Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×

Comment Re:Trends versus Data Points (Score 1) 360

There haven't been changes in the sun or in volcanic eruptions that are sufficient to account for the temperature trend.

The models also fail to account for the temperature trend (where or where did the predicted heat go?)

This is why you shouldnt be in bed with the modelers. But we see that you actually are...

Comment Re:wee little issue (Score 3, Interesting) 360

Let us know when we can download the raw unprocessed data to feed into their pet algorithm. Yeah... you are about to link to some place where you THINK the raw data is... but you are wrong... thats processed data ("adjusted") and they keep altering the old, already processed, data.... funny that.

(I am a witness to it - quite simple really, download their data... wait 4 weeks and download it again... do a difference.. note how old data keeps changing)

Comment Re:call me skeptical (Score 2) 360

Not data that contradicts his beliefs, I don't think.

No he clearly intended to "kill file" any site that contradicted his beliefs. he stated it quite clearly. if he meant something else then he should have said something else. But this is common with the AGW side of things .. they exaggerate like hell for the purposes of making their arguments seem stronger... but those with a keen eye see it the other way... their arguments seem weaker whenever they do it.

Comment Re:Benchmarks for that AMD chip look bad... (Score 1, Troll) 180

It will be slower and/or consume more energy than an Intel version

Unlikely at the same total price point. That only happens in the $300+ area of the desktop space.

The price of the Atoms people are comparing the E-series with doesn't include a GPU for the Atom... as if GPU's are magically free... dont use energy... conveniently ignored for the purpose of cheering on... shitty Atoms?

Comment Re:Virtualisation dates from the 1960's ! (Score 1) 180

The 8088 had an 8 bit external bus

..which has nothing to do with being 8-bit...

bus width
address lines
fastest word size

Which one of these has never been used to define the bitness of a machine? Yes, its the one you are using.

Anyone with a triple channel i7 has a 192-bit desktop right now (thats the width of the data bus of first gen i7's) according to your idea of what machines a machine 8-bit...

no IBM PC was ever 8-bit.. never.. they started with a 16-bit word size and 20-bit addresses.. some might argue they were 20-bit, but its pretty well accepted that 16 is the right description while 20 is the wrong description.

8 isnt even wrong, its just retarded... something someone could only think was right if they didnt know fucking anything at all about what they are talking about... thats you.. and you know it.. so why are you talking? You know you certainly shouldnt be pretending to be knowledgeable.. so why are you doing it?

Comment Re:Benchmarks for that AMD chip look bad... (Score -1, Troll) 180

Funny how in the desktop space the Intel fans always talk about single-threaded performance even when the multi-threaded performance is way behind, but then in the low-watt space they do an about face and ignore the single-threaded performance.. even when the multi-threaded performance is barely better at all while the single-threaded is way behind....

Let us know when for the price of that Atom, you at least get a GPU too. Funny how you ignored that too.

Comment Re:Caches, threading, SIMD/GPUs, and floating poin (Score 1) 180

Also note that rather recently Intel drastically dropped the accuracy of their FPU's in order to make the performance numbers look better.... dont expect 80-bit procession even when explicitly using the x87 instructions now... its now been documented that this is the case but for a few years Intel got away without publicly acknowledging the large drop in accuracy....

Comment Re:Should hardware even be a concern? (Score 1) 180

Keep in mind that the fastest implementations of just about any ultra-common standardized algorithms ((un)encryption, (de)compression, etc...) are in assembler. This fact doesnt shed as good light on compilers as some people like to disingenuously shine. Sure, most of the time you don't care that much about every last bit of performance, but when the CPU time for the algorithm in question easily totals billions of CPU hours... suddenly the arguments about assembler not being important start to look rather foolish.

Comment Re:1980s? (Score 0) 180

Depends on your definition of 8-bits. The 8088 in the original IBM PC had an external 8 bit data bus, unlike the 8086.

The size of the data bus has never mattered wrt low level programming, and everyone who was ever involved in said low lever programming knows that, meaning that you are not included in that group. Conveniently you became an anonymous coward to defend yourself with even more ignorance. Idiot.

The size of the address bus was 20-bits on both 8088 and 8086. Thats much wider than we see here in your last ditch effort to label these are 8-bit processors. Idiot.

Guess that means you are at least partially retarded.

Nope, it means that you dont know what matters and what does not. Idiot.

Comment Re:1980s? (Score 1) 180

You are a retarded idiot. The author states right at the beginning of the article that he's focusing on x86. In the (late) 80s, most people had an IBM PC, if they had anything.

..and there was never an 8-bit IBM PC. Talk about retarded... you're the idiot I guess.

Comment Re:Perfect? Really? (Score 1) 340

That's because the perfect strategy is suboptimal.

..and in all likelihood it is a losing strategy.

Consider a chess engine that sees that it will be mated in 13 moves, and that the only reason its 13 moves and not 12 moves is because it can sacrifice its queen right now delaying the mate by 1 additional move.

The minimax strategy is to play the queen sacrifice, but in practice that just increases the likelihood of a loss because all opponents now have an easy win, not just those that see these mates.

Now poker isnt a 2 player game, so the effects of collusion are to be considered. Clearly against players who are colluding the perfect strategy is a losing strategy (that maybe just happens to minimize the losses.) Note that even when your opponents are not colluding, that does not mean that the decisions that they are making aren't equivalent to players who are colluding (colluders would bet and raise here, and so coincidentally are these fools you are playing again), ergo the perfect 3+ payers strategy is almost certainly a losing strategy.

Slashdot Top Deals

Living on Earth may be expensive, but it includes an annual free trip around the Sun.

Working...