Note: And before you dismiss me as a dumb yank who knows nothing, I spent 6 years living in the UK (4 in Scotland, 2 in London), and originate from Dublin, Ireland. I even went through 6 months of police training in Scotland before deciding that the police wasn't for me.
The original post was about cultural differences. Do you not think that the fact that you were (by the sound of your post) born and brought up in the US may contribute to the fact that you feel this way?
Eg, I am english, and I believe in gun control. If I had been brought up in the states I would probably believe in the right to bear arms. But I wasn't so I don't.
I am not trying to say you are right or wrong - just that people in this country are different.
The right not to have the police following you around all the time, waiting for you so commite some crime so they can arrest you.
First of all, the police are not watching you - for the simple reason that it would cost too much. Councils contract private security companies to staff CCTV systems, since the police are already over streached - and the CCTV operators will only bother calling the police if they see a crime in progress.
Secondly, some people do want to be watched. For example, I heard of a pilot scheme in one city in the UK, where there is a phone number that a single woman walking home alone at night can ring. She can leave her description, a time, and roughly what route she will be following. Now, rather than walking home alone in the dark afraid of being attacked, every time she turns a corner she will be greeted by the sight of a CCTV camera turning to focus in on her. Having a big brother to watch over you is not always a bad thing.
used or posesed any illegal drugs
To quote the subject at the top of this thread, "CCTV is a reflection of cultural differences." Please bear in mind, that in this country, if you are caught smuggling 5 grams of pot into the country it is assumed to be for personal use and you will be given a £70 on the spot fine. Compare that to the US view on drugs smuggling. Cultural differences.
Note for non-Brits:
Mark Thomas is a politically motivated comedian-slash-borderline-terrorist (that's meant as a compliment) and probably one of the biggest pains in the government's ass.
One thing that he had great fun playing with in his recent series, (not what prev. poster was talking about, but relevant to CCTV), was the Data Protection Act.
This is a wonderful piece of UK legislation, which allows you to demand any company/organization which holds information about you to give you a copy (with certain exclusions ie some government agencies). So you can walk into MacDonalds, fill out a form while you eat your burger, giving the time, date, a description of yourself, the clothes you are wearing, etc, then hand it in before you leave forcing them to send you a copy of the footage of you sitting there filling out the form.
That anecdote about using/dev/hda instead of/dev/ttyS1 sounds like BS.
Yeah - I want to believe, but...
Reminds me of urban legend of the software company demonstrating voice recognition software for MS-DOS. One member of the audience viewing the demonstration calls out "format c colon enter", and in quick succession someone else shouts "y enter". The demonstration is brought to a premature end;-)
"early testers of beta versions of Windows XP already complain that the most popular MP3 recording applications -- which compete with Microsoft's format -- don't seem to function properly"
Surely the phrases "MP3 recording" and "which compete with Microsoft's format" are redundant. I think you meant to say:
"early testers of beta versions of Windows XP already complain that the most popular applications don't seem to function properly"
No. Maybe.
Remember that this is not legislation, only a dirtective, and it depends on how this is framed into law.
The dvd consortium attacked decss by claiming that it violated a trade secret. Cue cat hired a lot of lawyers, and fired off a lot of threatening letters, but at the end of the day didn't have a leg to stand on. Even their lawyers weren't mad enough to enter a courtroom and argue that a one byte xor was a trade secret.
It really is far to soon to judge whether a tool to crack such a simple obfuscation could be attacked under the legislation that may well not become law for another year and a half.
Technical copies on the net The Directive provides an obligatory exception for service providers, telecommunications operators and certain others in limited circumstances for particular acts of reproduction which are considered technical copies. A satisfactory balance has been found for what has been an extremely controversial issue. There are many conditions to be fulfilled before the exemption applies. In particular, those acts of reproduction have to form an essential part of a technological process and take place in the context of a transmission in a network. The Directive ensures therefore that there will be effective operation of the World Wide Web for those who place copyrighted material on the net and those who transmit or carry such material.
So caching web proxies are exempt - now this is a only an EU directive, not a piece of legislation, so we will have to wait and see how this is drafted into law in individual countries, but there might be a nice big loophole for freenet here:-)
And the violent crime rate went up something like 40-50% in the unsurveillanced neighbouring areas, right?
Not likely. One of the key problems in the area was to do with violence associated with drunk and dissorderly behaviour late in the evenings after pubs emptied out onto the streets, and as people made their way home. Not the type of problem liable to migrate elsewhere.
In the more general case, you make a fair point, but the kind of problem you describe would not occur if you similtaneously increase CCTV coverage at a reasonably uniform rate everywhere. Obviously, this would make implementation more difficult.
That's a so fucking typically anglo-saxon "solution"
I'd quite like to hear some explaination/justification for this statement.
I used to live in an area in North Wales where a fairly comprehensive network of security cameras were fitted. After one year, the violent crime figures in that area were down something like 40-50% on the previous year.
The kind of conclusion you seem to draw from the national rise in violent crime is not really valid unless you look at the corellation between regional crime figures, and relative numbers of security cameras in the areas.
One interesting note:
Two of the largest news stories in the UK in recent weeks have featured security camera footage as key evidence.
The first is the case of a dutch lorry driver, sentenced to, uh, 15 years, for the murder of 60 chinese imigrants, who suffocated in the back of his lorry, after he closed an air vent. He claimed that he didn't know what cargo he was carrying, and unless this could be proven, he would have got off (unless you can demonstrate that he knew the imigrants were in the back of the truck, then closing the air vent cannot constitute murder). His conviction depended entirely on Dutch CCTV footage showing his buying crates of tomatos that were used as a screen to hide the imigrants, showing that he was involved in loading the lorry.
The second is a trial still running, of a group of professional football players accused gbh on an asian student. Not too much information, since the jury is still out, but apparently key evidence is 12 clips of CCTV footage taken from Leeds city centre camaras.
hell, I don't like them, but as I understand it the results from the CCTV systems are very good.
I'd be careful. The article is originally talking about the availability of CD images - well who is to say that all of the software on the CD is GPL'ed?
It is true that you can grab a debian CD, rip it, and freely distribute the iso you produce. But this is not true for any distributions that contain copyright software, and you could find yourself violating copyright on the non-GPL software.
I think some moderators got blinded by the clever troll...
X is NOT bad. It's very well-designed.
Yup. X is a brilliant protocol for remote display. It's just a goddamn awful protocol for local display. Just because something is well well designed does not make it appropriate for the given task.
I see a future where X actually has most of the stuff running inside the video card, and programs on the local machine sending updates to it, similar to the way in which programs update remote displays over a network.
Not possible, X's primatives are simply too fine grained.
I look forward to using a gui that runs directly onto an accelerated local framebuffer, with an X server running on that. The network abstraction should run ontop of the gui. Not vice versa.
Yeah, the human genome project is probably going to have had the biggest impact, when we look back in ten years time. Of course, not necessarily in a good way...
SMT also doesn't save you from cache miss latency.
Please enlighten us.
I'm sat here working on the software side of an SMT project. This is exactly where SMT offers benefits. The processor switches threads on a cache miss. A 4-way SMT scheme can offer > 2x performance for 2x die size. A SMT procoessor cannot reduce the cache latency experienced by a particular thread of execution, but it can reduce the amount of time that execution units sit idle.
There was a wonderful place in Manchester (uk) where I used to love to hang out with fellow comp.sci students.
They had a dalek, and a costume from babylon 5 by the dj's booth. They had a cabinet full of cult tv/movie related toys/memorobilia.
They had original starwars & space invaders arcade machines, house of the dead, pinball, etc. Instead of tables they had some of the old fashioned table format arcade machines (ie, the screen horizontal so you put your drink down on top of the screen (sadly they weren't playable, but the demos ran.))
They played funky 70s music, while showing wierd, surreal 60s/70s films, thunderbirds, old cartoons, etc. They put on a screening of Episode 1 the weekend after it was released in the states. (uk movie releases are months behind the states. Yes, this was illegal.)
Sadly, the place has gone badly downhill. Rough, overcrowded, bad music, grrrrr.:-(
Ownership of a copyright, or of any of the exclusive rights under a copyright, is distinct from ownership of any material object in which the work is embodied.
IANAL, but this looks about the most innocuous part of the DMCA to me.
If you write a book, and sell a copy to me, then you retain copyright. I am bound by your copyright, and my actions are restricted in certain ways, e.g. I cannot print copies of your book and sell them myself.
If youwrite a piece of software / author a DVD, and sell a copy to me, then you retain copyright. I am bound by your copyright, and my actions are restricted in certain ways, e.g. I cannot cut copies of the software / DVD and sell them myself.
Are you trying to claim that it's the half of the web which isn't running on asp that is insecure;-)
It's not because Perl is insecure, it's because Perl makes it so damn easy to just "get something working", and clueless people are writing insecure scripts all over the place.
Uh... you miss the point. You run the script (root 'sploit, not cgi) on the perl interptreter on your local machine - not the remote machine that you are trying to hack into. Which err... was my point - the hacker uses perl as a tool, perl isn't part of the problem.
and if you are refering to cgi scripts that people have writen themselves - well when you write a cgi script you are writing a piece of software - and if the software that you write is insecure, then that is not a bug in perl, it's yer own damn fault.
I think he is refering to the fact that perl is a language commonly used for writing root exploit scripts - perl is very powerful in terms of the fact that you can do a lot in a few lines of code (hence obfuscated perl contests).
His mistake (or probably troll) is in the fact that the waekness is not in perl - perl is just usefull in expoilting the weaknesses in other software.
I'm involved in a project involving a SMT (well, more CMT) processor design. We do not yet have any silicon, but are getting good results in simulation.
We have a simulator that can be set to simulate a processor with any number of closely coupled cores, and any number of threads per core. We get good results at a 8 core * 4 threads setup (total up to 32 way parallel).
Using some basic automatic parallelization on a piece of code designed to run in a single thread, we have generated up to a 26X speedup, 8 core * 4 threads versus 1 core * 1 thread.
The advantage of SMT over a normal processor is that it makes use of clock cycles that would otherwise be wasted, eg waiting for the cache to fill. If your architchture spends half of its time stalled, and you can make use of these cycles by adding SMT, then you can increase your processor performance very efficiently.
SMT basically requires you to duplicate all of the processor's registers n times (n = #threads), + a little extra hardware ('little', relative to duplicating the entire core). So for ((1 * core) + (2 * registers) + SMT hardware) you are getting the performance of ((2 * core) + (2 * registers)). Good bang per buck ratio, when you count up the transistors.
But SMT naturally gives you diminishing returns for each thread you add - the whole point is that each new thread is using up wasted cycles - and once you reach ~4 threads there are very few cycles left over. At this point, if you have room left over on the die, you may as well start thinking about SMP on the same die.
Surpriesd the article didn't mention SMT & AMD. Check out this link.
Considering word size is different, there's obviously some sort of emulation going on.
Fair assumption, but not really true.
In short - your pentium 2/3 is internally a VLIW processor. it contains multiple RISC execution units running in parallel. It has a CISC decoder sat in front of this, to decode x86 instructions into internal mircocode that runs on these units. The itanium executes x86 code in exactly the same way. It just also has the ability to run VLIW instructions directly - skipping the decode stage.
As for word size - is the x86 (x >= 3) emulating a 286 when it runs 16 bit code? Not really - it just only uses 16 of the 32 bits available to it.
Linux/MS would not NEED to release a new version for it, would they?
OS developers have two options:
do a full port - more work but the who OS is compiled into 64bit code, make use of the larger memory space, etc.
do a partial port - just write code to load 64 bit binaries, switch into 64 bit mode before executing, set up task gates to switch the system back into 32 bit mode when the process calls the operating system (can leave the exception handling routines as 32 bit code).
The other reason that a port may be necessary is that although the processor may support the IA32 instruction set, there is nothing to say that the rest of the system architecture (e.g. motherboard, busses, bios) may not be backwards compatible with legacy systems. This would be invisible to user space programs, but OS developers would have to deal with the new architecture.
When wasn't the first release of any Intel processor greatly overpriced?
Anyways...
unproven processor that's incompatible with just about everything that's been built up in the PC-clone era of the last 19 years.
Uh.... how incompatible.... in so mach as it runs any x86 code, back so far as the 8086 back in the 70's? hmmm....
It remains to be seen if the Itanium is really where personal computing is headed.
IA64 is definitely not intended for the personal computing market - at least not for a long time yet. This is a server processor, and Intel already have SGI & Sun lined up with designs based around the IA64. MIPS & Sparc may not be dead yet, but there is more movement from Intel's most direct competitors, towards this product, than any other processor they have previously released.
Moving to an entirely new processor *family*, not just the next generation of what's currently available, is not to be taken lightly.
The IA64 architecture is a rare example of an easy transition between ISAs. The processor supports both the new VLIW/EPIC instruction set and the IA32 instruction set. For the OS writers, they can have a mix of 32 & 64 bit code running on the machine, to the extent that a 64 program can have its system calls serviced by 32 bit exception handlers, and vice-versa. So far as application software, a user can run new 64 bit application software alongside legacy applications that they cannot port to 64 bit. There are few easier ways to escape from the headache of the x86 instruction set.
This is doubly true when the benefits of such a change are not at all obvious.
In the server market, the 32bit address space is already becoming a problem. (You can buy yourself a linux box with 4gb ram today).
(Score:5, Informative) for the parent post? In reality it is probably a subtle Troll.
I am a final year CS student, in the UK. On my course, we were taught SML, lisp, mips assembly & C in the first year and Java from the second year (amongst other languages). From next year on, SML & C will be cut, and they will teach Java from day 1 of the course.
The C course was tough for people new to the language - and most people like me who knew the language well skipped every lecture & aced the exam. I can see an arguement for dropping C from the core compulsory syllabus - most of the language overlaps with Java - and there are many CS related jobs in the world for which you never need know what a pointer is.
But it is a shame that SML has to go. SML is a functional language, and few people had programmed in it prior to starting the course, so it brought everyone to the same level. It taught something new to everyone, and opened up the minds of a bunch of hardened C hackers to a different paradigm. I'm sure exactly the same could be said for lisp, but I hated the lousy stinking language myself;-)
I'm sure that most people reading this have probably never programmed in a functional language, and if you are a C hacker I recommend you do so - it will expand the way you approach programming. Or try something ever more, uh, different - go look up the language brainfuck, and get yourself stuck in a turing tarpit.
End rant. Just thought I'd relate this experience:-)
Slashdot Mirror
-
Note: And before you dismiss me as a dumb yank who knows nothing, I spent 6 years living in the UK (4 in Scotland, 2 in London), and originate from Dublin, Ireland. I even went through 6 months of police training in Scotland before deciding that the police wasn't for me.
The original post was about cultural differences. Do you not think that the fact that you were (by the sound of your post) born and brought up in the US may contribute to the fact that you feel this way?Eg, I am english, and I believe in gun control. If I had been brought up in the states I would probably believe in the right to bear arms. But I wasn't so I don't.
I am not trying to say you are right or wrong - just that people in this country are different.
- The right not to have the police following you around all the time, waiting for you so commite some crime so they can arrest you.
First of all, the police are not watching you - for the simple reason that it would cost too much. Councils contract private security companies to staff CCTV systems, since the police are already over streached - and the CCTV operators will only bother calling the police if they see a crime in progress.Secondly, some people do want to be watched. For example, I heard of a pilot scheme in one city in the UK, where there is a phone number that a single woman walking home alone at night can ring. She can leave her description, a time, and roughly what route she will be following. Now, rather than walking home alone in the dark afraid of being attacked, every time she turns a corner she will be greeted by the sight of a CCTV camera turning to focus in on her. Having a big brother to watch over you is not always a bad thing.
-
used or posesed any illegal drugs
To quote the subject at the top of this thread, "CCTV is a reflection of cultural differences." Please bear in mind, that in this country, if you are caught smuggling 5 grams of pot into the country it is assumed to be for personal use and you will be given a £70 on the spot fine. Compare that to the US view on drugs smuggling. Cultural differences.- Don't you ever watch any Mark Thomas?
Note for non-Brits:Mark Thomas is a politically motivated comedian-slash-borderline-terrorist (that's meant as a compliment) and probably one of the biggest pains in the government's ass.
One thing that he had great fun playing with in his recent series, (not what prev. poster was talking about, but relevant to CCTV), was the Data Protection Act.
This is a wonderful piece of UK legislation, which allows you to demand any company/organization which holds information about you to give you a copy (with certain exclusions ie some government agencies). So you can walk into MacDonalds, fill out a form while you eat your burger, giving the time, date, a description of yourself, the clothes you are wearing, etc, then hand it in before you leave forcing them to send you a copy of the footage of you sitting there filling out the form.
This is all wonderfully silly.
- That anecdote about using
/dev/hda instead of /dev/ttyS1 sounds like BS.
Yeah - I want to believe, but...Reminds me of urban legend of the software company demonstrating voice recognition software for MS-DOS. One member of the audience viewing the demonstration calls out "format c colon enter", and in quick succession someone else shouts "y enter". The demonstration is brought to a premature end ;-)
G.
- "early testers of beta versions of Windows XP already complain that the most popular MP3 recording applications -- which compete with Microsoft's format -- don't seem to function properly"
Surely the phrases "MP3 recording" and "which compete with Microsoft's format" are redundant.I think you meant to say:
- Yes
No. Maybe.Remember that this is not legislation, only a dirtective, and it depends on how this is framed into law.
The dvd consortium attacked decss by claiming that it violated a trade secret. Cue cat hired a lot of lawyers, and fired off a lot of threatening letters, but at the end of the day didn't have a leg to stand on. Even their lawyers weren't mad enough to enter a courtroom and argue that a one byte xor was a trade secret.
It really is far to soon to judge whether a tool to crack such a simple obfuscation could be attacked under the legislation that may well not become law for another year and a half.
G.
- Technical copies on the net
So caching web proxies are exempt - now this is a only an EU directive, not a piece of legislation, so we will have to wait and see how this is drafted into law in individual countries, but there might be a nice big loophole for freenet hereThe Directive provides an obligatory exception for service providers, telecommunications operators and certain others in limited circumstances for particular acts of reproduction which are considered technical copies. A satisfactory balance has been found for what has been an extremely controversial issue. There are many conditions to be fulfilled before the exemption applies. In particular, those acts of reproduction have to form an essential part of a technological process and take place in the context of a transmission in a network. The Directive ensures therefore that there will be effective operation of the World Wide Web for those who place copyrighted material on the net and those who transmit or carry such material.
G.
- And the violent crime rate went up something like 40-50% in the unsurveillanced neighbouring areas, right?
Not likely. One of the key problems in the area was to do with violence associated with drunk and dissorderly behaviour late in the evenings after pubs emptied out onto the streets, and as people made their way home. Not the type of problem liable to migrate elsewhere.In the more general case, you make a fair point, but the kind of problem you describe would not occur if you similtaneously increase CCTV coverage at a reasonably uniform rate everywhere. Obviously, this would make implementation more difficult.
- That's a so fucking typically anglo-saxon "solution"
I'd quite like to hear some explaination/justification for this statement.G
The kind of conclusion you seem to draw from the national rise in violent crime is not really valid unless you look at the corellation between regional crime figures, and relative numbers of security cameras in the areas.
One interesting note:
Two of the largest news stories in the UK in recent weeks have featured security camera footage as key evidence.
The first is the case of a dutch lorry driver, sentenced to, uh, 15 years, for the murder of 60 chinese imigrants, who suffocated in the back of his lorry, after he closed an air vent. He claimed that he didn't know what cargo he was carrying, and unless this could be proven, he would have got off (unless you can demonstrate that he knew the imigrants were in the back of the truck, then closing the air vent cannot constitute murder). His conviction depended entirely on Dutch CCTV footage showing his buying crates of tomatos that were used as a screen to hide the imigrants, showing that he was involved in loading the lorry.
The second is a trial still running, of a group of professional football players accused gbh on an asian student. Not too much information, since the jury is still out, but apparently key evidence is 12 clips of CCTV footage taken from Leeds city centre camaras.
hell, I don't like them, but as I understand it the results from the CCTV systems are very good.
just my £0.02,
G
- That's also permitted in the GPL.
I'd be careful. The article is originally talking about the availability of CD images - well who is to say that all of the software on the CD is GPL'ed?It is true that you can grab a debian CD, rip it, and freely distribute the iso you produce. But this is not true for any distributions that contain copyright software, and you could find yourself violating copyright on the non-GPL software.
cheers, G.
Yup. X is a brilliant protocol for remote display. It's just a goddamn awful protocol for local display. Just because something is well well designed does not make it appropriate for the given task.
Not possible, X's primatives are simply too fine grained.
I look forward to using a gui that runs directly onto an accelerated local framebuffer, with an X server running on that. The network abstraction should run ontop of the gui. Not vice versa.
- 2: genome
Nah, KDE rulez![ducks, runs for cover]
Yeah, the human genome project is probably going to have had the biggest impact, when we look back in ten years time. Of course, not necessarily in a good way...
- SMT also doesn't save you from cache miss latency.
Please enlighten us.I'm sat here working on the software side of an SMT project. This is exactly where SMT offers benefits. The processor switches threads on a cache miss. A 4-way SMT scheme can offer > 2x performance for 2x die size. A SMT procoessor cannot reduce the cache latency experienced by a particular thread of execution, but it can reduce the amount of time that execution units sit idle.
- It includes 2 tightly coupled processor units
In SMT, you have multiple threads of execution (eg. multiple PCs) feeding one CPU.There was a wonderful place in Manchester (uk) where I used to love to hang out with fellow comp.sci students.
:-(
They had a dalek, and a costume from babylon 5 by the dj's booth. They had a cabinet full of cult tv/movie related toys/memorobilia.
They had original starwars & space invaders arcade machines, house of the dead, pinball, etc. Instead of tables they had some of the old fashioned table format arcade machines (ie, the screen horizontal so you put your drink down on top of the screen (sadly they weren't playable, but the demos ran.))
They played funky 70s music, while showing wierd, surreal 60s/70s films, thunderbirds, old cartoons, etc. They put on a screening of Episode 1 the weekend after it was released in the states. (uk movie releases are months behind the states. Yes, this was illegal.)
Sadly, the place has gone badly downhill. Rough, overcrowded, bad music, grrrrr.
- Read section 202.
Why? Doesn't seem relevant. Example:- Ownership of a copyright, or of any of the exclusive rights under a copyright, is distinct from ownership of any material object in which the work is embodied.
IANAL, but this looks about the most innocuous part of the DMCA to me.If you write a book, and sell a copy to me, then you retain copyright. I am bound by your copyright, and my actions are restricted in certain ways, e.g. I cannot print copies of your book and sell them myself.
If you write a piece of software / author a DVD, and sell a copy to me, then you retain copyright. I am bound by your copyright, and my actions are restricted in certain ways, e.g. I cannot cut copies of the software / DVD and sell them myself.
- Especialy since there is a root exploit for every known version of perl in existance
I think it is fair to say that perl is a relatively secure & bug free piece of software.- web sites running Perl are notoriously insecure
Are you trying to claim that it's the half of the web which isn't running on asp that is insecure- It's not because Perl is insecure, it's because Perl makes it so damn easy to just "get something working", and clueless people are writing insecure scripts all over the place.
Uh... you miss the point. You run the script (root 'sploit, not cgi) on the perl interptreter on your local machine - not the remote machine that you are trying to hack into. Which err... was my point - the hacker uses perl as a tool, perl isn't part of the problem.and if you are refering to cgi scripts that people have writen themselves - well when you write a cgi script you are writing a piece of software - and if the software that you write is insecure, then that is not a bug in perl, it's yer own damn fault.
And anyway, there's nothing wrong with blunt object references ;-)
His mistake (or probably troll) is in the fact that the waekness is not in perl - perl is just usefull in expoilting the weaknesses in other software.
- Any admin that installs a server and leaves the r-services enabled should be tarred and feathered.
Any admin that installs a server and leaves the r-services enabled should be tarred and gzipped.I'm involved in a project involving a SMT (well, more CMT) processor design. We do not yet have any silicon, but are getting good results in simulation.
We have a simulator that can be set to simulate a processor with any number of closely coupled cores, and any number of threads per core. We get good results at a 8 core * 4 threads setup (total up to 32 way parallel).
Using some basic automatic parallelization on a piece of code designed to run in a single thread, we have generated up to a 26X speedup, 8 core * 4 threads versus 1 core * 1 thread.
The advantage of SMT over a normal processor is that it makes use of clock cycles that would otherwise be wasted, eg waiting for the cache to fill. If your architchture spends half of its time stalled, and you can make use of these cycles by adding SMT, then you can increase your processor performance very efficiently.
SMT basically requires you to duplicate all of the processor's registers n times (n = #threads), + a little extra hardware ('little', relative to duplicating the entire core). So for ((1 * core) + (2 * registers) + SMT hardware) you are getting the performance of ((2 * core) + (2 * registers)). Good bang per buck ratio, when you count up the transistors.
But SMT naturally gives you diminishing returns for each thread you add - the whole point is that each new thread is using up wasted cycles - and once you reach ~4 threads there are very few cycles left over. At this point, if you have room left over on the die, you may as well start thinking about SMP on the same die.
Surpriesd the article didn't mention SMT & AMD. Check out this link.
Fair assumption, but not really true.
In short - your pentium 2/3 is internally a VLIW processor. it contains multiple RISC execution units running in parallel. It has a CISC decoder sat in front of this, to decode x86 instructions into internal mircocode that runs on these units. The itanium executes x86 code in exactly the same way. It just also has the ability to run VLIW instructions directly - skipping the decode stage.
As for word size - is the x86 (x >= 3) emulating a 286 when it runs 16 bit code? Not really - it just only uses 16 of the 32 bits available to it.
OS developers have two options:
The other reason that a port may be necessary is that although the processor may support the IA32 instruction set, there is nothing to say that the rest of the system architecture (e.g. motherboard, busses, bios) may not be backwards compatible with legacy systems. This would be invisible to user space programs, but OS developers would have to deal with the new architecture.
cheers,
G
Anyways...
- unproven processor that's incompatible with just about everything that's been built up in the PC-clone era of the last 19 years.
Uh.... how incompatible.... in so mach as it runs any x86 code, back so far as the 8086 back in the 70's? hmmm....- It remains to be seen if the Itanium is really where personal computing is headed.
IA64 is definitely not intended for the personal computing market - at least not for a long time yet. This is a server processor, and Intel already have SGI & Sun lined up with designs based around the IA64. MIPS & Sparc may not be dead yet, but there is more movement from Intel's most direct competitors, towards this product, than any other processor they have previously released.-
Moving to an entirely new processor *family*, not just the next generation of what's currently available, is not to be taken lightly.
The IA64 architecture is a rare example of an easy transition between ISAs. The processor supports both the new VLIW/EPIC instruction set and the IA32 instruction set. For the OS writers, they can have a mix of 32 & 64 bit code running on the machine, to the extent that a 64 program can have its system calls serviced by 32 bit exception handlers, and vice-versa. So far as application software, a user can run new 64 bit application software alongside legacy applications that they cannot port to 64 bit. There are few easier ways to escape from the headache of the x86 instruction set.- This is doubly true when the benefits of such a change are not at all obvious.
In the server market, the 32bit address space is already becoming a problem. (You can buy yourself a linux box with 4gb ram today).(Score:5, Informative) for the parent post? In reality it is probably a subtle Troll.
cheers,
G
I am a final year CS student, in the UK. On my course, we were taught SML, lisp, mips assembly & C in the first year and Java from the second year (amongst other languages). From next year on, SML & C will be cut, and they will teach Java from day 1 of the course.
;-)
:-)
The C course was tough for people new to the language - and most people like me who knew the language well skipped every lecture & aced the exam. I can see an arguement for dropping C from the core compulsory syllabus - most of the language overlaps with Java - and there are many CS related jobs in the world for which you never need know what a pointer is.
But it is a shame that SML has to go. SML is a functional language, and few people had programmed in it prior to starting the course, so it brought everyone to the same level. It taught something new to everyone, and opened up the minds of a bunch of hardened C hackers to a different paradigm. I'm sure exactly the same could be said for lisp, but I hated the lousy stinking language myself
I'm sure that most people reading this have probably never programmed in a functional language, and if you are a C hacker I recommend you do so - it will expand the way you approach programming. Or try something ever more, uh, different - go look up the language brainfuck, and get yourself stuck in a turing tarpit.
End rant. Just thought I'd relate this experience
cheers,
G