I noticed they refiled the suit. Did the first one get thrown out?
I'm really lousy when it comes to legal terms, especially when translating from Dutch to English, knowing that legal systems vary significantly. I'll try to explain: the company first filed a 'kortgeding', meaning a lawsuit that's handled in a short time. The judge didn't accept the case, saying that it wasn't urgent enough to be handled as a 'kortgeding'. Therefore the company filed the case as a 'procedure ten gronde', a procedure that takes a longer time.
No, according to the article I read in a Belgian newspaper (ServersCheck is Belgian, as am I), the company first tried to stop Google from displaying said sites as suggestions via other ways. In each case, Google either ignored or denied their request. The only other option they saw was the legal one.
At work we sell navigation software (mainly for inland navigation) and we often do just that: a GPS on the bow and one on the stern; we just read the GPS positions and calculate the heading in software. There are better solutions though, but they are also more expensive. JRC for example sells a GPS compass which calculates the difference between the GPS positions in an earlier stage of the signal processing.
machine-and-human parseable - while not as easy as vi-ing a textfile this tool helps a lot. it's very elegant and you can always dump and edit the text.
It's the same thing with some of the movie theaters here in Belgium: you can (1) buy tickets via the Internet and print it yourself, (2) buy tickets on a machine with your debet or credit card or (3) buy it the old-fashioned way from a real human being. Last time I checked, (1) and (2) or more expensive than (3). I guess there are some things that I will never understand.
I am aware it is a descending sort, you'll notice the name of the proc in assembly is descendingSort32. I probably should have changed the variable name.
Ah yes, I should have seen that.
As far as the speed difference between us, I suspect the compiler you are using may be altering the sort, or is just damn good.
gcc 3.4.4 on cygwin, nothing special.
In the end, I think as I said in the OP the best bet is to compile optimized code into assembly and then edit the nested loops. I'm still fairly certain I could beat the compiler.
Very well possible. I'm not so sure I could do it in most cases in any acceptable period of time. I'm more into trying to use the most appropriate algorithms for the given case (macro-optimizations instead of micro-optimizations) and using efficient prebuilt implementations where possible (std::sort for instance in this case).
You do realize that you're sorting in descending order? Your variable 'small' actually determines the greatest number instead of the largest.
Anyway, on what kind of machine did you run the benchmarks? I tried the C++ version on a Pentium 4-M 2.2 GHz under Cygwin. Without optimizations it takes 19.4 seconds (for the whole program, including filling the array etc.). With -O3 it takes 7.5 seconds. I can't compare with the ASM, since I have no idea how to create a program for testing it. I don't even know how to turn assembly code into an executable.
CPU? Well, we now have 3.8GHz P4 processors (which perform about as well as much slower Athlons...), and while they're certainly much faster than my old 20MHz 286 in 1991, I really don't see much difference in regular desktop usage between my 3.6GHz P4 and my old 1GHz Duron from 3-4 years ago, or my 300MHz Celeron before that, unless I'm running transcode or something (i.e., not often).
You don't see much improvement in regular desktop usage precisely because all the other components haven't had performance increases as much as the CPU. Memory is notoriously falling more and more behind. And many tasks are I/O-bound rather than CPU-bound, which means that speed increases in the CPU won't help unless the I/O bottlenecks get faster. One of those bottlenecks is in many cases still the HD.
I used to type numbers on the numeric keypad too, but that habit is changing for two reasons: - The more I use vi(m), the more I'm learning the benefit of keeping my hands on the keyboard and the more I dislike moving them away from it. - About two years ago I bought a laptop and started using it as my main computer. Since it doesn't have a numeric keypad, I'm forced to use the digits on the top row.
It's a bit awkward though, since I use azerty which means that I have to press shift to access the digits on the top row. I've had typing classes when I was a kid; I guess that helps. I've been thinking about switching to qwerty though, since the qwerty layout provides easier access to { } [ ] ^ \ and I use those a lot when typing C/C++ code.
It's not geek speak; at least to me it sounds much more like marketing speak than anything a geek might say. There's nothing substantial, nothing technical, nothing concrete. Only some babbling about leveraging a new kind of browser to empower the user. Bah.
It indead enhances accuracy of military GPS too, but those atmospheric errors are smaller than those introduced by SA.
The company I work for sells navigation software for inland cargo ships and often we also supply the GPS. When SA was still in effect we often used DGPS devices, but nowadays we almost always use normal GPS's. Only survey or dredging boats or the like that demand higher precision use DGPS now.
DGPS is intended to cancel out errors introduced by the signals travelling trough the atmosphere. Cancelling out the selective availability (which by the way isn't normally used any more) is just a nice side-effect.
BTW, I doubt DGPS works all that well in the middle of the Atlantic anyway. You need ground stations in your neighbourhood, and I don't think there are any in the middle of the ocean. Not that you need 10m accuracy in the middle of the ocean.
Can't we have both at the same time? Both 'New tab' (or window) and 'Clone tab' (or window)? I know I would use both, depending on the needs of the moment, if they were available. Preferences don't give me that ability.
Any professional typographer will tell you the way Firefox interprets the site is much user-friendlier. Text lines should not be too wide or it will make reading more difficult.
That's my problem, not the site designer's.
This is a common problem with most liquid layouts and max-width would be the perfect solution to the problem if IE supported it.
The perfect solution would be for web designers to stop using max-width and other width constraints. I have a windowing system that lets my resize my browser to exactly the size I want it to be; web sites shouldn't think they know better than me. I don't know any liquid layout that causes me problems; I know lots of sites with width constraints that are either too small or too large. Most are too small.
Slashdot Mirror
And the suit is filed in Leuven, which is in Flanders, not in the Walloon region.
I'm really lousy when it comes to legal terms, especially when translating from Dutch to English, knowing that legal systems vary significantly. I'll try to explain: the company first filed a 'kortgeding', meaning a lawsuit that's handled in a short time. The judge didn't accept the case, saying that it wasn't urgent enough to be handled as a 'kortgeding'. Therefore the company filed the case as a 'procedure ten gronde', a procedure that takes a longer time.
No, according to the article I read in a Belgian newspaper (ServersCheck is Belgian, as am I), the company first tried to stop Google from displaying said sites as suggestions via other ways. In each case, Google either ignored or denied their request. The only other option they saw was the legal one.
Indeed, and the number of fingers we have is specified in the genes, so maybe it's not a coincidence after all?
At work we sell navigation software (mainly for inland navigation) and we often do just that: a GPS on the bow and one on the stern; we just read the GPS positions and calculate the heading in software. There are better solutions though, but they are also more expensive. JRC for example sells a GPS compass which calculates the difference between the GPS positions in an earlier stage of the signal processing.
Or http://sqlitebrowser.sourceforge.net/.
Instead of Internet2 we just got Web 2.0. Bweeh.
Have you tried XnView? I don't use it, but from the description it looks as if can do more or less the same thing as IrfanView.
I'd choose #4 without any doubt, since that shows what the link points to: an article about a decine in proper anchor texting.
CmdrTaco prefers #5, but I think that's suboptimal. It leaves out an important part of the story. It says what it's about, but not what it is.
Note that my excessive use of parentheses probably damaged your pipeline more than any badly spelled word ever could.
Actually it didn't. I guess exceptions cause more pipeline stalls than function calls and returns.
It's the same thing with some of the movie theaters here in Belgium: you can (1) buy tickets via the Internet and print it yourself, (2) buy tickets on a machine with your debet or credit card or (3) buy it the old-fashioned way from a real human being. Last time I checked, (1) and (2) or more expensive than (3). I guess there are some things that I will never understand.
gcc 3.4.4 on cygwin, nothing special.
Very well possible. I'm not so sure I could do it in most cases in any acceptable period of time. I'm more into trying to use the most appropriate algorithms for the given case (macro-optimizations instead of micro-optimizations) and using efficient prebuilt implementations where possible (std::sort for instance in this case).
The truck was made of lego bricks itself.
You do realize that you're sorting in descending order? Your variable 'small' actually determines the greatest number instead of the largest.
Anyway, on what kind of machine did you run the benchmarks? I tried the C++ version on a Pentium 4-M 2.2 GHz under Cygwin. Without optimizations it takes 19.4 seconds (for the whole program, including filling the array etc.). With -O3 it takes 7.5 seconds. I can't compare with the ASM, since I have no idea how to create a program for testing it. I don't even know how to turn assembly code into an executable.
You don't see much improvement in regular desktop usage precisely because all the other components haven't had performance increases as much as the CPU. Memory is notoriously falling more and more behind. And many tasks are I/O-bound rather than CPU-bound, which means that speed increases in the CPU won't help unless the I/O bottlenecks get faster. One of those bottlenecks is in many cases still the HD.
I was thinking the same thing. Not sure what that says about us.
I used to type numbers on the numeric keypad too, but that habit is changing for two reasons:
- The more I use vi(m), the more I'm learning the benefit of keeping my hands on the keyboard and the more I dislike moving them away from it.
- About two years ago I bought a laptop and started using it as my main computer. Since it doesn't have a numeric keypad, I'm forced to use the digits on the top row.
It's a bit awkward though, since I use azerty which means that I have to press shift to access the digits on the top row. I've had typing classes when I was a kid; I guess that helps. I've been thinking about switching to qwerty though, since the qwerty layout provides easier access to { } [ ] ^ \ and I use those a lot when typing C/C++ code.
It's not geek speak; at least to me it sounds much more like marketing speak than anything a geek might say. There's nothing substantial, nothing technical, nothing concrete. Only some babbling about leveraging a new kind of browser to empower the user. Bah.
Where are the Nazis when you need them? We should invoke Godwin, since every recursive thread needs a stop condition.
It indead enhances accuracy of military GPS too, but those atmospheric errors are smaller than those introduced by SA.
The company I work for sells navigation software for inland cargo ships and often we also supply the GPS. When SA was still in effect we often used DGPS devices, but nowadays we almost always use normal GPS's. Only survey or dredging boats or the like that demand higher precision use DGPS now.
DGPS is intended to cancel out errors introduced by the signals travelling trough the atmosphere. Cancelling out the selective availability (which by the way isn't normally used any more) is just a nice side-effect.
BTW, I doubt DGPS works all that well in the middle of the Atlantic anyway. You need ground stations in your neighbourhood, and I don't think there are any in the middle of the ocean. Not that you need 10m accuracy in the middle of the ocean.
I think you meant "... are rarely used without modifications and customization
Can't we have both at the same time? Both 'New tab' (or window) and 'Clone tab' (or window)? I know I would use both, depending on the needs of the moment, if they were available. Preferences don't give me that ability.
The perfect solution would be for web designers to stop using max-width and other width constraints. I have a windowing system that lets my resize my browser to exactly the size I want it to be; web sites shouldn't think they know better than me. I don't know any liquid layout that causes me problems; I know lots of sites with width constraints that are either too small or too large. Most are too small.
Yeah, because going 70mph in a convertible with the roof down is an excellent way to judge audio quality.