I have an idea.
How about -- we start allowing movies to end as early as possible, but we'll also write the actual completion time to the movie comments, where applicable (and where volunteers are willing to do that)?
This way everyone should be pleased.
Movie speed comparisons can be done on the movie length basis, but I'll reserve the right to choose the entertainment modifier (which might sometimes favor a slower movie over a faster one) based on the actual method of game completion. This should give a fair chance for movies that beat a game in an impressive way even if it causes in a longer input.
Edit: (grin) - I know this is not *my* idea. I just tasted other people's ideas and chose the combination that didn't feel bad :)
Edit 2: Here are some other thoughts I had - already partially discussed in IRC.
- I don't like the idea of punishing the movie author of displaying the game intro and cutscenes. They are entertaining content, and displaying them should be favorable.
- However, this contradicts with the fact that people like to think in numbers, when they compare movies.
Thinking of that, what will happen if I decide to kill Bowser with fireballs? I will probably do that in my warpless version since killing all Bowsers with them, is faster than grabbing all hammers. Also, Yy in his video stop his movie later after he got the hammer.
That rule is somewhat compromised since I will be the first one to use fireballs in a warpless version.
How about -- we start allowing movies to end as early as possible
So if I make an SMB speed run, how am I supposed to know if it's faster or slower than Phil's? It'll be hard to tell whether we stopped at exactly the same time. I would also have to jump higher if I wanted to be able to stop the movie sooner, and that goes against my style of jumping as low as possible. I also might get lucky with the positioning of the axes that he throws and be able to jump even higher. Does that mean that I beat the game faster?
Personally I am a fan of only stopping at the point where no more user input can speed up the onset of completion.
In SMW for instance its fine to end the movie when the final mecha koopa is thrown up as after that the game is out of your hands. Same would hold for a final shot in Megaman or the elevator in Metroid.
I am not a fan of the 'it ends as long as the viewer cant stop it from ending' idea again by the reasons people have said of maybe pausing the game or whatnot.
I am worried though that the point at which no more user input can speed up the completion may be put into question.... like what if holding start made the credits scroll more quickly? The end the input as soon as you can idea would avoid any dispute.
How about -- we start allowing movies to end as early as possible
So if I make an SMB speed run, how am I supposed to know if it's faster or slower than Phil's?
You take Phil's movie and see which frame something happens, and compare it to the frame number from your own movie. Just the same way as you've done to compare level times.
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
even after touching the axe, the watcher can still pause the game, i dont see why should we worry about watcher input, it's so much simpler to allow competition for shortest movies possible.
Joined: 11/27/2004
Posts: 29
Location: Calgary, AB
I have a question. Let's say I did a time attack of Cobra Triangle where I beat the currently published time by like 30 seconds. That is to say, I complete level 24 faster than the previous submission, but I don't stop my movie there. Let's say I keep it rolling so that I can input my initials after the ending sequence. Even though the movie is longer, the actual completion time would be shorter. Would such a submission be considered 'superior' or not?
This site makes it clear that the goal of time attacks is ENTERTAINMENT. The end result is supposed to be AVI files that are interesting and enjoyable to watch. As an end user, I gotta tell you guys: I would have been much more impressed watching Cobra Triangle if I had not had to wait for 60 seconds to elapse at the 'input initials' scene before the movie would end. It caught me off guard and I started wondering why on earth the 'author' refused to put in his initials at the end. After perusing this thread, I see his motivations.
Using early movie stop times is fine for competitions, but they shouldn't be used if it jeopardizes the entertainment value of the run.
As a final note, when I watch a SMB time attack, I like to know how it compares to the real world record on Twin Galaxies. If it blows that away, I'm more impressed. How on earth can I make that comparison if one time is for 'touching the axe' and the other time is for 'least frames used for input' or 'movie stop' time? I think all videos on the site should have the REAL completion time listed.
The above are my opinions and observations, no need to flame.
Joined: 4/16/2004
Posts: 1276
Location: Uppsala, Sweden
Out of curiosity I must ask, did you beat my record? Also, when someone (me in this case) stops a movie my input ends there. So In that case I couldn't have entered my initials because after I stop the movie it's up to the encoder (in most cases Bisqwit) to set the length of the avi. I guess he could have shortened it a bit. Like stopping the avi after 1 second on the high score screen.
It's however an interesting thought, I have done some more games with high score-input after the game ends. Maybe I should redo them and put my initials in BUT the movie might be longer?
Joined: 5/1/2004
Posts: 4096
Location: Rio, Brazil
agreed,
// The recording should be stopped as soon as player input can't make the avi shorter. //
this is the rule i think is the best, maybe in other words it would look cooler, but the idea is that.
I would have been much more impressed watching Cobra Triangle if I had not had to wait for 60 seconds to elapse at the 'input initials' scene before the movie would end.
I had no problem with that. It let me listen to the score entry song :)
As a final note, when I watch a SMB time attack, I like to know how it compares to the real world record on Twin Galaxies. If it blows that away, I'm more impressed. How on earth can I make that comparison if one time is for 'touching the axe' and the other time is for 'least frames used for input' or 'movie stop' time? I think all videos on the site should have the REAL completion time listed.
Well, in a game such as SMB, sure, but remember, even something as popular as SMB is a completely different game - comparing them isn't too fair. You're right that one is touching the axe, one isn't, but also one uses alternate pipes, solid object walking, and wall jumping, and one doesn't. The ending point is probably the smallest variation. And just about every video here that incorporates the slightest glitch will be different from Twin Galaxies in that regard.
So did we get a verdict on this? I'm about to finish my Kid Chameleon video (finally) and I can finish input a few seconds before the movie ends. Doing this means that I will touch the last goalpost a lot later though. Should I
1) Stop movie as early as possible and finish the level later? (other previous KC videos have done this)
2) Stop movie earlier only if the completion time of the level is the same?
3) Stop the movie when I finish the level? (my previous KC video did this)
Quick answers would be appreciated.
2
Though this brings up an issue. I'll use NG1 as an example. I chose to stop it after I killed the last boss, but technically I could have kept it going to hurry through the ending story sequence.
Do you stop it after there is no more gameplay, or after there is no more possible input?
Personally I think it should be stopped after the gameplay is stopped, and the game will continue with the ending story and/or credits on it's own. This is how most movies are done now.
Well, two of the games I've done so far (Wai Wai World and Monster Party) have endings which can be sped through with user input, and both times I have had to include these parts in the movie. But since yours is already published, I wonder how it'll be handled...
And I would vote 2 here as well.
-Josh
but then you take my 75 perchance chance of winning, if we was to go one-on-one, and then add 66 and two-thirds ch...percents...i got a 141 and two-thirds chance of winning at sacrifice
IMO, The game is complete when the final objective is finished.
In Dragon Warrior your objective is to defeat the Dragonlord and return to the castle. You cannot beat the Dragonlord and end the movie because you still have to make your way back to the castle and receive the King's congradulations to finish the game.
In Ninja Gaiden, your objective is to defeat the final boss. Speeding up the story at the end isn't part of your objective and you honestly aren't accomplishing anything.
Game objectives are usually given in the instruction manual or in the game itself. Usually it involves defeating some sort of final boss or something similar. In the case of Dragon Warrior you also have to make your way back to the castle, this is mentioned in the game, iirc. Taking Ninja Gaiden again, there is no stated objective that you must scroll through the story. You've already saved the girl and slew the beast, the rest is reward.
The best time should be judged upon which playing of the game completes the objective quickest, then the stated frame count is considered. (In other words, it's more impressive to coast to a finish from way back there than drive it all the way, but if the driver who drove the entire way gets there first he gets there first)
While this "stated objective" approach is broken by some games, (that new Wario puzzle game, where there are no stated objectives and you have to figure out what the hell your doing comes to mind, as well as Cheetamen II), it ought to hold for most, every rule has exceptions.
So there's my attempt at reconcilling the "clearly in game" vs. "clearly after the game."
So, number 2, with some modifications.
Build a man a fire, warm him for a day,
Set a man on fire, warm him for the rest of his life.
If I recall correctly, Monster Party needed user input to access the credits. The text boxes wouldn't close if they were just left open without user input. Which brings be to my thought on this matter.
I think you should end the movie as early as possible as long as it still accesses the credits, which is, essentially, choice #1 up there. It would be nice if we had a point in each game where we ended the movie, but it would be kind of hard to establish. For example, all Super Mario Bros. movies ends once the ax is touched, all Rockman movies end once the last attack hits Wily, etc. While this would be nice to have and it would make comparing movies easier, it is a lot of trouble to go through.
