And the answer is yes. Because we know that lim(x->0) -x / x3 = -infinity. So we just need to set g(x) to -x and we set 1/ln(f(x)) = x3 which gives f(x) = exp(x-3)
Is it possible to have two functions f(x) and g(x) such that
lim(x->a) f(x) = 0,
lim(x->a) g(x) = 0
lim(x->a) f(x)g(x) = 0
and f(x) is not just the constant 0 (nor reduces to the constant zero, ie. eg. "f(x)=x-x" would be too boring of an answer)?
I don't really understand. I thought all of the suggested runs were rejected, and there will effectively be no TAS block. Am I missing something, or are we talking about some further future GDQ event?
Alyosha wrote:
I don't know, the selection committee said they wanted typical TAS stuff, but I'm not sure they really actually do.
Given that they rejected all of the proposed regular runs, I have to agree. They think they want "normal" speedruns (just tool-assisted), and that they don't want "gimmicky" runs, but when we submitted regular runs they rejected them all, which sends a rather different message. I don't think they themselves know what they want.
A table tennis ball has a diameter of 40 millimeters, and a mass of 2.7 grams. Two of them are floating in empty space, at rest relative to each other, 10 meters apart, with no other significant forces affecting them.
How long does it take for them to collide, and what's their relative speed when they do so?
You don't need a wormhole to allow time travel. It is my understanding that according to the equations two rotating black holes close enough to each other (which pretty much certainly happens when two black holes, which existence is in very little doubt anymore, are in the process of colliding) will cause a situation where a particle could travel through their ergospheres in such a manner that it will arrive back at its starting point before it started.
I have no idea what the consensus is if a particle does indeed do that, in real life. For example, what happens if the particle does that, and collides with itself? Perhaps there even isn't a hypothesis about this.
Future time travel is easy to understand i.e. exceed the speed of light and time will speed up relative to you in order to compensate. However, I've never understood how time traveling into the past in theory works (never mind all the possible paradoxes that could bring). Despite physicists saying it's theoretically possible to do.
Try to get your head around the fact that, theoretically, according to general relativity, a particle could travel to the past (using the ergosphere of a black hole) and collide with itself.
Suddenly I started thinking: Are there any convex polyhedra with an odd number of faces, where each face is an n-gon, for a given fixed n? (In other words, every face is a triangle, or every face is a 4-gon, for example. Obviously they don't all need to be the same shape, only to have the same number of sides.)
If we remove the requirement for convexity, then there are examples (can you think of one?), but I couldn't think of a convex example.
The point is that there are several glitches that could be used in a TAS that aren't being used in unassisted runs.
You just contradicted yourself within the next paragraph..
Warp wrote:
(but I don't know if such glitches exist).
You took two different paragraphs, talking about different things, and are claiming that I'm contradicting myself?
The first paragraph is talking about known glitches that are not being used in unassisted any% runs.
The second paragraph is me theoretizing that there may be even more glitches out there to find (or perhaps even known) that cannot be even done by humans. I don't know if these glitches exist.
For what reason are you confusing these two statements?
As for "nearly identical", that makes absolutely no sense. Can you name even one single other game where people are unwilling to make an any% TAS because "it would be almost identical" to the unassisted version?
Also, your definition of "nearly identical" is very strange, given that, as said, there are several glitches that are not being used in the speedrun, which could be used in the TAS. How is that "nearly identical"?
Many forums even allow uploading images at will (to be embedded in posts). Not that I'm asking for that (it would probably be a bad idea), but having people's avatars hosted at the tasvideos server might be viable.
I think this has been suggested before, but could it perhaps be a good idea that if there are more than, let's say, three authors, they would agree on a team name, in order to shorten the title of the run. The members of the team would then of course be listed in the details.
All I said is that there are glitches that unassisted any% runners aren't using because they are too difficult. The one I mentioned is one. Mido skip is another.
But they aren't not used because they are too difficult. They are not used because the setup would take too much time and they barely save any time without the setup.
The point is that there are several glitches that could be used in a TAS that aren't being used in unassisted runs. These glitches would make the TAS different, and in a sense quite awesome (precisely because it would showcase how a speedrun of the game would look like if a perfect runner were playing the game, being able to get all the very difficult glitches perfectly right).
I wouldn't be surprised if there were a few glitches that are only possible with TASing because they are way too difficult for a human to perform (but I don't know if such glitches exist).
I find this whole situation a bit baffling. This is possibly the only game in existence for which there's a lot of speedrunning activity, and a lot of interest, but for which nobody seems interested in making an any% TAS. I'm not aware of any other such game. Just imagine if nobody were interested in making an any% TAS of Super Mario 64, for instance, because it would be "boring". How strange would that be. (SM64 and OoT are two of the most quintessential N64 games with a huge speedrunning community. But only one of them has this strange phenomenon that nobody is interesting in making an any% TAS.)
I think that the core idea of TASing has always been that the end result is a data file consisting of timed keypresses which, at least in theory, could be inputted to the controller port of a completely unmodified console, running a completely unmodified original version of the game. This shouldn't require anything special from the target hardware nor its software.
In order to achieve this (and, in a sense, retain the "purity" of the console and its software being completely unmodified), the data file is generated with an emulator that emulates the entire system within itself (the emulated system being an exact "copy" of the original unmodified hardware and its software).
Even if the TAS can't be run in practice using the actual physical console because of technical limitations (such as the simple lack of a "tasbot" for that particular console to do it, or other reasons), the next best thing is being able to run it using an emulator that emulates the entire unmodified hardware&software, without requiring anything special from the latter. This is the closest thing we can get to actually "verifying" that the TAS is genuine (in the sense that it, at least in theory, could be run on the original hardware).
Once you start modifying the hardware, or creating/running the TAS requires either special software in that hardware, or requires a modified version of the game itself, it breaks this level of abstraction.
I suppose that if the end result is still a file of timed inputs that could ostensibly be run on the original unmodified hardware&software by simply feeding the input to its controller port (assuming there existed a "tasbot" to do it), then in theory it's still within this driving principle. But lacking the means to verify the TAS this way, if the only alternative is to do it is a modified console (or its software), then it becomes questionable (and, as Mothrayas points out above, impractical).
If the TAS can only be replicated with a modified hardware or software, or a modified game, then I think it goes beyond acceptable.
I've talked to plenty of people outside of TASVideos about the TAS Blocks at GDQ's and I've heard plenty of people express various complaints about them. Many straight up no longer care about the TAS Blocks, think the ACE stuff has gotten stale, or think that there's no point to showing something in a marathon that could just be watched on youtube at any time. To them, a TAS Block would be a waste of schedule space.
Well, that sounds a lot like only one conclusion can be drawn.
andypanther wrote:
And don't even try to bring up the charity argument, you know exactly that GDQ is not a non-profit organization!
There is indeed a lot of criticism that many people raise against GDQ (especially AGDQ and the PCF), but perhaps this thread is not appropriate to delve into all that.
People are taking things way too personally around here and feel way too entitled.
Given that, as far as I understand (and please correct me if I'm wrong), the organizers do not pay the runners anything, and on the contrary it's the runners who suffer a monetary loss for participating (they have to travel there and live there for several days, which is bloody expensive), I would say that a bit of entitlement is justified. Talented people, with enormous amounts of experience and skill, are offering their contribution to the event for free, at their own expense, yet it's the event organizers who feel entitled to pick&choose which one of these contributions are accepted.
Yes, I'm probably taking this way too seriously (especially since I'm not involved myself in any way, nor are personally affected by any of this), but I can't help but feel just a bit annoyed by this. I hope you understand.
Of course a much more mundane and closer-to-life example would be: A photon is emitted from the Sun. At that exact moment it has an unobstructed path towards the Earth. However, while it's on its way, something gets in its way, blocking it. Thus it hits the blocking object rather than arriving at the Earth.
But from the photon's own perspective this all happened in 0 seconds, which seems paradoxical. The moment that it left the Sun, and the moment it hits the obstruction, are the exact same moment. (The distance between the Sun and the object is zero, no matter how far apart they are from an sub-lightspeed point of view...)
It could also happen the other way around: When the photon is emitted, there's an obstruction between it and the Earth, but right before it hits the obstruction, it moves and the photon doesn't hit it and arrives at the Earth instead. But from the photon's own perspective there wasn't any obstruction to begin with. Which seems paradoxical.
On another topic, it's estimated that the photons that arrive to Earth from the Sun are actually tens of thousands of years old, because it takes that much for those photons to reach the surface of the Sun from its core.
But this seems to imply that every single photon created in the Sun is created in its core. One would think that eg. the solar corona emits boatloads of photons itself. Is it really true that all photons from the sun are tens of thousands of years old, or would it be more accurate to say that some of them are?
It's curious how that's something like 77 orders of magnitude larger than the age of the universe, yet from the photon's own perspective the travel time is instantaneous (at least assuming it can travel in vacuum the whole distance).
This seems paradoxical. The universe could end before the photon reaches its target, yet from its own perspective it reaches the target in 0 seconds.
It's final, barring things like unforseen late drop-outs that mean they have to fill the time with something else.
I don't want to say that it's a bit insulting for them to reject all the TAS submissions, but I'm really tempted to.
I would certainly not blame anybody if they wanted to skip the TAS block altogether this time too.
Suppose the distance between two objects in space is 100 light-seconds, but space is expanded so that every second another 100 light-seconds is added to this distance. Would a photon sent from one object towards the other reach that other object?
Let's simplify the problem by making the expansion happen in discrete steps. In other words, space is static, and we wait for a second for the photon to travel, then we expand space in one discrete step so that the distance between the objects increases by 100 light-seconds, wait for another second for the photon to travel, and so on.
During the first second the photon travels 1% of the distance. During the next second it travels 1/2% of the distance. Then 1/3%, then 1/4% and so on.
As we know, the sum 1+1/2+1/3+1/4+... diverges, which means that the photon will eventually reach the other object. But how long does it take? (I know that "how long" is a complicated question in general relativity, but I suppose it's "from an external point of view".)
Of course in actuality the distance between objects (eg. galaxies) doesn't increase linearly. A constant rate of expansion of space causes the distance to grow in an accelerating manner (the distance between two objects doubles every n seconds, which means that the growth of distance is exponential).
Is there a limit for a constant rate of metric expansion of space that still allows light to reach receding objects that start at a given distance from each other? Is there a formula for this?
A lot of staples were not accepted this year - no Super Mario Maker, not even Super Metroid.
My concern is not about what has been accepted, but how many. If only one single run is accepted, I don't think it makes sense to bother (unless you, or somebody else, really wants to do it regardless).
But is this a final judgment or just some kind of preliminary one? Will they accept further submissions?
How does it make any sense to have a TAS "block" with one single run?
I agree with the sentiment of canceling the whole TAS block. If they are going to reject everything, except one "bonus" run, might just as well not go at all. It would be just a huge waste of time and especially money to go to the event for one single TAS demonstration. Just skip it.
I'm about 40 episodes into Ranma 1/2. It's mostly met my expectations
It starts very interesting, but in later seasons succumbs more and more into lack of new ideas, and introducing tons and tons of new characters who have little to no relevance (it's almost like they replaced coming up with new, original, interesting ideas with just introducing the thirtieth-or-so new character that has never been seen before.) Still some good episodes in the later seasons, but more rarely.
After the series proper, the OVAs are a good watch.
