 1024x768 and an external CRT at 1280x1024. Unfortunately, it only has 8MB of VRAM (a fault corrected in the latest generation of PowerBooks), so I have to give up 3d acceleration if I make use of this while in my 3d program... a very bad thing. This product - which, as far as my net searches have shown, is completely novel - would allow me to attach an extra graphics card and drive
3 monitors at once (believe me, with big art projects these will all come in handy).
Besides that, I could plug in a cheap ATA controller and create a repository for MP3s I dont want to store on my main hard drive all the time, as well as installs of alternate OS's and the like. The PowerBook G3's ATA controller uses the main hard drive as an ATA/66 master with no way to connect a slave, making this the only reasonable way to connect additional storage (although a $120 CardBus 20MB/s SCSI card and a SCSI hard drive, or a FireWire hard drive are also options).
So if I install an ATA controller and a Voodoo 3 PCI, I only have 1 slot left for a SCSI card attached to a scanner! (Although FireWire scanners are now more reasonably priced).
Compared to buying a new CPU (especially one on par with my Pismo G3), spending $600 isn't so bad. What's more, when I upgrade the processor daughtercard on my PB, my desktop and my laptop have been upgraded. If I buy this. Which I don't have the money to do. (Or, for that matter, to buy any of these peripherals I've been ranting about). But if this company stays around, I think I have an alternative to getting a desktop...
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
The MacOS has supported multiple monitors at the 2D-drawing-command level since color was introduced in 1988. So, a PowerBook with one of these could support the LCD, an additional monitor running of the motherboard VRAM, and 2 monitors per PCI slot (w/ a special, really expensive card) in this baby - 8 monitors for one huge desktop.
Digressing back onto your topic, the reason this won't work as an upgrade for non-3D-accelerated laptops is that most laptops new enough to use the riser probably have 3D acceleration. Aside from that, anyone willing to shell out $600 for the privilege of connecting a 3D card probably won't cringe at the cost of a decent monitor to go with it.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Just where have "these kinds of things" existed? CardBus 2.1 isn't yet supported on a lot of laptops (Apple, the inventor of CardBus, only introduced it along with FireWire on their laptops late last year).
Port replicators have existed, but a cross-platform way to add any PCI card to your laptop and use it at over 1.2GB/s is new.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Actually, this product caught my eye before when it was mentioned on MacNN. I have a compatible PowerBook which I use for programming and art. The computer supports driving 2 monitors to make an extended work area, running the LCD at 1024x768 and an external CRT at 1280x1024. Unfortunately, it only has 8MB of VRAM (a fault corrected in the latest generation of PowerBooks), so I have to give up 3d acceleration if I make use of this while in my 3d program... a very bad thing. This product - which, as far as my net searches have shown, is completely novel - would allow me to attach an extra graphics card and drive 3 monitors at once (believe me, with big art projects these will all come in handy).
Besides that, I could plug in a cheap ATA controller and create a repository for MP3s I dont want to store on my main hard drive all the time, as well as installs of alternate OS's and the like. The PowerBook G3's ATA controller uses the main hard drive as an ATA/66 master with no way to connect a slave, making this the only reasonable way to connect additional storage (although a $120 CardBus 20MB/s SCSI card and a SCSI hard drive, or a FireWire hard drive are also options).
So if I install an ATA controller and a Voodoo 3 PCI, I only have 1 slot left for a SCSI card attached to a scanner! (Although FireWire scanners are now more reasonably priced).
Compared to buying a new CPU (especially one on par with my Pismo G3), spending $600 isn't so bad. What's more, when I upgrade the processor daughtercard on my PB, my desktop and my laptop have been upgraded. If I buy this. Which I don't have the money to do. (Or, for that matter, to buy any of these peripherals I've been ranting about). But if this company stays around, I think I have an alternative to getting a desktop...
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Does anyone remember that part of the Foundation trilogy - I think it's the fourth book, Foundation and Earth, where they land on the deserted, ultra-desolate planet and discover that their ship has been contaminated with a single-celled life form that thrives on the tiniest bit of energy? Kinda scary to see that now we have these things that have evolved to survive in a vacuum... this has nasty implications for all long-term space missions.
I mean, this stuff has to survive being constantly irradiated in a vaccuum environment! How does it get any materials through the layers of protective shell it must have?
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Then what do you consider binary-coded-decimal, or BCD?
Base conversions are very complicated, and if it was easier to build a computer that was decimal all the way through than to divide and multiply by 10 (which is very computationally intensive) and subtract to get modulos... the instructions for this process can take a significant amount of punch-cards! Even the Motorola 680xx series (the not-too-distant anscestor of today's Palm s and many embedded DSPs) included add, subract, and conversions to and from BCD to aid driving numeric displays.
Actually, one of the main reasons for it being binary is probably the fact that floating-point arithmetic can hardly be done at all in any other base than 2. And binary isn't the same as digital: base-10 is still digital. The difference is that in digital computers all arithmetic results are infinitely precise and reproducible, as opposed to analog computers which add and subtract voltages, which are inherently irrational numbers and therefore cannot be represented as digits (i.e., not digital).
I'm sorry, it's late at night. I don't mean to be overly critical - this guy invented the first FPU, way ahead of his time - but if it didn't have conditional branching etc. you have to question if it really ran a program, per se. I just read this, a link I found elsewhere in this discussion. It goes into some detail about Zuse, you might like it.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Quit dissing the MacOS... the only thing at fault here is IE. I'm using iCab and the word is broken up fine. I'd like to see how IE on Windows breaks this up... but unfortunately there are no good screenshot capture utilities for Windows. (OK, I'm talkin' out my ass, but nothing as good as Snapz, and nothing included with the OS.)
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
There are people saying to use these in heatsinks (which would require large amounts of tubules), and people saying put these in the processor core (which is impossible w/ current fabrication processes). Why not use nanotubes in between, in the packaging of the chip? This would increase the efficiency of whatever 3lb. heatsink is being used on top, and would make the best use of the tubules per g (which is the largest scale of production we can even dream of anytime soon).
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
It turns out that, for the Itanium, we will need to buy new furniture and replace our home electrical wiring. The 4-microchip, 100-million-transistor behemoth will require a 600-pound heatsink to dissipate heat from its 1kW of electrical power consumption. This type of power cannot be carried by regular home wiring, of course, so it is expected that consumers who want an Itanium computer will replace this insufficient equipment to support their new computers.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
The computer created completely novel devices, given only knowledge of physics, to determine the effectiveness of each form. It came up with arrangements of the preprogrammed parts that had not been anticipated.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Yeah, the US and Israel are stealing Canadian military secrets. Soon, American tanks will be replaced by 50-ton armored horses with laser eyes and the ability to break the sound barrier... Israel will fill their already-strong air force with autonomous Pegasi, equipped with heavy machine guns and manure bombs... The Palestinians better not mess with themanymore.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I mentioned this elsewhere, but gotta say - FireWire is a standard for transmitting digital frame data from a camera to a display/computer/general device. AGP is a bus to carry information from the processor to a graphics card. As has been mentioned, you need several hundred MB/s to run a monitor at 60Hz (that is, transmitting the data from the processor). The article states that AGP8x will give about 2GB/s throughput (533MHz * 4 bytes wide) - how are you going to get anything like that through a *serial* bus? 1394b is already clocked at 400MHz!
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Slashdot Mirror
In other words, that system call had nothing to do with e.g. NX to begin with, although now they are related.
There's more to free software than free as in beer. Free as in speech allows disparate volunteer programmers to create a base of commonly-used code, to standardize the essential operating system and tools, so that nobody's unjustly at the mercy of a company that uses control of such tools as a weapon.
When it comes to free services, however, free as in beer is about the long & the short of it. Moreover, unlike Linux, SourceForge is not free to run, and each new "customer" costs OSDN $$.
So maybe it would have been better if he'd said, "you get what OSDN pays for." It's charity.
Hehehe - I'm being so preachy to a complete stranger who's prolly 4 years older than me. [S]he was just trying to blow off some serious frustration, and I scream uselessly at him/her. Been a while since I've posted on Slashdot...
There are two themes included with this install, Modern (the default) and Classic. Modern isn't too ugly, it's very round. Not Mac-y, but I can put up with it.
Classic is a knockoff of Apple's standard theme. It's supposed to be indiscriminable, but it's obviously a fake. Scroll bars have an arrow on each end and a proportional tab - an impossibility unless you've used a shareware tool to customize your theme. The accent color (used for scroll tabs etc.) is PURPLE. Yuck.
And, dashing my only real hope for it, it doesn't restore standard Mac behavior to the text boxes. Damn.
I forgot one thing in the previous post: the page source display doesn't format the HTML based on tags or anything. It's all plain black. Got some catchin' up with iCab to do there....
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
The first impression was from the installer, which doesn't use Apple's APIs to draw the controls on the screen. Easily one of the worst bastardizations of the Mac interface to be seen on the Mac. It seems that none of it uses Mac APIs for any control-drawing or standard interaction. Right now, if I double-click a word and drag, first the word is selected but when I drag it switches to selecting by character: nonstandard, crappier behavior. Expect that.
I haven't noticed any glaring visual glitches in my usual sites. However, when I clicked the Slashdot Login nutty little box (it forgot all my cookies, apparently), the entire table element that contained it jumped a little to the left. Unexpected, really incredibly strange (I can't imagine how a bug could cause elements to *jump around* without some kind of aliasing - it erased the space left behind when it moved), but I guess within tolerances. Which sums up my NS6 experience till now - still no crashes.
Page rendering times, btw, are incredible. It beats iCab (and iCab beats IE).
But compared to iCab in other ways, this is falling short. There's no standard Mac interface to be seen, and going back to the old Preferences dialog is pretty sad. Although there's no ad filtering, they were thoughtful enough to add an "image blocking" feature, which gets an entire Preferences tab to itself (and there aren't very many tabs), whose sole purpose is to stop loading of images. Shouldn't there be a menu item for that? Practically no cookie control either - it's either "load" or "don't load". You can tell it to accept all cookies from a certain site by clicking "remember this decision" when it asks you the first time (I think), but I haven't seen where to undo that operation.
Of course, it has the IE-style popup sidebar that iCab (semi-)lacks, but I'm pretty happy without one.
Other Netscape-standard behavior is also a pain. If I type "arstechnica" into the navigation bar, it takes me not to "http://www.arstechnica.com" but to a Netscape Search page that doesn't even list Ars' homepage in the first page of search results. Don't expect any overhauls of this kinda thing, it's mainly a facelift.
But it's an update, and it hasn't crashed on me yet. Although being (relatively) independent of the Mac APIs causes nonstandard behavior, which is a pain, that means that it'll probably be maintained better on our platform than before, also.
And it DOESN'T INSTALL ANY EXTENSIONS INTO THE SYSTEM FOLDER! This is the reason I'll be keeping it on my HD as an update to NS4.7 - IE is not for me. And iCab (although less frequently than ever) sometimes doesn't cut it.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
What on the Internet is more evil than the alarming number (thousands!) of people hawking scams to technologically illiterate people? They should be protecting people from themselves!
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Boeing gave them one meeleeon dollars <put pinky finger to mouth>!
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
This should be under Humor!
The Onion should do a piece on this... "Hoax receives millions of dollars from scared multinational corporation".
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
No offense, but this should be -1 offtopic -1 misinformative. The article compares several processors at various speeds using an Intel benchmark, and it's apparent that they perform differently. That's the point of this article. And the point of this post is?
This post couldn't possibly be informative because the sole nugget of information is already prerequisite to understanding the article it's posted to in the first place.
Not only that, but once the author Informed us that MHz mean nothing (which it doesn't), he laments that "they" won't know the real performance of various chips until they start looking at the "real" simple quantitative properties of processors (at least as he seems to be seeing it), the "mips and flops". I have news for you, Anaxagoras. There is no way to place a general performance score on any processor. How many FLOPS a processor can do makes no difference if you're using a word processor. The maximum number of instructions per second it can do can't be compared across processors that don't run the same ISA. And what's the representative program that says what instructions will determine the MIPS of the processor? Hence the science of benchmarking... but the author of the article used a benchmark!
And what's more, this post is *not* in response to anything. I went and reloaded the thread at -1 and this remained, totally unsolicited.
I'm sorry, I don't normally beat up on people like this. But it irks me when I see a comment with the length and value of a pedagogical sig get more moderation points than so many other people (e.g. moi) get for hard research, typing, and thought.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I'm not sure why I'm responding to an offtopic AC with no reply e-mail address, but...
The Cube is built on the same motherboard as every other Mac currently shipping. It uses the same video, sound, USB, firewire, etc. circutry, so Linux should work fine on it.
It is true that the PowerBook w/ FireWire didn't boot Linux when it first came out, but that was (as far as I could tell by trying to install it on one) due to the trackpad not behaving like a generic mouse, and it was before the new motherboards were well supported.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
An abridged repost of something I wrote before, to someone who couldn't see the use of this product:
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Digressing back onto your topic, the reason this won't work as an upgrade for non-3D-accelerated laptops is that most laptops new enough to use the riser probably have 3D acceleration. Aside from that, anyone willing to shell out $600 for the privilege of connecting a 3D card probably won't cringe at the cost of a decent monitor to go with it.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Port replicators have existed, but a cross-platform way to add any PCI card to your laptop and use it at over 1.2GB/s is new.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Besides that, I could plug in a cheap ATA controller and create a repository for MP3s I dont want to store on my main hard drive all the time, as well as installs of alternate OS's and the like. The PowerBook G3's ATA controller uses the main hard drive as an ATA/66 master with no way to connect a slave, making this the only reasonable way to connect additional storage (although a $120 CardBus 20MB/s SCSI card and a SCSI hard drive, or a FireWire hard drive are also options).
So if I install an ATA controller and a Voodoo 3 PCI, I only have 1 slot left for a SCSI card attached to a scanner! (Although FireWire scanners are now more reasonably priced).
Compared to buying a new CPU (especially one on par with my Pismo G3), spending $600 isn't so bad. What's more, when I upgrade the processor daughtercard on my PB, my desktop and my laptop have been upgraded. If I buy this. Which I don't have the money to do. (Or, for that matter, to buy any of these peripherals I've been ranting about). But if this company stays around, I think I have an alternative to getting a desktop...
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I mean, this stuff has to survive being constantly irradiated in a vaccuum environment! How does it get any materials through the layers of protective shell it must have?
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I'm impressed. Who manufactured them?
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Base conversions are very complicated, and if it was easier to build a computer that was decimal all the way through than to divide and multiply by 10 (which is very computationally intensive) and subtract to get modulos... the instructions for this process can take a significant amount of punch-cards! Even the Motorola 680xx series (the not-too-distant anscestor of today's Palm s and many embedded DSPs) included add, subract, and conversions to and from BCD to aid driving numeric displays.
Actually, one of the main reasons for it being binary is probably the fact that floating-point arithmetic can hardly be done at all in any other base than 2. And binary isn't the same as digital: base-10 is still digital. The difference is that in digital computers all arithmetic results are infinitely precise and reproducible, as opposed to analog computers which add and subtract voltages, which are inherently irrational numbers and therefore cannot be represented as digits (i.e., not digital).
I'm sorry, it's late at night. I don't mean to be overly critical - this guy invented the first FPU, way ahead of his time - but if it didn't have conditional branching etc. you have to question if it really ran a program, per se. I just read this, a link I found elsewhere in this discussion. It goes into some detail about Zuse, you might like it.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Quit dissing the MacOS... the only thing at fault here is IE. I'm using iCab and the word is broken up fine. I'd like to see how IE on Windows breaks this up... but unfortunately there are no good screenshot capture utilities for Windows. (OK, I'm talkin' out my ass, but nothing as good as Snapz, and nothing included with the OS.)
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
There are people saying to use these in heatsinks (which would require large amounts of tubules), and people saying put these in the processor core (which is impossible w/ current fabrication processes). Why not use nanotubes in between, in the packaging of the chip? This would increase the efficiency of whatever 3lb. heatsink is being used on top, and would make the best use of the tubules per g (which is the largest scale of production we can even dream of anytime soon).
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Nanotubes are prohibitively expensive. They have a ton of uses, there's just no good way to manufacture mass quantities.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
It turns out that, for the Itanium, we will need to buy new furniture and replace our home electrical wiring. The 4-microchip, 100-million-transistor behemoth will require a 600-pound heatsink to dissipate heat from its 1kW of electrical power consumption.
This type of power cannot be carried by regular home wiring, of course, so it is expected that consumers who want an Itanium computer will replace this insufficient equipment to support their new computers.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
The computer created completely novel devices, given only knowledge of physics, to determine the effectiveness of each form. It came up with arrangements of the preprogrammed parts that had not been anticipated.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I don't think the computer was given wheels to play with.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
Yeah, the US and Israel are stealing Canadian military secrets.
Soon, American tanks will be replaced by 50-ton armored horses with laser eyes and the ability to break the sound barrier...
Israel will fill their already-strong air force with autonomous Pegasi, equipped with heavy machine guns and manure bombs...
The Palestinians better not mess with themanymore.
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;
I mentioned this elsewhere, but gotta say -
FireWire is a standard for transmitting digital frame data from a camera to a display/computer/general device. AGP is a bus to carry information from the processor to a graphics card. As has been mentioned, you need several hundred MB/s to run a monitor at 60Hz (that is, transmitting the data from the processor). The article states that AGP8x will give about 2GB/s throughput (533MHz * 4 bytes wide) - how are you going to get anything like that through a *serial* bus? 1394b is already clocked at 400MHz!
Fsck this hard drive! Although it probably won't work...
foo = bar/*myPtr;