If the emulation bug indeed allows a shortcut that's impossible on the console, then I think we should rescind the unwritten "no unpublications" policy in this particular case.
I do not think that a run that's based in an emulator bug should be admissible in any way, shape or form, and no grandfather clause (in this case "it was published, it thus has the privilege of being protected forever") ought to trump this principle.
Now you left me wondering what the other thread you are referring to is. I honestly have no idea.
"Derailing the thread" is nonsense, to be honest. I made a completely on-topic post, someone acted like an asshole towards me, and I responded in kind. Sure, this results in a couple of off-topic posts, which will probably quickly be forgotten (I for one am not going to make it into any kind of prolonged flamewar even in the unlikely case that he would be ready to.) That's hardly "derailing the thread". The thread isn't going to get "derailed" because of an exchange of unpleasantries.
You are right. I didn't think that by making the substitution the terms in the sum might not be unique anymore because F(n-1) or F(n-2) may already be there. (Or rather, the latter, because as was established earlier, the sum can consist of non-consecutive fibonacci numbers.)
Thanks for pointing that out.
When someone asks a question about the problem at hand, a normal person would simply show the error in the deduction. An asshole acts all smug and simply says "I already proved that you can only remove one number" and not indicate what the error was, as if the objection were beneath him.
At least you admit acting like a smug asshole. Well, either start acting like a normal sensible person or just fuck off.
Even if speed is not the main goal of a run like this, it should nevertheless always be a secondary goal. In other words, even an entertainment-based run should never waste time for no reason, and should get to its goal (whatever it might be) as soon as possible, with no frames wasted. Even an entertainment-oriented run can have what we would call "sloppy play" (which is most often precisely wasted frames that serve no purpose.)
If the same level of entertainment and/or the same objective (eg. maximum score) can be achieved with another playable character that's faster, then I would say that the faster character should be used (unless there's a very good reason to not to.)
I think it's reasonable to expect some effort to be put into making a question intelligible and understandable. The question in one's brain cannot be transferred properly to another, if the medium is a garbled mess of random words.
A person cannot move at the speed of light. A person can move close to the speed of light, and time passes at different speeds at different frames of reference.
If a spaceship leaves Earth at a very high speed, travels for a time, then stops and turns back, and travels back to Earth, then more time will have passed on Earth than in the spaceship. If the spaceship could accelerate to near the speed of light, then the time difference would become quite significant. Eg. a person in the spaceship could experience one year and when he comes back to Earth decades or centuries could have passed on Earth.
I don't quite understand your first question. Please explain it in more detail.
I could posit a counter-argument.
Any fibonacci number F(n) in the sum can be substituted with F(n-1)+F(n-2). This means that any F(n) can be removed from the set as long as F(n-1) and F(n-2) are not removed. This sounds to me like you could remove every third fibonacci number, and still be able to express any natural number as a sum of fibonacci numbers.
(Or did you mean under the condition of "a sum of non-consecutive fibonacci numbers"? You didn't make that completely clear.)
In short, a "branch" in this context means the same as "alternative category". It's alternative to the default category, which is the fastest "world record" completion of the game (and which according to recent policy is an unnamed "branch".)
(Note that not every game can have a default "fastest completion" main category, if the game is such that speed completion makes no sense. In that case it's ok to have named branches only.)
"Moon" is one of the three publication tiers.
"Vault" is where every submission (that doesn't break site rules) of "fastest completion" and "fastest 100% completion" go regardless of quality or entertainment (unless they are deemed "moon-worthy".)
"Moon" is the tier for alternative branches that are deemed good enough for publication. Also well-received "fastest completion" and "fastest 100% completion" submissions can go there.
"Star" is a limited collection of representative Moon runs that showcase the art of tool-assisted speedrunning. (Every starred run doesn't necessarily have to be a top-rated run, although many of them tend to be.)
Disregarding the first fibonacci number (which is just the same as the second one, ie. 1), can numbers be removed from the sequence while still retaining the property that any natural number can be expressed as the sum of unique values in the sequence?
If the answer is yes, is there a formula that gives all the "unneeded" fibonacci numbers?
I don't understand how purely random input can become even near completion of the game in less time than the estimated age of the universe, no matter how fast you emulate.
There's an equal chance of going backwards than going forwards, and than going off-rails. Just getting out of the starting town should be near impossible (and even if it gets out of it, there's an about 50% chance that it will go right back, destroying any progress.) It just sounds like trying to solve the game via brute force, except worse, because it tries things at random rather than meticulously going through all possibilities.
If the rng has made any significant progress, I suspect foul play (unless someone can explain to me how it can be possible.)
But it got me thinking about a more interesting project: A bot that uses some heuristics to try to beat the game. Something like pre-programmed intermediate goals that the bot should try to achieve (ie. like "reach this point", "do this", and so on), and then it uses semi-random movements that prefer movements that go in the particular direction of the next goal. (Just purely going to that direction isn't going to work unless it has a full path-finding algorithm and other logic that would be quite complicated.)
This would, in fact, kind of simulate TPP.
That gives me an idea. Since I own Pokemon White as well, all I should need is to find someone who owns a DS, so it should be possible to transfer pokemon between them. I think I know at least someone who does.
With one participant, sure. With two participants, if they cooperate or make a deal that only one of them inputs moves, sure.
With 50 thousand participants? Theoretically possible, but I'd rather have something more interesting happen in anarchy mode than the stream spending weeks at one place and the viewership dropping to single-digits, after which nobody else is interested anymore.
It would be really cool to get a Steelix in Pokemon White 2, but the options to achieve this seem pretty limited. It seems that the only way to get it is to trade an Onix that's holding a Metal Coat. The Onix I have, the metal coat I can get, but the trading is the problem. I don't know of anybody who has this game.
Any suggestions?
People keep mentioning the "original purpose" of the stream. (It's not a term used only here, but constantly on the stream chat, the irc channels and other chats.)
What exactly was this mystical "original purpose" that people keep mentioning, but which doesn't seem to be specified anywhere?
I have hard time believing that this "original purpose" was for the stream to die at the Safari Zone.
What is the sparsest sequence of distinct natural numbers for which it can be said "any natural number can be written as a sum of unique numbers in this sequence"?