Furthermore, I am not saying that an Alpha chip is like a Pinto and an AMD chips is like a Porsche
I surely hope you don't... An Alpha at 150 MHz is roughly equivalent to a PPro/PII at 800 - 1000 MHz, for floating-point intensive programs. Partly because the Alpha is just blazingly fast for FP operations, and partly because the x86 FP architecture sucks. When they developed the 8087 originally, they chose a design which is flawed from beginning to end, but they couldn't get it to work any other way... And now we're stuck with it.
You save the time and effort, don't ask the user the year, and the user has to make a minor correction once every four years (roughly). Is choice #2 really that ugly?
I have to make a correction to my watch every other month (almost), and I'm perfectly happy with that arrangement...
Computers are a different story, but I trust my timezone files with that. They even get daylight savings time on and off at the right dates and hours (which my watch doesn't do either).
Name 1 - 1 task - that requires a supercomputer that can't be broken down into nodes well
Sure, I can name several. Just some examples: Weather simulation. Ocean simulation. Molecular simulations. Simulation of astronomical bodies. All of which are very real problems.
In short, any problem which is not trivially parallel will get a much poorer speedup on a NOW (network of workstations) versus a real supercomputer. Many of the problems above will generate many MB/s of data per processor (60 - 200 MB/s is not uncommon).
What you fail to realize is that many problems run for many iterations, and for each iteration you need to distribute the global dataset to all worker nodes. Take the Barnes-Hut program, for example. In that program, each node get a set of close-by astral bodies (stars and planets), and calculates their new positions for the next time step. To do that you need the positions of all other stars. For the next time step, you need to a) collect the calculated positions from all worker nodes, and b) distribute them back for the next iteration. When trying to run that on a NOW, you will very soon find that doubling the size of the cluster will not give any speedup at all, since they will spend most of the time chatting with each other on the network. On a supercomputer, you can run many more worker nodes before this happens.
You need a pointer to the first entry in the list. Assuming that the last "next"-pointer and the first "prev"-pointer is NULL, it's easy to go from there. Sample:
p = first; prev = NULL; while (p) { (do stuff with *p) tmp = p->xoredstuff ^ prev; prev = p; p = tmp; }
Traversing in the other direction is left as an exercise for the reader.
Argh! Moving stuff to hardware, when we have long struggled to get rid of esoteric hardware instructions?!? The whole idea behind RISC is to remove this kind of stupidness.
History is full of examples where this kind of hardware "features" made things faster originally, but has since become bottlenecks instead of improvements. "rep movsb" made sense on the 8086, but on a Pentium Pro it's just another thing which makes the chip more complicated, big, and power-comsuming.
And as for garbage collection, those of you who say that "GC is slow" are just plain wrong. There is a lot of research on garbage collection, and nowadays a good GC may be faster than manual malloc/freeing. It does not lock up your programs either - there are incremental GC's which you won't notice. They are even used in real-time environments today.
A major hindrance for good GC's in commonplace programs is that people usually write in C(++), and you can't have a good GC in a programming language which allows pointers. Just to give an example: I once saw a program which used a doubly-linked list. But instead of storing "next" and "prev" pointers, it only used one value - the XOR of prev and next! It was possible to traverse, since you always knew the address of either the previous or the next entry, so getting the other was just an XOR operation. But please tell me, how the h* is a GC algorithm supposed to know this? Even if you use the loosest approach, "treat every single word in memory as a possible pointer", it will fail - since there are no real pointers at all!
So the result is this: There are lots of good GC's around, but they can't be used in C - so people don't know that they exist. More or less.
What's greatly needed for larger systems is often a lot of main memory. The standard PC limit of 1 GB main memory is a major pain in the @$$ when you need some serious computation done.
At my department we need machines with lots and lots of memory. Currently we have a 4-CPU Sun with 4 GB memory, but that's really too small. Around 24 GB would be more along our needs. (Coding and crypto research, eats lots of MB's).
You can get a lot of memory by using clustered machines (or an Origin 2000, SP/2 or whatever), but it's kind-of silly to use a parallel computer just to get the memory, but not the parallelism...
Anyway, the sooner we can get more than 1-4 GB into a standard PC the better. The need is already here.
I don't have an IPX, but I do have a nice little LX at home. And the other day I got hold of an SS5 too:-) Our sysadm is trying to get rid of a lot of older Suns, and you just can't let them go to waste... A few Ultra 1's are next to go, but I don't think I can get one of those. They are still too useful - they will go to other departments instead:(
do you know know about the -K (that's a capital) switch for man?
No, and it doesn't seem like I'll ever find out what it's supposed to do:
~>man -K man: invalid option -- K
But I think manpages rule anyway. Always the same sections in the same order, and everything on one page, easily searched in less. Compare that to the absolutely horrible "info" format, where the information is split up into myriads of small sections, and you have to navigate yourself to death before you find what you're looking for. And when you've found it, there's no easy way to get to the other stuff you wanted to know, which means even more frustrating navigation.
And as a special bonus, its very easy to get a nice-looking hardcopy of a manpage, just use "man -Tps proggie | lpr". (I am definately not a big fan of info, does that show?:))
No, I saw the words "Nano Logo", and thought that Hemos wanted a new logo for the nanotech postings. I was almost aiming for the Gimp until I read the rest of the letters in the story... And it wasn't even posted by Hemos.
Why would anyone want to add newlines, tabs or whitespace?
Have you seen the previous entries? A great example is the pi program from 1988 (westley.c). Without whitespace the whole charm of the program would be lost.
Creative indentions can often make a program much sweeter in its source code appearance.
About this games4linux link, couldn't you put up something more interesting than the Apache default page? Won't take more than a minute to write a page... ~ ~ ~ ~ "index.html" 2 lines, 118 characters
Assuming Norways legal system works as in Sweden, this is simply not possible. "Deals" are explicitly disallowed in Swedish law. You can't get any "favours" by cooperating, because it is assumed that the same crime should always give the same punishment ("everybody is equal to the law", something like that).
Although I hear lately that the courts have been found to do this kind of stuff in semi-secret, even though it's illegal. There is an investigation going on, don't know what they will find, though.
As a programmer, I am always disturbed by "closing" tags without an "opening" tag (or vice versa). A C compiler would complain if you tried to use only }'s and never any {'s, just as an example. So I always "balance" my HTML (not that I write much html anyway, webpage creation is something I think other people can do) by using both p and/p, tr and/tr, td and/td, etc. The BR tag still disturbs me, since it doesn't have any matching closing tag. Not that it would make much sense, but anyway.
This gives me the occasion to ask something I've been wondering about a long time - How come all american radio stations have these funny four-letter names? I mean, KXYZ doesn't mean a thing. Over here, they are usually called something along the lines of "Megapol", or "Radio Kristianstad", or something with the frequency in the name, like "City 107" or "Gold 105". I find that a wee bit easier to remember than "something with four letters, beginning with a K, and, oh yes, there was a W in there somewhere"...
Not intended as any kind of flamebait, I'm just curious about these strange letters.
Just an addendum: Most TV:s have four pages in memory, so if you view page 601 and want to flip to 602, that one is already loaded (mostly).
There are also (more expensive) TV:s which can cache several hundred pages. With one of those, you seldom have to wait for a page to be broadcast before you can view it.
And, there are always web gateways to use if you have a web browser handy.
Just a small point: The main reason they can't release the console everywhere at once is that it is simply not possible to manufacture the required amount of consoles (you would need millions of them in stock). So, they sell them at home first, and when the rush declines there, they will have some to send to the US (and we Europeans get it last, as always:()
The Dreamcast hasn't even been advertised in Sweden, and they are still sold out until March or so (they figured it would be stupid to make ads for something already sold out). If it was software they had to deal with, it's easy enough to press a few million CD:s, but hardware needs to be built, which can only be done so fast.
But, I think (and hope) that the PS2 will be a success - I know I will buy one.
Strange, I have the same problem, but the other way around: My (old, admittedly) monitor works perfectly in X, but not in Windows. The reason being that to run 800x600 and higher it requires an interlaced video mode, and there are no windows drivers which will give me that. With XF86Setup, I just enter "30-38 kHz, 50-100 Hz" and it just works, giving me 1024x768 in 16-bit color. In Windows, on the other hand, I can at most get a distorted 800x600, at the very edge of the monitor's capabilities.
The only way I can think of to get 1024x768 in Windows is to switch in my trusty old Trident 8900C, which is only capable of interlaced video in high resolutions, but then I can't get an acceptable colordepth... And that's one of the reasons I never run Windows on my box, except for emergencies (maybe once a year or so).
Slashdot Mirror
For some obscure reason, this sounds like an OS designed for men with vests and Harley-Davidsons...
Furthermore, I am not saying that an Alpha chip is like a Pinto and an AMD chips is like a Porsche
I surely hope you don't... An Alpha at 150 MHz is roughly equivalent to a PPro/PII at 800 - 1000 MHz, for floating-point intensive programs. Partly because the Alpha is just blazingly fast for FP operations, and partly because the x86 FP architecture sucks. When they developed the 8087 originally, they chose a design which is flawed from beginning to end, but they couldn't get it to work any other way... And now we're stuck with it.
I really can't be arsed to do an x86 assembler version :)
Will MIPS do? :-)
la t0,year
lw t1,0(t0)
add t5,zero,zero
andi t2,t1,3
bnz t2,1f
addi t5,zero,1
addi t0,zero,100
div t1,t0
mflo t0
bnz t0,1f
addi t0,zero,400
div t1,t0
mflo t0
beqz t0,1f
add t5,zero,zero
1:
la t0,isleap
sw t5,0(t0)
I could give you sparc, M68k or x86 code also, but this will have to do for now...
You save the time and effort, don't ask the user the year, and the user has to make a minor correction once every four years (roughly). Is choice #2 really that ugly?
I have to make a correction to my watch every other month (almost), and I'm perfectly happy with that arrangement...
Computers are a different story, but I trust my timezone files with that. They even get daylight savings time on and off at the right dates and hours (which my watch doesn't do either).
Name 1 - 1 task - that requires a supercomputer that can't be broken down into nodes well
Sure, I can name several. Just some examples: Weather simulation. Ocean simulation. Molecular simulations. Simulation of astronomical bodies. All of which are very real problems.
In short, any problem which is not trivially parallel will get a much poorer speedup on a NOW (network of workstations) versus a real supercomputer. Many of the problems above will generate many MB/s of data per processor (60 - 200 MB/s is not uncommon).
What you fail to realize is that many problems run for many iterations, and for each iteration you need to distribute the global dataset to all worker nodes. Take the Barnes-Hut program, for example. In that program, each node get a set of close-by astral bodies (stars and planets), and calculates their new positions for the next time step. To do that you need the positions of all other stars. For the next time step, you need to a) collect the calculated positions from all worker nodes, and b) distribute them back for the next iteration. When trying to run that on a NOW, you will very soon find that doubling the size of the cluster will not give any speedup at all, since they will spend most of the time chatting with each other on the network. On a supercomputer, you can run many more worker nodes before this happens.
You need a pointer to the first entry in the list. Assuming that the last "next"-pointer and the first "prev"-pointer is NULL, it's easy to go from there. Sample:
p = first;
prev = NULL;
while (p) {
(do stuff with *p)
tmp = p->xoredstuff ^ prev;
prev = p;
p = tmp;
}
Traversing in the other direction is left as an exercise for the reader.
Argh! Moving stuff to hardware, when we have long struggled to get rid of esoteric hardware instructions?!? The whole idea behind RISC is to remove this kind of stupidness.
History is full of examples where this kind of hardware "features" made things faster originally, but has since become bottlenecks instead of improvements. "rep movsb" made sense on the 8086, but on a Pentium Pro it's just another thing which makes the chip more complicated, big, and power-comsuming.
And as for garbage collection, those of you who say that "GC is slow" are just plain wrong. There is a lot of research on garbage collection, and nowadays a good GC may be faster than manual malloc/freeing. It does not lock up your programs either - there are incremental GC's which you won't notice. They are even used in real-time environments today.
A major hindrance for good GC's in commonplace programs is that people usually write in C(++), and you can't have a good GC in a programming language which allows pointers. Just to give an example: I once saw a program which used a doubly-linked list. But instead of storing "next" and "prev" pointers, it only used one value - the XOR of prev and next! It was possible to traverse, since you always knew the address of either the previous or the next entry, so getting the other was just an XOR operation. But please tell me, how the h* is a GC algorithm supposed to know this? Even if you use the loosest approach, "treat every single word in memory as a possible pointer", it will fail - since there are no real pointers at all!
So the result is this: There are lots of good GC's around, but they can't be used in C - so people don't know that they exist. More or less.
What's greatly needed for larger systems is often a lot of main memory. The standard PC limit of 1 GB main memory is a major pain in the @$$ when you need some serious computation done.
At my department we need machines with lots and lots of memory. Currently we have a 4-CPU Sun with 4 GB memory, but that's really too small. Around 24 GB would be more along our needs. (Coding and crypto research, eats lots of MB's).
You can get a lot of memory by using clustered machines (or an Origin 2000, SP/2 or whatever), but it's kind-of silly to use a parallel computer just to get the memory, but not the parallelism...
Anyway, the sooner we can get more than 1-4 GB into a standard PC the better. The need is already here.
I don't have an IPX, but I do have a nice little LX at home. And the other day I got hold of an SS5 too :-) Our sysadm is trying to get rid of a lot of older Suns, and you just can't let them go to waste... A few Ultra 1's are next to go, but I don't think I can get one of those. They are still too useful - they will go to other departments instead :(
do you know know about the -K (that's a capital) switch for man?
No, and it doesn't seem like I'll ever find out what it's supposed to do:
~>man -K
man: invalid option -- K
But I think manpages rule anyway. Always the same sections in the same order, and everything on one page, easily searched in less. Compare that to the absolutely horrible "info" format, where the information is split up into myriads of small sections, and you have to navigate yourself to death before you find what you're looking for. And when you've found it, there's no easy way to get to the other stuff you wanted to know, which means even more frustrating navigation.
And as a special bonus, its very easy to get a nice-looking hardcopy of a manpage, just use "man -Tps proggie | lpr". (I am definately not a big fan of info, does that show? :))
No, I saw the words "Nano Logo", and thought that Hemos wanted a new logo for the nanotech postings. I was almost aiming for the Gimp until I read the rest of the letters in the story... And it wasn't even posted by Hemos.
Well, since it's called America instead of Colombia, at least someone must have remembered Amerigo Vespucci...
Why would anyone want to add newlines, tabs or whitespace?
Have you seen the previous entries? A great example is the pi program from 1988 (westley.c). Without whitespace the whole charm of the program would be lost.
Creative indentions can often make a program much sweeter in its source code appearance.
Oh yeah, so where's the obfuscated Intercal code contest?
Oh, maybe we should drop that...
The way I've seen it is
"C program run
C program crash
C programmer cry"...
This is the standard behaviour in my empiric tests, anyway.
About this games4linux link, couldn't you put up something more interesting than the Apache default page? Won't take more than a minute to write a page...
~
~
~
~
"index.html" 2 lines, 118 characters
Assuming Norways legal system works as in Sweden, this is simply not possible. "Deals" are explicitly disallowed in Swedish law. You can't get any "favours" by cooperating, because it is assumed that the same crime should always give the same punishment ("everybody is equal to the law", something like that).
Although I hear lately that the courts have been found to do this kind of stuff in semi-secret, even though it's illegal. There is an investigation going on, don't know what they will find, though.
Thank you! Although the "3myst.htm" document (in the same place) explained a bit more.
As a programmer, I am always disturbed by "closing" tags without an "opening" tag (or vice versa). A C compiler would complain if you tried to use only }'s and never any {'s, just as an example. So I always "balance" my HTML (not that I write much html anyway, webpage creation is something I think other people can do) by using both p and /p, tr and /tr, td and /td, etc. The BR tag still disturbs me, since it doesn't have any matching closing tag. Not that it would make much sense, but anyway.
This gives me the occasion to ask something I've been wondering about a long time - How come all american radio stations have these funny four-letter names? I mean, KXYZ doesn't mean a thing. Over here, they are usually called something along the lines of "Megapol", or "Radio Kristianstad", or something with the frequency in the name, like "City 107" or "Gold 105". I find that a wee bit easier to remember than "something with four letters, beginning with a K, and, oh yes, there was a W in there somewhere"...
Not intended as any kind of flamebait, I'm just curious about these strange letters.
I wouldn't be surprised if it did. IBM supports Linux, and it's already running on RS/6000, so why shouldn't it run on an SP cluster?
For the moment, though, I would recommend you to use AIX instead, since it has been in the playground a while longer for this kind of systems.
You haven't studied your INTERCAL manual... ABSTAIN FROM is a keyword in that wonderful language.
No, I don't say that you should try to write anything useful in intercal, but "he who knows the most programming languages when he dies wins" :-)
Just an addendum: Most TV:s have four pages in memory, so if you view page 601 and want to flip to 602, that one is already loaded (mostly).
There are also (more expensive) TV:s which can cache several hundred pages. With one of those, you seldom have to wait for a page to be broadcast before you can view it.
And, there are always web gateways to use if you have a web browser handy.
Just a small point: The main reason they can't release the console everywhere at once is that it is simply not possible to manufacture the required amount of consoles (you would need millions of them in stock). So, they sell them at home first, and when the rush declines there, they will have some to send to the US (and we Europeans get it last, as always :()
The Dreamcast hasn't even been advertised in Sweden, and they are still sold out until March or so (they figured it would be stupid to make ads for something already sold out). If it was software they had to deal with, it's easy enough to press a few million CD:s, but hardware needs to be built, which can only be done so fast.
But, I think (and hope) that the PS2 will be a success - I know I will buy one.
Strange, I have the same problem, but the other way around: My (old, admittedly) monitor works perfectly in X, but not in Windows. The reason being that to run 800x600 and higher it requires an interlaced video mode, and there are no windows drivers which will give me that. With XF86Setup, I just enter "30-38 kHz, 50-100 Hz" and it just works, giving me 1024x768 in 16-bit color. In Windows, on the other hand, I can at most get a distorted 800x600, at the very edge of the monitor's capabilities.
The only way I can think of to get 1024x768 in Windows is to switch in my trusty old Trident 8900C, which is only capable of interlaced video in high resolutions, but then I can't get an acceptable colordepth... And that's one of the reasons I never run Windows on my box, except for emergencies (maybe once a year or so).