I've had terabytes of data stored on hundreds of ext3 filesystems for years. The only data loss we've had has been due to hardware failure or user error.
Erm, he wrote fetchmail, bogofilter, hexdump. The original sed distributed with the GNU utilities was written by ESR. He's got code in Gnome, Gnuplot, libpng, emacs, screen, etc.
Regardless of what you think about him personally, it's hard to dispute that he's an "actual programmer".
They do include a 3-axis accelerometer (ST Electronics MEMS), a high resolution image sensor (PixArt Multi-Object Tracking engine), rumble, speaker, wireless, and add-on interface. Eight out of nine ain't bad. Parent didn't mention that there will be embedded memory as well.
I expect heat won't be an issue since they're specifically designing for low power consumption.
For the most part, I don't forsee too many issues. The chips are derivitives of a proven design manufactured using mature technology at a quality fab plant (IBM's fishkill facility). Most of the other components are off the shelf - Broadcom ethernet and bluetooth chips, MoSys SRAM, etc.
Actually, doing a bit of looking around, appears they've finally got a point to point serial interconnect (CSI) in the works to replace the aging AGTL+ system. Somehow that approach sounds familiar.:-)
The two companies are going to be leapfrogging each other for quite some time to come. Frankly was glad to see Intel swing back with a decent offering after their disappointing offerings of the last few years. If they had released the Core 2 chips before I built my last two machines at home, I may well have gone that route. Intel has a lot of very compelling chipsets these days.
The rumours of yorkfield being eight core have largely fallen by the wayside. See here for more recent speculation. (The link you provided points to rumours from 2005-12, mine are from 2006-09.)
I actually harbor no enmity toward Intel. That statement was meant more tongue-in-cheek than it came off, though.
However, there is a reason to care about how quad core is being achieved. In both the first generation design (where they have two disctint does on an MCM) and the second generation design (the Yorkfield, with two dual core blocks on one die) - the cache. Not only is there likely to be data duplicated in the two L2 cache sets (thus reducing the "effective" amount of cache), any cache traffic has to travel over the FSB.
They made a lot of smart decisions in their design. Svartalf is likely right in one sense though - the actual execution pipelines of the Opteron are theoretically more capable than those of the Core architecture. Of course that doesn't mean anything if the pipeline isn't kept fed.
Something that I might be concerned about at Intel is that some of their optimizations aren't necessarily coupled to their design decision, such as the intelligent pre-fetch in their memory controller. It was more necessary in their design, since they have inherently higher latency to overcome. There's no reason the idea couldn't be adapted to do pre-fetch into L3 cache on the new AMD design, though, to further cut down on latency (especially when they add support for FB-DIMMs).
There is a ton of money to be made in the high end of the market. Fewer boxes, but much higher revenue (and profit margin) per box.
Whether it makes sense... depends. Some applications really do benefit from big boxes - OLTP, ERP, datamining, application servers, directory servers. Commodity virtualization has made consolidation a buzzword again.
The other thing to keep in mind is that when you move into a large environment the initial purchase price is not necessarily the primary cost - rack space costs money, as do power, switch ports, battery backed power, power distribution, cooling.
Some of the pitfalls of horizontal scaling are addressed by blade centers, but then you're looking at a cost of more like $2-3k for a dual processor system, depending on the vendor and architecture. You're also going to pay more for storage (2.5" disks), and be limited in capacity (typically no more than 2 x 74GB). There's a fair number of blades out that that will only take one disk, which precludes mirrored storage without external storage of some sort. The cost of the chassis also needs to be taken into account. Usually only a few grand, but then the switch module is a few grand more, redundant power supplies another grand or two, storage module if you need it another grand or two. If you're not already running high power equipment, getting new power circuits and equipment will cost thousands more.
Then of course there's the soft cost - aggregate support costs are likely lower with multiple boxes, operating system licensing, software licensing (expecially when we're talking about stuff that runs into six figures per machine), system administrators, etc.
Not knocking horizontal scaling, mind you. We should be picking up at least seventy more blades before the end of the year. The demand for big boxes isn't going to disappear anytime soon, though.
(Not to mention that $20k is still firmly in the "volume server" category. They may sound expensive next to a $1-2k x86 machine, but compare it to $200k for an eight-way Sparc or PA-RISC system and it looks like a mighty fine bargain.:)
Hrm, hadn't caught that change in the winds of rumour. Interesting. This looks a twist in the rumour progression. So far as I can tell it went from Yorkfield being mostly a die shrink four core MCM, to an eight core MCM, to a single die with unified cache, and now to the single die with split cache.
The K8L design has been pretty concrete for a while now. AMD makes a point of talking up the benefits of having seperate cores with regards to contention. Of course nothing is really universal. It seems a win for AMD since inter-core communication happens on-die, and so far Intel has that all happen over the FSB.
True, though they're to be lauded for creating an architecture that can take advantage of a ginormous cache as well as it does. They really did learn to "work smarter, not harder" in their current chip. And of course, in the end, the consumer doesn't care how performance is achieved.:)
In the case of Intel's quad core solution, they seem to be achieving higher overall performance, as expected, but at the expense of pushing their thermal envelope back up to 130W even after the die shrink. AMD, on the other hand, has commited to shipping a 68W quad part and that's including the integrated memory controller.
Still, it's not only a few months of bragging rights that Intel is buying - Woodcrest only supports dual socket configs which means four cores at most unless you fall back to Dempsey, which simply isn't price or performance competitive with the Opteron. This at least scales them to eight core configurations. Of course on the AMD side of the house you can pick up an eight-way board, which gets you to sixteen cores today, and will scale up to a whopping thirty-two cores next year. Between that and Intel's lackluster bus technology, they've got serious problems at the high end of the server market.
The specifications list bandwidth for the 1.0, 2.0, and 3.0 specs as 6.4, 11.2, and now 20.8GB/s respectively. AMD is jumping from 1.0 to 3.0. They're actually pushing a bit more than the original spec on the current processors though, since the spec originally only included bus speeds of up to 800MHz and they've got it running at 1000MHz which bumps throughput to 8.0GB/s. So, assuming they max bus speed, it'll be about two and a half times faster.
This is where I think AMD gets themselves a big win. Intel's FSB, even clocked at 1333MHz (actually it's 333MHz QDR, but we'll not quibble) pushes only 10.6GB/s. And that's not accounting for the off-die memory controller. Even with dual buses (like the 5000 series chipsets tout) they only just barely have enough aggregate throughput to handle memory transfers.
That's because Intel is cheating. They don't have a quad-core die, they have two dual core dies shoved onto a multi-chip package. Each die has a shared 6MB cache.
Oh, duh. Forgot they added a power button to the controller. That's even on my list of rationalizations on why I really should buy a Wii as soon as possible.
If this thing is going to be plugged in always, and running always, doesn't it consume enormous amounts of power?
The article states the power consumption is "from one-third to as little as one-fourth that of current hardware". Since the Gamecube drew only about 20W, that comes to 5-7W. That would make a full days consumption about the same as having an XBox 360 on for an hour. A years worth of power, assuming 0.14/kWh comes to a whopping $8.58. And that's assuming it draws full power the entire time. They have a power button which presumably drops consumption further. My guesss is it powers down the video chip, bluetooth, usb ports, and possibly clocks down the CPU.
My Gamecube was cheap ($229 including a game), low power, quiet, plugs directly into the TV, and saves me from having to bother looking to see if I meet the system requirements for a given game. Seems a reasonable value to me.:)
I'd be very surprised is "the grand percentage" of people blow $600 on a video cards. Among the people I know (mostly geeks) I know of only one person who likely has something in that range (wife bought him a ludicrously priced Alienware box for their 10th anniversary), a handful of people who bought in the $200-300 range, a bunch of people who bought in the $100-200 range, and a ton of people who use some form of on-board video because they game on consoles exclusively or not at all.
I fall into the last category personally. I rarely spend as much as $600 on a full machine, much less a video card.
It had sold out, and you couldn't buy it anymore. A quick google search shows several vendors with both the game and the special edition bundle in stock.
Slashdot Mirror
I've had terabytes of data stored on hundreds of ext3 filesystems for years. The only data loss we've had has been due to hardware failure or user error.
Erm, he wrote fetchmail, bogofilter, hexdump. The original sed distributed with the GNU utilities was written by ESR. He's got code in Gnome, Gnuplot, libpng, emacs, screen, etc.
Regardless of what you think about him personally, it's hard to dispute that he's an "actual programmer".
I was referring to the iPod, not audio players in general. :)
However I do see cell phones as convergance devices becoming a strong trend for a few reasons:
* Fewer devices actually is a win in the portable market, where it means less to carry around and few things to charge.
* Takes advantage of existing processing power that otherwise will go unused.
* Allows for useful integreation, like pausing songs automatically when a call comes in.
* The purchase price of a cell phone is generally spread over the course of a contract.
* Wires are lame. Bluetooth headphones which can double as headsets are cool.
They all have Ipods, possibly for no better reason than the white earbuds are a fad.
Thing about fads... they end.
Those are categories of products. Maybe you meant "Will the Ford Model T ever die, will General Electric D-12 [toaster, circa 1909] ever die"? :)
They do include a 3-axis accelerometer (ST Electronics MEMS), a high resolution image sensor (PixArt Multi-Object Tracking engine), rumble, speaker, wireless, and add-on interface. Eight out of nine ain't bad. Parent didn't mention that there will be embedded memory as well.
I expect heat won't be an issue since they're specifically designing for low power consumption.
For the most part, I don't forsee too many issues. The chips are derivitives of a proven design manufactured using mature technology at a quality fab plant (IBM's fishkill facility). Most of the other components are off the shelf - Broadcom ethernet and bluetooth chips, MoSys SRAM, etc.
box office gross != profit... just saying. :)
The IMDB grosses are a bit out of date, though. Current figures from Box Office Mojo peg it at $431M domestic + $493M foreign.
Just means he'd have to make the first episode deal with the crew of serenity making first contact with an alien species. :)
I'd take Howard the Duck over Episodes I-III, myself.
Actually, doing a bit of looking around, appears they've finally got a point to point serial interconnect (CSI) in the works to replace the aging AGTL+ system. Somehow that approach sounds familiar. :-)
The two companies are going to be leapfrogging each other for quite some time to come. Frankly was glad to see Intel swing back with a decent offering after their disappointing offerings of the last few years. If they had released the Core 2 chips before I built my last two machines at home, I may well have gone that route. Intel has a lot of very compelling chipsets these days.
The rumours of yorkfield being eight core have largely fallen by the wayside. See here for more recent speculation. (The link you provided points to rumours from 2005-12, mine are from 2006-09.)
I actually harbor no enmity toward Intel. That statement was meant more tongue-in-cheek than it came off, though.
However, there is a reason to care about how quad core is being achieved. In both the first generation design (where they have two disctint does on an MCM) and the second generation design (the Yorkfield, with two dual core blocks on one die) - the cache. Not only is there likely to be data duplicated in the two L2 cache sets (thus reducing the "effective" amount of cache), any cache traffic has to travel over the FSB.
They made a lot of smart decisions in their design. Svartalf is likely right in one sense though - the actual execution pipelines of the Opteron are theoretically more capable than those of the Core architecture. Of course that doesn't mean anything if the pipeline isn't kept fed.
Something that I might be concerned about at Intel is that some of their optimizations aren't necessarily coupled to their design decision, such as the intelligent pre-fetch in their memory controller. It was more necessary in their design, since they have inherently higher latency to overcome. There's no reason the idea couldn't be adapted to do pre-fetch into L3 cache on the new AMD design, though, to further cut down on latency (especially when they add support for FB-DIMMs).
There is a ton of money to be made in the high end of the market. Fewer boxes, but much higher revenue (and profit margin) per box.
:)
Whether it makes sense... depends. Some applications really do benefit from big boxes - OLTP, ERP, datamining, application servers, directory servers. Commodity virtualization has made consolidation a buzzword again.
The other thing to keep in mind is that when you move into a large environment the initial purchase price is not necessarily the primary cost - rack space costs money, as do power, switch ports, battery backed power, power distribution, cooling.
Some of the pitfalls of horizontal scaling are addressed by blade centers, but then you're looking at a cost of more like $2-3k for a dual processor system, depending on the vendor and architecture. You're also going to pay more for storage (2.5" disks), and be limited in capacity (typically no more than 2 x 74GB). There's a fair number of blades out that that will only take one disk, which precludes mirrored storage without external storage of some sort. The cost of the chassis also needs to be taken into account. Usually only a few grand, but then the switch module is a few grand more, redundant power supplies another grand or two, storage module if you need it another grand or two. If you're not already running high power equipment, getting new power circuits and equipment will cost thousands more.
Then of course there's the soft cost - aggregate support costs are likely lower with multiple boxes, operating system licensing, software licensing (expecially when we're talking about stuff that runs into six figures per machine), system administrators, etc.
Not knocking horizontal scaling, mind you. We should be picking up at least seventy more blades before the end of the year. The demand for big boxes isn't going to disappear anytime soon, though.
(Not to mention that $20k is still firmly in the "volume server" category. They may sound expensive next to a $1-2k x86 machine, but compare it to $200k for an eight-way Sparc or PA-RISC system and it looks like a mighty fine bargain.
Hrm, hadn't caught that change in the winds of rumour. Interesting. This looks a twist in the rumour progression. So far as I can tell it went from Yorkfield being mostly a die shrink four core MCM, to an eight core MCM, to a single die with unified cache, and now to the single die with split cache.
The K8L design has been pretty concrete for a while now. AMD makes a point of talking up the benefits of having seperate cores with regards to contention. Of course nothing is really universal. It seems a win for AMD since inter-core communication happens on-die, and so far Intel has that all happen over the FSB.
True, though they're to be lauded for creating an architecture that can take advantage of a ginormous cache as well as it does. They really did learn to "work smarter, not harder" in their current chip. And of course, in the end, the consumer doesn't care how performance is achieved. :)
In the case of Intel's quad core solution, they seem to be achieving higher overall performance, as expected, but at the expense of pushing their thermal envelope back up to 130W even after the die shrink. AMD, on the other hand, has commited to shipping a 68W quad part and that's including the integrated memory controller.
Still, it's not only a few months of bragging rights that Intel is buying - Woodcrest only supports dual socket configs which means four cores at most unless you fall back to Dempsey, which simply isn't price or performance competitive with the Opteron. This at least scales them to eight core configurations. Of course on the AMD side of the house you can pick up an eight-way board, which gets you to sixteen cores today, and will scale up to a whopping thirty-two cores next year. Between that and Intel's lackluster bus technology, they've got serious problems at the high end of the server market.
The specifications list bandwidth for the 1.0, 2.0, and 3.0 specs as 6.4, 11.2, and now 20.8GB/s respectively. AMD is jumping from 1.0 to 3.0. They're actually pushing a bit more than the original spec on the current processors though, since the spec originally only included bus speeds of up to 800MHz and they've got it running at 1000MHz which bumps throughput to 8.0GB/s. So, assuming they max bus speed, it'll be about two and a half times faster.
This is where I think AMD gets themselves a big win. Intel's FSB, even clocked at 1333MHz (actually it's 333MHz QDR, but we'll not quibble) pushes only 10.6GB/s. And that's not accounting for the off-die memory controller. Even with dual buses (like the 5000 series chipsets tout) they only just barely have enough aggregate throughput to handle memory transfers.
That's because Intel is cheating. They don't have a quad-core die, they have two dual core dies shoved onto a multi-chip package. Each die has a shared 6MB cache.
Oh, duh. Forgot they added a power button to the controller. That's even on my list of rationalizations on why I really should buy a Wii as soon as possible.
If this thing is going to be plugged in always, and running always, doesn't it consume enormous amounts of power?
The article states the power consumption is "from one-third to as little as one-fourth that of current hardware". Since the Gamecube drew only about 20W, that comes to 5-7W. That would make a full days consumption about the same as having an XBox 360 on for an hour. A years worth of power, assuming 0.14/kWh comes to a whopping $8.58. And that's assuming it draws full power the entire time. They have a power button which presumably drops consumption further. My guesss is it powers down the video chip, bluetooth, usb ports, and possibly clocks down the CPU.
Multiple reports released since 9/11 have clearly demonstrated we had more than enough intelligence to know what was going to happen
s/what/something/;s/$/ at some unknown time, in some unknown location/
My Gamecube was cheap ($229 including a game), low power, quiet, plugs directly into the TV, and saves me from having to bother looking to see if I meet the system requirements for a given game. Seems a reasonable value to me. :)
I'd be very surprised is "the grand percentage" of people blow $600 on a video cards. Among the people I know (mostly geeks) I know of only one person who likely has something in that range (wife bought him a ludicrously priced Alienware box for their 10th anniversary), a handful of people who bought in the $200-300 range, a bunch of people who bought in the $100-200 range, and a ton of people who use some form of on-board video because they game on consoles exclusively or not at all.
I fall into the last category personally. I rarely spend as much as $600 on a full machine, much less a video card.
It had sold out, and you couldn't buy it anymore. A quick google search shows several vendors with both the game and the special edition bundle in stock.