It's a fundamental limitation in how computers do floating point math. Floating-point addition is not associative. Admittedly, the spec says it's supposed to be that way, so that's fine. The FDIV problem was really that the CPU wasn't performing according to spec.
You're on the right track, but what we really need is this:
First, you organize the vote normally. This is phase one. If any of the candidates gets over 50% of votes, he is banned from the presidency, forever. Next, you organize a new vote, with the only two candidates being the two people (excluding anyone thrown out) who got the most votes in phase one. This is phase two; whoever wins it gets the presidency.
Yeah, they also use bombs to make the core start spinning one way, while conveniently not making something else spin the other way so as to conserve angular momentum. Brilliant.
But how do you know the binary you downloaded came from the same source, and that your compiler doesn't insert the backdoor while compiling it, and insert the backdoor generator when compiling itself?
I agree completely. I routinely drive from the bay area to LA and back and there's a spot southbound on I-5 where the speed limit drops from 70 to 55 and then to 45 in the span of just a few miles. There's also a giant flashing sign that says "45 MPH SPEED LIMIT WILL BE ENFORCED". Traffic in this area varies from about 60 mph to 75 or higher. Going 45 in any lane except the truck lane would be very dangerous. I have no idea why California decided to do this.
I'm not convinced that classes with low average grades are always a bad idea. One professor at my $SMARTYPANTS_U, from whom I took a whole lot of classes, was notorious for giving out problems he thought were "interesting". Often this meant unsolvable, but everyone in the course knew what to expect. Solving these problems completely often took collaboration between half the students in the class, which lead to giant collaboration sessions in the computer lab, in which everyone ended up learning things much faster than if we'd all worked alone. Yeah, it was tough, but it was definitely worth it.
Yes, but NAT breaks a lot of other things like, say, incoming connections. It'll be really nice to not have to connect to the router, forward a port, and then lose all of your existing connections while the router reboots.
b) Nowhere did you say float var = 0.0f; System.println(var);//prints infinity.
Dude...
If you try to print a float who's value is exactly 0.0f, it would print "infinity" instead.
I knew it was exactly 0.0f because somewhere earlier in the program I set it to that. The bug was definitely in the Metrowerks runtime. The code I posted worked fine because if the var was 0.0f it would print "0", and if it wasn't, then the runtime worked fine. I'm going to stop replying to this pointless thread now.
Metrowerks taught me to never trust anything from them, even for correctness.
While working on an absolutely trivial Java program for a class in high school, I ran into this nice bug in the Metrowerks Java runtime that was available at the time: If you try to print a float who's value is exactly 0.0f, it would print "infinity" instead. Lead to code like this:
MS couldn't take over SCO even if they wanted to because it would be a clear violation of anti-trust laws and probably a violation of the DOJ settlement.
Slashdot Mirror
Higher CRT refresh rates are also good if you're using the sync signal to drive stereo shutter glasses.
Market pressure. Somebody has to want to buy your products.
It's a fundamental limitation in how computers do floating point math. Floating-point addition is not associative. Admittedly, the spec says it's supposed to be that way, so that's fine. The FDIV problem was really that the CPU wasn't performing according to spec.
You mean like how 1.0 + 1e-18 - 1.0 = 0 while 1.0 - 1.0 + 1e-18 = 1e-18? Computers perform incorrect math all the time.
IPv6 does indeed have fragments, it just doesn't allow fragmentation to occur anywhere but at the originating host.
You're on the right track, but what we really need is this:
First, you organize the vote normally. This is phase one. If any of the candidates gets over 50% of votes, he is banned from the presidency, forever. Next, you organize a new vote, with the only two candidates being the two people (excluding anyone thrown out) who got the most votes in phase one. This is phase two; whoever wins it gets the presidency.
I had to do that recently due to some very heavy fog. Try 52 mpg.
Yeah, they also use bombs to make the core start spinning one way, while conveniently not making something else spin the other way so as to conserve angular momentum. Brilliant.
But how do you know the binary you downloaded came from the same source, and that your compiler doesn't insert the backdoor while compiling it, and insert the backdoor generator when compiling itself?
I agree completely. I routinely drive from the bay area to LA and back and there's a spot southbound on I-5 where the speed limit drops from 70 to 55 and then to 45 in the span of just a few miles. There's also a giant flashing sign that says "45 MPH SPEED LIMIT WILL BE ENFORCED". Traffic in this area varies from about 60 mph to 75 or higher. Going 45 in any lane except the truck lane would be very dangerous. I have no idea why California decided to do this.
ping6 ff02::1 -I eth0
http://www.archive.org/audio/etree.php
I'm not convinced that classes with low average grades are always a bad idea. One professor at my $SMARTYPANTS_U, from whom I took a whole lot of classes, was notorious for giving out problems he thought were "interesting". Often this meant unsolvable, but everyone in the course knew what to expect. Solving these problems completely often took collaboration between half the students in the class, which lead to giant collaboration sessions in the computer lab, in which everyone ended up learning things much faster than if we'd all worked alone. Yeah, it was tough, but it was definitely worth it.
Yes, but NAT breaks a lot of other things like, say, incoming connections. It'll be really nice to not have to connect to the router, forward a port, and then lose all of your existing connections while the router reboots.
He declined, for the "hey, we trusted you until ten minutes ago" reason above.
To be fair, they were also paying you until ten minutes ago.
Yeah, but then you can't have rest mass, and I like my mass where it is, thank you very much. ;)
G-force is caused by acceleration. Assuming you accelerate slowly enough, you can get up to $VERY_FAST without dying.
You do realize that more than one person uses Slashdot, right?
My piano teacher was blind from birth, and she could play just fine, although she had a little trouble "sightreading" Braille music.
Dude, chill, it's also an adjective!
"Steve located Dave's VCR" -- past participle of locate
"Dave's VCR is located at Steve's cafe" -- adjective
Try using a dictionary sometime.
b) Nowhere did you say //prints infinity.
float var = 0.0f;
System.println(var);
Dude...
If you try to print a float who's value is exactly 0.0f, it would print "infinity" instead.
I knew it was exactly 0.0f because somewhere earlier in the program I set it to that. The bug was definitely in the Metrowerks runtime. The code I posted worked fine because if the var was 0.0f it would print "0", and if it wasn't, then the runtime worked fine. I'm going to stop replying to this pointless thread now.
a) Writing 0f or 0 doesn't matter if var is a float, because the compiler will infer the right type for 0.
b) Did you even read what I wrote? I said it prints the wrong thing if the number is exactly 0.0f. Everything else worked fine.
In vim, you can search for \<i\>. Other editors probably have something similar.
Metrowerks taught me to never trust anything from them, even for correctness.
While working on an absolutely trivial Java program for a class in high school, I ran into this nice bug in the Metrowerks Java runtime that was available at the time: If you try to print a float who's value is exactly 0.0f, it would print "infinity" instead. Lead to code like this:Frustrating!
MS couldn't take over SCO even if they wanted to because it would be a clear violation of anti-trust laws and probably a violation of the DOJ settlement.
Since when has that ever stopped them?