High-resolution timers in Linux are a joke. The early POSIX-RT patches simply multiplied gettimeofday() by 1000.
The best way to get performance data on linux or windows is via the Intel chip's time-stamp counter; here's some example gcc code to do it:
static unsigned long long rdtsc(void) {
register unsigned long long d;
__asm__ __volatile__ ("rdtsc" : "=A"(d));
return d;
}
The previous method takes about 13 cycles on an Athlon 750. (DO NOT try and make it inline -- or gcc might optimize your to-be-timed code out from between the rdtsc() calls.) It is a straightforward manner to read the cpu clock speed from/proc/cpuinfo and use that to convert cycles into milli- (or micro-) seconds.
As with any timing method, take care to execute it a few times before you gather any information, to prime the i-cache.
Apparently the lameness filter believes that this is a "junk character post", so I'll type some more. Intel has a useful whitepaper that describes how to do this in an M$ compiler, available here: http://developer.intel.com/software/idap/resources/technical_collateral/pentiumii/RDTSCPM1.HTM
Avoiding any possible M$ jokes? You must be kidding. M$ is capable of producing software that will do nothing, unsuccessfully, a la "An unknown program has terminated in an unknown module with an unknown error. Please reboot your system."
Any "clueful geek" who "knows quite a bit" is not trying hard enough if he can't get a "geek job", even if he is "self-taught". One of the best consultants I've ever known was completely self-taught and pulls down probably 15 times the theoretical maximum wage in a factory job. Eric S. Raymond actually brags about having no computer science education in his resume.
Why not get your name attached to some high-profile projects? Why not even get a menial "web page design" job through a local temp agency, and then work your way up? High-school students turn jobs like that down! Even "PC technicians" at mass-makret retail stores can find employment in a more interesting and profitable sector of IT eventually.
Honestly, it's very hard to believe that -- if you're really that good -- you're stuck gutting fish or operating a forklift all day as your "geek satyagraha" seeps out from your atrophied wrists.
You don't need "friend" for this -- "friend" is merely a bit of syntactic sugar that people who don't know how to use the Visitor Pattern or double dispatch must resort to to get bidirectional relationships.
In your particular example, you'd want to use a seperate object to represent the borrower-book relationship (likely a State pattern object), have a method in the Book class which registers itself with a Borrower argument (or a BookVisitor argument, more generally), establishing the relationship thusly.
As if that isn't bad enough, for certain specific sites Yahoo charges $600 just to look at your site.
Like what certain specific sites? Like your "watch me take off my shirt and program html" site? You got into that business about five years too late. Sorry, but I can't bring myself to have sympathy for a smut site operator who complains about the rest of the net treating her unfairly.
I guess it's too bad that you can't just use your breasts to make money on the 'net without any initial investment, but even the world's oldest profession requires you to spend money to make it.
I have a masters degree (Mathematics). For as much as it was a fun way to spend the time, it has had no bearing on my career whatsoever. In fact, apart from
getting me in the door at a few short-sighted companies, my undergraduate degree was pretty worthless, too.
I don't think you can make that comparison. A masters' in maths is basically the consolation prize for quitting (or failing out of) a Ph.D. Sure, you're not using it, because a masters' in maths is designed to equip you to do maths research. On the other hand, a traditional masters' in CS can take one of two paths: as prepatory for research or as prepatory for industry.
Add to this the fact that many schools are also offering the a terminal CS masters' (sometimes called a MCS instead of an MS), and a masters' in CS is a clear advantage. Fact is that many students never learn design skills or many critical big-picture CS skills, and a masters' can provide valuable training as well as providing experience working on a large project.
It really depends on what kind of software development you want to do. Do you want to crank out perl scripts for some dot-con, or do you want to work on large, critical systems? I'd say that in the case of the former, you're wasting your time in college. If it's the latter, though, then find a grad program with a strong engineering and methodologies emphasis as well as strong fundamentals and do the MS. It'll only take you a year-and-a-half if you're supporting yourself, but you ought to try to get a research job if they offer you support.
Ever read Kafka? I'd suggest you start there. Not for the conspiracies, but for the soul-numbing bureaucracy of it all. These are the same people who've sponsored the proliferation of Ada, for Pete's sake.
In all seriousness, working for the government is an easy gig if you're not the sort to want to work terribly hard (you won't, and they won't encourage you to) and if you don't take your job home with you -- or particularly seriously. If you are terribly career-driven or are a perfectionist, you will be driven mad.
Also: you cannot accept gifts of any kind while acting as a government employee. That means no free dinners, no fancy hotels at conferences where you're giving a talk, and (worst of all) no frequent flyer miles.
If you honestly can't see why statements such as "X has to be written in N instructions" are inherently platform specific, then I can only conclude that you
are, in fact, mildly retarded.
As much as I regret to feed this troll again, allow me to rebut. Firstly, my "mild retardation" must make my D. Phil (that's "Ph.D." for you Yanks) in computer science all the more impressive. Secondly, neither Gray nor I said "X has to be written in N instructions"; rather, the statement was "less than 100 instructions" and "less than 10 instructions". For those of you who are having trouble following, those numbers are what people who have at least high school maths experience refer to as "orders of magnitude". So that's "a lock must be set 'on the order of' 100 instructions, as opposed to a latch, which must be an order of magnitude faster.
A blanket statement such as the above can't possibly be what he actually said. Now if he were talking about some specific RDBMS on some specific platform it might be a perfectly reasonable thing to say.
I was there, "Pete". You weren't. You also apparently do not have any computer science training, or you would know that a "specific RDBMS on a specific platform" is not only likely incredibly far from the state of the art (if it is a commercial system), but is also not interesting in general to actual researchers, unless they are writing a benchmarking paper to show why they aren't interested in commercial systems.
In the RDBMS case, the maximum size of the locking code would depend on the computer, OS, and lots o' other stuff.
Actually, you're quite wrong. First, note that I said "instructions", not "cycles". All modern processors provide a test-and-set or compare-and-swap instruction. So it does not depend at all the architecture. RDBMS locking is not done at the OS level -- that is FAR too much overhead. An RDBMS maintains its own locks, its own file structure, etc. Typical instruction-count costs for seeking through a highly-optimized B+-tree, doing a hash join, or marking a page as "dirty" have not changed since the introduction of virtual memory support in hardware, and neither has the cost of setting and releasing locks. Furthermore, for a scalable system, the cost times for updating those locks need to be as Gray said.
Kindly find someone else to troll, or perhaps ask your parents to buy you the newest Quake game if you're really frustrated.
Are these some fundamental laws of computer science?
No, they're remarks from a SIGMOD talk given by Jim Gray -- perhaps you've heard of him. While the vast majority of RDBMS code is written in C, the locking routines are ALL assembly, because they are so time-critical.
The Solaris kernel is between 10-15% fine-tuned assembly language. You think Duff's Device is odd? Wait until you see the Solaris bcopy() routine. It's like a switch statement without labels, except harder to understand.
Also, RDBMS vendors will likely always use assembly. The locking code for an RDBMS needs to operate in fewer than 100 instructions to guarantee the possibility of concurrency -- and that includes deadlock detection. Furthermore, the "latches", or internal locks on the buffer pool, etc., must execute in under 10 instructions.
Assembly language is important, especially since highly-optimized C code is faster than unoptimized C code by a factor of 5 or more. How do you think that those optimizations get into the compiler? Sure, no one writes applications -- or even, at this point, games -- in assembly, but today's processors are so complex that
You NEED to hand-optimize some speed-critical things
A simple syntax-directed translation scheme is more inadequate than ever for today's compilers -- if you're writing compilers, you really get to worry about speed.
do slashdogs ALWAYS reply without reading links?
on
MP3 Recorders?
·
· Score: 2
Expect to see products using this chip within 6-9 months.... As for products available at present, Idon't believe there are any, though the Nomad II will record voice using a built in mic at 32kbps.
Look, no one is saying that such a product is not feasible. However, the industry will keep such products from ever coming to market in the US or EU. Because of the SDMI failure, the music industry cannot control MP3, and thus will do everything to prevent it. Look at how they killed DAT if you need an example.
As for your assertion that the Nomad will record voice at 32 kbps, that is just plain ridiculous. You can get better fidelity than a 32k mp3 with a knitting needle and a wax record on a hot day. The question specifically states that the biologist needs at least 96kbps, and I'm willing to bet that the "built-in mic" is going to be horribly inadequate as well.
If there is a hi-fidelity MP3 recorder that allows ordinary consumers to produce non-watermarked mp3s and transfer them to computer within the next year, I will publically recant. However, there is no chance of this. Read the cryptome article if you don't believe me.
MP3 recorders will never exist
on
MP3 Recorders?
·
· Score: 3
Look at this excellent rant for why mp3 recorders will never exist. Minidisc is your best bet. You can get up to 149 minutes of monaural on a single minidisc, and the results will be very high quality. I personally used a minidisc for ornithology research back at Uni, with great success.
OpenLink Software has released JBench, their TPC-A and TPC-C implementation under the GPL. The catch? You have to fill out a form to download it as part of their (non-free) Virtuoso db suite.
Perhaps he was interested in royalties from the to-be-released and oft-anticipated "Hellmouth Team Tournament" series, pitting pimply, maladjusted malcontents against unarmed, God-fearing jocks and shrieking, well-endowed cheerleaders.
And: "already favors expensive seminars sold to clueless PHB's"
Just doing my part to keep the bullshit level down around here.
Actually, no, it's "favours". Show respect to the nation who created the language which you so readily butcher.
As an engineer who was recently "demoted" to management, I take extreme umbrage at your comments. XP works because developers love it, not because management bought a seminar (for the record, I have never attended a "management seminar" and I do quite well, because I have the engineer's perspective) These benefits to the developer "trickle-up" in the form of more reliable code and fewer missed deadlines. If it didn't work, it would fall into disuse.
No development methodology is a panacaea, but XP has been invaluable in my group. That's all I'm saying.
Well, ESR certainly has lived a charmed life. He says "I gave up having a job 15 years ago...if I do what I want to do, then the money comes." Does anyone know how he is able to subsist on "doing what he wants to do?" You guessed it. His wife is a lawyer, and she supports him.
ESR is a one-trick pony, desperate to hold on to his "status" by publishing diatribes which are merely "n+1" from his previous screeds. He needs to get out of the way. The community doesn't need him anymore.
As Suck said in their brilliant parody of Slashdot, "RMS, GNU, ESR: FOAD".
Extreme Programming, as advocated by refactoring and OO gurus, already favors some similar approaches, especially programming in pairs and functionality-oriented design and testing through collaborative meetings between developers who discuss "User Stories", or anecdotal program requirements from users.
It's worked wonders in my organization, and I suspect that the "war room" approach lends itself to similar types of productivity gains.
What fantasy world do you live in? I can think of three examples of companies whose budgets I know off the top of my head (all Fortune Global 100, two pharmas and a computer company), and they each spend less than 1% of the amount that they spend on research on lawyers. Your imaginary world of "companies which hire lawyers to extort and steal instead of plunking the vast majority of their revenue into developing products" does not exist.
Let's think about this: in a research-driven industry, when you have time-limited patent protection, you need to (gasp!) keep doing research in order to retain profitability.
Most/. readers need not fear -- the list of creditcards.com affiliates reads like a veritable "Who's not who" of "e-tailing". (The full list is available at their site.)
The list includes such e-nobodies as "iKnowledge", "eCashier", "SpyGate", and the "Christian Concert Authority." And those are the more plausible-sounding of the bunch...
How on earth is this post "overrated"? It was at score 1. It hadn't been "rated" at all. This post was not inflammatory, insulting, or factually incorrect -- it was just an unpopular view. I'd really like an explanation from the moderator here.
Why is it that Microsoft shills always resort to ad hominem attacks to respond to allegations about the types of programmers that Microsoft's inferior wares attract? Show me a single VB programmer who is truly capable of architecting a complex software system (other than Steve McConnell, whose behavior of late is inexplicable), and I'll recant.
However, I strictly follow a reader-response theory of language critcism -- I judge languages by the sorts of programmers who are most attracted to them. Calling VB "Venereal Basic" is not a "childish insult"; it's funny, and it's an apt metaphor, considering that VB is a painful disease and most well-to-do people are embarassed to discuss it in polite company.
Do you think that someday, a project like plex86 and a project like Bochs might combine in kernel space to produce a lightweight virtual machine that could run x86 code?
Did you actually read that Wiki page that you linked to? Here's a quote that demonstrates that you're the flamer, not I:
The last pair of these are only issues with static type-checking, and the first can be done quite nicely with delegation idioms (especially based around
#doesNotUnderstand:), so we don't much need MI in Smalltalk. Java covers the last pair with interfaces but loses out sorely on the first. It has nothing
"multiple" there. The other languages need MI because they don't have decent delegation and they do have static types.
How is the fact that Disney still has a copyright on Mickey Mouse harming you? Oh, you can't sell DivX:-) rips of "Steamboat Willie"? That's a shame. Oh wait, that's it. Have I missed anything?
Slashdot Mirror
High-resolution timers in Linux are a joke. The early POSIX-RT patches simply multiplied gettimeofday() by 1000.
/proc/cpuinfo and use that to convert cycles into milli- (or micro-) seconds.
s /technical_collateral/pentiumii/RDTSCPM1.HTM
The best way to get performance data on linux or windows is via the Intel chip's time-stamp counter; here's some example gcc code to do it:
static unsigned long long rdtsc(void) {
register unsigned long long d;
__asm__ __volatile__ ("rdtsc" : "=A"(d));
return d;
}
The previous method takes about 13 cycles on an Athlon 750. (DO NOT try and make it inline -- or gcc might optimize your to-be-timed code out from between the rdtsc() calls.) It is a straightforward manner to read the cpu clock speed from
As with any timing method, take care to execute it a few times before you gather any information, to prime the i-cache.
Apparently the lameness filter believes that this is a "junk character post", so I'll type some more. Intel has a useful whitepaper that describes how to do this in an M$ compiler, available here: http://developer.intel.com/software/idap/resource
Avoiding any possible M$ jokes? You must be kidding. M$ is capable of producing software that will do nothing, unsuccessfully, a la "An unknown program has terminated in an unknown module with an unknown error. Please reboot your system."
Why not get your name attached to some high-profile projects? Why not even get a menial "web page design" job through a local temp agency, and then work your way up? High-school students turn jobs like that down! Even "PC technicians" at mass-makret retail stores can find employment in a more interesting and profitable sector of IT eventually.
Honestly, it's very hard to believe that -- if you're really that good -- you're stuck gutting fish or operating a forklift all day as your "geek satyagraha" seeps out from your atrophied wrists.
You don't need "friend" for this -- "friend" is merely a bit of syntactic sugar that people who don't know how to use the Visitor Pattern or double dispatch must resort to to get bidirectional relationships.
In your particular example, you'd want to use a seperate object to represent the borrower-book relationship (likely a State pattern object), have a method in the Book class which registers itself with a Borrower argument (or a BookVisitor argument, more generally), establishing the relationship thusly.
Private members are private for a reason.
Like what certain specific sites? Like your "watch me take off my shirt and program html" site? You got into that business about five years too late. Sorry, but I can't bring myself to have sympathy for a smut site operator who complains about the rest of the net treating her unfairly.
I guess it's too bad that you can't just use your breasts to make money on the 'net without any initial investment, but even the world's oldest profession requires you to spend money to make it.
I don't think you can make that comparison. A masters' in maths is basically the consolation prize for quitting (or failing out of) a Ph.D. Sure, you're not using it, because a masters' in maths is designed to equip you to do maths research. On the other hand, a traditional masters' in CS can take one of two paths: as prepatory for research or as prepatory for industry.
Add to this the fact that many schools are also offering the a terminal CS masters' (sometimes called a MCS instead of an MS), and a masters' in CS is a clear advantage. Fact is that many students never learn design skills or many critical big-picture CS skills, and a masters' can provide valuable training as well as providing experience working on a large project.
It really depends on what kind of software development you want to do. Do you want to crank out perl scripts for some dot-con, or do you want to work on large, critical systems? I'd say that in the case of the former, you're wasting your time in college. If it's the latter, though, then find a grad program with a strong engineering and methodologies emphasis as well as strong fundamentals and do the MS. It'll only take you a year-and-a-half if you're supporting yourself, but you ought to try to get a research job if they offer you support.
In all seriousness, working for the government is an easy gig if you're not the sort to want to work terribly hard (you won't, and they won't encourage you to) and if you don't take your job home with you -- or particularly seriously. If you are terribly career-driven or are a perfectionist, you will be driven mad.
Also: you cannot accept gifts of any kind while acting as a government employee. That means no free dinners, no fancy hotels at conferences where you're giving a talk, and (worst of all) no frequent flyer miles.
Good luck!
As much as I regret to feed this troll again, allow me to rebut. Firstly, my "mild retardation" must make my D. Phil (that's "Ph.D." for you Yanks) in computer science all the more impressive. Secondly, neither Gray nor I said "X has to be written in N instructions"; rather, the statement was "less than 100 instructions" and "less than 10 instructions". For those of you who are having trouble following, those numbers are what people who have at least high school maths experience refer to as "orders of magnitude". So that's "a lock must be set 'on the order of' 100 instructions, as opposed to a latch, which must be an order of magnitude faster.
Hopefully, this won't burst your small head.
Actually, you're quite wrong. First, note that I said "instructions", not "cycles". All modern processors provide a test-and-set or compare-and-swap instruction. So it does not depend at all the architecture. RDBMS locking is not done at the OS level -- that is FAR too much overhead. An RDBMS maintains its own locks, its own file structure, etc. Typical instruction-count costs for seeking through a highly-optimized B+-tree, doing a hash join, or marking a page as "dirty" have not changed since the introduction of virtual memory support in hardware, and neither has the cost of setting and releasing locks. Furthermore, for a scalable system, the cost times for updating those locks need to be as Gray said.
Kindly find someone else to troll, or perhaps ask your parents to buy you the newest Quake game if you're really frustrated.
Also, RDBMS vendors will likely always use assembly. The locking code for an RDBMS needs to operate in fewer than 100 instructions to guarantee the possibility of concurrency -- and that includes deadlock detection. Furthermore, the "latches", or internal locks on the buffer pool, etc., must execute in under 10 instructions.
Assembly language is important, especially since highly-optimized C code is faster than unoptimized C code by a factor of 5 or more. How do you think that those optimizations get into the compiler? Sure, no one writes applications -- or even, at this point, games -- in assembly, but today's processors are so complex that
As for your assertion that the Nomad will record voice at 32 kbps, that is just plain ridiculous. You can get better fidelity than a 32k mp3 with a knitting needle and a wax record on a hot day. The question specifically states that the biologist needs at least 96kbps, and I'm willing to bet that the "built-in mic" is going to be horribly inadequate as well.
If there is a hi-fidelity MP3 recorder that allows ordinary consumers to produce non-watermarked mp3s and transfer them to computer within the next year, I will publically recant. However, there is no chance of this. Read the cryptome article if you don't believe me.
Look at this excellent rant for why mp3 recorders will never exist. Minidisc is your best bet. You can get up to 149 minutes of monaural on a single minidisc, and the results will be very high quality. I personally used a minidisc for ornithology research back at Uni, with great success.
Check here for more info.
Perhaps he was interested in royalties from the to-be-released and oft-anticipated "Hellmouth Team Tournament" series, pitting pimply, maladjusted malcontents against unarmed, God-fearing jocks and shrieking, well-endowed cheerleaders.
As an engineer who was recently "demoted" to management, I take extreme umbrage at your comments. XP works because developers love it, not because management bought a seminar (for the record, I have never attended a "management seminar" and I do quite well, because I have the engineer's perspective) These benefits to the developer "trickle-up" in the form of more reliable code and fewer missed deadlines. If it didn't work, it would fall into disuse.
No development methodology is a panacaea, but XP has been invaluable in my group. That's all I'm saying.
ESR is a one-trick pony, desperate to hold on to his "status" by publishing diatribes which are merely "n+1" from his previous screeds. He needs to get out of the way. The community doesn't need him anymore.
As Suck said in their brilliant parody of Slashdot, "RMS, GNU, ESR: FOAD".
It's worked wonders in my organization, and I suspect that the "war room" approach lends itself to similar types of productivity gains.
Let's think about this: in a research-driven industry, when you have time-limited patent protection, you need to (gasp!) keep doing research in order to retain profitability.
The list includes such e-nobodies as "iKnowledge", "eCashier", "SpyGate", and the "Christian Concert Authority." And those are the more plausible-sounding of the bunch...
How on earth is this post "overrated"? It was at score 1. It hadn't been "rated" at all. This post was not inflammatory, insulting, or factually incorrect -- it was just an unpopular view. I'd really like an explanation from the moderator here.
However, I strictly follow a reader-response theory of language critcism -- I judge languages by the sorts of programmers who are most attracted to them. Calling VB "Venereal Basic" is not a "childish insult"; it's funny, and it's an apt metaphor, considering that VB is a painful disease and most well-to-do people are embarassed to discuss it in polite company.
Do you think that someday, a project like plex86 and a project like Bochs might combine in kernel space to produce a lightweight virtual machine that could run x86 code?
How is the fact that Disney still has a copyright on Mickey Mouse harming you? Oh, you can't sell DivX :-) rips of "Steamboat Willie"? That's a shame. Oh wait, that's it. Have I missed anything?