Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
Easily worthy of publication to the Vault.
With that said, I'd like to know two things:
1) What is the basic idea behind the bot? Surely there are some basic concepts in there. What was your strategy in writing it? (Also, are these puzzles NP-complete, as I suspect?)
2) Why does it look like you trace twice over every path? I realize this may be an automatic process by the game, but I found it kind of distracting. What's going on here?
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
I don't know if this is a nontrivial question.
I went for a bicycle ride around my neighborhood the other day. My neighborhood is shaped roughly like a 3x1 grid (two north-south streets crossed by four east-west streets). I wanted to go on a ride that wouldn't be boring, so I wanted to see every street both forward and backward. That got me thinking: Is it possible to go on a bike ride that does this without traveling the same direction down the same road twice? To make things more interesting, I'll add one more condition: you cannot immediately double-back down a road you just traversed. (My justification for this rule is that my bicycle has momentum, so it is inconvenient to make a 180o turn.)
My suspicion is that it is possible, based on my quick survey of theorems relating to Eulerian paths. We can simply construct a symmetric directed graph from the streets, after which the fifth theorem would apply if 180o turns were allowed. So does disallowing 180o turns alter the validity of the theorem?
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
No love for the Wii's Classic Controller? All of my latest TASing projects have been botted, so I haven't had reason to use it. Nevertheless, whenever I need a controller, I use the Classic Controller, interfaced using GlovePIE.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
I think this should be published to the Moon tier. It's not vastly different from the NES and SNES versions, but it's different enough.
I was entertained.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
I'm watching this run for the first time and I noticed something. Wouldn't it be faster to teach Blastoise both Surf and Strength at the same time (as opposed to about five minutes apart)? You don't use Surf until after Blastoise has learned Strength.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
As far as judging goes, I expect the poll should be applied something like this:
•Solid Yes votes? Publish immediately after clearing basic formalities. (Does it beat all records? Are all authors properly credited? Etc.)
•Solid No votes? Reject immediately.
•Anything in the gray area? Read the comments, seek further input from the author(s), judge based on that.
What I'm saying is that the votes don't (or shouldn't) matter. If people vote No because a movie was boring while still technically impressive, judges should recognize that and at least publish to the Vault.
This is something I like about TASVideos: it isn't a democracy.
If anything, votes are more useful to viewers, allowing them to get a quick sense of the quality of a movie. (I've skipped a lot of movies because they were soundly rejected and even watched some movies I normally wouldn't have because feedback was so positive.)
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
hegyak wrote:
I cried at the end of Grave of the Fireflies.
I made a deal with my friend that if I watched 10 or so of his favorite movies, he would play Mother 3. He recommended Grave of the Fireflies, warning me that it was extremely depressing. In his own words: "The happiest parts are watching people slowly die." Likewise, I warned him that Mother 3 was very depressing.
I watched the movie, was touched but not moved to tears. I called him up and said, "Is that the best you've got?"
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
AngerFist wrote:
This topic is long overdue. Feel free to post your favourite anime movies or show. My favourite anime films and shows are:
1. Shigurui. Tremendous show. I owe moozooh my life when he recommended it.
2. Ninja Scroll the movie.
3. Kurozuka. Again, moozooh..
4. Vampire Hunter D - Bloodlust.
5. Basilisk.
6. Jojo's Bizarre Adventure.
If you have any anime show or movie that is similar to my above taste, please do not hesitate to post.
As per creaothceann's suggestion, don't read the spoilers below if you have any interest in watching the ultra-violent anime Shigurui. The show's only 12 episodes long so just watch the damn thing!
Wow, I tend to strongly dislike manga and I usually shy away from violence, but Shigurui is one of my all time favorites. It's so beautifully animated and somehow both very realistic and ridiculous and fantastic. On its surface, it's just a Tiger vs. Dragon story, but I think it's an excellent period piece as well. It's amazing how bad the men have it, and yet the women have it far worse throughout the story.
For anyone interested, the entire show is free online via Hulu. It's just 12 episodes, so you can easily knock it out in a day or two. As graphically violent as it was, I actually found myself perversely enthralled by all the (non-specific spoilers) decapitations, dismemberment, bifurcations, partial decapitations, eyeball eating, and so on, but the series ended with (another non-specific spoiler) one of the most disgusting, discomforting things I've ever seen on television. (somewhat more specific spoiler) "Whoa, he's messed up. Dang. Wait, what are you doing, Fujiki? ACK! STOP THAT!"
The full manga is also available online. I can't say I recommend it quite as strongly as the anime, which has a tighter story, but it does bring closure. The middle third of the manga is kind of dumb except (more non-specific spoilers) even though it is the bloodiest part of the entire series, I found myself bored because hardly anything happens. Then it ends with another part just as discomforting as the end of the anime.
Whew! Enough spoilers for you?
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
franpa wrote:
Couldn't stand the game over text, so I'll vote no. Other then that it looked fine. I'd definitely vote yes for a play through that avoids the glitches that cause the Game Over prompts though.
It disappears after 4 minutes and only makes a brief reappearance. It was not too distracting for my tastes.
Also, couldn't this movie end a few seconds earlier? I know you have to dodge the delicious fruit at the end, but it looks like you do so only at the last moment. I've only seen runs of this game on YouTube, but i was under the impression that you regain control of the character sooner than what we see here.
Otherwise, great work!
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
Warp wrote:
Regarding a question I have asked previously in this thread, namely: "Is the set of all points inside a unit square as big as the set of all points on a unit line?"
If the two sets are equal in size, that means that there exists an unambiguous one-to-one mapping between all the points on the line and all the points in the square. Would this be a valid such mapping?
Take the decimal representation of the x and y coordinates of the point inside the square, and interleave each decimal digit to get a value between 0 and 1 (which would therefore correspond to a value on the unit line).
For example, if the point is (0.1234, 0.5678) then its mapping onto the unit line would be 0.15263748. Or for example (0.12, 0.3456) would map to 0.13240506.
Does that work? Or is it ambiguous (ie. more than one point in the square would map to the same point on the unit line)?
I've heard of that method before, and yes it works. Both sets have the same cardinality of C. Give me a minute to look it up on Wikipedia...
(Nope, can't find it. However, I did find this article on space-filling curves.)
I'd be interested in whether mapping R to R2 causes any "difficulties". From what I can tell, you may sacrifice the notion of a metric since the curve must necessarily be of infinite length. Any mathematicians want to weigh in?
As for your previous post, you are correct that the distinction between the rationals or reals and the irrationals is very sticky to pin down. I think your assertion needs refinement, though: every real number can be represented to arbitrary precision by a finite string of natural numbers. Once you want to represent a number by its "true" decimal expansion, the number of digits necessary blows up too much and you end up unable to represent all the reals. Note that these so-called "true" representations (infinite decimal expansions) are essential to Cantor's diagonal argument-- you need the infinitude of digits to construct a new number that is still real but not on the list of real numbers.
This reminds me of a conversation I had with a friend in high school. I had learned enough at that point to know that the digits of pi are "random" (which I now know hasn't actually been proven...) but my friend said that it has to settle down at some point into some kind of repeating decimal or pattern. I said no, it doesn't, but he just couldn't wrap his head around the idea that pi was transcendental.
As a physicist, I actually don't care much about real vs. rational numbers or greater-than vs. greater-than-or-equal-to signs, but if you put up with the convoluted real analysis long enough, you just learn to accept it as already proven.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
Warp wrote:
I reached the limits of my mathematical knowledge when a question came up whether you can represent all real numbers using decimal notation.
On one hand, the set of reals is uncountable while the set of integers is countable, and therefore it would seem to me that you cannot represent all possible reals with integral digits.
On the other hand, are there real numbers that cannot be represented with integral digits, when we allow an infinite amount of digits? My intuition would say yes, because a countably infinite amount of digits (from a finite set of them) doesn't somehow turn uncountable. Or does it? If it doesn't, then there would be real numbers that cannot be expressed even with infinitely many digits, but if there are, what would they be?
Strangely, I don't know the answer to this, even though I really should.
Aren't (finite) decimal expansions essentially just successively better approximations to the number in question? I think your question is answered in that sense: all real numbers can be represented by an infinite number of digits because there is no countable way to cycle through them.
For example, if you tried to represent all numbers by way of
1) 0.1
2) 0.2
3) 0.3
...
9) 0.9
10) 0.10
11) 0.11
12) 0.12
...
then you would never reach the decimal expansion for the square root of 2. That is, there is no finite number that encodes the square root of 2's decimal expansion. This would seem to indicate that when we switch to a countably infinite number of digits, the set does indeed become uncountable (see Cantor's diagonal argument). I guess what I'm trying to say is that even though the decimal expansions themselves are countably infinite, the set of all such decimal expansions is uncountably infinite. For more information, research the power set.
(I later decided to bold this because I think it is the most succinct answer to your question.)
However, my guess is that there are hyperreal numbers that cannot be expressed through decimal expansion. I don't know for sure, since I don't have much expertise in them.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
I'll keep it brief: I'm voting No.
It's for the same reasons I voted No last time and the same reasons that are being discussed at length above. I just want my vote to be known by name, not given anonymously.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
Not that I'd be one to lead the crusade (I'm apathetic to the cause and who in their right mind would copy the ending to Carmen Sandiego?), but hasn't YouTube gotten extremely conservative about copyright infringement lately? Couldn't TASVideos users claim unfair use of their copyrighted material and spook YouTube into removing the videos and possibly even suspending or banning the user? By the time any copyright claims are challenged and overturned, the user should be reeling from several infractions.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
I don't really have anything to weigh in with here except to say that I too have been studying assembly language (Z80 and 6502) and reverse engineering RNGs. Great minds think alike, eh, MetalStorm?
If you'd like to collaborate and/or compare notes, talk to me on IRC.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
So... uh... now that I'm not going to be the one who sticks out his neck and call "troll"...
Does anyone else think WiiUltimate is our old friend Billy? This looks like classic Billy to me.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
VanillaCoke wrote:
Bobo the King wrote:
No, I'm afraid I still don't get it. You seem to be hung up on the fact that the rule is arbitrary. So it is. But that isn't a reason to fight the rule, it just gives you the means to fight it, should you choose to do so.
So you've chosen to do so. Great. Request that the video of your run link to your YouTube encode, submit your run, and bask in all your glory. You've done great work and should be rewarded for it. Why are we arguing about all this?
I reiterate my agreement with AngerFist.
Edit: I like big fonts and I cannot lie...
Let's assume that SoulCal submits his .dtm and requests in the submission text that he wants his channel as the sole host for his TAS. Oh no, that request is denied. Now SoulCal has now relinquished his only bargaining chip. Nooooot the smartest move from his position.
Also, here's a thought. Quite commonly, a TAS on the workbench is rejected for an arbitrary goal. So why couldn't a TAS reject the workbench for an arbitrary rule?
Well, that's why I said he should make the request before submitting the run...
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
SoulCal wrote:
Bobo the King wrote:
SoulCal, I have no idea why you would turn down more recognition. Basically, you're willing to limit yourself to a much smaller audience as long as they all understand it's you, rather than a larger audience among whom a small proportion will mistake your work for another's.
Again, recognition is only 1 reason out of about 5 I've listed. One of the bigger ones though is we have exceptions totherule, making the rule arbitrary. I've listed my other reasons in past posts, so I'm not going to sound like a broken record any longer by repeating myself. I agree with AngerFist this discussion should move on, and be more productive.
Bobo the King wrote:
Suppose keeping your TAS unpublished and exclusive to your own video channel means you get approximately 1,000 viewers (I checked how many views your videos have gotten). Now suppose that publishing your video means that it gets 10,000 viewers, 2,000 of whom fail to recognize that it's your work while 8,000 say, "Hey, this guy SoulCal has made something really great." On what planet is that a bad thing?
TVC doesn't get but 500-1000 views with 1/2 the content it posts, and it has 5000 subs. And I couldn't care less about my view count anyway.
"AngerFist wrote:
SoulCal, if you want your run to solely be uploaded on your channel then fine.
If someone high up on the TASVideos ladder can agree to this, then I will gladly submit my dtm the moment it happens. New runs bring new people to the site, so we all benefit. My encode also has annotations which explain many common questions people have (sorta similar to the audio commentaries on some TASes). I've had quite a few people say they're very helpful when trying to follow along when they're confused, but there'd still be a run for download w/o them.
No, I'm afraid I still don't get it. You seem to be hung up on the fact that the rule is arbitrary. So it is. But that isn't a reason to fight the rule, it just gives you the means to fight it, should you choose to do so.
So you've chosen to do so. Great. Request that the video of your run link to your YouTube encode, submit your run, and bask in all your glory. You've done great work and should be rewarded for it. Why are we arguing about all this?
I reiterate my agreement with AngerFist.
Edit: I like big fonts and I cannot lie...
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
SoulCal, I have no idea why you would turn down more recognition. Basically, you're willing to limit yourself to a much smaller audience as long as they all understand it's you, rather than a larger audience among whom a small proportion will mistake your work for another's.
Suppose keeping your TAS unpublished and exclusive to your own video channel means you get approximately 1,000 viewers (I checked how many views your videos have gotten). Now suppose that publishing your video means that it gets 10,000 viewers, 2,000 of whom fail to recognize that it's your work while 8,000 say, "Hey, this guy SoulCal has made something really great." On what planet is that a bad thing?
Edit: And I'd like to echo AngerFist's above sentiment.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
Spikestuff wrote:
Nintendo3DSandWiiFan wrote:
The DEsMuMe is not compatible with 3DS games
(including Lego the Lord Of the Rings and Scribblenauts Unlimited)
I'm sorry but if you wanna record videos you are going to need a Capture Board for 3DS.
Hi Captain Obvious! (had to)
A Capture Board is what we aren't looking for that's something for SDA. (Oh and to let you know, they already know about this)
Here at TASVideos we use emulators to do what the human cannot... Go through each frame of the game for a period of time to perfect and get completed in the fastest amount of time, then learning a glitch then using that to beat it faster and faster and than it becomes like Super Mario Bros. by Happy Lee where the game is done to perfection.
3ds is something that won't be in the works for maybe another 6 months maybe more or maybe less and it can probably be like the psp emulators, 3 dev teams later we finally get one that is the most accurate and can run a wide range of psp games (ppsspp).
Desmume team is most likely working hard for 3ds emulation probably even other teams.
And to clarify, here at TASVideos we do not use consoles unless it's for verification.
"Here at TASVideos, we strive to push games to their limits. The emulators we use allow for undoing mistakes, slow-motion gameplay, and even in some cases utilizing robots to do our bidding."
While his post was a little obvious, it was his first post. Cut him a little slack and allow him to introduce himself to the board.
I know next to jack about the inner-workings of emulators, but I'd conjecture the question here is along the lines of whether the 3DS uses the same assembly language and overall architecture as the DSi, no?
I mean, ultimately the question is meaningless because DeSmuME does not support 3DS games, period. That is the only "objective" answer. "Okay," you respond, "But how similar is it to the established DS line of systems' coding?" The answer is tautological: it is exactly as similar as it is. Whether DeSmuME can be adapted to work with 3DS games is entirely a judgment call on the part of its developers.
It reminds me of this Shakespeare quote:
LEPIDUSWhat manner o' thing is your crocodile?MARK ANTONYIt is shaped, sir, like itself; and it is as broad
as it hath breadth: it is just so high as it is,
and moves with its own organs: it lives by that
which nourisheth it; and the elements once out of
it, it transmigrates.LEPIDUSWhat colour is it of?MARK ANTONYOf it own colour too.LEPIDUS'Tis a strange serpent.
Experienced Forum User, Published Author, Player
(79)
Joined: 8/5/2007
Posts: 865
SineSwiper wrote:
Hmmm, after looking at that and some older post about BisqBot, I wonder if the BisqBot concept can be taken to a more extreme level. Call it the DeepBlue of TASing.
Some ideas of improvements:
1. Make it really think ahead, like 10 seconds ahead.
2. Give it more (weighted) variables as to what is considered "better". Make things like losing bigness a negative.
3. Have some form of smaller loops that save good inputs, so that the full 10 seconds isn't completely random. For example, as soon as it kills a Goompa, consider that input pattern to be a "good one" and keep duplicating that for a while.
I guess item #3 would be explained as almost like some sort of multi-looped binary division:
10 second loop --> 2 5s loops --> 4 2.5s loops --> etc.
I'm sure there is some form of genetic programming that would help out here, too.
I was thinking of something along the lines of "trajectories", by which I mean, "If I were to execute the following button combinations over these frames, how would I expect relevant RAM values to change?" This would allow the bot to peer into the future predictively and relatively efficiently. In principle, it shouldn't be too difficult and should closely model human play. In practice... I'm in over my head.
