Pointers are twice the size (yes only 48 bits of the addresses are actually used right now, but all the standards say pointers should be 64 bit and dealing with odd sized pointers would probablly have an even higher pointer cost). This can be a major issue for some apps (java apps tend to suffer particularlly badly because java basically forces the coder to use far more pointers than they really want). However you can get arround this and still get some of the advantages of 64 bit by running 32 bit apps on a 64 bit OS. This lets you have a full 4GB per process while keeping the size of pointers down.
but the main issue is that more than 4GB of ram is rare at the moment(and will probably remain rare until 64 bit takes off, kind of a chicken and egg situation)and for most people the difference between 3.somethingGB of usable ram and 4GB of usable ram is frankly not that great. Hell it is only recently that ordinary desktop/laptop chipsets started supporting more than 4GB of address space.
I belive XP added sound support for DOS games but iirc it was quite laggy. It also made the compatibility options more accessible to users. I don't think there were any other major compatibility improvements.
XP SP2 did make some real imrpovements but they still had the feeling of too little too late.
One might argue that WinXP brought along some additional support like WiFi XP may have had more OS level support for WI-FI but that didn't stop network card vendors producing win2K drivers for thier cards.
and afaik was IPv6 not part of Win2k either. MS offered a IPV6 stack for 2K though they consider it experimental
there are a lot of advantages to being on PC architecture, your machine can run windows if the user wishes, your machine can run wine, your machine can run flash without having to come to a special arrangement with adobe for a port. You can use a regular desktop linux distro with no (or only minor) modifications.
Yes MS has been abusive to it's OEMs in the past but afaict various court judgements have made it much harder for them to get away with that.
btw do you have any info on when exactly that changed? Did original release XP count sockets not cores? Did windows 2000 ever get updated to count sockets not cores?
I don't think MS has shown any intentions of dropping the downgrade rights from corp or academic licenses. IIRC they still come with downgrade rights all the way down to windows 95.
Even for tech savvy home users it isn't too bad as vista buisness comes with downgrade rights to XP pro and even if it did they can pirate.
The people who will be most affected by the XP availibility descisions is the ordinary home and small buisness users. MS wants to drag those users into upgrading but at the same time they don't want to lose this new emerging market (small cheap laptops) to linux.
What about XP's current TCP/IP stack limitations? Do Microsoft intend to add IPv6 in a service pack (which would, if i understand correctly) require the replacement of the whole networking system? Winsock has always supported multiple protocols including third party ones.
MS implemented IPV6 as a completely seperate protocol. This means it was relatively non disruptive but is somewhat annoying for developers of server apps.
also iirc the dns servers have to be on IPV4 but I don't see that being too much of a problem in the near future.
there is also the issue of a rather lacking configuration interface but i'm sure third party apps can be developed if that becomes a significant problem (most ipv6 systems will probablly use autoconfiguration anyway)
IIRC MS claimed they would release XP sp3 in the first half of this year. The release candidate is already out so it doesn't seem like an empty promise.
Vista SP1 is due to hit automatic updates towards the end of this month. I suspect they will wait a month or so for problems related to that to die down and then push out XP sp3.
then imagine trying to make the whole Microsoft Office thing work in 8GB of flash memory. Umm even office 2007 should fit provided there isn't too much other crap on there, 2003 or 2000 should fit easilly.
in 2000 there were not many computers having multiple cores, which became a standard nowadays. They were high end but they DID exist and they WERE supported by windows 2000 and probablly earlier versions of the NT line too. Application programmers taking advantage of them is another story but that is not really an OS issue.
IIRC they had to patch the crippling at some point to count processor units rather than cores but that was really a pretty minor change (not sure if this one filtered down to win2K or not, finding good infornation is hard)
And, I at least expect to FULLY SWITCH to 64-bit architecture, since that's several years old innovation by now. True, a port to a new architecture is a big reason to upgrade, however MS did a x64 bit desktop OS based on the server 2003 codebase (they call it XP but under the hood it is really a cut down server 2003) which is apparently very stable. There is unlikely to be another major architecture change like that now for many years.
in the years leading up to 2000 there were major advances in the windows/PC world every couple of years
* 1993 - windows NT, a proper 32 bit version of windows. * 1995 - windows 95, introduced plug and play allowing users to easilly add devices. Unfortunately based on a rather crummy 16/32 bit hybrid codebase that gave better support for older apps but limited stability and security. * 1998 - windows 98, introduced decent support for USB (there was some support in the last OEM service releases of 95 but USB seriously got going with 98) allowing much easier addition of arbitary external devices. * 2000 - windows 2000, brought together the stability of the NT line with support for critical things like plug and play and USB. Since then the windows world has really stagnated. MS is adding new features but by and large they just aren't that significant to most users particually when the performance cost is considered.
Meanwhile linux has as you say been really catching up and even surpassing windows in many areas. For people with no apps tying them to windows (or who are buying a machine they don't plan to run such apps on) linux is now a very viable choice.
Trivial example with RGB values...you will probably not notice the difference between #FFFFFF and #FEFEFE or #FFFEFF, so if you simply overwrite the lowest bit of each R, G and B value with the data you wish to hide, you can store 3 bits per RGB pixel without visibly changing the appearance of the image. now suppose you took an image with basically no intrinsic noise (e.g. computer generated line-art) and did that too it. It would look suspicious to anyone who looked at the byte values.
also such a simple stegonographic system would not resist lossy encoding.
Combine theese two and that pretty much means if you use such a simple schme you have to hide your data in uncompressed or losslessly compressed photos or audio recordings. Given that for both those types of media lossy formats dominate this may be suspicious.
somewhat amusing the first time I saw a troll of this style. But after you have seen multiple copies of the same troll which is itself a minor modification to another worn out troll it quickly gets wearing.
but lets be honest if they found an unlicensed device with that much radioactives in it i'm sure they could stop it for that reason alone, no need for the patents.
In addtion to my previous post another problem is while you can add more transistors by making the chip bigger that is not the only reason for moving to finer processes.
I don't know what the velocity of signals in the tracks on a silicon chip are but I would expect it is somewhere arround 10^8 m/s. It will definitely be lower than 3x10^8 m/s .
10^8 m/s is about 10mm/ns . When you consider that modern processor clocks have a cycle time arround 0.3ns and that modern processors often already have fairly big dies this becomes something of a limiting factor.
There are two issues with making chips bigger. 1: obviously a bigger chip costs more to make, you need more semiconductor grade silicon, more space on the masks, more space in the furnaces etc. 2: a larger chip has a higher possibility of being struck by a defect increasing the reject rate and pushing up costs still further.
Making the wafers bigger helps will some of problem 1 and indeed wafers have to bigger over the years. But not all of it and brings problems of it's own (the larger a wafer is the more chance there is of accidently breaking it and you have to grow bigger ingots of silicon to make bigger wafers)
I'm pretty sure the cost of the raw material is a negliable part of the costs of making semiconductor grade silicon. Most of the costs are in the very energy intensive purfification processes.
The real advantage of silicon for many years was that SiO2 was/is a decent gate materal for mosfets and insulator for insulating the metal from the main body of the IC and could be grown easilly on the surface of silicon. But afaict this advantage has dwindled as we need CVD deposited insulators for insulating between multiple metal layers anyway and as processes have got smaller there is a push to switch to other gate materials for better performance.
The main advantage of silicon right now is probablly just that we are very used to it and know what does and doesn't work with it. Other semiconductors are more of an unknown.
Even if silicon gets displaced from things like the desktop/server CPU market though I suspect it will stick arround in lower performance chips.
Years later, after the disc proved to be the better in terms of reproduction costs and storage and all-around convenience Later variants may have been more conviniant but that is not the main point.
What edison achived was a sound recording system that was usefull. The reason it was usefull was because he had a way to convert those recordings back into audio form so that humans could interpret them.
Unfortunately for this french guys idea of automatic transcription hearing and vision are very different. Two waveforms that appear totally different on a trace can sound pretty much identical to human ears. Recoding something in a format noone could interpret until after over a centuary of technology advances wasn't usefull.
The problem is in a word processor there are two things going on which are somewhat at odds with each other. Semantics and layout.
Semantics are important because they are what editors use, editors expect to be able to change any text and have things reflow automatically within constraints like page breaks and fixed location images.
Layout is important because when you have your semantics right you then tweak it to look nice, no almost empty pages, no page changes in bad places etc and once this work is done you want those who receive the document to see it the same as you did.
If you just want to preserve semantics that is easy: see html. If you just want to preserve layout that is easy: see pdf
The problem is keeping the file properly editable while also making it open in a predicatable way. So far the only method that seems to work is to use the exact same code on every machine. Even different versions of the same package often break the layout.
Unfortunately afaict neither ODF or OOXML has A soloution to this problem (feel free to correct me if I am wrong).
right, I imagine with the help of bots one person could "monitor" several machines doing enough to make the accounts look human. If you can get your farmers up from say 2 accounts to 5 that is a big saving.
I get very annoyed when I can't complete a game on the easiest difficulty setting and failing to complete a game is liable to put me off the sequal.
Sometimes I cheat but often it doesn't get any easier and once you get into cheating constantly the game starts to feel pointless.
Sadly as a casual gamer I find I get stuck on far too many games and that puts me off gaming in general. The only series I've played so far where I haven't got really stuck (I've had to ask my brother for help occasionally in some of the games but mostly it has been ok) is ratchet and clank (recently completed gladiator, waiting for size matters to come out on the PS2 in the uk) and pokemon (I completed the stories and my brother and I got all but a few event only pokemon and the pokemon from pokemon channel in the gba/gamecube generation but i've mostly got bored of it now).
Slashdot Mirror
Pointers are twice the size (yes only 48 bits of the addresses are actually used right now, but all the standards say pointers should be 64 bit and dealing with odd sized pointers would probablly have an even higher pointer cost). This can be a major issue for some apps (java apps tend to suffer particularlly badly because java basically forces the coder to use far more pointers than they really want). However you can get arround this and still get some of the advantages of 64 bit by running 32 bit apps on a 64 bit OS. This lets you have a full 4GB per process while keeping the size of pointers down.
but the main issue is that more than 4GB of ram is rare at the moment(and will probably remain rare until 64 bit takes off, kind of a chicken and egg situation)and for most people the difference between 3.somethingGB of usable ram and 4GB of usable ram is frankly not that great. Hell it is only recently that ordinary desktop/laptop chipsets started supporting more than 4GB of address space.
I belive XP added sound support for DOS games but iirc it was quite laggy. It also made the compatibility options more accessible to users. I don't think there were any other major compatibility improvements.
XP SP2 did make some real imrpovements but they still had the feeling of too little too late.
One might argue that WinXP brought along some additional support like WiFi
XP may have had more OS level support for WI-FI but that didn't stop network card vendors producing win2K drivers for thier cards.
and afaik was IPv6 not part of Win2k either.
MS offered a IPV6 stack for 2K though they consider it experimental
there are a lot of advantages to being on PC architecture, your machine can run windows if the user wishes, your machine can run wine, your machine can run flash without having to come to a special arrangement with adobe for a port. You can use a regular desktop linux distro with no (or only minor) modifications.
Yes MS has been abusive to it's OEMs in the past but afaict various court judgements have made it much harder for them to get away with that.
btw do you have any info on when exactly that changed? Did original release XP count sockets not cores? Did windows 2000 ever get updated to count sockets not cores?
I don't think MS has shown any intentions of dropping the downgrade rights from corp or academic licenses. IIRC they still come with downgrade rights all the way down to windows 95.
Even for tech savvy home users it isn't too bad as vista buisness comes with downgrade rights to XP pro and even if it did they can pirate.
The people who will be most affected by the XP availibility descisions is the ordinary home and small buisness users. MS wants to drag those users into upgrading but at the same time they don't want to lose this new emerging market (small cheap laptops) to linux.
what isn't being made clear is exactly what microsofts definition of "ultra-low-cost laptop" will be.
What about XP's current TCP/IP stack limitations? Do Microsoft intend to add IPv6 in a service pack (which would, if i understand correctly) require the replacement of the whole networking system?
Winsock has always supported multiple protocols including third party ones.
MS implemented IPV6 as a completely seperate protocol. This means it was relatively non disruptive but is somewhat annoying for developers of server apps.
also iirc the dns servers have to be on IPV4 but I don't see that being too much of a problem in the near future.
there is also the issue of a rather lacking configuration interface but i'm sure third party apps can be developed if that becomes a significant problem (most ipv6 systems will probablly use autoconfiguration anyway)
I dunno about the XO but ASUS supported the EEEPC with XP from the start and is either offering or planning to offer EEEPCs with XP preinstalled.
MS probablly knows they can't kill linux in this space but they really really don't want linux to be the only preinstalled option for such machines.
IIRC MS claimed they would release XP sp3 in the first half of this year. The release candidate is already out so it doesn't seem like an empty promise.
Vista SP1 is due to hit automatic updates towards the end of this month. I suspect they will wait a month or so for problems related to that to die down and then push out XP sp3.
then imagine trying to make the whole Microsoft Office thing work in 8GB of flash memory.
Umm even office 2007 should fit provided there isn't too much other crap on there, 2003 or 2000 should fit easilly.
in 2000 there were not many computers having multiple cores, which became a standard nowadays.
They were high end but they DID exist and they WERE supported by windows 2000 and probablly earlier versions of the NT line too. Application programmers taking advantage of them is another story but that is not really an OS issue.
IIRC they had to patch the crippling at some point to count processor units rather than cores but that was really a pretty minor change (not sure if this one filtered down to win2K or not, finding good infornation is hard)
And, I at least expect to FULLY SWITCH to 64-bit architecture, since that's several years old innovation by now.
True, a port to a new architecture is a big reason to upgrade, however MS did a x64 bit desktop OS based on the server 2003 codebase (they call it XP but under the hood it is really a cut down server 2003) which is apparently very stable. There is unlikely to be another major architecture change like that now for many years.
indeed
in the years leading up to 2000 there were major advances in the windows/PC world every couple of years
* 1993 - windows NT, a proper 32 bit version of windows.
* 1995 - windows 95, introduced plug and play allowing users to easilly add devices. Unfortunately based on a rather crummy 16/32 bit hybrid codebase that gave better support for older apps but limited stability and security.
* 1998 - windows 98, introduced decent support for USB (there was some support in the last OEM service releases of 95 but USB seriously got going with 98) allowing much easier addition of arbitary external devices.
* 2000 - windows 2000, brought together the stability of the NT line with support for critical things like plug and play and USB.
Since then the windows world has really stagnated. MS is adding new features but by and large they just aren't that significant to most users particually when the performance cost is considered.
Meanwhile linux has as you say been really catching up and even surpassing windows in many areas. For people with no apps tying them to windows (or who are buying a machine they don't plan to run such apps on) linux is now a very viable choice.
Trivial example with RGB values...you will probably not notice the difference between #FFFFFF and #FEFEFE or #FFFEFF, so if you simply overwrite the lowest bit of each R, G and B value with the data you wish to hide, you can store 3 bits per RGB pixel without visibly changing the appearance of the image.
now suppose you took an image with basically no intrinsic noise (e.g. computer generated line-art) and did that too it. It would look suspicious to anyone who looked at the byte values.
also such a simple stegonographic system would not resist lossy encoding.
Combine theese two and that pretty much means if you use such a simple schme you have to hide your data in uncompressed or losslessly compressed photos or audio recordings. Given that for both those types of media lossy formats dominate this may be suspicious.
somewhat amusing the first time I saw a troll of this style. But after you have seen multiple copies of the same troll which is itself a minor modification to another worn out troll it quickly gets wearing.
but lets be honest if they found an unlicensed device with that much radioactives in it i'm sure they could stop it for that reason alone, no need for the patents.
In addtion to my previous post another problem is while you can add more transistors by making the chip bigger that is not the only reason for moving to finer processes.
I don't know what the velocity of signals in the tracks on a silicon chip are but I would expect it is somewhere arround 10^8 m/s. It will definitely be lower than 3x10^8 m/s .
10^8 m/s is about 10mm/ns . When you consider that modern processor clocks have a cycle time arround 0.3ns and that modern processors often already have fairly big dies this becomes something of a limiting factor.
There are two issues with making chips bigger.
1: obviously a bigger chip costs more to make, you need more semiconductor grade silicon, more space on the masks, more space in the furnaces etc.
2: a larger chip has a higher possibility of being struck by a defect increasing the reject rate and pushing up costs still further.
Making the wafers bigger helps will some of problem 1 and indeed wafers have to bigger over the years. But not all of it and brings problems of it's own (the larger a wafer is the more chance there is of accidently breaking it and you have to grow bigger ingots of silicon to make bigger wafers)
I'm pretty sure the cost of the raw material is a negliable part of the costs of making semiconductor grade silicon. Most of the costs are in the very energy intensive purfification processes.
The real advantage of silicon for many years was that SiO2 was/is a decent gate materal for mosfets and insulator for insulating the metal from the main body of the IC and could be grown easilly on the surface of silicon. But afaict this advantage has dwindled as we need CVD deposited insulators for insulating between multiple metal layers anyway and as processes have got smaller there is a push to switch to other gate materials for better performance.
The main advantage of silicon right now is probablly just that we are very used to it and know what does and doesn't work with it. Other semiconductors are more of an unknown.
Even if silicon gets displaced from things like the desktop/server CPU market though I suspect it will stick arround in lower performance chips.
Years later, after the disc proved to be the better in terms of reproduction costs and storage and all-around convenience
Later variants may have been more conviniant but that is not the main point.
What edison achived was a sound recording system that was usefull. The reason it was usefull was because he had a way to convert those recordings back into audio form so that humans could interpret them.
Unfortunately for this french guys idea of automatic transcription hearing and vision are very different. Two waveforms that appear totally different on a trace can sound pretty much identical to human ears. Recoding something in a format noone could interpret until after over a centuary of technology advances wasn't usefull.
could you please provide evidence of why you think I am wrong.
http://www.apple.com/quicktime/download/ seems to offer me options both with and without itunes.
The problem is in a word processor there are two things going on which are somewhat at odds with each other. Semantics and layout.
Semantics are important because they are what editors use, editors expect to be able to change any text and have things reflow automatically within constraints like page breaks and fixed location images.
Layout is important because when you have your semantics right you then tweak it to look nice, no almost empty pages, no page changes in bad places etc and once this work is done you want those who receive the document to see it the same as you did.
If you just want to preserve semantics that is easy: see html.
If you just want to preserve layout that is easy: see pdf
The problem is keeping the file properly editable while also making it open in a predicatable way. So far the only method that seems to work is to use the exact same code on every machine. Even different versions of the same package often break the layout.
Unfortunately afaict neither ODF or OOXML has A soloution to this problem (feel free to correct me if I am wrong).
right, I imagine with the help of bots one person could "monitor" several machines doing enough to make the accounts look human. If you can get your farmers up from say 2 accounts to 5 that is a big saving.
I get very annoyed when I can't complete a game on the easiest difficulty setting and failing to complete a game is liable to put me off the sequal.
Sometimes I cheat but often it doesn't get any easier and once you get into cheating constantly the game starts to feel pointless.
Sadly as a casual gamer I find I get stuck on far too many games and that puts me off gaming in general. The only series I've played so far where I haven't got really stuck (I've had to ask my brother for help occasionally in some of the games but mostly it has been ok) is ratchet and clank (recently completed gladiator, waiting for size matters to come out on the PS2 in the uk) and pokemon (I completed the stories and my brother and I got all but a few event only pokemon and the pokemon from pokemon channel in the gba/gamecube generation but i've mostly got bored of it now).