This is a Any% TAS of Super Metroid, the aim for the movie is lowest possible ingame time thus some realtime sacrifices are made to lower ingame time such as collecting certain items and using a route with more door entry's. The movie is 1477 frames or 24.6 seconds slower in realtime then the currently published any% run which aims for realtime but it is 5271 frames or 87.5 seconds faster in ingame time. The ingame endtime is 24:04:37 and item collection is 22%.
The reason why I choose to aim for the lowest ingame time is because in my opinion it is a little more entertaining to watch both because of the route and more variety in item use. The majority of the improvement came from the route change and the troizo skip, the rest simply comes from new strategies and optimization. It was hard to make this TAS, especially since it is my first one (aside from 3 other "test TASes" I made of this game before this) and Super Metroid is known to be quite a hard game to TAS. Without the help of other released TASes which more in depth learned me the uses of various tricks I would not have been able to make this.
There is (I think) a good chance this movie will not be accepted because I put a higher priority on ingame time then realtime but if that is the case I still hope that as many as possible will be able to enjoy it through SMV anyway.
Thanks to various persons (in no particular order)
Tonski: He was the one whos WIP's made me want to TAS this game, and he has always supported me.
Hero of the day: Have since the release of my first test TAS provided me with tips and help.
Moozooh: His two WIPs and his discovery of the torizo skip helped me alot.
JXQ: I used many strategies from his 100% run.
Saturn: Has invented many tricks which i have used, the shortest charge for an example.
Michale flately: The inventor of many tricks, the armpump and phantoon 1 round most notably.
Terimakashi: He raised the standard for SM TASes and his 00:27 TAS was what first made me interested in TASing.
Frenom: The first (from what i know) SM TASer.
Everyone (I don't know them all by name) who have produced and improved the tools that I have used, the memory watcher and all the snes9x improvements.
White Angel for having a godlike patience with me playing SM all the time.
Everyone whom I have forgotten are also Thanked, contact me if you feel that you deserve to be thanked.
Truncated: I think it is unfortunate that this movie aims for in-game time instead of real time. It cannot be published along with the current any%, because I do not think real/in-game time should be a valid differentiation between runs of the same game.
Therefore we are reinstating an old tradition at nes...tasvideos, to not publish an AVI or make a separate entry for a movie, but to link it from the current entry in emulator movie format only. Like I said to moozooh, don't think of it as a rejection, think of it as a non-publication.
Bisqwit: Setting the state to "cancelled" instead of "published" for technical and consistency reasons. It will still continue to be linked from the publication of that other movie.
Joined: 11/14/2005
Posts: 1058
Location: United States
This isn't exactly relevant to the conversation, but I would like to hear some opinions on this. There are actually 2 internal clocks built into the Super Metroid engine. 1 clock stops during door transitions and item acquisitions, but the 2nd clock is a continuously running clock which starts ticking after the save file is selected. Now is this 2nd clock any less valid than the other? It is a essentially a real-time clock built right into the game itself. It is pretty interesting when you think about it. If it were a goal to minimize that clock, you would be aiming for real-time exclusively but at the same time it would also be an in-game timer. This 2nd built in clock syncs up very well with the actually frame count of the movie, give or take a couple hundred frames.
Question for all in-game clock fans, why is this 2nd clock less important? Is it only because it does not determine the minutes shown at the end screen?
I am not trying to make an argument or anything, just want to hear some feedback on this.
This isn't entirely true -- the second clock does not start at least until after the game finishes booting and enters the main loop for the first time. But more importantly, this second clock probably does not increment during lag frames.
Though I am not an in-game clock fan where Super Metroid is concerned, I think I can answer this.
When people say "in-game clock" what they generally mean is "timer that the game displays". So yeah, the second clock is less important because its end result is never displayed in normal gameplay.
How fleeting are all human passions compared with the massive continuity of ducks.
Given that the clock displays only the minutes that argument is kind of moot. Unless you can improve the movie so that this shown minute count is smaller, then arguing "this is faster by the game's internal clock" is more or less pointless. If you need to use a ram watcher to see the "real" value of that clock, you may as well use the second clock too.
For me (and probably for most Super Metroid fans), the by far most important goal in any kind of SM-run is to get a as low as possible game completion timer to display at the end of the game. This is the only factor that truly indicates the speed quality, and officially gives a quality-mark to the run. The seconds are irrelevant, it's all about the rounded minute displaying. Because of this reason, the 2nd clock that shows realtime is obviously less important (note: not a opinion, just a fact), as it's:
1. of no help in any way to plan out the ingame timer to get the lowest game completion to display at the end
2. not even to 100% in sync to the realtime emulator framecounter (according to hero)
Hope the point is clear.
I know and agree that realtime is the more important timer to aim when TASing most of the games, but Super Metroid is just a rare exception here, for many reasons stated many times in the past.
If I understood correctly the only difference can be caused by lag frames. However, lag frames cannot be "exploited" to get a smaller timer value (at the expense of a longer movie), and thus it's not relevant if that counter is not completely in sync with the emulator's frame counter. Optimizing the number of frames will automatically optimize that internal counter too. In that way they are "in sync".
Weird, I first suggested this decision for Saturns future Any% TAS and then it happen to me. Well at least this is something and the only thing I'll miss is that there will be no rating...
I wasn't even aware that this movie was "published" until someone else told me, and I check the front page rather often. Publication, at the lowest form on this site, has always occured with the creation of a movie page on the main page and a posting to the front page, provided the movie is "new" (i.e., not a recreation from an older movie).
I fully disagree with the joke of a "publication" this movie recieved. That does not mean I am for or against publication on the whole, but I do not believe that such a useless middle ground should be accepted.
Why is it useless? If the movie is not published at all because of the "too many movie entries of one game" rule, who does it serve? If it's a good movie and is otherwise worthy of publication, why is "half-publishing" it a negative thing? I do agree that there should probably be some note on the front page/published movies list about it.
Because having its status "upped" like that serves no useful purpose. Keep in mind that making a note in the published movie description would have been perfectly possible without having to "half-publish" this movie; we don't have to have a special permission for altering the description to include that kind of information, anyway.
Even if it serves no useful purpose, it still doesn't explain why it's a negative thing. Are you saying you'd be happier if the status of Cpadolf's submission was changed to cancelled or rejected instead of published but nothing else changed?
Joined: 4/21/2004
Posts: 3517
Location: Stockholm, Sweden
Well thats the whole idea of submitting something is it not? Among your goals when you submit a movie are to provide a fast and entertaining run which hopefully will eventually be published into a .avi file. Not some semipublished .smv file. I certainly did not notice Adolfs run was apparently published. Even when I did, I got confused. I checked for the .avi, which I bet 99% of our members do when some movie gets published. I dont think there are other movies who are "semi-published". I think the idea is pretty silly, confusing and serves no purpose. Therefore, count me in the group of people who does not understand this whole idea and disagrees with "semi-publication". However, this does not mean I support the idea of letting this movie get published (you know, the old drill; .avi, rank it etc). Adolfs effort was fine but unfortunately, I do not support his choice of goals.
Yes, since a rejection would be far less hypocritical. Publishing a movie means creating a separate page for it, with separate rating and links and other stuff, making it "published" in the site's database. "Half-publication", or as Truncated himself put it, "non-publication", is in all practical senses a rejection, just called differently not to needlessly aggravate the fans without having to state actual reasons for rejection (hence, hypocrisy).
As I understand it, the judges (at least those who had opinion on the matter) hesitated accepting it due to a different ("unwanted") goal set which made the movie longer in frame count than the current one. Not that it's a bad reason by itself, but provided the fact that it's, by far:
1) not the first movie with different goals to obsolete the preceding one;
2) not the first longer movie to obsolete the preceding one;
3) more optimized and up to date than the preceding one;
4) doesn't prevent a new movie with different goals to obsolete it,
it would be much preferred if the decision was justified. I don't have any problems with justified decisions. It wasn't. I've resigned myself, so you won't see me bickering with Truncated about it, but it doesn't make it any better in my book.
Does that answer your question?
You see, that really isn't important given that its a different catagory, you could easily invest all that time and effort into a real time run. Having two different runs would be too confusing for non-fans of the game. I don't have a problem with justified descisions as long as people don't completely refuse to accept that its not justified just becuase the outcome was not what they wanted.
Erm, wait a minute, why isn't it important? Yes, it's a different category and it's possible to implement the timesavers in a realtime-oriented run, but there is no new realtime-oriented run, and won't be for the time being. It doesn't make this submission of any less play quality.
And if you read the previous pages carefully, I didn't suggest having two runs. I suggested this one obsoleting the previous on grounds on being more optimized (and different, not simply "do all the same actions, just better").
Also, it seems you've ignored the other three points; they're supposed to work as a whole.
Then why not implement these known improvements? The Metroid TASing community is large, I don't see how there could be a shortage of people incapable of TASing it. Also we can wait patiently until a new one arrives.
Refer to the highlighted text in the upcoming statements.
Not a justified reason and "hyhocrisy" as you would call it, a slower run cannot obselete the current one by having different goals, there must be a solid base for all future improvements
.
All the other three points were unjustified reasons to justify your third point which was your primary argument.
I can't believe what I'm reading. You are willing to spend an incredible amount of energy on making a mountain of this molehill. Well, it's your energy.
If by "Metroid TASing community" you mean the list of people directly involved in making tasvideos-quality runs, then I have to disappoint you: as of now, it consists of about 5 players, with one of them (Hero of the day) working on an improvement to an any% TAS. All the others are already busy with their own projects.
That being said, "why not to implement these known improvements" is a good concern, but most (if not all) considerably large projects had known improvements in them at time of submission, including all the previous Super Metroid runs, including the current any%. That doesn't mean they were of bad quality, it's just that the game is versatile enough so that new tricks and strategies are found all the time. SM64 is another example of such game, with 6 any% runs being submitted over a year.
All the SM any% runs preceding Hero's aimed for ingame time. How the hell should a real-time run have obsoleted them, judging by your words?
Not to mention that similar situations happen all the time; here's a little selection out of the more recent examples:
movie #958: different difficulty, no damage run obsoleted the previous (the successive run could have been faster with the goal set identical to the previous run);
movie #948: different ROMs where characters have different abilities;
movie #945: Longer run with 2-players that contained a lot of speed/entertainment tradeoffs obsoleted a shorter run without them;
movie #859: a longer 100% run obsoleted any%;
movie #852: I think you remember the debates around this and the preceding movies yourself.
Where did the solid base go with those runs, huh? The goals were obviously different in them, yet the improvements were published, the common ground there being, from what I remember, higher technical quality (and in some cases entertainment). Indisputable, and fully understandable. Be aware that SM is no different from those games when it comes to appraising optimization quality, except it's even somewhat more conventient with there being a dedicated timer for in-game events. The solid base you are reffering to didn't change.
Even though it was primary, those points were merely illustrating how the case in question doesn't prevent publication in other cases, with the solid base and all the other things.
Do you want to continue?
I expected this decision to be controversial no matter what.
If setting a movie to "published" when it hasn't been given it's own AVI and movie number feels wrong, it could easily be set to "rejected" instead. This is a new situation since when we last did emulator movie only publishing, since there were neither submissions nor statuses when it last happened, only text that Bisqwit updated himself.
I see now that Bisqwit has set it to "cancelled". I thought this state was reserved for movie authors when they wanted to remove their own movie from the list, but if everyone think that's the best option we could do it like that in the future.
Hope that made sense.
As promised, I finally encoded a top quality AVI of this run:
SuperMetroid_TAS_24_Cpadolf.avi
It would be good to add this link to the submission or the currently published any% run, so that fans can watch the fastest run available on the net so far, without needing to bother about how to playback smv files.
Enjoy the AVI!