Yep. I believe this one one of the original arguments as to why Wikipedia would ultimately fail. Go dig up some of those and see what responses you get... Many arguments for it were optimistic/utopian... for such a good project the goodness of mankind will prevail! Many of the arguments against it were realistic/pessimistic... human nature will turn it into a cesspool.
I wouldn't say that entirely. I agree, though, that the ship more or less determines your role as you fly it (particularly in gangs). However, you can specialize in EW or just offensive damage or in logistics, for example. Of course, specializing in EW means you'll tend to fly the ships of each size class that fill that role (Heron, BlackBird, Scorpion, for example). Once you learn the skills, however, you can apply them to any ship... the DomiNos for example... while not spec'd for it, is a mean EW + EnergyVamp + Drone (which it is spec'd for) PvP ship, even though it's bonuses are for drones and hybrid guns (Many DomiNos setups don't even mount guns, thus ignoring that bonus).
You disable virtual memory? That can't be good. I'm assuming you mean the swap file, which is not the same thing as virtual memory. In any case, conventional/modern recommendations is to leave it alone and leave it set at letting Windows manage the size. There are oodles of postings as to why (search for posts by DriverGuru, particularly on ArsTechnica) you should do this. It's a question that comes up frequently.
Of course, if you look in the short term and not the long term, the game is horribly unbalanced. Someone who doesn't have T2 guns and armor will never win 1v1 vs one who does
It's a good thing to have friends and not to run around by yourself in lowsec... especially not being an agressor:) In any case, you always have to pick your fights. If you're low skills and in a Rifter, you probably don't want to run up by yourself and attempt to fight that HAC...;)
However... a new person can still have fun and do well even with fairly low skills, considering. You can specialize in frigates, say, and only have 5 million in skill points and still be about as good as someone with twice your skill points (or more) in frigates. The main difference is that while you may only be good in frigates, the other person would be more versatile and be not only good in frigates but good in cruisers as well. But many cruiser skills have no bearing on frigate play... cruiser weapons, for example, aren't fit on frigates (neither are battleship weapons) so all those skill points the more experienced person has in cruiser and battleship weapons mean absolutely nothing when he's flying a frigate.
You don't need to pick between skills because you eventually get them all.
I guess... but at last count, optimizing for speed in training, it'd take you over 20 years to get all the skills to maximum levels. So, for practical game play, you have to pick a path you want to go.
This actually works OK because you start off in frigates and advancement to other combat ships generally builds on that skill (you have to have Frigates trained to level 4 before you can fly a cruiser, and you have to have cruiser to level 4 before you can train to battleship, for example).
Then you have the tech2 tree with is kind of the same thing but more powerful... Assault Ships are frigates on steroids and require frigate to skill 5 (max) along with certain other support skills (mechanic to 5, engineering to 5). Heavy Assault ships are cruisers on steroids (have to have cruiser to 5, assault ships to 4, and all the dependences of those, plus a few more support skills).
That's a good point. I imagine that most of Philadelphia has broadband of some kind. I live in a city of about 200,000 or so and I'd say the majority of it now has access to broadband of some kind (DSL/Cable). The major issue there is whether or not to use existing infrastructure for broadband. Cable television was made available in many cities in the USA many years ago (30 years, give or take) and put in using materials available at the time without even a thought for 'futuristic' computer communication to the home.
So, the cables put in aren't the greatest (think: 10Base2) for computer communication. Broadband providers are now putting down new cable in lots of areas. In many places, fiber is being put down. A friend of mine in one large city has the option of getting fiber to his home, even, and getting 100Mb. So, while all that infrastructure is getting replaced (all over the country), the existing 1970s era infrastructure runs typically 1.5Mb to 6Mb broadband... but only in the cities because that's where the infrastructure actually is in place at all. There are many people who live outside of cities who still have to use dial-up because cable infrastructure was never in place. In those areas (still a huge portion of the area of the USA), television is still received over aerial antenna, for example.
It's also a matter of pure area. The USA is just over 22x the size of Sweden, for example, in both land area and population. In fact, the USA is larger than all of Europe put together. Europe, as a whole, has a much higher population density than the USA. I imagine that Sweden's population is highly concentrated around the southern portion of the country with it being very sparsely populated to the north, and then typically in isolated pockets of towns. How many people who live in Sweden do not live in a large city? Wikipedia link Coincidentally, large cities are much more apt to have the infrastructure and market to install higher speed networks. The 15th largest city in Sweden has about 100,000 people living in it. The 10th largest city in the USA is larger than the 1st largest in Sweden. I'm not throwing rocks, just pointing out facts and how/why it may be easier to give most of the people in Sweden better Internet access than most of the people in the USA.
Me too. TA has a great interface but even it could be a little better... for example, I'd like that no unit in a group move faster than the slowest one, that way you could make combined unit forces all stay together as they travel across the map for an attack, instead of getting strung out and chewed up piecemeal.
True, but the fact remains that you can have 4x4 now as long as you go dual-dual-core Opteron with dual-SLI. It's just really expensive (and still will be a year from now). 4x4 offers nothing except the use of unbuffered DIMMs. XBox360 has already been out since last Christmas and the PS3, well... is supposed to come out by Christmas but it's still a paper release for now, not to mention that the PS3 will be a bit different from your normal multithreaded application (in concept it's similar but in practice it's going to be a bit more complicated and different). Multiplatform games have been around for a long time so I assume you mean "multithreaded games" and those are already just starting to come out and hopefully there will be quite a few more of them by the time 4x4 comes on the scene a year from now. On a tangent, it'll be easier to port an XBox360 game to 4x4 than a PS3 game. Personally, I think 4x4 will be a flop and a dead-end and those who get one of those systems will be mocked for having more money than sense, but that's just my opinion.
Not at all. I love my X2 3800+ even for games that are single threaded. One core can handle OS overhead issues while the other plays the game, at the minimum. However, the fact remains that few games are multithreaded so even two cores isn't a huge benefit. If games can't make use of two cores other than passively (interrupts/OS/etc can use one while the game burns on the other) then four cores, regardless of who makes it, is kind of silly, wouldn't you think? And AMD is pushing 4x4 as their super-duper high-end solution for gaming?
Don't forget also that there are plenty of people who are getting 3.6GHz and higher on Core2Duo on standard air coolers (some hit 4GHz) and all this is on 65nm processes. Intel has clearly and obviously launched these at competitive speeds (just enough to trump AMD parts) while still having lots of headroom. As soon as AMD launches their 3.0GHz or 3.2GHz parts, Intel will release 3.33GHz Core2Duos... just enough to stay a little ahead of AMD. Intel has enough headroom to do this for the next year with current cores on their current process (65nm). The move to 45nm will just give them a little more headroom to continue the game.
Don't forget that Core 2 Duo's SSE unit can retire an instruction per clock now (as opposed to every other clock as previous Intel chips (and AMD ones)). I'm not sure if gcc 'knows' about that, though, so it may not schedule SSE instructions back-to-back as it should (resulting in lower SSE/fpu performance than it could achieve).
There have been exceptions... P4 with RDRAM? expensive (more expensive than other solutions) and dead end. Maybe you're too young to remember the DRAM shortage from a decade or so ago where DRAM of all kinds got expensive for a little while (until the resultant production ramp up caused the glut). That was not a good time to buy either.
But generally, yes, over longer periods of time, prices on mainstream kit drops.
You realize that a 4x4 machine with cost several thousands of dollars (USD) and that there are only a couple/three games that are multithreaded, right? Not only that but from everything that has been released so far, the 4x4 looks like a panic stricken kneejerk reaction to Core 2 Duo and seems to be nothing more than a (potentially cheaper) dual socket, dual core Opteron that uses unregistered memory on a motherboard with two PCIe x16 slots on it... these already exist except for the 2xx(x) Opterons that use unregistered memory. In other words, 4x4 is just a marketing brand for something that is very expensive to entice rich fanbois to buy their kit.
My wife wanted a Palm for her birthday a couple weeks ago and she wanted a PalmTX, so I got her one.
Why not to get Palm: -somewhat limited specs compared to same-priced HP/Dell handhelds (but specs rarely matter). No cheap wireless option.
The PalmTX supports 802.11b.
-most of the newer Palms cannot run Linux (yet). You are stuck with PalmOS
We didn't care what it ran. It's not like either of us will be trying to hack the kernel on it. She needed something with the functionality the PalmTX had, not something to hack on and we have no OS/device religion.
-Slow Bluetooth/USB transfers. Sending large files takes forever.
It seems pretty fast and there's 802.11b transfers as well (probably slower than the USB, but I haven't actually tested it).
-Backlight is not bright enough in full sunlight.
The TX seems pretty bright, but I admit I haven't seen it outside.
What if other projects adopt "no military" clauses like we've seen lately? This certainly has to be in the list of risks that the DoD will face.
Anyway, other than toolkits and general systems (a Linux based workstation to compile code on, use OpenOffice to write documents, and such) there's not going to be a lot of OSS that will be reusable for the developers since they will be writing software for missile guidance systems and interfacing to hardware not generally available to the public. Some GUI toolkits, maybe, and GCC, of course.
Plus, how will GPL's clauses about not having to release code for things you do on-site relate to the contractor/subcontractor relationships that are present in DoD projects and if parts are sold to other countries (like selling an F-16 to Israel, for example)?
I'm obviously not talking much about office productivity and listening to mp3s and stuff because I'm pretty sure that's not what the DoD is talking about here.
The increased number and width of the registers on x86-64 is transparent unless you do assembly programming or assembly level programming such as that for making compilers. For the rest of us, it doesn't really mean anything.
If you really like ISAs and register architectures, you'll most likely like CPUs other than x86, anyway (any of the load/store ones and/or those that are more orthagonal like the 68k and/or VAX, for example).
I was in Hawaii and stopped at a Dairy Queen for a Blizzard. I decided to get a "Hawaiian" Blizzard, since I was in Hawaii and all... They said they couldn't make one for me... they were out of coconut and pineapple... There was a coconut tree outside and a pineapple plantation about a mile down the road.
Yeah... somehow the Linux on the card reaches out across the Internet and makes my wire signals travel faster and makes sure my packets are prioritized across all routers. These guys are betting on advertising that their card has Linux on it being a selling point where the likes of/. readers will buy it simply because it has Linux on it (so they can brag to their similarly clueless friends). Of course, you could probably take the card and get the source and do other neat things with it if you had the time and the interest to do such things.
I keep seeing this and have to ask.... why does having the source to the Linux kernel matter for robotics? It seems that many people post saying this is a big deal. Unless there are kernel modifications required to support something, the closest to dorking around in the kernel that any of the groups would have to do is write a device driver. There is ample documentation on how to do this in both Linux and Windows and neither is exactly rocket science from the API point of view, particularly if your device plugs in via a 'standard' interface like a parallel/serial port, USB, or something. Additionally, as some have already posted, many of the devices that were used already have Windows drivers where on Linux, your team would likely have to spend a fair amount of time writing the driver, time spent not working on other things like writing the AI or whatever. Alternatively, suppose the source for the Linux driver were available, then the situation is the same for both. Unless the drivers don't work right, you won't be dorking around with the source for the drivers, either. You'll be spending your time elsewhere unless there is something wrong with the driver.
The source isn't necessarily the issue. One of the main reasons why Linux is used in robotics is because it's a bit like Un*x (and other embedded OSs that have tried to mimic the Un*x design) and the OS and the tools are well understood by the people who are developing controllers and devices so it's an easy tool to use to create their controllers. Also, Linux has a few 'nice' features for robotics such as there are process schedulers that are more responsive for those types of things. I'm guessing that an embedded systems version of Windows would have a similar scheduler (I've never worked with an embedded Windows but I've worked with a number of other embedded OSs including an embedded Linux and 'regular' Linux on an embedded system).
Depends on what you mean by 4P. Is that four cores or four CPUs (two dual-cores counts). There have been benchmarks already performed in both Windows and Linux (both 32-bit and 64-bit versions) which show quad-core Woodcrest systems (two dual-cores) perform very well... better than two dual-core Opterons on average, even with the shared FSB design of Woodcrest. However, if you're meaning that four socket boards are the sweet spot (four dual-core Opterons, for example, 8 cores total), then I'd agree that Intel is short in that department right now. I would think four-core systems, right now, are the sweet spot though (two dual-core CPUs).
Are you going to sacrifice the slots in the raid, though, for yet-another support class? It was already hard enough to get into a raid as a Shaman as there were only so many slots for support classes in it. Now there will be more competition because Paladins are unarguably the superior of the two classes on many raids. I had a 60 Shaman and it was hard to get into raids because there are only so many support slots available and you want to take better geared Shaman if one is available.
Yeah, and even with the upcoming price cuts, that's a bloody fortune to pay to play. A system like that is bought only by those with much more money than sense (and those who feel they have to make up for something else by having a huge system). The number of people who will buy into that fanboism market is tiny compared to what even the $316 Core 2 Duo with a reasonable graphics card (X1800 or single 7800GT type) or even a simple AMD X2 3800+ with a similar graphics card is. Even then, those are smaller than the group of people who will buy $200 graphics cards like the X1800GTO and the 7600GS range.
It's called "flailing about just in order to be seen on a day when the tide turns".
Slashdot Mirror
Yep. I believe this one one of the original arguments as to why Wikipedia would ultimately fail. Go dig up some of those and see what responses you get... Many arguments for it were optimistic/utopian... for such a good project the goodness of mankind will prevail! Many of the arguments against it were realistic/pessimistic... human nature will turn it into a cesspool.
I wouldn't say that entirely. I agree, though, that the ship more or less determines your role as you fly it (particularly in gangs). However, you can specialize in EW or just offensive damage or in logistics, for example. Of course, specializing in EW means you'll tend to fly the ships of each size class that fill that role (Heron, BlackBird, Scorpion, for example). Once you learn the skills, however, you can apply them to any ship... the DomiNos for example... while not spec'd for it, is a mean EW + EnergyVamp + Drone (which it is spec'd for) PvP ship, even though it's bonuses are for drones and hybrid guns (Many DomiNos setups don't even mount guns, thus ignoring that bonus).
You disable virtual memory? That can't be good. I'm assuming you mean the swap file, which is not the same thing as virtual memory. In any case, conventional/modern recommendations is to leave it alone and leave it set at letting Windows manage the size. There are oodles of postings as to why (search for posts by DriverGuru, particularly on ArsTechnica) you should do this. It's a question that comes up frequently.
Of course, if you look in the short term and not the long term, the game is horribly unbalanced. Someone who doesn't have T2 guns and armor will never win 1v1 vs one who does
:) In any case, you always have to pick your fights. If you're low skills and in a Rifter, you probably don't want to run up by yourself and attempt to fight that HAC... ;)
It's a good thing to have friends and not to run around by yourself in lowsec... especially not being an agressor
However... a new person can still have fun and do well even with fairly low skills, considering. You can specialize in frigates, say, and only have 5 million in skill points and still be about as good as someone with twice your skill points (or more) in frigates. The main difference is that while you may only be good in frigates, the other person would be more versatile and be not only good in frigates but good in cruisers as well. But many cruiser skills have no bearing on frigate play... cruiser weapons, for example, aren't fit on frigates (neither are battleship weapons) so all those skill points the more experienced person has in cruiser and battleship weapons mean absolutely nothing when he's flying a frigate.
You don't need to pick between skills because you eventually get them all.
I guess... but at last count, optimizing for speed in training, it'd take you over 20 years to get all the skills to maximum levels. So, for practical game play, you have to pick a path you want to go.
This actually works OK because you start off in frigates and advancement to other combat ships generally builds on that skill (you have to have Frigates trained to level 4 before you can fly a cruiser, and you have to have cruiser to level 4 before you can train to battleship, for example).
Then you have the tech2 tree with is kind of the same thing but more powerful... Assault Ships are frigates on steroids and require frigate to skill 5 (max) along with certain other support skills (mechanic to 5, engineering to 5). Heavy Assault ships are cruisers on steroids (have to have cruiser to 5, assault ships to 4, and all the dependences of those, plus a few more support skills).
That's a good point. I imagine that most of Philadelphia has broadband of some kind. I live in a city of about 200,000 or so and I'd say the majority of it now has access to broadband of some kind (DSL/Cable). The major issue there is whether or not to use existing infrastructure for broadband. Cable television was made available in many cities in the USA many years ago (30 years, give or take) and put in using materials available at the time without even a thought for 'futuristic' computer communication to the home.
So, the cables put in aren't the greatest (think: 10Base2) for computer communication. Broadband providers are now putting down new cable in lots of areas. In many places, fiber is being put down. A friend of mine in one large city has the option of getting fiber to his home, even, and getting 100Mb. So, while all that infrastructure is getting replaced (all over the country), the existing 1970s era infrastructure runs typically 1.5Mb to 6Mb broadband... but only in the cities because that's where the infrastructure actually is in place at all. There are many people who live outside of cities who still have to use dial-up because cable infrastructure was never in place. In those areas (still a huge portion of the area of the USA), television is still received over aerial antenna, for example.
It's also a matter of pure area. The USA is just over 22x the size of Sweden, for example, in both land area and population. In fact, the USA is larger than all of Europe put together. Europe, as a whole, has a much higher population density than the USA. I imagine that Sweden's population is highly concentrated around the southern portion of the country with it being very sparsely populated to the north, and then typically in isolated pockets of towns. How many people who live in Sweden do not live in a large city? Wikipedia link Coincidentally, large cities are much more apt to have the infrastructure and market to install higher speed networks. The 15th largest city in Sweden has about 100,000 people living in it. The 10th largest city in the USA is larger than the 1st largest in Sweden. I'm not throwing rocks, just pointing out facts and how/why it may be easier to give most of the people in Sweden better Internet access than most of the people in the USA.
Me too. TA has a great interface but even it could be a little better... for example, I'd like that no unit in a group move faster than the slowest one, that way you could make combined unit forces all stay together as they travel across the map for an attack, instead of getting strung out and chewed up piecemeal.
True, but the fact remains that you can have 4x4 now as long as you go dual-dual-core Opteron with dual-SLI. It's just really expensive (and still will be a year from now). 4x4 offers nothing except the use of unbuffered DIMMs. XBox360 has already been out since last Christmas and the PS3, well... is supposed to come out by Christmas but it's still a paper release for now, not to mention that the PS3 will be a bit different from your normal multithreaded application (in concept it's similar but in practice it's going to be a bit more complicated and different). Multiplatform games have been around for a long time so I assume you mean "multithreaded games" and those are already just starting to come out and hopefully there will be quite a few more of them by the time 4x4 comes on the scene a year from now. On a tangent, it'll be easier to port an XBox360 game to 4x4 than a PS3 game. Personally, I think 4x4 will be a flop and a dead-end and those who get one of those systems will be mocked for having more money than sense, but that's just my opinion.
Not at all. I love my X2 3800+ even for games that are single threaded. One core can handle OS overhead issues while the other plays the game, at the minimum. However, the fact remains that few games are multithreaded so even two cores isn't a huge benefit. If games can't make use of two cores other than passively (interrupts/OS/etc can use one while the game burns on the other) then four cores, regardless of who makes it, is kind of silly, wouldn't you think? And AMD is pushing 4x4 as their super-duper high-end solution for gaming?
Don't forget also that there are plenty of people who are getting 3.6GHz and higher on Core2Duo on standard air coolers (some hit 4GHz) and all this is on 65nm processes. Intel has clearly and obviously launched these at competitive speeds (just enough to trump AMD parts) while still having lots of headroom. As soon as AMD launches their 3.0GHz or 3.2GHz parts, Intel will release 3.33GHz Core2Duos... just enough to stay a little ahead of AMD. Intel has enough headroom to do this for the next year with current cores on their current process (65nm). The move to 45nm will just give them a little more headroom to continue the game.
Don't forget that Core 2 Duo's SSE unit can retire an instruction per clock now (as opposed to every other clock as previous Intel chips (and AMD ones)). I'm not sure if gcc 'knows' about that, though, so it may not schedule SSE instructions back-to-back as it should (resulting in lower SSE/fpu performance than it could achieve).
There have been exceptions... P4 with RDRAM? expensive (more expensive than other solutions) and dead end. Maybe you're too young to remember the DRAM shortage from a decade or so ago where DRAM of all kinds got expensive for a little while (until the resultant production ramp up caused the glut). That was not a good time to buy either.
But generally, yes, over longer periods of time, prices on mainstream kit drops.
You realize that a 4x4 machine with cost several thousands of dollars (USD) and that there are only a couple/three games that are multithreaded, right? Not only that but from everything that has been released so far, the 4x4 looks like a panic stricken kneejerk reaction to Core 2 Duo and seems to be nothing more than a (potentially cheaper) dual socket, dual core Opteron that uses unregistered memory on a motherboard with two PCIe x16 slots on it... these already exist except for the 2xx(x) Opterons that use unregistered memory. In other words, 4x4 is just a marketing brand for something that is very expensive to entice rich fanbois to buy their kit.
My wife wanted a Palm for her birthday a couple weeks ago and she wanted a PalmTX, so I got her one.
Why not to get Palm:
-somewhat limited specs compared to same-priced HP/Dell handhelds (but specs rarely matter). No cheap wireless option.
The PalmTX supports 802.11b.
-most of the newer Palms cannot run Linux (yet). You are stuck with PalmOS
We didn't care what it ran. It's not like either of us will be trying to hack the kernel on it. She needed something with the functionality the PalmTX had, not something to hack on and we have no OS/device religion.
-Slow Bluetooth/USB transfers. Sending large files takes forever.
It seems pretty fast and there's 802.11b transfers as well (probably slower than the USB, but I haven't actually tested it).
-Backlight is not bright enough in full sunlight.
The TX seems pretty bright, but I admit I haven't seen it outside.
What if other projects adopt "no military" clauses like we've seen lately? This certainly has to be in the list of risks that the DoD will face.
Anyway, other than toolkits and general systems (a Linux based workstation to compile code on, use OpenOffice to write documents, and such) there's not going to be a lot of OSS that will be reusable for the developers since they will be writing software for missile guidance systems and interfacing to hardware not generally available to the public. Some GUI toolkits, maybe, and GCC, of course.
Plus, how will GPL's clauses about not having to release code for things you do on-site relate to the contractor/subcontractor relationships that are present in DoD projects and if parts are sold to other countries (like selling an F-16 to Israel, for example)?
I'm obviously not talking much about office productivity and listening to mp3s and stuff because I'm pretty sure that's not what the DoD is talking about here.
The increased number and width of the registers on x86-64 is transparent unless you do assembly programming or assembly level programming such as that for making compilers. For the rest of us, it doesn't really mean anything.
If you really like ISAs and register architectures, you'll most likely like CPUs other than x86, anyway (any of the load/store ones and/or those that are more orthagonal like the 68k and/or VAX, for example).
Neither would OSX or it's lineage...
I was in Hawaii and stopped at a Dairy Queen for a Blizzard. I decided to get a "Hawaiian" Blizzard, since I was in Hawaii and all... They said they couldn't make one for me... they were out of coconut and pineapple... There was a coconut tree outside and a pineapple plantation about a mile down the road.
Yeah... somehow the Linux on the card reaches out across the Internet and makes my wire signals travel faster and makes sure my packets are prioritized across all routers. These guys are betting on advertising that their card has Linux on it being a selling point where the likes of /. readers will buy it simply because it has Linux on it (so they can brag to their similarly clueless friends). Of course, you could probably take the card and get the source and do other neat things with it if you had the time and the interest to do such things.
I keep seeing this and have to ask.... why does having the source to the Linux kernel matter for robotics? It seems that many people post saying this is a big deal. Unless there are kernel modifications required to support something, the closest to dorking around in the kernel that any of the groups would have to do is write a device driver. There is ample documentation on how to do this in both Linux and Windows and neither is exactly rocket science from the API point of view, particularly if your device plugs in via a 'standard' interface like a parallel/serial port, USB, or something. Additionally, as some have already posted, many of the devices that were used already have Windows drivers where on Linux, your team would likely have to spend a fair amount of time writing the driver, time spent not working on other things like writing the AI or whatever. Alternatively, suppose the source for the Linux driver were available, then the situation is the same for both. Unless the drivers don't work right, you won't be dorking around with the source for the drivers, either. You'll be spending your time elsewhere unless there is something wrong with the driver.
The source isn't necessarily the issue. One of the main reasons why Linux is used in robotics is because it's a bit like Un*x (and other embedded OSs that have tried to mimic the Un*x design) and the OS and the tools are well understood by the people who are developing controllers and devices so it's an easy tool to use to create their controllers. Also, Linux has a few 'nice' features for robotics such as there are process schedulers that are more responsive for those types of things. I'm guessing that an embedded systems version of Windows would have a similar scheduler (I've never worked with an embedded Windows but I've worked with a number of other embedded OSs including an embedded Linux and 'regular' Linux on an embedded system).
Depends on what you mean by 4P. Is that four cores or four CPUs (two dual-cores counts). There have been benchmarks already performed in both Windows and Linux (both 32-bit and 64-bit versions) which show quad-core Woodcrest systems (two dual-cores) perform very well... better than two dual-core Opterons on average, even with the shared FSB design of Woodcrest. However, if you're meaning that four socket boards are the sweet spot (four dual-core Opterons, for example, 8 cores total), then I'd agree that Intel is short in that department right now. I would think four-core systems, right now, are the sweet spot though (two dual-core CPUs).
Or, it could be SteakOut.
Are you going to sacrifice the slots in the raid, though, for yet-another support class? It was already hard enough to get into a raid as a Shaman as there were only so many slots for support classes in it. Now there will be more competition because Paladins are unarguably the superior of the two classes on many raids. I had a 60 Shaman and it was hard to get into raids because there are only so many support slots available and you want to take better geared Shaman if one is available.
Yeah, and even with the upcoming price cuts, that's a bloody fortune to pay to play. A system like that is bought only by those with much more money than sense (and those who feel they have to make up for something else by having a huge system). The number of people who will buy into that fanboism market is tiny compared to what even the $316 Core 2 Duo with a reasonable graphics card (X1800 or single 7800GT type) or even a simple AMD X2 3800+ with a similar graphics card is. Even then, those are smaller than the group of people who will buy $200 graphics cards like the X1800GTO and the 7600GS range.
It's called "flailing about just in order to be seen on a day when the tide turns".