Posts for Randil

Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Was there anyone beside me who found this game scary? I mean, that monkey looks so weird (in a scary way) with his arms waving around, and the strange music doesn't make things better... It actually took me a while to see that it was indeed a monkey, at first it looked like... some kind of monster of something. Perhaps I have too wild of an imagination. Hmm, well, as for the run itself, I decided to vote meh. I could clearly see that it was tool-assisted, and it was somewhat entertaining. However, I found it a bit too repetitive, and the music hurt my ears.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Really great run, I enjoyed every second of it. The technical quality looked very good too, so here's a strong yes vote from me. Great job!
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
I did it this way, and I got a really bad (but different) approximation, but it still holds: Throughout this whole argument, I will assume x>1. exp(x) = Sum[(X^k)/k!, k=0,1,...)] > (pick out the n+1:th term from the taylor series) > (x^(n+1))/(n+1)! For the polynomial we have: P(x) = Sum(a(i)*x^i) < Sum(abs[a(i)]*x^i) < Sum(abs[a(i)])*x^n Combining this we get exp(x) >(x^(n+1))/(n+1)! > Sum(abs[a(i)])*x^n > P(x) We want to find an x such that the middle inequality holds. Dividing it by x^n, we get that exp(x)>P(x) if x > (n+1)!*Sum(abs[a(i)]) =: x0 I would be interested in seeing more approximations.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
This topic has been discussed in this here thread.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
One advantage of fm2 is that the movie file is saved in text format. This means that you can open the movie file in a text editor, such as notepad, and edit the input you've recorded.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
As many of you probably know, the exponential function is larger than any polynomial for sufficiently large x values. To illustrate this, let's say we have an arbitrary polynomial of degree n, P(x) = a(0) + a(1)*x + a(2)*x^2 + ... + a(n)*x^n. Find an x0 such that exp(x)>P(x) for all x>=x0.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
TASVideoAgent really grinds my gears, my vote goes to him, no doubt about it.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
I agree with what zidanax said last, that pausing the game is probably the best way to go. That way you can take all the time you need to read each text box, and then just press the pause button to continue watching. I've been working on doing a lua script like this for my story book TASes, but I'm not finished with it.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Yeah, very entertaining to watch, thanks for making it!
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Yeah, very well written post. I think 2010 will be a great year too. I'll try and write some of the game tricks pages I've been planning on for quite some time, I have quite a few on my mind, just NES games though (surprise). And as always, you can expect more TASers from me this year (don't worry, it won't be duck hunt again) :).
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
moozooh wrote:
Wasn't there a thread about the exact same thing a couple years ago?
Yep. (although that wasn't the main purpose of that topic). I pronounce it like ass but with a t.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
I think that's a nice idea, I proposed it in the submission thread, so let's see what everyone else thinks.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Mr. Kelly R. Flewin had this idea: How about a run that starts from a savestate on 1 duck mode round 99, beats this round, shows off round 0, the kill screen (i.e. not kill the birds as fast as possible but rather as late as possible, to show how fast the birds fly), then resets the game, and plays 1 round each of 2 ducks and clay shooting? The delay times between birds and clay pigeons are variable, but they don't seem to be manipulatable.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Well, allright then, I will uncancel it, I guess it was a rash decision to cancel it. I had doubts about how this would be received, but now it seems more people liked it than I thought. Even if this gets rejected in the end, I'm happy to see that quite a few of you actually liked it. :) I'll uncancel this and leave it to its fate! Nach: Yeah, you can save time by avoiding the perfect screen, but I think that would look a bit too sloppy, even if it does indeed save time.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
samurai goroh wrote:
I would like to see this movie, so I hope someone encode it :)
Youtube'd for your viewing pleasure.
Post subject: Re: TOP GEAR INFO YOU'RE SO WELCOME
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
alden wrote:
adelikat wrote:
Does anyone really want to see a full run of top-gear?
I was working on Top Gear a bit before this all came up. I never tested, but several resources claim that you can go much much faster if you use manual rather than automatic transmission in the red car (the current movie uses automatic). There's 8 countries/regions with 4 tracks each, somewhat increasing in difficulty, but they have to be unlocked through normal play or passwords. There is also a glitch that theoretically allows you to get both 1st and 2nd place. Other than that it's however many minutes of cornering well. I was working on simply beating the existing movie, using the same savestate. TASing racing games is tedious in case anyone is curious.
With enough information about the game's mechanics and level layout, I don't think it would be too hard to make a lua bot that TASes the game for you. Tedious and repetitive TASing is exactly what bots are for!
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Yeah, I agree with what you're saying. So 1 round in "1 duck" mode, reset, 1 round in "2 ducks" mode, reset, and then 1 round of clay shooting? It should be about 1-1.5 minutes long, so it certainly wouldn't be too long. Hmm, I'll see what I can do.
Post subject: Re: Duck Hunt
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Nach wrote:
But in Duck Hunt, would a video show any more than some birds dropping as soon as they appear on screen?
This is exactly what a "fastest possible" run would look like (see my youtube video here to get the idea). Let's say that, in theory, I submit a Duck Hunt TAS, what do you think the goals should be? Fastest time to beat X levels? What should X be in this case? Should it perhaps sacrifice time to kill birds as late as possible, just for variaton?
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Dwedit wrote:
Can you play Duck Hunt with a joystick with TAS? Just send the correct patterns to indicate a hit.
You mean like with a gamepad? Not to my knowledge, though it would be cool if it worked. I have done a 100 rounds TAS of the 1 duck mode. It clocks in at 1h 15min 12sec, and ends with a score of 4 664 100 (27 400 more than in the youtube video). I recorded it with a lua script, and thanks to that this movie only took me about 30 min to make. Not many TASes take shorter to make than to watch! It seems microstorage can't handle files of this size, so I uploaded it to Media Fire, and you can get it here if you actually want to watch it. Unfortunately, since this is a TAS, it doesn't show off the speed of the bird on round 0, but rather kills it just as fast as on the other rounds - you can watch the youtube video Flewin pointed to if you want to see just how fast the bird moves. You can now all rest assured that a 100 rounds movie has now been done.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Mr. Kelly R. Flewin wrote:
Actually, there's a kill screen for Duck Hunt, go figure. http://www.youtube.com/watch?v=sWfoUmCf0OI and no KS for 2 birds http://www.youtube.com/watch?v=wXoWn_e_7Ss Mr. Kelly R. Flewin
Huh, I didn't know that. This got me interested in doing a complete 1 duck run, just to reach (and beat if possible) this kill screen.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
I have made a Duck Hunt topic in the NES section here where I answer these question and where we can continue discussing Duck hunt.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
Don't worry, I won't submit a Duck Hunt run or anything - I realize and agree that it's really not publish-worthy. Still, perhaps some people will enjoy seeing a Duck Hunt TAS - either way, you can't stop from making one and posting it in the forum! :) I have made a "2 ducks mode" TAS that aims for fastest possible time to get maximum score. Here is the .fm2. It gets 999999 points in 15 minutes. The score you get for each bird depends both on what round you are on and the color of the bird you shot. The bonus you get at the end of the round depends on what round you are on. Perhaps I'll do a 1 million points run of clay shooting later if I'm feeling up for it. In all game modes, after getting 999999 points, the score counter goes back to 0, as you can see in the movie. There are 99 rounds in all modes, and after that, the rounds counter turns to 0. So the only way to "beat" the game (in any game mode) is to beat 100 rounds so that you return to round 1. Any TAS that does this would be around 40 minutes long.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
There are also cases where the real time speedrun might be slower than published TAS, but uses a new route or trick that would save a significant amount of time if applied to the TAS. Goonies 2 for the NES is one example, the run on SDA uses a new route that would save many seconds in the TAS, although it times in at a higher time. There are also many games that have short and incomplete WIPs in the forum that show big potential improvements in the published TAS. Rockin' Kats for NES is one example. Perhaps it all boils down to how much the run is known to be improvable by. This raises the question, when is a run "very improvable" and would fit such a list? If it has 20 seconds of known improvements? 10? 5? 1 minute? This discussion raises many deep questions... :) I'm not taking any stand here, I'm just bringing up some points.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
I personally think that the only way to TAS Solomon's Key efficiently, (i.e. not having to spend 20+ hours on one level due to horrible randomness, which has happened to me), you need to fully understand the game's randomness - both in terms of what affects randomness, how it affects it, and what factors are affectable by randomness. I think this feat will take some ROM hacking skills (Bisqwit has made some progress, see this post.) I think it will also be necessary to somehow track the RNG of the game, so you know the RNG-state of the game at any frame in the movie. The feat described above is beyond my level, but if someone were to do the above and explain it to me, I might consider (at least helping out) TASing the game. I can't really stress the importance of cracking the game's RNG before starting a Solomon's Key TAS. Just jumping into the game, even if you have some new strategies at hand, will likely end in despair due to the game's randomness taunting you. With all this said, I would like to see a Solomon's Key run! I think Solomon's Key would make a nice team project/competition for a group of TASers.
Experienced Forum User, Published Author, Skilled player (1890)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
That Duck Hunt TAS isn't very optimized, in fact, you can usually shoot the ducks before they appear. Since this game is unsuitable to TAS, it's right up my alley, and I made a TAS of it: .fm2 and youtube. It's actually not that bad, and kinda funny if you're in the right mood. I think there are worse games to TAS than Duck Hunt, although I wouldn't say it makes for a great TAS.