Isn't organizing a bet somewhat risky? If everybody bets eg. for team 1, and it happens to win, you'll have to pay them the total amount of bets from your own pocket...
I notice now that the original problem may have the ambiguity that it doesn't specify whether the program has to print only one number, or whether it's allowed to output an arbitrary amount of numbers (all of which are counted as being printed by said program, and thus count towards the solution to the posed problem).
This makes a radical difference. If we require that the program only print one single number (even if it internally calculates any amount of them), it limits the result quite a lot (because only that one number will count in terms of the problem, but not any number smaller than that).
If, however, we allow the program to output any amount of numbers, then such a program could print all the numbers from 0 to whichever large number it calculated.
I suppose that for the problem to be more sensible, it would make more sense to require that it outputs only one number. (Of course internally it can calculate as many numbers as it wants, but it has to "mark" only one of them as the "result" of the calculations, which it outputs.)
The alternative version of the problem (ie. what's the largest number that can be printed by such a program) obviously doesn't require that restriction (because we don't care if it prints smaller numbers as well).
(Of course there are other technical aspects to the problem that have not been addressed. For example, are ints of unlimited size, or do they have a maximum size? If they are of unlimited size, then what does something like "unsigned(-1)" mean?
Maybe the problem would be more sensible by replacing "C++" with a language that's completely agnostic to how big the variables can be, and has no trickery like "unsigned(-1)".)
I don't see what is point to think about it.
You can output any decimal number from 0 to ~1806 digits.
I don't really understand what you are trying to say.
You can certainly output a lot more (consecutive) numbers than that with a (terminating) C++ program of 1000 characters or less, so the answer to both problems would be enormously larger than that.
Either way (and once again), my question was not what the numbers are, but the nature of the problem itself: Is it unsolvable or just very hard?
I think you're right, however, that we could simply list all programs with 1000 or fewer characters and analyze whether they halt or not. I see no reason why that shouldn't be possible.
Actually, I'm not sure it's that simple. I think it depends on whether the program has an unlimited amount of memory (and addressing space) available to it.
I suppose the original problem was insufficiently defined because it didn't specify whether the program has unlimited memory addressing or not. I'd say that with this kind of problems one can assume that by default the amount of memory is unlimited (and the program can address all of it).
So if we make that assumption, I wouldn't be so hasty to proclaim that the halting problem is solvable even if we limit the length of the program's source code.
What the limit on the length of the program's source code does, however, is that it limits the amount of possible ending conditions (because the ending condition must have been somehow expressed in the source code, no matter how convoluted and complex that condition might be).
But (and that's a big but), if there is indeed an unsolvable halting problem, this might mean that it's likewise impossible to determine the answer to the original problem (ie. "what's the smallest positive integer that can't be computed") or its converse (ie. "what's the largest integer that can be computed (the program must terminate)").
I forgot to mention that, obviously, for the problem to make sense, the program has to be a terminating one. In other words, there has to be an ending condition. (Else it could simply print increasing integers forever.)
An interesting variant of the problem is this:
"What is the largest integer that can be computed by a terminating C++ program of at most 1000 characters?"
The condition that the program must terminate actually puts an upper limit to how big the computer number can be. Since the program has a limited size, the computed number also must have a limited size.
Thus the core question is whether this problem is unsolvable or just very hard.
Since, for starters, all ~1000 digit numbers are computable under that criterion, I imagine the answer is very large. To throw out a guess, perhaps one million or so digits.
I'm not asking what the number is (because it's practically, perhaps even provably, impossible to calculate). I'm asking whether the problem is unsolvable or simply very hard.
One could naively think "it's not unsolvable; simply go through every possible valid C++ program of 1000 characters or less, and for each one see which numbers it outputs, if any". However, that method might not be usable because you may encounter the halting problem: You cannot know if a given program will ever halt.
Of course the halting problem is related to programs of any arbitrary size. Here we are talking about programs of a limited size, so perhaps the halting problem doesn't apply? Perhaps it's possible to simply list which programs halt and which won't?
Of course then there's the problem of programs that don't halt but do print numbers. If they just keep printing numbers, can you ever find out which ones?
An unbounded version of the problem is probably unsolvable. In other words, if we modify the problem like this:
"For any given n, what is the smallest positive integer that's not computable by a C++ program of at most n characters?"
Here we probably stumble across the halting problem outright.
I was thinking whether a problem like this is provably unsolvable, or simply extremely hard:
"What is the smallest positive integer that cannot be computed with a C++ program of at most 1000 characters?"
(Note that this doesn't mean that there aren't larger integers than the answer that can't be computed by such a program. The problem is asking for the smallest positive integer that cannot be, even if there exist larger ones that can.)
I know this is (probably) related to Kolmogorov complexity, but that subject is a bit too complex for my knowledge. (Also, I'm not sure if the concept of Kolmogorov complexity actually answers my question, ie. whether that problem is unsolvable or just very hard.)
The submission text simply assumes that the reader knows what "Super Metroid Spore Spawn RTA" means. Sure, one could try googling it, but I think it would be good form to briefly explain what it means in the submission text. This way the reader will understand what the goal of the run is and what to expect.
Not any kind of speedrunning per se, but vaguely related.
When I was playing Hitman: Absolution, at some point in the later parts of the game I noticed that many levels could be completed laughably easily.
The game is very open-ended and there is no one single path nor one single strategy that has to be used to complete levels. Levels are quite free to be traversed, and there are many paths you can take, and many tactics you can use, ranging all the way from complete stealth to killing everybody.
One tool available for this is the "instinct meter". When you are disguised, if enemies start becoming suspicious, you can use your instinct meter to blend in. This of course depletes the meter so you only have a limited amount of it.
When I reached this particular level for the first time, I just tried for the sake of it if I could just walk to the end disguised. The level itself is quite large and complicated, and there are quite many options (for instance, there's a sniper rifle at the beginning of the level, giving you one possible option). Playing it "normally" would surely take quite some time. However, I tried just "bluffing" myself to the end of the level, and I succeeded, even on my first playthrough.
This video isn't from that playthrough. I recorded a completion of the level using that tactic later, for demonstration purposes. Could be useful for speedrunning.
Link to video
Not as old as me. In fact, thinking about my age makes me somewhat depressed.
Let's just say I haven't seen a single post in this thread by a person older than me.
For some reason I can't switch to HD in either that Mario Circuit nor the flicker video. It just shows the spinning wheel forever.
(Btw, how about making something less annoying / potentially dangerous for some people than flickering the entire image?)
Depends on the TV you're using though. A CRT renders images completely differently from an LCD. Hell, as I understand it different CRTs will have vastly different displays depending e.g. on how quickly their phosphors "decay".
Since there's no objectively right answer, I'm inclined to say that whoever's going to the effort of making the screenshot should use the approach that looks best to them, and if someone disagrees then the two of them can hash things out.
OTOH things like bilinear filtering of textures has nothing to do with the display, so shouldn't the filtering be the one that was actually used on the console?
As for LCD, I don't think the N64 had anything else than analog video output (composite and S-video).
Shouldn't they, in principle, look as close as possible to the original game played on the original console on a TV? Thus eg. bilinear interpolation of textures should be used only if the original console supported it?
I'm still of the opinion that there should be a valid "uses intended route" Moon category, for those who want to see highly-glitched games being actually played through in a "normal" way.
To avoid situations where somebody makes such a run and spends a lot of effort on it just to have it rejected because their own view of what consists the "intended route" differs from other people's, there could be a system to make sure that nobody wastes so much effort:
If a runner wants to make a run of a game using that category, first he should post a proposal for the rules that he will be using to achieve said goal, and listen to people's opinions and constructive criticism. If a consensus is achieved for a good set of rules for that particular game, there will be some kind of assurance that the run will not be rejected because of disagreements in the definition of "intended route", and thus he can proceed to make the run, and thus the only remaining question is whether it will be entertaining enough to deserve publication (which ought to be assessed with WIPs).
Before the site even existed, Bisqwit drew my attention to Morimoto's runs and Famtasia. He actually challenged me to make a faster SMB run with Famtasia than he did (but I never made anything complete). As the community grew from just a couple of people to something like ten or twenty, he started collecting those runs into his own server (under something like http://bisqwit.iki.fi/nesvideos/ which was the name of the old site for a couple of years).
This forum was added at some point, as the community grew, so I joined pretty much from the beginning.
Yes, many women are also very competitive, and there's also some competitive hobbies/sports that are dominated by women (roller derby), but I think the majority of competitions for the sake of competition is male dominated.
Many people do indeed present the hypothesis that most hobbies (that involve groups of people) are male-dominant because most hobbies tend to be competitive.
OTOH, sports seem to go somewhat contrary to that hypothesis. Although it depends on the sport, of course. For example I have the impression that football is male-dominant (there are of course female teams and leagues, but I think they are a minority). Athletic sports seem to be somewhat more evenly distributed, even when we are talking about competitive sports (such as track&field).
There are many non-competitive fields that are also likewise male-dominant, or at least are mostly represented by males in their top positions. Think of things like film production in all of its different subcategories (scriptwriting, directing, music composition, etc.) For example if you look at the list of greatest film directors of all time, you'll hardly find even one single woman there. (Feminists will quickly jump to the conclusion that this is because the film industry is inherently sexist, but I don't know if that's just a hasty conclusion based on assumptions or if it's actually so significant.) Film production is but one example, though.
TheAxeMan wrote:
Also, I started playing the PC version of Final Fantasy XIII after never having a console that could play it. My first thoughts were that Lightning was pretty badass, but is it really necessary for her to wear a miniskirt and sleeveless top while the guys are all wearing long pants and jackets?
The reason is mostly pragmatic in terms of economy: Games with scantily clad ladies sell. That might tell something about the average consumer, but one shouldn't hastily jump to any conclusions about the creators, other than that they want to make money. If scantily clad male characters would sell more games, they would put those there. If characters dressed in eskimo wear would sell more games, their characters would be like that. It's just economy. There usually is no ideological or political agenda behind it.
Bisqwit noticed something about the divisibility of numbers of the form 2x*y-1, where x is an integer > 0 and y is an integer constant, so out of curiosity I investigated a bit further and noticed that (although with the caveat that I only tested with 64-bit integers, so there's room for error):
All 2x*2-1 are divisible by 3.
All 2x*3-1 are divisible by 7.
All 2x*5-1 are divisible by 31.
All 2x*7-1 are divisible by 127.
All 2x*13-1 are divisible by 8191.
All 2x*17-1 are divisible by 131071.
All 2x*19-1 are divisible by 524287.
Moreover:
Only numbers where y is a prime have a divisor that's different from all previous divisors.
All numbers of that form where y is even are divisible by 3. All such numbers where y is a multiple of 3 (and not even) are divisible by 7. All such numbers where y is a multiple of 5 (and not divisible by either 2 or 3) are divisible by 31. And so on.
All those divisors seem to be prime. Moreover (and this is only a conjecture because of the limits of 64-bit arithmetic), it seems that the divisors always increase, and they seem to themselves by of the form 2z-1 where z is a prime.
Even more curiously, the z in that formula is the same as the y in the original numbers
2x*2-1 are divisible by 3; 3 = 22-1
2x*3-1 are divisible by 7; 7 = 23-1
2x*5-1 are divisible by 31; 31 = 25-1
2x*7-1 are divisible by 127; 127 = 27-1
2x*13-1 are divisible by 8191; 8191 = 213-1
2x*17-1 are divisible by 131071; 131071 = 217-1
etc.
Thus a conjecture can be formulated from the above:
All numbers of the form 2x*y-1, where x and y are integers, x>0 and y is prime, are divisible by 2y-1.
I don't know, however, if it's reasonable to conjecture that 2y-1 is always prime (because it would always be a Mersenne prime, which seems unlikely).
Edit: I actually noticed that I inadvertently skipped a crucial value:
All 2x*11-1 are divisible by 23.
There doesn't seem to be a clear connection between 11 and 23. I suppose the conjecture holds for Mersenne primes, but not others.
However, it could still be conjectured that the divisor itself is always prime. Its form is more complex than simply 2y-1, though.
In my experience, nearly all variants of dancing are done mostly by women. The only dances I've come across where this isn't true are breakdance, salsa and lindy hop.
Line dancing!
Nach wrote:
By any chance are the majority of these hobbies competitive hobbies?
I have noticed an overwhelming predominance of males in the vast majority of hobbies, and I have been wondering why. Typically you go to, like, a tournament of some tabletop game or something, and at the very least 95% of the participants will be male. This is true for almost all other hobbies as well.
In some hobbies the distribution may be slightly more equal. For example this is often so with many sports.
However, I have been wondering for some time now if there's any hobby or sport where the distribution is reversed, iow. the overwhelming majority of participants are female. And I'm not talking about hobbies that have traditionally been deemed extremely "girly" (like knitting or ballet), but something that has no such historical cultural burden, nor any obvious reason for it to be so female-dominant.
After years of wondering this, I finally stumbled across such a thing: http://en.wikipedia.org/wiki/Roller_derby
