What in the name of god is bible study?

[Asher]

Originally posted by Kidicious


The difference is that we are much more critical of all of our politicians. You seem to think you have all the smart politicians, just like you think MS is so great, when in fact it's just another corporation.

You have to be ****ing kidding me that we're not critical of our politicians. We're so critical of our politicians that most Canadians can't stand any of them.

Do some ****ing research, Ben Jr.

[Felch]

Originally posted by Asher
You have to be ****ing kidding me that we're not critical of our politicians. We're so critical of our politicians that most Canadians can't stand any of them.

Do some ****ing research, Ben Jr.

Asher, you're sorely lacking in what we call, "a sense of humor." Smoke more weed. Seriously.

[Asher]

I haven't done that in years. I need to stay sharp at work.

[Felch]

That's terrible. I'd rather be poor and stoned than rich and aware of how awful life really is.

[Asher]

I think it takes a tremendous sense of humour not to go absolutely crazy (borderline is okay!) after reading to and responding to as many Ben Kenobi, Agathon, Urban Ranger, and Kidicious posts that I have.

Asher

[Kidlicious]

Originally posted by Asher
I haven't done that in years. I need to stay sharp at work.

Getting mad is worse than smoking weed.

[Jaguar]

Asher, your thing about Chess and Go seems to just be one of your typical "I like suchandsuch thing better so it's superior in every way" trolls. Go and Chess and Checkers are all perfect information games that computers can solve completely when they get powerful enough. Go will assuredly take the longest. That doesn't mean it's necessarily more fun for people to play. (Although it's certainly more fun than Checkers, which sucks.)

[Ben Kenobi]

If you can make a concrete estimate of the increases in speeds, then you can do the calculation yourself.

That's all I'm saying. The speeds that will be necessary to crack chess, you are looking at quantum computing. Those are the chip sizes that would be required to pull it off in any reasonable amount of time.

[Asher]

Originally posted by Jaguar
Asher, your thing about Chess and Go seems to just be one of your typical "I like suchandsuch thing better so it's superior in every way" trolls. Go and Chess and Checkers are all perfect information games that computers can solve completely when they get powerful enough. Go will assuredly take the longest. That doesn't mean it's necessarily more fun for people to play. (Although it's certainly more fun than Checkers, which sucks.)

This seems to be your typical "I like suchandsuch thing better so it's superior in every way" trolls.

Chess is just boring. I've tried so many times to get interested in it, but I just can't do it. That's why I don't like it. Go > Chess > Checkers.

[Asher]

Originally posted by Ben Kenobi
If you can make a concrete estimate of the increases in speeds, then you can do the calculation yourself.

That's all I'm saying. The speeds that will be necessary to crack chess, you are looking at quantum computing. Those are the chip sizes that would be required to pull it off in any reasonable amount of time.

So what you are saying is that the Bible, like Chess, is a completely solvable problem?

Just to be clear.

[Felch]

I think you guys are deviating from the actual purpose behind studying a subject.

Most studying isn't original research. When my nephew studies arithmetic, it doesn't matter that other people already know the sum of 43+62. It matters that he doesn't, and he has to study. So comparing Bible study to solving a board game is kind of missing the point. Even if Chess were solved, people would still have to study it to know the strategies.

[Lul Thyme]

Originally posted by Asher
I'm referring about the progress towards these games being "solved", which has seen tremendous progress. I've made that very clear, I thought.

I was referring to comments such as the following:

Originally posted by Asher
This is actually an area in game theory & computer science that is making tremendous progress.

Originally posted by Asher
I'd be very interested to know what kind of person made that estimate and if they took into account the standard Moore's law (as Ben has), or if they actually knew about scientific computation and understand projected future advancements and even current advancements (eg, Larrabee/GPGPU/Cell) which provide orders of magnitude of performance improvements over the period of a year or two.

While I agree that Ben's estimates using Moore's law are naively pessimistic, his estimates based on the number of different positions are naively optimistic. I believe the complexity of the problem grows much faster than linearly with the number of positions.

Of course, any predictions of this kind are going to be at least somewhat wild, but if you look, for example, at the time it took to go from 5 to 6 to 7 pieces endgame analysis in chess, and think about the problem a bit, it seems hard to imagine this increase being much faster than linear, if even that.

6-pieces were done in 2006 and 7-pieces is on its way for sometimes in 2015 according to
(http://www.gilith.com/research/papers/chess.pdf)

Originally posted by Asher
I make the claim that chess isn't far behind not terms of computational power to compute it all (though I think it's far closer than 100 years), but in terms of understanding how we can approach solving it. We pretty much know how to do it, it just needs the computational horsepower to do it, which really is only a matter of time.

But then this isn't really saying much at all. Even for checkers, relatively few deep new ideas were involved. It was mostly a matter of computing power. You can make the same claim about go.

Originally posted by Asher
Yes, and we've already had several fundamental breakthroughs since he's made this comment years ago. We no longer have single-core CPUs like the Pentium 4 that are emphasizing MHz over parallel processing. Retrograde analysis massively benefits from parallel processing, and the strives made in recent years do genuinely end up as being considered breakthroughs. Schaeffer used 200 PCs to calculate checkers, most of which were Pentium or Pentium IIs. There are far better ways to do this now -- look at Folding@Home. Look at their initial estimates for how long it'd take to simulate protein folding, and then look how much they have been absolutely shattered by having a popular distributed computing application, then look how much they've been further shattered by the introduction of PS3 clients (~30 times faster than PCs) and GPGPU clients (~110 times faster than PCs). The growth in computational horsepower for such applications -- of which retrograde analysis for chess is an ideal candidate -- is absolutely astounding and far beyond what people had predicted.

This is both very impressive and not that much. Even these "unexpected" supra-technologies have come in at a somewhat regular pace in the past and can be expected to do so in the future.

Taking these types of "breakthrough" in account, the growth of computing power available for the problem under consideration is probably supra-exponential yet the solution is still decades away at the very least.

For example, I'd be very surprised if interviewing Schaeffer about this again today would yield a very different answer.

I think the type of improvements you mention are somewhat expected (not individually, but in the sense that some improvements of this kind will crop up) by researchers in the field and are (roughly) taken into account in predictions.

[Lul Thyme]

Originally posted by Ben Kenobi

That's all I'm saying. The speeds that will be necessary to crack chess, you are looking at quantum computing. Those are the chip sizes that would be required to pull it off in any reasonable amount of time.

Quantum computing is not a chip size.

[Elok]

No, but "teeny-weeny" is.

[Lul Thyme]

Asher :

I find the two following comments slightly at odds, especially in the context of your argument with Ben.

Originally posted by Asher
Checkers has been "solved", Chess is not far behind.

Originally posted by Asher
I make the claim that chess isn't far behind not terms of computational power to compute it all (though I think it's far closer than 100 years), but in terms of understanding how we can approach solving it. We pretty much know how to do it, it just needs the computational horsepower to do it, which really is only a matter of time.

If you really meant that chess was not far behind in terms of understanding how to solve it, then the fact that checkers has been "solved" recently (or at all) is irrelevant. In terms of understanding, both these games were solved in the 50s at least.

We pretty much knew how to do it, all that was needed was computational horsepower to do it, which was really just a matter of time.

In fact, if this understanding is really what you meant, then it goes against what you were trying to argue about with Ben, since understanding the problem and the problem being finite doesn't make it solvable in practice, which is roughly what Ben was saying.

For your original sentence to be an argument, you need to be arguing that chess is in fact nearly solved IN PRACTICE, a claim which you seem to backing from in the second quote.