And another problem with that is that it's physically impossible under any theory of time travel (that I know of).
I'm not so sure of that. The equations of General Relativity seem to show that in certain circumstances odd things can happen. For example, the math seems to show that if you put two rotating black holes close enough to each other, there's an 8-shaped path around them that if you traversed it, you would end up in your own past. (An ergosphere is a region around a rotating black hole, outside of its event horizon, where weird things happen...) I assume that it's currently unknown if this results in a real time travel or if something else kicks in, but I also assume that it would be a rather infeasible way of doing it anyways.
Also in quantum mechanics time (like everything else) becomes a complicated subject. For example there are situations where decisions made in the future seem to affect quantum effects in the past (see for example the delayed choice quantum eraser experiment). It seems that time doesn't work at quantum levels in the same way as in everyday life. ("Most people think time is like a river, that flows swift and sure in one direction. But I have seen the face of time, and I can tell you - they are wrong. Time is an ocean in a storm.")
Surely he wanted to use "utilization" instead. That has many more letters, so it must be right!
But perhaps he wants to learn British English instead of American, in which case it would be "utilisation". (Although I heard some rumors that Britons are slowly shifting towards American spelling as well... Nothing is sacred anymore.)
We generally ask that the version of the game most familiar to the audience (largely North American viewers) be used unless there are significant glitches present in another version which significantly aid gameplay.
Isn't that bit of a US-centric attitude? It's not like Europe (or Japan, for that matter) is a niche market for games. The official reason for the preference ought to perhaps be a bit more neutral (eg. related to consistency).
Another possibility would be to use Visual Studio Express (which is free) to compile the project. libgd can be compiled with it, and animmerger can probably too, although you'll probably have to create the project yourself. (Shouldn't be too hard: Just add the cc files to the project and get libgd to compile and link with the project. Make sure you are making a console app rather than a windowed one.)
Sorry but I don't see the problem with this. Disc based system also don't emulate load times. They just load instantly.
It's not that simple with cassette-based systems. With cassettes the data is modulated into an analog carrier signal sound. The routine that reads the data simply samples the input sound at certain intervals and interprets the levels. While some speed adjustement is performed, there are limits, which is why you can't simply speed up the sound to get the data to load faster.
The ROM of the computer (basically it's "operating system") offers a routine for loading data from a cassette tape. If the game just calls it to load the data, what the emulator can do is to load the requested data in a microsecond and then return to the calling code. However, if the game uses its own custom loading routine (as many games did), it becomes more complicated. The emulation speed could be increased at this point by several orders of magnitude, but whether this can be done automatically by the emulator or whether it has to be done manually, I'm not sure. (Also if the game shows something on the loading screen while it loads using its own loading routine, which was a pretty common reason to use such a routine, it would be superfast and thus the speeded up emulation is easily visible.)
It may be different with a console using CDs if games always call the ROM/OS routines to load data from it because then the emulator can simply load the data as fast as the hosting system allows. (Alternatively the emulator can simply pretend that the CD drive has a speed of one millionX, if the loading routines are based eg. on some kind of DMA system.)
I have wondered for quite long why this hasn't been implemented a long time ago.
There are other systems from the 80's where TASing would be theoretically possible, but have one major dilemma: Mainly, if the games were distributed on cassettes (and even if they were on disquettes). How should a TAS deal with loading the game?
With a ROM cartdridge system the question is trivial: Just reset the system with the ROM cartridge in (all this emulated, of course), and it just starts in a second (and in a deterministic way at that) and that's it. However, what if the system only supports programs in cassettes? You can't just start up the system and make it load the game from a cassette because it would take at least 5 minutes to do so, 10 minutes at worst. (Yes, it really was like that back then with those systems. We had to wait at worst 10 minutes before we could start playing a game. Ah the times...)
Many emulators of such systems support speed-loading: They detect when a game is being loaded (using the system's ROM routines) and switch to turbo mode in which they load the game as fast as the hosting computer allows (in modern PCs that means the game will load in a fraction of a second instead of 10 minutes). However, this is basically "non-standard" in that it's not doing what the original system is doing. (Also there exist a few games which bypass the loading routines in the system's ROM and use their own, in which case it's more difficult for the emulator to automatically do this.)
Another solution commonly used is to use a ROM dump instead of the original game data. This is done by simply loading the game as usual and then at the startscreen saving a snapshot of the entire ROM with the emulator, and using that to load the game instead of the original data. Again, doesn't emulate the original system where this obviously wasn't possible. (It also means the game is not started from reset, and the exact point in time when the ROM dump was done is up to the person who did it rather than it being standardized in any way. Also games that load in multiple parts, iow. are not loaded all at once, cannot use this, obviously.)
However, AFAIK the Atari 2600 used ROM cartridges, so this shouldn't be a problem.
Warp, nobody here cares about that distinction but you. If someone says "programming errors", I'm sure you know what they mean by it. Putting enough effort into every post to prevent you from jumping on every little thing that isn't worded properly hinders productivity. Please encourage productivity by suppressing your need for optimal semantics. That's how I see it, anyway.
Last time I checked, expressing one's opinion about a subject was not forbidden, nor even a faux pas, but in fact a completely normal form of social interaction between people. You are seeing something in my post that isn't there.
Also, I know this is going to sound hypocritical due to the above paragraph, but I don't think the Sega Tetris "TAS" described by zaphod is a "TAS", since actual real-time gameplay was necessary. That sort of run is simply an unassisted speed run, considering you have to play it in real time. This sequence mapping is the similar to the route mapping that's done by the players of Speed Demos Archive. It's not like andrewg got his 5-minute SMB run on his first time playing those levels, only knowing the game's physics mechanics.
I do not consider it a tool-assisted performance either, for many reasons: Tools are not being used to do the run (unless you consider the human brain a "tool", which in this context is usually not), the run is not repeatable by independent parties (because there is no keypress sequence file) and there is room for human imperfection, among others.
Thanks. I've thought about trying my own hand at TASing, but I am not very good beating games the normal way, I don't think I'd be cut out to replicate any of the amazing feats I see on here.
Well, one of the key tenets of TASing is to remove the human skill factor from the equation, which means that even if you are not good at playing the game, it's still perfectly possible to make a good TAS of it. You should at least give it a try. It can be great fun.
you say tomayto, I say tomahto. :)
(lazy and incompetent programming often leads to errors)
I consider a programming error something that was made by mistake rather than deliberately (regardless of how crappy and incompetent that deliberate implementation might be). It's not an error if the program works exactly as designed.
Abuses programming errors. (regardless of sequence picked for the game, there were safe times to try, even without foreknowledge of the arrangement chosen for that episode)
It's questionable if it was a programming error or simply lazy/incompetent programming. (I'm betting it was the latter.)
Only a very small number of PAL discs has been produced.
I didn't know they made different versions of the game disc depending on which continent it's intended for. Does that mean that one should be very careful about buying Xbox360 games from outside Europe because the game might not actually work? (Does it make a difference that I'm using the Xbox360 with a VGA monitor and not a TV?)
I don't remember ever seeing an online store specifying which version of the Xbox360 is required to play the game...
As for Tales of Vesperia, I'd love to get my hands on it, but it's RIDICULOUSLY priced (for example at play.com you can get a used version for a "mere" 143 euros, and new at 200 euros; I'm not paying that kind of money for a 3-years-old game that would normally cost about 20 or 30 euros at this point).
What is it with this game in particular? I have been watching the situation regularly to see if anything changes, but no. There seems to be no rational way of getting hold of this game at a reasonable price.
My two regular online shop sources of games are play.com and cdon.com. In the former the game is still listed as 150 euros used (200 euros new). In the latter it's more moderately priced (66 euros) but not in stock, hence not available.
As a curiosity, I checked on amazon.co.uk. Here's a list of prices for several JRPG's for the Xbox360 (all for new condition games):
Blue Dragon: £8.76
Final Fantasy XIII: £12.69
Infinite Undiscovery: £8.22
The Last Remnant: £11.20
Lost Odyssey: £11.96
Star Ocean: The Last Hope: £12.97
Eternal Sonata: £11.76
Resonance of Fate: £12.02
Tales of Vesperia: £145.95
Whoa. What makes the last game different from all the others?
Does anyone have any idea where I could get this game for a reasonable price?
With the idea of fractal dimensions and the Cantor set, I could create a set with pi-dimension.
Even if a non-integral dimension can be defined mathematically, it still leaves open the question if a non-integral-dimensional unit cube is possible, given how a cube can be defined. One can specify a definition of "unit cube" such that it will define any such shape in any amount of dimensions from 2 upwards. (In 2 dimensions it would be a unit square, in 2 dimensions a unit cube, in 4 dimensions a unit hypercube, and so on.) The definition would be a combination of the shape having 1 unit of "n-dimensional volume", 2*n faces (in parallel pairs which are orthogonal to all the others), right angles, etc.
However, could a definition of "unit cube" be given that gives a sensible meaning to one in non-integral dimensions?
Btw, thinking about it, how could we have an uncountably infinite amount of dimensions? Dimensionality, at least in cartesian coordinates, seems to be a purely integral concept because each dimension is represented by its own number, and there can obviously be only an integral amount of numbers: You can have one number (representing coordinates in one-dimensional space), two numbers (representing coordinates in two-dimensional space) and so on. By this logic you can only have countably infinite dimensions (ie. the set of natural numbers). How could one even pose the possibility of having uncountably infinite dimensions? It's not like we could have a rational, much less an irrational amount of dimensions. (What would it even mean to have a "pi-dimensional cube"?)
On a different note: Thinking about the question of "is the set of reals in the range [0,1] the same size as the set of reals in the range [0,2]?", the answer is trivial: Yes, because we can formulate a simple one-to-one mapping between them: Just take a number in the former set, multiply it by 2, and you have unambiguously mapped each value in the former with the latter. Likewise for any set of reals in the range [0,x]: Simply multiply by x and you have a simple one-to-one mapping.
However, what happens if x is infinite? How do you create a one-to-one mapping then? It's not like you can multiply by infinity. (This is, basically, the case with the unit square: We have an uncountably infinite amount of different one-dimensional ranges of size 1, and we have to come up with a one-to-one mapping between the reals [0,1] and the points inside the unit square.)
That's the hard part to understand.
(If that became clear, then it would be trivial to understand why the same applies to a cube, a hypercube, and a cube in any amount of dimensions, because each is just an infinite expansion of [0,1] ranges.)
The other case is when the base is uncountable. The cardinality will be the same as the set of all real functions from R to R (R^R) . Using the same diagonal argument Cantor used to prove R is not enumerable, we can prove that the cardinality of R^R is 2^card(R), bigger than card(R). The generalized continuum hypothesis states that this cardinality is aleph-2, but it's unprovable in the most common axiomatization of mathematics. If you reject these axioms, a lot of things I wrote will no longer hold.
Do I understand correctly from what you are saying (and from what I have read at wikipedia about the continuum hypothesis) that whether there exists a one-to-one mapping between the set of reals in the range [0,1] (or, for that matter, the entire set of reals, it doesn't really make a difference AFAIK) and the set of points inside a unit hypercube of uncountably infinite dimensions or not cannot be either proven or disproven with "regular" set theory mathematics (for a definition of "regular" that goes well beyond my understanding of set theory)?
(Wow, that was one big sentence. I tried to split it up, but couldn't figure out a natural way of doing it without it becoming even more contrived...)
This is not the case if we can fill the unit square by mapping points from the unit line. Let x be a number on the unit line and let [X,Y] be a point inside the unit square. Let r(x) be the rational part of x and i(x) the irrational part of x. Consider the mapping:
[X,Y]=[2*r(x),2*i(x)]. This maps the unit line to a right angled triangle with corners at [0,0],[2,0] and [0,2]. This triangle covers the unit square, and we can fill the entire triangle since both the rational and irrational numbers are dense on [0,1]. Only the rational numbers are filled on the x-axis, and only irrational on the y-axis, but since both these sets are dense, we have filled (more than) the unit square mapping only points from the unit line. This means that the unit square does not have more points than the unit line.
I'm not sure I'm allowed to do what I just did, but it was worth a try!
This is not really my field of expertise, but it sounds to me like you are attempting to enumerate all the vertical coordinates of the square with rational numbers. Since rational numbers are countable, it would mean that the vertical coordinates (which are expressed with reals) would be countable, but that's not possible because the amount of coordinates is uncountable.
p4wn3r wrote:
By your phrasing, I think you're asking whether the cardinality of [0,1] x [0,1] is larger than [0,1]. The answer is no. Dimensionality (or more strictly, the cartesian product of two sets) doesn't increase the cardinality of a set.
I assume from this that the set of points inside a unit cube would also be the same as those on a unit line, and likewise for a 4-dimensional unit hypercube, and so on. Is there any amount of dimensions that would make the set larger?
Is the set of points inside a unit square larger than the set of points on a unit line?
(Before someone instinctively answers "of course", remember that eg. the set of rational numbers is actually exactly as large as the set of integers. Adding a dimension doesn't necessarily increase the size of an infinite set.)
Given that the emulator runs the game in a completely deterministic way, it cannot desync no matter how slow the computer (or virtual environment). If the computer (or environment) is way too slow, it might be unable to emulate the game at full speed (ie. at the same speed as the original console), but that doesn't affect syncing. It just means that the emulation runs slowly.
