Right now, it's the only way to get more than 16gb of memory on a board, because regular memory modules are quite demanding on the motherboard's controller, while FB-DIMMs provide their own "smart" interface. Not entirely true, there are definately boards that can take 32GB of ordinary ram, I dunno if any of them are intel based though.
The real problem though is that if what you are after is a normal desktop apples range sucks.
For the midrange buyer on the one hand you have the imac which has no expansion room, only two cores and locks you into apples choice of monitor (which also pushes up the effective cost if you already have an acceptable monitor). On the other hand you have the mac pro which is built out of server parts with a pricetag to match.
When it comes to low end desktops things look even worse for apple. During a previous argument over whether apple was expensive I specced out a vostro 420 and a mac mini as close as possible, the vostro was both considerablly cheaper and considerablly better specced. http://yro.slashdot.org/comments.pl?sid=1245805&cid=28115689
In fact on debian at least they have made it more hell recently. They only generate a "minimal" config which is fine if it works but if you are say using a KVM switch or an old monitor you basically end up writing an old style config from scratch without any tools to help you.
They don't really have much competition. OS-X is tied to apples weird hardware (which isn't expensive for what it is but forces you into buying machines that are often a poor fit), linux is viewed (rightly or wrongly) as too geeky and both OS-X and linux aren't compatible with a lot of software that buisnesses are tied into.
Besides few people will be paying theese prices, like many companies MS make the full price high so they can discount more.
The trouble is that requires a lot of very expensive dismantling and reassembly. Also with the winding down of the shuttle program the facility that would normally have done such a procedure is no longer operational.
Your math is wrong, you are forgetting that you are dealing with sqare units (the common notation for square units is rather confusing which doesn't help). There are 10000 square centimeters in a square meter and 1000000 square meters in a square kilometer. You also seem to be in a bit of a mess with the units of various figures.
The trouble is that most applications are designed for public computer labs not private homes. If you tell users where they can and can't use thier laptops or that they can't log in from machines in open plan offices/university clusters/whatever then (unless you have a huge ammount of organisational or legal clout) you will most likely be ignored.
Given that when making an application or website for general use that requires login your only reasonable option is to err on the side of caution.
I would contend that well written java that has been jited is probablly a bit slower than well written code in a native compiled language but there isn't that much in it and it's hard to compare because the "best" coding methods in the languages are not the same.
However it does have high memory use because of the fact that jited code can't be shared between instances, the large bloat of the standard library, the limited availibility of efficiant data structures (all object types are object references) and the fact that the java garbage collectors don't realease memory back to the OS. This means that if a java app gets swapped out it takes a lot longer to swap back in. It also takes a while to start and get up to full performance, especially if it's the first time a java app has been run this OS session.
In summary on the desktop (where startup and swapin times matter) java is slow, on the server not so much.
now try getting say a wi-fi signal with the remote end having only a normal wi-fi antenna and transmit power at a hundred miles (bottom end of LEO). Maybe you could with a sufficiantly big dish but launching and pointing that dish from a sattelite is going to get expensive and the coverage area is likely to be pretty small (remember antenna gain trades off with antenna directionality). Not to mention that LEO satellites move pretty quick with relation to earth so you would need a load of them all with motorised dishes.
The second would be a very invasive regulation of private individuals, that as far as I know is not practised in any countries with socialized medicine either. I was under the impression that canada didn't allow public hospitals though canadians can go out of the country for treatment (and afaict many richer canadians do)
Nobody really believes that the wireline maintenance is going to be covered by the revenue generated by selling naked DSL service. As long as people need fixed line (whether telco twisted pair, cableco coax busses or fiber) last mile communication services somone will be there to charge them for it. Most likely in a lot of areas it will be the same providers that do it now since those providers have been taking steps to actively crush any attempts at starting a new independent service.
If naked DSL is currently underpriced and lots of people take it then some readjustment of prices may be needed but I can't see that killing the telcos.
Truly effective at stopping shit dead in it's tracks. It's useful to point out, that it is not just web browsers either. It's EVERYTHING. Unless your program is written to somehow work without using the communications stack in Windows, it will have to go through the hosts file. If you want to bypass the hosts file you can quite easilly directly connect to an IP or ship your own dns lookup code (dns isn't that hard to implement) either using hardcoded dns servers or getting the machines dns server addresses through iphlpapi.
I don't think MS has any trouble with you installing XP if you have a valid full license for XP or a volume license with suffiicant downgrade rights for the machine.
All MS is saying is that OEM copies of 7 buisness sold after a certain date won't come with downgrade rights to XP so if you want to run XP you will have to purchase it some other way (and if you want to get new licenses that basically means software assurance).
There are a number of problems with your analysis, not least that the Pentium III is faster clock-for-clock than the Pentium IV at almost all workloads; It is but IIRC intel throught at the time that they would be able to push the P4 to crazy clock speeds which would more than make up for the lower performance per clock.
Unfortunately they didn't get the clock speeds they had hoped for and the high clock speeds they did get required very high power consuption.
note that testing migration is handled based on source packages. So if any binary package generated from the openoffice.org source package depends on mono it will be considered a dependency for testing migration purposes.
The default is to install the base system (the really important stuff) standard system task (a load of old unixy stuff). Depending on the language you select some language specific stuff will be installed and if you are on a laptop some power management related stuff.
TFA is titled "default desktop install" which probablly reffers to the above plus selecting teh "desktop environment" task when the installer brings up tasksel. IIRC that will (among other things) drag in the gnome metapackage and anything it depends on.
Ultimately the capstone course for the degree was nothing but a single group project that was a "fly or die" course. Design a plane that has certain capabilites as outlined at the beginning of the Fall semester. In the Spring semester you have to build and flight test the plane. If it doesn't fly, you don't graduate. The big problem I see with this is that an incompetant student on a competent team could be carried though while a competent student on an incompetent team could be failed.
What if any mechanisms did they have in place to try and prevent this?
Integration by parts isn't... tricky? What I found very tricky (I kinda got it in the end but then forgot most of it again since when I went on to do EE at uni integrals that required substituion seemd to mostly dissapear) was figuring out how you were supposed to do an integration question. Parts or substitution, if the former how to split the equation, which half to integrate and which to differentiation, if the latter which of the loads of possible substitutions to use.
I also don't think it would have helped... I always found the ability to see several calculations worth of history on screen hugely helpfull for keeping mistakes down, far more usefull than the actual graphing functionality.
It was rather a pain when I went to uni (to do electronic systems engineering) and was forced to drop down to a far more basic calculator for my exams.
You don't even need any hardware to do a simple serial interface on an embedded controller. For USB you need to slap on a whole expensive dedicated chip. Or use a microcontroller with the functionality built in, for example the PIC18F2550.
and we all know, users really install third party drivers that didn't come with their computer. Surely if the motherboard has USB 3 support the drivers would come with the computer (at least if it's from a decent manufacturer).
As you mention MS, something comes to my mind... There is nothing stopping MS from _never_ releasing drivers There is also nothing stopping intel writing them themselves.
ClamAV has terrible detection rates It's good enough that I have seen virtually no emailed viruses get into my mailbox since i set up clamav/procmail to route them to the bit bucket.
Slashdot Mirror
Right now, it's the only way to get more than 16gb of memory on a board, because regular memory modules are quite demanding on the motherboard's controller, while FB-DIMMs provide their own "smart" interface.
Not entirely true, there are definately boards that can take 32GB of ordinary ram, I dunno if any of them are intel based though.
The real problem though is that if what you are after is a normal desktop apples range sucks.
For the midrange buyer on the one hand you have the imac which has no expansion room, only two cores and locks you into apples choice of monitor (which also pushes up the effective cost if you already have an acceptable monitor). On the other hand you have the mac pro which is built out of server parts with a pricetag to match.
When it comes to low end desktops things look even worse for apple. During a previous argument over whether apple was expensive I specced out a vostro 420 and a mac mini as close as possible, the vostro was both considerablly cheaper and considerablly better specced. http://yro.slashdot.org/comments.pl?sid=1245805&cid=28115689
I did very recently actually.
In fact on debian at least they have made it more hell recently. They only generate a "minimal" config which is fine if it works but if you are say using a KVM switch or an old monitor you basically end up writing an old style config from scratch without any tools to help you.
They don't really have much competition. OS-X is tied to apples weird hardware (which isn't expensive for what it is but forces you into buying machines that are often a poor fit), linux is viewed (rightly or wrongly) as too geeky and both OS-X and linux aren't compatible with a lot of software that buisnesses are tied into.
Besides few people will be paying theese prices, like many companies MS make the full price high so they can discount more.
The trouble is that requires a lot of very expensive dismantling and reassembly. Also with the winding down of the shuttle program the facility that would normally have done such a procedure is no longer operational.
you do but because your exhaust velocity is so high you need far less of it than with conventional rockets.
Your math is wrong, you are forgetting that you are dealing with sqare units (the common notation for square units is rather confusing which doesn't help). There are 10000 square centimeters in a square meter and 1000000 square meters in a square kilometer. You also seem to be in a bit of a mess with the units of various figures.
200,000 W = 200,000,000 mW
200,000,000 mW / 50 (mW/cm^2 )= 4,000,000 cm^2
4,000,000 cm^2 = 400 m^2
sqrt(400 m^2/3.14) ~= 128m
so a 128m diameter receiver, not small but not massive either.
The trouble is that most applications are designed for public computer labs not private homes.
If you tell users where they can and can't use thier laptops or that they can't log in from machines in open plan offices/university clusters/whatever then (unless you have a huge ammount of organisational or legal clout) you will most likely be ignored.
Given that when making an application or website for general use that requires login your only reasonable option is to err on the side of caution.
under what law? afaict iran doesn't have a copyright treaty with the USA.....
I would contend that well written java that has been jited is probablly a bit slower than well written code in a native compiled language but there isn't that much in it and it's hard to compare because the "best" coding methods in the languages are not the same.
However it does have high memory use because of the fact that jited code can't be shared between instances, the large bloat of the standard library, the limited availibility of efficiant data structures (all object types are object references) and the fact that the java garbage collectors don't realease memory back to the OS. This means that if a java app gets swapped out it takes a lot longer to swap back in. It also takes a while to start and get up to full performance, especially if it's the first time a java app has been run this OS session.
In summary on the desktop (where startup and swapin times matter) java is slow, on the server not so much.
now try getting say a wi-fi signal with the remote end having only a normal wi-fi antenna and transmit power at a hundred miles (bottom end of LEO). Maybe you could with a sufficiantly big dish but launching and pointing that dish from a sattelite is going to get expensive and the coverage area is likely to be pretty small (remember antenna gain trades off with antenna directionality). Not to mention that LEO satellites move pretty quick with relation to earth so you would need a load of them all with motorised dishes.
The second would be a very invasive regulation of private individuals, that as far as I know is not practised in any countries with socialized medicine either.
I was under the impression that canada didn't allow public hospitals though canadians can go out of the country for treatment (and afaict many richer canadians do)
Nobody really believes that the wireline maintenance is going to be covered by the revenue generated by selling naked DSL service.
As long as people need fixed line (whether telco twisted pair, cableco coax busses or fiber) last mile communication services somone will be there to charge them for it. Most likely in a lot of areas it will be the same providers that do it now since those providers have been taking steps to actively crush any attempts at starting a new independent service.
If naked DSL is currently underpriced and lots of people take it then some readjustment of prices may be needed but I can't see that killing the telcos.
Truly effective at stopping shit dead in it's tracks. It's useful to point out, that it is not just web browsers either. It's EVERYTHING. Unless your program is written to somehow work without using the communications stack in Windows, it will have to go through the hosts file.
If you want to bypass the hosts file you can quite easilly directly connect to an IP or ship your own dns lookup code (dns isn't that hard to implement) either using hardcoded dns servers or getting the machines dns server addresses through iphlpapi.
I don't think MS has any trouble with you installing XP if you have a valid full license for XP or a volume license with suffiicant downgrade rights for the machine.
All MS is saying is that OEM copies of 7 buisness sold after a certain date won't come with downgrade rights to XP so if you want to run XP you will have to purchase it some other way (and if you want to get new licenses that basically means software assurance).
There are a number of problems with your analysis, not least that the Pentium III is faster clock-for-clock than the Pentium IV at almost all workloads;
It is but IIRC intel throught at the time that they would be able to push the P4 to crazy clock speeds which would more than make up for the lower performance per clock.
Unfortunately they didn't get the clock speeds they had hoped for and the high clock speeds they did get required very high power consuption.
note that testing migration is handled based on source packages. So if any binary package generated from the openoffice.org source package depends on mono it will be considered a dependency for testing migration purposes.
The default is to install the base system (the really important stuff) standard system task (a load of old unixy stuff). Depending on the language you select some language specific stuff will be installed and if you are on a laptop some power management related stuff.
TFA is titled "default desktop install" which probablly reffers to the above plus selecting teh "desktop environment" task when the installer brings up tasksel. IIRC that will (among other things) drag in the gnome metapackage and anything it depends on.
Exactly, assignments and labs are too easy to cheat on. Projects can also be easy to cheat on and/or difficult to mark fairly.
So exams are left as the easy and relatively cheat resistant way to test students.
Ultimately the capstone course for the degree was nothing but a single group project that was a "fly or die" course. Design a plane that has certain capabilites as outlined at the beginning of the Fall semester. In the Spring semester you have to build and flight test the plane. If it doesn't fly, you don't graduate.
The big problem I see with this is that an incompetant student on a competent team could be carried though while a competent student on an incompetent team could be failed.
What if any mechanisms did they have in place to try and prevent this?
Integration by parts isn't... tricky?
What I found very tricky (I kinda got it in the end but then forgot most of it again since when I went on to do EE at uni integrals that required substituion seemd to mostly dissapear) was figuring out how you were supposed to do an integration question. Parts or substitution, if the former how to split the equation, which half to integrate and which to differentiation, if the latter which of the loads of possible substitutions to use.
I also don't think it would have helped...
I always found the ability to see several calculations worth of history on screen hugely helpfull for keeping mistakes down, far more usefull than the actual graphing functionality.
It was rather a pain when I went to uni (to do electronic systems engineering) and was forced to drop down to a far more basic calculator for my exams.
You don't even need any hardware to do a simple serial interface on an embedded controller. For USB you need to slap on a whole expensive dedicated chip.
Or use a microcontroller with the functionality built in, for example the PIC18F2550.
and we all know, users really install third party drivers that didn't come with their computer.
Surely if the motherboard has USB 3 support the drivers would come with the computer (at least if it's from a decent manufacturer).
As you mention MS, something comes to my mind... There is nothing stopping MS from _never_ releasing drivers
There is also nothing stopping intel writing them themselves.
ClamAV has terrible detection rates
It's good enough that I have seen virtually no emailed viruses get into my mailbox since i set up clamav/procmail to route them to the bit bucket.