Problem: While you say you strive for entertaining TASes on this site, TASes can leave out some of the most entertaining parts of a game in exchange for being as fast as possible. This might include skipped dialogue, cutscenes, unused tricks or glitches, among other things. What I'm proposing is a small and simple addition to emulator movie formats. I've heard talk about adding subtitles and such, but I think this idea would be much easier to implement, and it would make watching a TAS on the emulator a much more fun experience.
The idea is a "movie stop" command to be placed inside of the emulator movie format as an input. Upon executing this command, the emulator would take a savestate then stop the movie input, allowing the player to temporarily take over. This could be used for allowing the user to read dialogue or watch cutscenes without hurting the movie's framecount. When finished, the watcher could hit a resume button, which would reload the movie and the state, and continue playing from that point.
It might also have other uses, such as "Why not explore this glitched area for a while before I zip through it?" or "Try beating this boss yourself and see how hard it is before you watch me decimate it in two seconds!"
Of course, if you just wanted to watch the movie without having the whole interactive experience, you could disable an option and cause the emulator to completely ignore all these commands and just play the TAS. Also, it might be fun to allow the watcher to stop the movie themselves and play around for a bit, without having to wait for a place the movie creator suggests they do this.
An extension on this idea: Replay branches. This might be more difficult to implement, but would exponentially increase the entertainment level of TASes. A similar command would be issued and a savestate would be taken, but instead of stopping the movie, it would start playing a series of input which is flagged off from the main movie. Upon finishing these inputs, it would either stop the movie and wait until the watcher presses the resume button, or it would automatically resume after a specified amount of frames. Resuming would load the state and cause the movie to continue from the end of the flagged area. I'd imagine it would be difficult to accurately measure the movie's framecount with all the extra data in the file, but it sounds like it could be done with some extra effort. Also, the emulator should show the watcher some kind of indication that it is playing a branch instead of the main movie.
To recap, here are the applications these ideas might have:
-Allow watchers to view parts of the game which the TAS would normally skip, such as dialogue or cutscenes, without damaging the movie's time
-Allow watchers to play parts of the game for themselves to better appreciate the TAS
-Allow movie makers to show off tricks and glitches that are entertaining but not fast as possible
-Create instant replays of cool tricks
-Create blooper reels
A downside to this would be the difficulty in programming the tools necessary to create this extra content, not to mention the difficulty on the movie makers to actually record the extra content. Besides that, it sounds like a neat idea that would add a new level of satisfaction and interactivity to the whole TAS experience.
Thoughts?
Movies here no longer aim to be "entertaining" when it conflicts with the goal of being "fast". Same reason it was changed from "Tool-assisted superplay movies" to "tool-assisted speedruns".
The only time people actively work for entertainment anymore is when they can't speed it up, and while fast can be entertaining, when a lot of the main game is just sliced and diced away, lots of the nostalagia and fun of the old one is lost too.
Emulator Coder, Site Developer, Site Owner, Expert player
(3573)
Joined: 11/3/2004
Posts: 4754
Location: Tennessee
like zurreco said, you can create this feature by good use of read only mode. I use it often when judging submissions. I keep a save state at the beginning of each level. If I see something I want to investigate further, I load state and get to that point and stop the movie and test whatever. Then I reopen the movie and loadstate again. This is hardly less convenient than the tools you propose.
I have no idea how you even want the branch thing to be done but for the other thing, it is not difficult to make your own savestate, mess around, and then restart the movie and load the state.
I think the suggestion involved that the creator of the movie could set those points, to tell the viewer "hey, this is a good area to look around".
If you don't know the game, you won't be able to judge where to place your savestates. On dialoges, you'll realize you should have placed a savestate right after the dialog has been skipped, when it's too late.
Movies here no longer aim to be "entertaining" when it conflicts with the goal of being "fast". Same reason it was changed from "Tool-assisted superplay movies" to "tool-assisted speedruns".
I think that's as silly as saying "movies at SDA no longer aim to be "entertaining" when it conflicts with the goal of being "fast"."
Speedruns have always had one goal: To complete the game as fast as possible. Tool-assisted speedruns (starting all the way back from Doom TASes in the end of the 90's) have always had the same goal. I have been following Bisqwit's work on this site from the very beginning, from his very first attempts at TASing supermario bros, from his very first site draft. The goal has always been the same: Complete games as fast as possible (by any means possible, including and even preferably abusing game bugs). You make it sound like the original purpose of the site was to make machinima videos, and somehow inexplicably it shifted to make speedrun videos. That just isn't so.
If what you want is a machinima video, then go ahead and do it. Heck, you might even succeed in having it accepted here. You will, however, probably have more success in a forum dedicated to machinima videos (try googling for some). Or youtube.
If you can't find speed for the sake of speed entertaining, then nobody forces you to watch the videos.
You make it sound like the original purpose of the site was to make machinima videos, and somehow inexplicably it shifted to make speedrun videos. That just isn't so.
XKeeper isn't wrong. The original purpose was indeed to provide "machinima" videos. Video game movies that are entertaining to watch. That the very first movies happened to be speedruns, and every other movie after that, doesn't conflict with that goal. :P
The problem is, with movies that aren't speedruns, it is very difficult to judge which movies are worth publishing, and which ones help the site to live and which ones help it to die.
And apparently, it is also very difficult to create interesting movies that aren't speedruns.
The guidelines of this site are written to welcome submissions that make entertainment-speed tradeoffs, i.e. sacrificing some of the competitiveness for the sake of more entertaining video, and I'm also trying to do the best to protect the freedom to do so without having to fear frame-aggressive competition ― but without excessively annoying those who are in favor of frame-aggressive competions.
Still, we don't see many of those. Foda, with his Mario64 movie, played this edge very well. I congratulate him.
EDIT:
Yes, the guidelines also are strict about wasting time and looking sloppy. The master key is to be entertaining without looking like you're wasting time. The looks are the importance. When I judge a movie, I don't care about the clock (unless it competes a previous publication of the same game). I care only about how it looks like.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Entertainment was the goal in the beginning, I was around since day 2 as well.
If not for entertainment we wouldn't have seen the 90 lives attained in World 8 Tank of SMB3.
Gradius was basically a work in entertainment.
Same can be said for River City Ransom one of the first videos by a contributor.
I remember Bisqwit's early vids, one of the things that striked me the most about them was that at times he purposely ran right next to an enemy before jumping, this is clearly about entertainment and not speed.
Then the SMB2 2 way fight was off, then it became a 3 way fight, and most notably Genisto brought meaningless glitch usage to the table.
Anyone else remember all of Phil's early vids and the fight over the wobble?
From our start I see an entertainment a main ingredient, it was also stated very clearly in the Why and How page.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
I think you understood my post a bit wrongly. I didn't say "entertainment is *not* a thing to aim for, only pure speed". Of course the SMB3 video gets 99 lives because it can do so without sacrificing any speed, and because it's more entertaining to do so. Gradius doesn't sacrifice any speed for entertainment either (boss fights are still done as fast as possible). There's nothing mutually exclusive in these two things. As I have said countless times, speed and entertainment are not contradictory goals.
Also having restrictions for the sake of entertainment is not contradictory with speed either. Just look at all the different Quake done Quick speedruns out there: They don't become less of a speedrun even though they may have some restrictions or different goals (such at getting all kills and secrets, or using the axe only).
However, what I was responding to was a post which, at least to me, sounded like speed was actually not a goal originally, and it has only become a main goal lately.
A machinima video is something where entertainment is the only goal. It might be eg. a music video made out of little sequences of a game with a music track added. Or it might be some story "filmed" using a game (such as the Red vs. Blue videos). In machinima videos there are no rules, no restrictions, no nothing. You can do whatever you want for the sake of entertainment.
That is quite far from TASes (and regular speedruns too), which have definite goals and strict rules, and where not obeying the rules may be seen as cheating. The concept of "cheating" would be just ridiculous when talking about purely entertainment-oriented videos (ie. machinima).
Perhaps one could say that this is one thing where TASes (and also regular speedruns) and entertainment contradict: The former has strict rules and the concept of cheating, concepts which would be ridiculous in the latter.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Yes you are right.
However speed wasn't focused on to such a degree as it is now. Back then seeing a tiny trade off in speed for more entertainment happened a couple of times, you rarely see it anymore though.
Back then a video that was less than a second faster was also frowned on. Bisqwit even refused to publish any new sizable movie unless it was several minutes faster than the previous one. Remember Zelda?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
I actually happen to agree that a 30-minute movie should not lightly be obsoleted with a movie which is 2 frames faster. However, a movie which is 10 seconds long (such as the current record) may perfectly be obsoleted by one which is 0.5 seconds faster.
While speed is certainly a big goal (and the main one, basically), we should also think about the amount of hard work that people have put into making the videos. Nothing is more appalling and unrewarding that you spending 2 months making a 1-hour video and a week later someone else obsoleting it with another which is just 1 second faster. I agree that this should not be done unless there's a really good reason for it. Another purely practical thing is that publishing a new video which doesn't add anything relevant to the old one is only going to annoy people who downloaded the first one and would have to download it again just for a 1 second improvement.
I suppose that there's an unwritten rule like "a video should be x% faster than the previous one for it to be acceptable, except if it's from the same author, in which case a y% improvement is ok, with y < x". No exact percentages, but left to the judgement of the judges. I agree with this because it's practical.
This doesn't negate the goal of speed. It's just a question of practicality and resource management. Sometimes it's not practical nor interesting to accept a video which doesn't add anything to the previous one.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
I like that philosophy.
Although I'd think we should accept a movie which is faster if the improvement is <x% if the game is played radically different.
Edit:
But still the amount of time beats the z% threshold.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
The last time, I did such video, it was refused. It was Double Dragon 1, today, I admit it is imperfect but the goal was to show some glitches and oddities.
Like you, I thought it was entertaining but now, I change my mind and follow SPEED=ENTERTAINMENT.
Fortunately for you, in most cases, SPEED is also = to USES OF GLITCHES.
I don't agree with the "You must beat current movie by x seconds to obsolete current runs". It's stupid and it discourages people to TAS games. Worst than that, it encourages people to do new games that aren't so interesting at all.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
One problem with just showing glitches and oddities that it's hard to judge, and practically impossible which must go if there's more than one of them. Although I would like to see one of these, especially for a game like Super Mario Bros. where there's dozens of glitches which can be showed off.
And if you don't beat an old movie by x% it should be discouraging to the author them self. Will I really download a 100MB file which is 10 frames faster than the AVI I already have? I can't even see a difference.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Don't exaggerate, A 100 MB AVI means an 1 hour video. I don't think someone will only do a 10 frames improvement.
Personally, I am congratulating the person that is patient to do that.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
In other words you agree with me that there's something wrong with publishing a video which is 1 hour long and only 10 frames faster than the previous.
Of course the threshold changes with the length of the video.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
I think the "concept demo" category was created precisely to show "non-TASes", ie. videos which just demonstrate some concept (eg. glitches or such). It's hardly ever used.
But I don't blame the judges for this. They are overworked as they are (how long is the accepted movies queue right now? ;) ), so I don't blame them for not taking even more work, especially when it is about something which so hard to judge as "concept demos".
In other words you agree with me that there's something wrong with publishing a video which is 1 hour long and only 10 frames faster than the previous.
Of course the threshold changes with the length of the video.
No, I think that this situation had never happened and won't happen but if it happens, :P, then I am congratulating him/her for his/her patience.
Evidently, if the precedent was clumsy and the new one is still clumsy then NO.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Phil wrote:
Nach wrote:
In other words you agree with me that there's something wrong with publishing a video which is 1 hour long and only 10 frames faster than the previous.
Of course the threshold changes with the length of the video.
No, I think that this situation had never happened and won't happen but if it happens, :P, then I am congratulating him/her for his/her patience.
Evidently, if the precedent was clumsy and the new one is still clumsy then NO.
Well take this for example: http://tasvideos.org/172M.html
File is >100MB and it's for only 14 seconds faster, I personally didn't download it. Of course it's not as absurd as being faster measured in frames.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
A similar idea I once had would be a togglable "commentary" option, which would just be a bit of text at the bottom of the screen to explain certain things.
But then, I guess just including notes and savestates is a decent enough system as it is right now.
"Big-endian guys drive like THIS, and little-endian guys drive like ISTH!"
Being obsoleted by someone is just the way TASing goes. People who don't like that fact have left over it. Perhaps they shouldn't have been here in the first place if they were to get so upset over an obsoletion. Play for entertainment, not ego.
<Swordless> Go hug a tree, you vegetarian (I bet you really are one)
Wow, less than 12 hours and this thread really gets off-topic.
Tub wrote:
I think the suggestion involved that the creator of the movie could set those points, to tell the viewer "hey, this is a good area to look around".
If you don't know the game, you won't be able to judge where to place your savestates. On dialoges, you'll realize you should have placed a savestate right after the dialog has been skipped, when it's too late.
That's exactly the idea.
For those of you who say "Just take your own save states!" I say, heck, why not just do the whole TAS yourself? The idea is to introduce a new way to allow the movie maker to entertain the watcher, not for the watcher to entertain themselves. TASes would become like DVDs with built-in behind the scenes footage and video commentary.
Being obsoleted by someone is just the way TASing goes. People who don't like that fact have left over it. Perhaps they shouldn't have been here in the first place if they were to get so upset over an obsoletion. Play for entertainment, not ego.
Agreed, if someone was to obsolete all my TASes on here I wouldn't bitch or moan. I actually would encourage people to because I just want to see my favorite games destroyed.
Having said that I hope somebody improves my Actraiser run because it needs it badly! Get to it!
If I would make a TAS as entertaining as possible I'd add these functions:
Subtitles.
Slowdown programmed into the movie. Just before you do a cool stunt the movie slows down, matrix-style, and you can see Megaman/Mario/Sonic perfectly dodge that enemy, a detail that otherwise wouldn't have been noticed. Then it speeds up again while you sit down, gasping for air. Or something like that.
Replays (needs a blinking R in the corner).
And all of these functions should of course be possible to toggle.
