If you increase the amount of money the value of each unit of money goes down. A system with unlimited money would mean each unit of money becomes worthless.
the PCIe bus is only fast in one direction (data into the card) PCIe (unlike AGP) was designed to be a general purpose interface and therefore fast in both directions. If getting data out of the GPU fast is a problem that's an issue with the GPUs design.
I don't think many have even scratched the surface of what a PPU (Parallel Processing Unit) can do or how it can improve the quality of just about any application... I think this is going to be Hott. Maybe so but the fact is that MOST PCs use intel CPUs and i'm pretty sure they will do for the forseeable future (even IF AMD can overcome Intel's marketing machine and find a large source of capital it will still take them quite a lot of time to scale up thier production capacity) so something that is AMD specific is likely to see very limited support.
If ((not realistic for server marker) Afaict other than not having dual socket support the i7 980x is pretty much the same design as the xeon 5600 series. So it seems they are selling this design into both the very high end desktop market and the server market.
Where is Intel's budget 6-core design? At least according to wikipedia a cheaper model (though still pretty expensive) is coming Q3 of this year. I don't know how reliable that info is though.
BTW unlike most other extreme editions the i7-980x isn't that bad value when compared to other relatively high end stuff.
With most extreme edition chips you pay a lot more for a small ammount of extra performance. That isn't the case here. Yes you pay $1000 for the i7-980x vs $570 for the i7-960 (the top non-extreme desktop chip) but you get arround half as much again on the performance! Given that you will probablly be spending arround $500 on the rest of such a system (just the motherboard and 6GB of ram alone will run you arround $300) the price/performance of a system using this extreme chip is likely to be similar to the price/performance of a system using the top non-extreme intel chip!
Alternatively compare the cost to a pair of xeon E5530's (which I would expect to have similar agregate performance in highly multithreded processor bound applications) and the price is about the same. Add in the fact you don't have to buy an expensive workstation/server board/case and this chip looks pretty attractive for those who want lots of CPU horsepower and can live with only 24GB of ram.
Either way, the proof is in the pudding. They are not targeting the highly parallel market either by choice ("ignoring that market" scenario) or by mistake ("caught with pants down" scenario) When you have chips far better than the best your competitor can offer (afaict the new AMD 6-cores are comparable to high end intel quad cores in highly paralell tests) AND you have far better marketing than your competitor why sell them cheap?
You can assume they will have a high likelyhood of working fine but just because two browsers use gecko or webkit doesn't mean the version of browsers you have promised to support are using the same version of it. And afaict at least with webkit based browsers some of them have done things like swapping out the JS engine which has the potential to break stuff.
The IE engine is an unusual case as afaict the version of it used depends on the version of IE installed rather than the version of the "browser" the user is using.
Sure it is not good to overload everything up to the eyeballs, but one could say that in C it is easy to make millions of silly macros or functions. Lets first consider why we use functions/methods in the first place. There are several reasons but one of the important ones is to let humans reading the code read the program at a higher level without getting bogged down in the low level details.
With functions the programmer can choose arbitary names for ordinary methods. Decent programmers choose ones that at least somewhat reflect what the function does and try not to use the same name for two different things. So the person reading the code just has to remember what each function does.
With operator overloading (at least the forms of it I have seen) you can't create your own operators, you can only redefine existing ones. So when coders find something that it would be nice to make an operator (because it makes a common operation more concise) but doesn't have a standard operator they tend to reuse (many would say abuse) a standard operator. That in turn means that operator now has two very different meanings in the same code depending on the types involved. This in turn means that every time I want to know whether a statement is performing a shift or writing something to a stream I have to check up on the types of the variables involved.
Besides, GCC makes quite heave use of GMP which is the *IDEAL* candidate for operator overloading. Agreed, that is the sort of place where operator overloading definately should be used.
Bitmaps in pdf files can be any resolution so the device will likely have to scale thle bitmaps even if it isn't scaling the entire page.
Though in my experience most of the major electronics vendors seem to be competent enough to prepare their datasheets with vector graphics anyway. When they are bitmaps they are also often quite high resoloution ones with fairly thick lines so they scale well to a variety of sizes.
In summary i've rarely if ever noticed issues with pdfs scaling badly when viewing them onscreen at various zoom levels or printing them on A4 paper and I don't see why this would be any more of a problem for a dedicated reader device.
meh as long as the DPI is decent the small ammount of scaling down from reading a datasheet prepared in A4 on a letter device or vice-versa shouldn't really be noticable.
I just hope it ends up going in the direction the music industry did in the end where they essentially decided that breaking apples monopoly on legit online sales for ipod users was worth giving up drm.
Death is not actually required for a darwin award. Inability to further reproduce is sufficiant for that part of the critera. This can be achived without death for example by blowing your balls off
For those still able to breed they have a category "at risk survivours" (previously termed "honourable mentions")
mmm, a related problem is that if less water flows then the polloution is less diluted.
So unless you put in place a cleanup program at the same time a diversion or drought is likely to render previously usable water supplies unusable even if it doesn't make any supplies dry up.
But so is splurging, if you splurge big and ride that splurge for many years the cost can tend to average out to be the same or less as the 2nd hand buyer. But does "splurging big" on a computer really extend it's lifetime all that much. Is a $1000 computer really so much better than a $500 computer that it will have double the useful life.
I would get a decent (but not insane) CPU up front though, most other components (graphics, ram, storage) can be easilly upgraded later but CPUs can be a PITA due to frequently changing sockets (AMD have been better about this than intel recently but there is no gaurantee that will continue).
He said "chip X running at frequency Y vs chip X running at frequency Z" (emphasis mine). Generally that is true, if you take the same chip or even two chips from the same family and run at different clocks the higher clock will almost certainly bring better performance.
Sometimes cache comes into play but generally within a familily processors that are clocked faster will also have the same or more cache.
For example, I have an app that runs 50% faster on a 2GHz Phenom II than it does on a 3GHz Pentium D, because the app has a working set size that fits in the Phenom II's L1 cache (which is four times as big as the Pentium D's). Assuming you really mean a pentium D and not a pentium dual-core are you sure the preformance difference is down to the smaller L1 cache and not just due to the pentium 4's general suckiness.
As I understand the statute (17 USC 512(g) [copyright.gov]), fourteen days days after Google receives the counter-notice, any contract that Google might have already had with you requiring Google to offer your product for download is reinstated. The question would then become does the app author have a contract with google requiring them to offer thier product or do they have a contract that can be terminated at any time for any or no reason whatsoever?
The biggest differentiators in their higher end models are number of cores and cache size. I'd add the platform to that.
The low and midrange desktop chips (i3, i5 and i7 8xx) are on the LGA1156 platform. That means dual channel memory, a max memory of 16GB* and only 16 channels of fast PCIe**. What that means is that any addition storage controllers etc end up either stuck behind the PCH or stealing channels from the graphics card.
The high end desktop chips (i7 9xx) are on the x58 single socket LGA1366 platform, that means triple channel memory, a max memory of 24GB* and 36 channels of fast PCIe**.
The xeon 5500/5600 workstation/server chips are also on LGA1366 but it's a dual socket variant. That means six channels of memory (three per CPU), a max memory of 144GB* and 36 or 72 channels (depending on whether the motherboard vendor uses one or two IO hubs) of fast PCIe** (and yes there are boards that use all 72 e.g. http://www.supermicro.co.uk/products/motherboard/QPI/5500/X8DAH_.cfm ).
The higher end xeon chips support even more.
* Max memory figures for desktop chips assume 4GB modules and two modules per channel. I haven't seen a desktop board that claims to support higher configurations than this though it may technically be possible. Max memory for dual socket workstation chips assumes 8GB modules and three modules per channel, more is possible in theory if you can find 16GB modules which supposedly exist but i've never seen for sale.
** PCIe channels that run at the 2.0 speeds and are taken from the processor or IOH. Not the channels that are taken from the ICH/PCH that only run at 1.0 speeds and are potentially bottlenecked by the DMI/ESI connection.
There's no 'citizen number' that can be used as a common key for a join There is the "national insurance number" which will let them tie together anything tax or benifits related, probablly other stuff too (I can't remember if passport applications ask for the national insurance number or not).
At one point in the late '90s, the shares that Acorn held in ARM were actually worth more than the market capitalisation of Acorn. As I understand it this was what finally killed acorn. IIRC their arm holdings were worth more than the rest of the company combined but selling them would have left them with a massive capital gains tax bill whereas breaking up the company wouldn't or something like that (this is my memory of stuff I read at the time, it may be flawed).
Lets assume a machines lifecycle is five years. It used to be more like three but machines lifecycles have got stretched in recent times.
With windows the manufacturer usually installs the latest version unless specifically requested not to. Given that microsofts lifecycle policy promises 7 years of security support overlap between releases and that third party app vendors generally support older releases (it's only in the last few years that i've seen apps dropping support for 9x and 2K) it is quite practical to keep running the same version of windows for the lifecycle of the machine.
With ubuntu LTS if you want to keep security support you will most likely have to upgrade twice over a computers five year lifecycle. Also most software comes via the distro so if you want more recent application software you have to upgrade the whole OS to get it.
I thought some of nvidia's integrated graphics for core 2 processors (which afaict are the best integrated graphics out there for intel processors at the moment) were compararable to nvidia's bottom end cards but i've failed to find any benchmarks confirming or refuting this.
Afaict graphics for PCs come into two main categories*. Discrete graphics chips have their own memory and communicate with the rest of the system over standard busses (PCI AGP or PCIe depending on age). Integrated graphics are integrated with some part of the chipset and don't usually have any dedicated ram.
I think the term originated in the laptop market. Many laptops have discrete graphics soloutions soldered to the motherboard.
*There are graphics setups that don't fall nicely into either category. Servers often have graphics integrated with the out of band management chipset, so they aren't really discrete (they are integrated with something) but they aren't integrated with the main chipset either.
So Intel's next cpu will the same suck video build in? Intel are already building in their sucky (though not quite as sucky as it used to be) video into their dual core i3 and i5 chips (technically it's a multi-die module ATM but from the system integrators POV that doesn't really make any difference). With the next gen I believe they are planning to put it on-die on all their low and mid range chips (maybe the high end too, information on the next gen high end stuff seems very sketchy at the moment)
IMO Integration of graphics onto the CPU was a pretty inevitable result of Intel's decision to put the memory and fast PCIe controllers on the CPU. Putting the graphics elsewhere in such a system would either require dedicated graphics memory or having a very fast link between the CPU and the device containing the graphics with carefully designed prioritisation.
AMD will kill them. BS, the only market segment intels integration of video into the CPU will really impact are those who currently explicitly buy laptops with nvidia chipsets to get a bit better graphics performance without the size and battery life sacrifices of a fully independent graphics soloution. Afaict those are a fairly small proportion of the laptop market.
From a desktop perspective this is no big deal. Graphics cards with performance comparable to the best integrated graphics aren't exactly expensive. Gamers will go from leaving the graphics integrated in the northbridge disabled to leaving the graphics integrated in the CPU disabled.
Plus even if AMD had a technically better solution than Intel afaict they simply don't have the production capacity to kill Intel any time soon. Nor it seems do they have the marketing.
Some of us for whatever reason want a lot of storage on our laptops. Generally on laptops the fastest HDDs you can fit are 7200 RPM and you can only fit one drive. Yeah messing arround with external drives is a possibility but it's one I'd rather avoid for a number of reasons.
That is what I see as the niche for these drives. Those who want a faster drive in their laptop without sacrificing storage capacity and without spending $1400 to get a SSD that is approaching the largest laptop HDDs in size.
Slashdot Mirror
If you increase the amount of money the value of each unit of money goes down. A system with unlimited money would mean each unit of money becomes worthless.
the PCIe bus is only fast in one direction (data into the card)
PCIe (unlike AGP) was designed to be a general purpose interface and therefore fast in both directions. If getting data out of the GPU fast is a problem that's an issue with the GPUs design.
I don't think many have even scratched the surface of what a PPU (Parallel Processing Unit) can do or how it can improve the quality of just about any application ... I think this is going to be Hott.
Maybe so but the fact is that MOST PCs use intel CPUs and i'm pretty sure they will do for the forseeable future (even IF AMD can overcome Intel's marketing machine and find a large source of capital it will still take them quite a lot of time to scale up thier production capacity) so something that is AMD specific is likely to see very limited support.
If ((not realistic for server marker)
Afaict other than not having dual socket support the i7 980x is pretty much the same design as the xeon 5600 series. So it seems they are selling this design into both the very high end desktop market and the server market.
Where is Intel's budget 6-core design?
At least according to wikipedia a cheaper model (though still pretty expensive) is coming Q3 of this year. I don't know how reliable that info is though.
BTW unlike most other extreme editions the i7-980x isn't that bad value when compared to other relatively high end stuff.
With most extreme edition chips you pay a lot more for a small ammount of extra performance. That isn't the case here. Yes you pay $1000 for the i7-980x vs $570 for the i7-960 (the top non-extreme desktop chip) but you get arround half as much again on the performance! Given that you will probablly be spending arround $500 on the rest of such a system (just the motherboard and 6GB of ram alone will run you arround $300) the price/performance of a system using this extreme chip is likely to be similar to the price/performance of a system using the top non-extreme intel chip!
Alternatively compare the cost to a pair of xeon E5530's (which I would expect to have similar agregate performance in highly multithreded processor bound applications) and the price is about the same. Add in the fact you don't have to buy an expensive workstation/server board/case and this chip looks pretty attractive for those who want lots of CPU horsepower and can live with only 24GB of ram.
Either way, the proof is in the pudding. They are not targeting the highly parallel market either by choice ("ignoring that market" scenario) or by mistake ("caught with pants down" scenario)
When you have chips far better than the best your competitor can offer (afaict the new AMD 6-cores are comparable to high end intel quad cores in highly paralell tests) AND you have far better marketing than your competitor why sell them cheap?
Note that site tests things that aren't actally required by the current drafts of the spec. e.g. support for particular audio/video codecs.
You can assume they will have a high likelyhood of working fine but just because two browsers use gecko or webkit doesn't mean the version of browsers you have promised to support are using the same version of it. And afaict at least with webkit based browsers some of them have done things like swapping out the JS engine which has the potential to break stuff.
The IE engine is an unusual case as afaict the version of it used depends on the version of IE installed rather than the version of the "browser" the user is using.
Sure it is not good to overload everything up to the eyeballs, but one could say that in C it is easy to make millions of silly macros or functions.
Lets first consider why we use functions/methods in the first place. There are several reasons but one of the important ones is to let humans reading the code read the program at a higher level without getting bogged down in the low level details.
With functions the programmer can choose arbitary names for ordinary methods. Decent programmers choose ones that at least somewhat reflect what the function does and try not to use the same name for two different things. So the person reading the code just has to remember what each function does.
With operator overloading (at least the forms of it I have seen) you can't create your own operators, you can only redefine existing ones. So when coders find something that it would be nice to make an operator (because it makes a common operation more concise) but doesn't have a standard operator they tend to reuse (many would say abuse) a standard operator. That in turn means that operator now has two very different meanings in the same code depending on the types involved. This in turn means that every time I want to know whether a statement is performing a shift or writing something to a stream I have to check up on the types of the variables involved.
Besides, GCC makes quite heave use of GMP which is the *IDEAL* candidate for operator overloading.
Agreed, that is the sort of place where operator overloading definately should be used.
Bitmaps in pdf files can be any resolution so the device will likely have to scale thle bitmaps even if it isn't scaling the entire page.
Though in my experience most of the major electronics vendors seem to be competent enough to prepare their datasheets with vector graphics anyway. When they are bitmaps they are also often quite high resoloution ones with fairly thick lines so they scale well to a variety of sizes.
In summary i've rarely if ever noticed issues with pdfs scaling badly when viewing them onscreen at various zoom levels or printing them on A4 paper and I don't see why this would be any more of a problem for a dedicated reader device.
A Darwin awards winner will never sue anybody
Death is not strictly a requirement for a darwin award.
meh as long as the DPI is decent the small ammount of scaling down from reading a datasheet prepared in A4 on a letter device or vice-versa shouldn't really be noticable.
I just hope it ends up going in the direction the music industry did in the end where they essentially decided that breaking apples monopoly on legit online sales for ipod users was worth giving up drm.
Maybe the kindle will do the same for ebooks.
Death is not actually required for a darwin award. Inability to further reproduce is sufficiant for that part of the critera. This can be achived without death for example by blowing your balls off
For those still able to breed they have a category "at risk survivours" (previously termed "honourable mentions")
mmm, a related problem is that if less water flows then the polloution is less diluted.
So unless you put in place a cleanup program at the same time a diversion or drought is likely to render previously usable water supplies unusable even if it doesn't make any supplies dry up.
But so is splurging, if you splurge big and ride that splurge for many years the cost can tend to average out to be the same or less as the 2nd hand buyer.
But does "splurging big" on a computer really extend it's lifetime all that much. Is a $1000 computer really so much better than a $500 computer that it will have double the useful life.
I would get a decent (but not insane) CPU up front though, most other components (graphics, ram, storage) can be easilly upgraded later but CPUs can be a PITA due to frequently changing sockets (AMD have been better about this than intel recently but there is no gaurantee that will continue).
He said "chip X running at frequency Y vs chip X running at frequency Z" (emphasis mine). Generally that is true, if you take the same chip or even two chips from the same family and run at different clocks the higher clock will almost certainly bring better performance.
Sometimes cache comes into play but generally within a familily processors that are clocked faster will also have the same or more cache.
For example, I have an app that runs 50% faster on a 2GHz Phenom II than it does on a 3GHz Pentium D, because the app has a working set size that fits in the Phenom II's L1 cache (which is four times as big as the Pentium D's).
Assuming you really mean a pentium D and not a pentium dual-core are you sure the preformance difference is down to the smaller L1 cache and not just due to the pentium 4's general suckiness.
As I understand the statute (17 USC 512(g) [copyright.gov]), fourteen days days after Google receives the counter-notice, any contract that Google might have already had with you requiring Google to offer your product for download is reinstated.
The question would then become does the app author have a contract with google requiring them to offer thier product or do they have a contract that can be terminated at any time for any or no reason whatsoever?
The biggest differentiators in their higher end models are number of cores and cache size.
I'd add the platform to that.
The low and midrange desktop chips (i3, i5 and i7 8xx) are on the LGA1156 platform. That means dual channel memory, a max memory of 16GB* and only 16 channels of fast PCIe**. What that means is that any addition storage controllers etc end up either stuck behind the PCH or stealing channels from the graphics card.
The high end desktop chips (i7 9xx) are on the x58 single socket LGA1366 platform, that means triple channel memory, a max memory of 24GB* and 36 channels of fast PCIe**.
The xeon 5500/5600 workstation/server chips are also on LGA1366 but it's a dual socket variant. That means six channels of memory (three per CPU), a max memory of 144GB* and 36 or 72 channels (depending on whether the motherboard vendor uses one or two IO hubs) of fast PCIe** (and yes there are boards that use all 72 e.g. http://www.supermicro.co.uk/products/motherboard/QPI/5500/X8DAH_.cfm ).
The higher end xeon chips support even more.
* Max memory figures for desktop chips assume 4GB modules and two modules per channel. I haven't seen a desktop board that claims to support higher configurations than this though it may technically be possible. Max memory for dual socket workstation chips assumes 8GB modules and three modules per channel, more is possible in theory if you can find 16GB modules which supposedly exist but i've never seen for sale.
** PCIe channels that run at the 2.0 speeds and are taken from the processor or IOH. Not the channels that are taken from the ICH/PCH that only run at 1.0 speeds and are potentially bottlenecked by the DMI/ESI connection.
There's no 'citizen number' that can be used as a common key for a join
There is the "national insurance number" which will let them tie together anything tax or benifits related, probablly other stuff too (I can't remember if passport applications ask for the national insurance number or not).
At one point in the late '90s, the shares that Acorn held in ARM were actually worth more than the market capitalisation of Acorn.
As I understand it this was what finally killed acorn. IIRC their arm holdings were worth more than the rest of the company combined but selling them would have left them with a massive capital gains tax bill whereas breaking up the company wouldn't or something like that (this is my memory of stuff I read at the time, it may be flawed).
Was that value ever real though or was it just based on unrealistic expectations?
Afaict market cap is only ever a rough approximation of a companies true value (which is virtually impossible to measure)
Lets assume a machines lifecycle is five years. It used to be more like three but machines lifecycles have got stretched in recent times.
With windows the manufacturer usually installs the latest version unless specifically requested not to. Given that microsofts lifecycle policy promises 7 years of security support overlap between releases and that third party app vendors generally support older releases (it's only in the last few years that i've seen apps dropping support for 9x and 2K) it is quite practical to keep running the same version of windows for the lifecycle of the machine.
With ubuntu LTS if you want to keep security support you will most likely have to upgrade twice over a computers five year lifecycle. Also most software comes via the distro so if you want more recent application software you have to upgrade the whole OS to get it.
Actually it's maverick meerkat.
I thought some of nvidia's integrated graphics for core 2 processors (which afaict are the best integrated graphics out there for intel processors at the moment) were compararable to nvidia's bottom end cards but i've failed to find any benchmarks confirming or refuting this.
Afaict graphics for PCs come into two main categories*. Discrete graphics chips have their own memory and communicate with the rest of the system over standard busses (PCI AGP or PCIe depending on age). Integrated graphics are integrated with some part of the chipset and don't usually have any dedicated ram.
I think the term originated in the laptop market. Many laptops have discrete graphics soloutions soldered to the motherboard.
*There are graphics setups that don't fall nicely into either category. Servers often have graphics integrated with the out of band management chipset, so they aren't really discrete (they are integrated with something) but they aren't integrated with the main chipset either.
So Intel's next cpu will the same suck video build in?
Intel are already building in their sucky (though not quite as sucky as it used to be) video into their dual core i3 and i5 chips (technically it's a multi-die module ATM but from the system integrators POV that doesn't really make any difference). With the next gen I believe they are planning to put it on-die on all their low and mid range chips (maybe the high end too, information on the next gen high end stuff seems very sketchy at the moment)
IMO Integration of graphics onto the CPU was a pretty inevitable result of Intel's decision to put the memory and fast PCIe controllers on the CPU. Putting the graphics elsewhere in such a system would either require dedicated graphics memory or having a very fast link between the CPU and the device containing the graphics with carefully designed prioritisation.
AMD will kill them.
BS, the only market segment intels integration of video into the CPU will really impact are those who currently explicitly buy laptops with nvidia chipsets to get a bit better graphics performance without the size and battery life sacrifices of a fully independent graphics soloution. Afaict those are a fairly small proportion of the laptop market.
From a desktop perspective this is no big deal. Graphics cards with performance comparable to the best integrated graphics aren't exactly expensive. Gamers will go from leaving the graphics integrated in the northbridge disabled to leaving the graphics integrated in the CPU disabled.
Plus even if AMD had a technically better solution than Intel afaict they simply don't have the production capacity to kill Intel any time soon. Nor it seems do they have the marketing.
Some of us for whatever reason want a lot of storage on our laptops. Generally on laptops the fastest HDDs you can fit are 7200 RPM and you can only fit one drive. Yeah messing arround with external drives is a possibility but it's one I'd rather avoid for a number of reasons.
That is what I see as the niche for these drives. Those who want a faster drive in their laptop without sacrificing storage capacity and without spending $1400 to get a SSD that is approaching the largest laptop HDDs in size.