It'd be nice if you could finish it before you consider it not worth finishing at all anymore for whatever reason (f.e. new improvements). Would be a pitty.
Enjoyed watching it.
Your choice of making the new challengers lose once kind of bugged me. I was looking forward to see their combos, but they just got knocked out by the old team. Maybe they don't have many combos to showcase? It still was kind of irritating to me.
I'm not sure whether I should vote yes or meh on this. I was entertained, but think a more entertaining movie could have been produced with slightly increased effort. Maybe it has to do with personal taste though.
It contained a floating Hadouken glitch and some crazy glitchy combo with Chun Li that I wouldn't like to spoil, so I'll vote yes. I personally hope to see it obsoleted somewhen though. :X
I didn't get all of it, but it looked pretty cool. Very high standard for a first TAS, or at least you managed to fake it well. ;)
Only complaints I have are the balloon flights. Although it's a nice idea not to just fly straight to the goal, the wobbling was too repetitive for my tastes. I guess you could have maked it look better. I'm also kind of sad I couldn't read the ending dialogue, but oh well, I guess it can't be helped. :X
Yes. And thanks for the encode. ^^
At the starting speed walking straight to the next "intersection" takes 16 frames, turning takes 14 frames. At the next speed level walking to the next intersection takes between 12 and 13 frames, turning still costs 14.
At the starting speed your proposal is 6 frames faster by my calculations.
Current route = 8*16 + 14 + 2*16 + 14 + 8*16 + 14 + 16 + 14 + 5*16 = 440
Your route = 8*16 + 14 + 2*16 + 14 + 3*16 + 14 + 16 + 14 + 2*16 + 14 + 16 + 14 + 3*16 + 14 + 16 = 434
He was already going at a faster speed though.
Actually, moving speed walks somewhat like this:
Between two interesection there are 8 steps Sonic has to move.
At his initial speed he takes a step every second frame. That's 4 steps in 8 frames.
At his second speed he pauses 3 frames for 5 steps. (pause, move, pause, move, move, pause, move, move). After having Sonic turn the chain doesn't always start from the same point. That's 5 steps in 8 frames.
At his third he puases every fourth frame. That's 6 steps in 8 frames.
At his fourth he pauses every eigth frame. That's 7 steps in 8 frames.
At his final speed he moves 8 steps in 8 frames.
There are 5 speed levels. Here is how many frames it takes Sonic to go from intersection to intersection on average for any of them:
lvl 1: 16 frames
lvl 2: 12,8 frames
lvl 3: 10,67 frames (10 2/3)
lvl 4: 9,143 frames (9 1/7)
lvl 5: 8 frames
turning: 14 frames
Edit: First input is accepted 3 frames before Sonic's shadow appears. That marks the start of the bonus level. From there on every 1800 frames the speed level increases.
Jumping doesn't affect speed.
Walking into a bumper and getting reversed:
lvl 1: 16 frames
lvl 2&3: ~14 frames (only got 14 frames by testing, but it feels strange)
lvl 4: 12 frames
lvl 5: 11 frames
Turning around by pressing forward at lvl 1:
18 frames till you're at the same position
By turning around 6 frames earlier, then immediately turning right or left only 8 frames are wasted for turning around.
Instead of 8 steps you need to go to the next intersection, some events seem to be triggered earlier than that. Blue spheres turning into rings or the level ending happen 3 steps early for speed lvl 1 and 4 steps early for lvl 2.
Collecting a ring costs 1 frame at the score screen.
If you time it right you can jump over the rings the blue spheres turn into, saving one frame.
If he selects at random from the other 2 doors, the game master will knock you out in 50% of the cases where you'd normally win by switching. That limits the problem to only 4 cases in which you even have a choice. In 2 of them you get the car by not switching, in the other 2 by switching. Chances of winning are 33.33% no matter what.
As the amount of available doors (with goats in them) increases, it stays the same. Switching pays off for the friendly game master and it doesn't matter for the random one. If they open more than one door at a time, it still stays that way. Not sure what happens if there are more cars though.
If he thinks Grunty's graph proves his point he should probably look at a graph that plots how many more/less tails than 50% have been thrown for increasing coin tosses. The diversion should be printed in numbers, not in percentage. The resulting curves should look pretty random for good rngs. >_>
If he's so sure he's right he should try and get rich with roulette.
If the program is written in C/C++, I seem to recall that its rand function is just a repeated long sequence of numbers, so they aren't really random. I have written a program that graphically illustrates this years ago iirc. Nevertheless, it shouldn't be possible to predict the outcome of a coin toss the way he did.
Maybe he's playing a trick on you. Or his understanding of randomness is that off that he wrote a piece of code that's more akin to his slightly twisted understandings of it than to the definition of randomness.
I suspect that his code simulates random distribution or that he uses a strange way of generating random numbers.
Or, at least, that's the point of all this. Except it depend of the game, the authors and the viewer, but in a different way.
Take, for example, Phil's Combatribes movie. I really hope this doesn't attract any drama. This movie was published here just because he could, no one actually voted on it.
About the games, some genres are negectled. See fighting games. Some people vote "no" just for disling the genre itself, without even watching the movie. And how many full-game racing games we have, besides Mario Kart and clones? And yet, we have really really long RPGs runs, and no one complains about it.
And the viewers... Well, search for "Ocarina of Time" rejected movies, but with oh so many votes.
Well, the reason why fighting games are neglacted is because with most games you can only choose one character for the whole movie. The movie is trying to finish the game fast with the most amount of entertainment that's somehow still possible. Many people agree that this concept fails to deliver enough entertainment though because it's just one character's moves and I agree too. Shang Tsung is an exception of course.
For fighting games I'd like to see movies that can display more crazy combos by the whole cast and not just one fighter. Concepts that I can imagine doing this would be fighting in 2-player mode. I imagine it to be ridiculously hard to come up with good ideas of what actually should happen during the fights to make them most entertaining though. It'd be fun if they also involved two authors.
I can see how racing games can also be too boring. I suspect some would actually do better with time trails as opposed to full game runs. Mario Kart f.e. is an exception imo because it profits from the entertainment provided by bouncing off from other karts. For the majority of racing games, however, it'd make more sense to me to have every track recorded seperately, starting from save states, and encoded into seperate files then. That way single tracks could be obsoleted by different authors easily. The problem would be how to make this downloadable as it would require an entirely different format than the rest of the content of this site. Bringing out a new collective torrent whenever a single track has been obsoleted comes to my mind first. It would feature all the tracks of the previous release with the exact same file names, except that one (or more) files are different.
I realize this would be troublesome to do, but I can't think of any other format that would suit the majority of racing games. As it is, lots of their TASes seem to prosper on youtube instead of here where they'd belong. ;) Sorry, if this suggestion doesn't seem useful, but imo the problem with racing game submissions is their format, not the genre.
As for rejected movies with huge amounts of yes-votes. They are mostly aiming for speed as primary goal. It makes sense to still reject them if they aren't fastest as a new movie will arrive sooner or later anyway and it'd be an unnecessary step in between. For OoT there has even been made an exception to this because making a run used to last even longer than it does now. I'd say most of the time, the judges' decision seems reasonable. But of course you can't satisfy everybody.
For entertainment focused movies, the only valid reason I see for their rejection would be if they fail their goal and aren't entertaining enough, just like speed focused ones get rejected for failing to achieve their goal. Votes would have more influence in the judjing of these. Obsoletion would have to work different, but it would. I can imagine a poll where viewers vote for the least entertaining scenes, then they are redone completely. The most entertaining ones would keep unchanged for the most part.
Obsoletion only becomes problematic if there are two entirely different runs. If they both follow a different theme (basic idea where the entertainment comes from is different (f.e. lots of glitches, strange routes)) and are done well enough I can imagine both being published. If that's not an option maybe one would have to be dumped. As there'd be high risk for the newer movie to be dumped, I don't see many authors attempting to create problematic movies like that. So the problem might occur only rarely.
"I'm just doing some sort of freeplay." would be problematic because that's a totally arbitrary goal and hard to obsolete. Movies like these should have goals like "most surprising strategies", "most beautifully looking" and so forth instead. It'd be still hard to find out which movie beats the other, but that's what polls are for.
What would you dislike about the SM64 run I gave as an example? It wouldn't repeat any basic strategy and not just use BLJs, but surprise you anew with every collected star. "I didn't think that was possible!", "Ah, that's why he did that, it was to set up this!", "Wow, I admit that looked kind of cool!" and so forth would be reactions that hypothetical movie commonly provoked.
I get your general argument and agree, but I don't see how it affects my proposal. You were quoting it which led me to think you'd think it would.
I guess this misbelief often comes from a misunderstanding what randomness means. The thinking process can work like this:
The chance is 50/50, so most of the times the count of heads and tails will be even. Therefore, if we throw a coin for 20 times, there should almost always be 10 heads and 10 coins. So, if there are already 5 heads on the first 7 throws, the next ones have to be tails because it will make it even and most of the time it gets even.
That thinking process falsly assumes random distribution (of an equal amount of outcomes) instead of randomness.
In reality, the only case where it's sure that results will be even is if you throw a coin for infinite times. Wait, that's also not realistic. >_>
And if anything, previous results are more likely to repeat under realistic conditions. The coin is almost definitely slightly heavier on one side.
how many of the currently-published runs would benefit from "free play"?
Not many if any. Most of the existing ones focus on speed and adding in a bit of free play would make it half-assed. Those would have to be independent runs.
Warp wrote:
Note that what you are describing is something which could entertain for a minute or so. It would be foolish to try to make an entire 20-minute run which completes the game and just fools around such a concept.
That convinced me. I noted it, it truly was foolish of me to suggest such a thing. Really though, just because it wouldn't entertain you for more than a minute doesn't mean everybody else woud feel the same way.
Warp wrote:
And there are only so many "funny tricks" you can pull off in one game.
Since you took the "drunken master" like concept thing as an example, I'll say that there are games where the amount of funny tricks is high enough. For example, I bet that coming up with 70 hilarious and surprising strategies for getting a star is possible in SM64. Such a run would be hard to do and require lots of creative planning though.
Warp wrote:
If you repeat it too much, it quickly becomes boring,
Yep, that would be bad execution or bad game choice.
Warp wrote:
especially because it just wastes time rather than completing the game sooner, making the video longer than it could be.
This is basically saying it's boring, especially because it's not fastest. I don't get the causal relation here.
Warp wrote:
(In other words, rather than a 20-minute speed completion you get 30 minutes of fooling around; is that really going to be entertaining?)
This is saying it'd be boring because it's longer, I also don't get that argument. Whether it's really going to be entertaining depends on the effort the author puts into the movie.
Warp wrote:
Your idea might work for some very short games, those which can be completed in less than 5 minutes, and which lend themselves to such "free play".
Of course it isn't going to work for any game. Note that finishing as quickly as possible also doesn't.
Warp wrote:
However, the average length of the published videos is probably somewhere in the 20 minutes. Making them longer is usually only going to make them more boring to watch.
I'm starting to think what you're talking about is a run that does everything a normal speedrun would do, except for adding derails. That's not what I meant, I was talking about totally different and surprising routes.
Note how I said "the game is finished in totally obscure and surprising ways" and not "the game is finished just like normal with a bit of additional (time wasting) fooling around squeezed in".
Warp wrote:
The first 5 minutes might be funny, but it gets old very fast.
You seem to assume the run would be full of repetition. I don't see why it has to be.
Warp wrote:
There isn't that much variation in games. I don't think "free playing" is a panacea which will suddenly make long videos more entertaining, if you are bored of watching them.
I also don't. Your initial post was about how squeezing in additional entertainment into already existing runs wouldn't be going to make them more entertaining in the end. I agree with that. With my post I wasn't expressing disagreement about that. I was talking about entirely different concepts because I thought bringing them up would fit the topic title, not to start an argument.
Warp wrote:
As I said in my original post, if you want such a "free play" run, then make one, put it on youtube or somewhere, and post links.
Why not put it on this site if it's good?
Warp wrote:
It's one thing to talk about it, and another to actually do something about it. Maybe if you succeed, it will catch on.
I know. Unfortunately I don't feel skilled enough in any game to do such a run on my own. Taking the time to get skilled enough also wouldn't be worth it for me.
I'd really like to see some more entertainment focused concepts for runs.
I think multigame runs are a nice idea, though I miserably failed to finish the one I started in the past and it doesn't seem like I'll ever have the time to do so. But I think they are nice examples of when entertainment should be prioritized over speed. Having multiple characters move in unison isn't necessarily the fastest option, but it just looks so much better. Also with multiple games, finding the ultimately fastest input is quite problematic, so to me aiming for speed as a primary goal makes less sense in those. The big problems with those runs are the game choice, choice of game quantity and the music.
There's a glitchfest run on this site, showing off as many glitches as possible. Although it's quite a nice idea as well, I ended up fast forwarding through a lot of it. Making those runs not finish the game would feel lame, but doing so inevitably makes them filled with boring glitchless sections.
I'd like to see other creative ideas for runs. Some "drunken master" like play could maybe be interesting where the game is finished in totally obscure and surprising ways. The game would have to provide a high degree of freedom to the player and the player would have to have loads of knowledge about it though.
Maybe a run where every action has to happen so it fits the music? This has been brought up as an idea for a certain game featuring Michael Jackson before.
I honestly think that most runs using such concepts would be easily accepted on this site if they were executed well enough. Whether they are more entertaining than the fastest completion would depend on the authors, the game and the viewer.
It might a good idea to do so then.
This is a tweaked version of my previous submission. Stylistically it is mine, not HL's.
I apologize then. I knew I could be a bit off with what I'm saying which is why I included the part in the brackets. I didn't want to take the time to get what's necessary to watch the runs, but still felt the need to speak out, just in case. :X
The idea for improvement didn't come from the author of this movie and it seems to be rather identical to the other submission other than that (without having watched either). I think the author of the earlier submission would have improved the movie on his own if he had been told the idea for improvement and was given enough time. In one of his posts in his submission thread he asks for details on how his run would be improvable, but it seems he received no answer. I think he should be the (co-)author of this movie.
Thanks, alden and Derakon! ^^
I've watched the run. It looked seemingly perfectly executed and the game is really well suited for a TAS. I'm glad that so many different forms had to be used.
I was being kind of sarcastic. Imo it's the users' responsibility not to watch non-entertaining runs. Ratings are displayed anyway and if it's the fastest completion of a game they like, there might be people who still enjoy it.
We could just add another symbol similar to the special star to mark bad runs. It should somewhat look like a danger sign and protect visitors of this site from boring content. "Watch at own risk. Wasted time won't be refunded."
Low ratings are displayed for those movies anyway, but careless users might miss them. Then again, the danger sign could be missed as well. So maybe taking down those movies would be safest after all..
The danger in not using your brain is that you could believe absolutely ridiculous things that only benefit some certain spiritual leaders and neither you nor your friends. Of course you can also use it too much and that can actually cause the exact same problems.
Btw, the first part of your post seems familiar. Why not just post a link to your youtube video? ;)
Thanks. ^^
However, it desynced for me when Link crawled back out of the hole. He then went to slash some bushes after which he read the big sign instead of cutting it in half. I used the same plugins and my version of the game has the correct checksum.
Edit: Tried to activate the MemPak option which only made it desync earlier.
