> Any "solution" in the OS (including the current MS one for Excel) is going to be a nasty kludge anyway. If I see a "." key on the keypad, hit the key in Excel, and a "," appears - now *that* is a bug:)
Calling it a bug is a bit too harsh IMHO. I think that it should be customisable in the locale settings. Or even better, to avoid the possibility of customising the keyboard in two different location, have a redirection to a special subwindow of 'keyboard customisation'.
I think having three posibilities for the numeric dot would be interesting: - always a dot. - always the "," in the locale. - the dot in normal application but the "," in numeric fields.
It is a bit complicated and I don't see how to make it work correctly with xmodmap and the like, but it seems to me the best solution.
>Turning off hyperthreading on a dual-processor system is FLAT CHEATING.
Actually on some benchmark turning off hyperthreading has IMPROVED the score.
I doubt that it is the case here, but I just wanted to tell you that hyperthreading is not necessarily a magic thing which always improves performance.
>If its the size of 1 mm, shouldn't it be called a "milli-engine"?
You're wrong: nanotechnlogy is not about the size of the object produced but about the accuracy used to create the object. If you created a very big object, by controling precisely the position of each molecules of the object, it would be nanotechnology, even if the object is very big.
But you're right: these object are not nano at all: these are MEMS.
I've looked at the EPIC instruction set and I didn't find it more complicated than the ugly x68 instruction set.
And as for coding in VLIW assembly language complicated enough to code an OS: an OS usually is mostly coded in C, so it isn't really a problem.
And if the assembly language is too complicated, and C too rigid for some parts, nothing prevents the use of some kind of "in-between" language: a C extended with all the operation offered by the CPU for example.
Well, it IS a problem when your CV that you carefully edited to fit on 1 page, doesn't fit on this page for the people you sent it: it make you look like a fool!
It happened to me once: next time I sent a CV by email, I'll use both a doc and a pdf!
Well I hope that this extension ask before loading the pages! Because if it loads the page automatically and it is the content of a page which crash the browser..
I wonder why it hasn't been put directly into the browser instead of an extension.
You're guessing too much, nobody said the apps restarted at the exact place playing movies or sound. It is just: 1) an intelligent design from apps: from time to time they record to the HDD where they are 2) a journalised FS (not something common at the time) so no fsck. 3) a FAST boot time: around 10sec to boot the PC under BeOS. 4) intelligent design of the OS which restarts what was running before the crash.
About the fast boot time, everybody here shrug saying well I never reboot my PC, I say that you're lucky: I have the PC next to my bed and I must stop it at night, I get quite annoyed by the boot time under WindowsXP and it is twice faster to boot under XP than under Mandrake: and that is when I've removed every demon (both under XP and Mandrake).
And about the design I'd like that Mozilla would copy something from Opera: if it has crashed Opera would ask you if you wanted to reopen the web pages previously opened, Mozilla crashes not too often these day, but it still happen, why Mozilla cannot do something similar?
I agree Ati is not cheating: if they were doing this for a game, nobody would complain.
Still I own a Radeon 9800 and I'm pissed off by ATI: there are still bugs in their driver (SplinterCell, and IL2 Forgotten Battles) and they used some developper's time to optimise for 3DMark instead of debugging the driver!!
>I highly doubt that anybody at nVidia or ATI has ever officially stated that the benchmark numbers are ground truths about how their hardware performs relative to other hardware in other programs.
And I don't beleive it either, but benchmarks are here to give an idea of the performance of the card to the consumer. So I call it cheating on their consumer, and I wrote to ATI asking them for an explanation for the difference seen in 3DMark, and telling them I'd like that they focus on correcting their drivers instead of cheating in benchmarks. (I don't care that the difference in 3DMark is lower, I care that they're trying to cheat)
The problem is my veiled threat "go shopping with someone else" is a bit hollow, what card could I buy? Nvidia? The company we're sure has cheated?
Anyway, I hope that the pressure put on ATI & NVidia will cause them to stop trying to deceive their consumer.
>I'd probably write the companies before posting here. I did both.
>NVidia improved their driver so that a certain set of operations runs faster. >There is nothing deceitful about this.
I disagree, if the optimisation were usefull to improve games, then it would be a honest optimisation.
These optimisations were made solely to inflate artifically the score of the benchmark so they are trying to cheat on their customers.
Apparently ATI made the same thing, but it's not sure right now.
The right things to do IMHO for both ATI and NVidia is to promise to stop trying to optimise for benchmarks and to concentrate more on optimising their drivers for games.
I have a Radeon9800Pro and I can tell you that there are still enough bugs in the driver that I would be quite pissed off if I learn that someone at ATI is payed to improve artificially the scores on synthetic benchmarks instead of correcting real bugs in games..
> In France, students learn Caml during their first year before school of engineer.
Since when? When I was in Math Sup/Spe (88/89), we used Pascal to do our computing lesson, I'd like to know when it changed.
Anyway I doubt that functionnal programs will become popular: I learned Lisp during my third years of enginer's school but I never liked it: it looks like noise to me.
Recently I tried to lear OCaml as it seems very fast, but I didn't like it too.. I think that I'm going to use Python or Ruby the next time I need to do programming: they are much more easy to use and they do have some functionnal constructs if needed.
Re:Can't beat my filtering
on
I, Spammer
·
· Score: 1
Why retrain only from time to time?
Each time, the filter fails to mark spam as junk, I mark it manually then delete it. The result? Very, very little spam and only one false positive.
>how the hell does one create and come up with this stuff?
Special Relativity was really "in the air" when Einstein made his paper: I bet that if Einstein wasn't here someone else would have made the discovery.
General Relativity on the other hand, is IMHO the real masterpiece of Einstein: he tried to "apply relativity" to gravity and it didn't work then he thought that there is really no difference between acceleration and gravity and he tried to translate this into equations. If I remember correctly the "translation into equation" phasis took him about 10 years and he had to took math lessons (about tensors).
So the answer to your question is: an imaginative mind and hard work!
The discussion is about using GPU for doing vector processing tasks: the GPU has only 32 bits floatting points whereas usually for scientific tasks, scientists use 64-bit floating point values.
I'm not even sure that the 32-bit floating points value used in the GPU are real IEEE 32-bit floating points value (must have specific handling of very small values for example), I suspect videocard makers cut corners in their implementations which lower precision..
>With that kind of first contact I bet they will understand the nature of humanity quite well and respond appropriately.
They'll send us back blinking light? What is a very strong laser here, would be nothing more than a harmless blinking light when it reaches a remote star.
No! That's part of the problem: if the name were explicit, there would be much less confusion about where to place what..
>/bin bare essential utilities needed to manipulate the system once booted or before multi-user mode Nice to learn it and if it would be moved to/system/single/bin or something like to this, this explanation would have been unnecessary.
The mere fact that you have to explain the meanings of these name to/. users proves that those names should be replaced.
I'm not saying that everything should be in/bin, there is a reason why there are several locations, but the names are too confusing: people places things in the wrong directories because the name or the purpose of these directory is unclear.
Hell, if I remember correctly even the LSB don't say if kde must go in/opt or under/usr..
Usually this internal RISC form is more like and "expanded form" and it is used also by RISC: it isn't truly an instruction set as its intructions has no real length..
>> if you look at all the other CPU except the 80x86, they are all RISC CPUs, or VLIW. >Wrong. Ever heard about ColdFire, AKA M68K?
OK, my point is: each "new CPU" is RISC or VLIW, there is no "new CISC": both x86 and Coldfire exists because of backward compatibility..
>> The fact that 80x86 is CISC is irrelevant >Not for me. I actually enjoy having a fanless, energy-efficient iBook that runs quite fast for its clock
It would be interesting to compare the power consumption of Pentium-M vs PPC, it is not obvious at all that the PPC would win: Intel has one of the best process here.
>> being able to run the cheapest OS (Microsoft OSs) >So is now MS giving away its source code?
No, but back then MS-DOS was cheap and the BSD OS didn't manage to get popular, I'd say because of 1) a lack of marketing --> lack of commercial applications (spreadsheet, games) 2) the need of more expensive CPU to run *BSD instead of MS-DOS which would run happily on a 80286..
Now it is a different games, but MS has an enormous installed base, we'll see..
> In 64-bit mode, do they finally use constant-width instructions or at least limit themselves to 2-byte and 4-byte instructions?
No, they just added yet another prefix, this way they can mix regular x86 instructions and x86-64 instructions. Yes it is another kludge, which will increase the code size of x86-64 code, but at least it gives you 16 registers.
3DNow! may be much cleaner than MMX (it isn't hard, MMX suck so much it isn't funny), but it is also dying a slow death: SSE and SSE2 will replace it eventually.
There is no such thing as being internally RISC: RISC is about a style of Instruction Set (Reduced Instruction Set Cpu).
Still, if you look at all the other CPU except the 80x86, they are all RISC CPUs, or VLIW.
The fact that 80x86 is CISC is irrelevant, 80x86 has won by being 1) cheap 2) being able to run the cheapest OS (Microsoft OSs), with lots of software.
> Apple's market share is stagnant or decreasing and PC sales are flat which implies that for every new user to the platform Apple is loosing an established user.
Uh? I maybe misunderstanding but you seem to imply that the fact that PC sales are flat means that the number of PC user is constant, which isn't true of course..
Slashdot Mirror
> Any "solution" in the OS (including the current MS one for Excel) is going to be a nasty kludge anyway. If I see a "." key on the keypad, hit the key in Excel, and a "," appears - now *that* is a bug :)
Calling it a bug is a bit too harsh IMHO.
I think that it should be customisable in the locale settings.
Or even better, to avoid the possibility of customising the keyboard in two different location, have a redirection to a special subwindow of 'keyboard customisation'.
I think having three posibilities for the numeric dot would be interesting:
- always a dot.
- always the "," in the locale.
- the dot in normal application but the "," in numeric fields.
It is a bit complicated and I don't see how to make it work correctly with xmodmap and the like, but it seems to me the best solution.
>Turning off hyperthreading on a dual-processor system is FLAT CHEATING.
Actually on some benchmark turning off hyperthreading has IMPROVED the score.
I doubt that it is the case here, but I just wanted to tell you that hyperthreading is not necessarily a magic thing which always improves performance.
>If its the size of 1 mm, shouldn't it be called a "milli-engine"?
You're wrong: nanotechnlogy is not about the size of the object produced but about the accuracy used to create the object.
If you created a very big object, by controling precisely the position of each molecules of the object, it would be nanotechnology, even if the object is very big.
But you're right: these object are not nano at all: these are MEMS.
I've looked at the EPIC instruction set and I didn't find it more complicated than the ugly x68 instruction set.
And as for coding in VLIW assembly language complicated enough to code an OS: an OS usually is mostly coded in C, so it isn't really a problem.
And if the assembly language is too complicated, and C too rigid for some parts, nothing prevents the use of some kind of "in-between" language: a C extended with all the operation offered by the CPU for example.
Easy to understand: when Lisp was fashionable when computers where really slow and it wasn't rare to use assembly language because C was too slow.
Now that computers are fast, Python and Perl speed doesn't matter very much.
Well, it IS a problem when your CV that you carefully edited to fit on 1 page, doesn't fit on this page for the people you sent it: it make you look like a fool!
It happened to me once: next time I sent a CV by email, I'll use both a doc and a pdf!
Well I hope that this extension ask before loading the pages!
Because if it loads the page automatically and it is the content of a page which crash the browser..
I wonder why it hasn't been put directly into the browser instead of an extension.
You're guessing too much, nobody said the apps restarted at the exact place playing movies or sound. :
It is just
1) an intelligent design from apps: from time to time they record to the HDD where they are
2) a journalised FS (not something common at the time) so no fsck.
3) a FAST boot time: around 10sec to boot the PC under BeOS.
4) intelligent design of the OS which restarts what was running before the crash.
About the fast boot time, everybody here shrug saying well I never reboot my PC, I say that you're lucky: I have the PC next to my bed and I must stop it at night, I get quite annoyed by the boot time under WindowsXP and it is twice faster to boot under XP than under Mandrake: and that is when I've removed every demon (both under XP and Mandrake).
And about the design I'd like that Mozilla would copy something from Opera: if it has crashed Opera would ask you if you wanted to reopen the web pages previously opened, Mozilla crashes not too often these day, but it still happen, why Mozilla cannot do something similar?
I noticed it, but the problem that I have is different : sometimes some lightsources (lamps or windows for example) are reflected behind the player.
It happens only on some parts of the games: for example on the garage where we must look for Ivan.
When this weird reflection happens it really block a huge part of the view which is annoying.
Bah, this game is quite fun still, but by the time the developpers or ATI corrects all their bugs I will problably already have finished it..
Sometimes a "reflection" of a light appears behind the character "in the air", I don't know if it is a problem in the game or in the driver though.
I'll have a look on the web for the antialiasing bug as I haven't noticed it.
I agree Ati is not cheating: if they were doing this for a game, nobody would complain.
Still I own a Radeon 9800 and I'm pissed off by ATI: there are still bugs in their driver (SplinterCell, and IL2 Forgotten Battles) and they used some developper's time to optimise for 3DMark instead of debugging the driver!!
>I highly doubt that anybody at nVidia or ATI has ever officially stated that the benchmark numbers are ground truths about how their hardware performs relative to other hardware in other programs.
And I don't beleive it either, but benchmarks are here to give an idea of the performance of the card to the consumer.
So I call it cheating on their consumer, and I wrote to ATI asking them for an explanation for the difference seen in 3DMark, and telling them I'd like that they focus on correcting their drivers instead of cheating in benchmarks.
(I don't care that the difference in 3DMark is lower, I care that they're trying to cheat)
The problem is my veiled threat "go shopping with someone else" is a bit hollow, what card could I buy?
Nvidia? The company we're sure has cheated?
Anyway, I hope that the pressure put on ATI & NVidia will cause them to stop trying to deceive their consumer.
>I'd probably write the companies before posting here.
I did both.
>NVidia improved their driver so that a certain set of operations runs faster.
>There is nothing deceitful about this.
I disagree, if the optimisation were usefull to improve games, then it would be a honest optimisation.
These optimisations were made solely to inflate artifically the score of the benchmark so they are trying to cheat on their customers.
Apparently ATI made the same thing, but it's not sure right now.
The right things to do IMHO for both ATI and NVidia is to promise to stop trying to optimise for benchmarks and to concentrate more on optimising their drivers for games.
I have a Radeon9800Pro and I can tell you that there are still enough bugs in the driver that I would be quite pissed off if I learn that someone at ATI is payed to improve artificially the scores on synthetic benchmarks instead of correcting real bugs in games..
> In France, students learn Caml during their first year before school of engineer.
Since when?
When I was in Math Sup/Spe (88/89), we used Pascal to do our computing lesson, I'd like to know when it changed.
Anyway I doubt that functionnal programs will become popular: I learned Lisp during my third years of enginer's school but I never liked it: it looks like noise to me.
Recently I tried to lear OCaml as it seems very fast, but I didn't like it too..
I think that I'm going to use Python or Ruby the next time I need to do programming: they are much more easy to use and they do have some functionnal constructs if needed.
Why retrain only from time to time?
Each time, the filter fails to mark spam as junk, I mark it manually then delete it.
The result?
Very, very little spam and only one false positive.
>how the hell does one create and come up with this stuff?
Special Relativity was really "in the air" when Einstein made his paper: I bet that if Einstein wasn't here someone else would have made the discovery.
General Relativity on the other hand, is IMHO the real masterpiece of Einstein: he tried to "apply relativity" to gravity and it didn't work then he thought that there is really no difference between acceleration and gravity and he tried to translate this into equations.
If I remember correctly the "translation into equation" phasis took him about 10 years and he had to took math lessons (about tensors).
So the answer to your question is: an imaginative mind and hard work!
The discussion is about using GPU for doing vector processing tasks: the GPU has only 32 bits floatting points whereas usually for scientific tasks, scientists use 64-bit floating point values.
I'm not even sure that the 32-bit floating points value used in the GPU are real IEEE 32-bit floating points value (must have specific handling of very small values for example), I suspect videocard makers cut corners in their implementations which lower precision..
>With that kind of first contact I bet they will understand the nature of humanity quite well and respond appropriately.
They'll send us back blinking light?
What is a very strong laser here, would be nothing more than a harmless blinking light when it reaches a remote star.
And do you think those name are meaningfull?
/bin bare essential utilities needed to manipulate the system once booted or before multi-user mode /system/single/bin or something like to this, this explanation would have been unnecessary.
/. users proves that those names should be replaced.
/bin, there is a reason why there are several locations, but the names are too confusing: people places things in the wrong directories because the name or the purpose of these directory is unclear.
/opt or under /usr..
No! That's part of the problem: if the name were explicit, there would be much less confusion about where to place what..
>
Nice to learn it and if it would be moved to
The mere fact that you have to explain the meanings of these name to
I'm not saying that everything should be in
Hell, if I remember correctly even the LSB don't say if kde must go in
>You translate external CISC into internal RISC
Usually this internal RISC form is more like and "expanded form" and it is used also by RISC: it isn't truly an instruction set as its intructions has no real length..
>> if you look at all the other CPU except the 80x86, they are all RISC CPUs, or VLIW.
>Wrong. Ever heard about ColdFire, AKA M68K?
OK, my point is: each "new CPU" is RISC or VLIW, there is no "new CISC": both x86 and Coldfire exists because of backward compatibility..
>> The fact that 80x86 is CISC is irrelevant
>Not for me. I actually enjoy having a fanless, energy-efficient iBook that runs quite fast for its clock
It would be interesting to compare the power consumption of Pentium-M vs PPC, it is not obvious at all that the PPC would win: Intel has one of the best process here.
>> being able to run the cheapest OS (Microsoft OSs)
>So is now MS giving away its source code?
No, but back then MS-DOS was cheap and the BSD OS didn't manage to get popular, I'd say because of
1) a lack of marketing --> lack of commercial applications (spreadsheet, games)
2) the need of more expensive CPU to run *BSD instead of MS-DOS which would run happily on a 80286..
Now it is a different games, but MS has an enormous installed base, we'll see..
> In 64-bit mode, do they finally use constant-width instructions or at least limit themselves to 2-byte and 4-byte instructions?
No, they just added yet another prefix, this way they can mix regular x86 instructions and x86-64 instructions.
Yes it is another kludge, which will increase the code size of x86-64 code, but at least it gives you 16 registers.
3DNow! may be much cleaner than MMX (it isn't hard, MMX suck so much it isn't funny), but it is also dying a slow death: SSE and SSE2 will replace it eventually.
There is no such thing as being internally RISC: RISC is about a style of Instruction Set (Reduced Instruction Set Cpu).
Still, if you look at all the other CPU except the 80x86, they are all RISC CPUs, or VLIW.
The fact that 80x86 is CISC is irrelevant, 80x86 has won by being 1) cheap 2) being able to run the cheapest OS (Microsoft OSs), with lots of software.
I agree that performance per $$ is the only interesting benchmark but the thing is: Athlon64 are not sold at the moment!
So in the meantime, benchmarks which show the performance per clock of the Athlon64 have their place..
Sorry but Java has already produced a child: C#.
So for Java to be an evolutionary dead-end both Java and C# should die: C# is pushed by Microsoft, Java by all the other (Sun,IBM..).
Both won't die soon..
> Apple's market share is stagnant or decreasing and PC sales are flat which implies that for every new user to the platform Apple is loosing an established user.
Uh? I maybe misunderstanding but you seem to imply that the fact that PC sales are flat means that the number of PC user is constant, which isn't true of course..