Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
jlun2 wrote:
So as long as that is still the case, there will still be people recording them on an old emulator because......nostalgia? User-friendliness? (Idk, I never TASed SNES games before)
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
creaothceann wrote:
Resyncing does not change the actual content (gameplay) of the TAS, it just adjusts it to a slightly different timing situation. Therefore I don't see why TAS authors should feel bothered about it at all.
I think we were talking about the hypothetical situation where the author of resync wants to remove names of authors of the original run.
Anyway, the whole situation is absurd. There should be no resyncs, except for truly frame-perfect runs. People should just improve old runs (while recording them for new emulators) and that's it.
1) New emulators are supposed to provide better tools (as the general progress goes), so that it's easier to find new strategies.
2) New people are supposed to have a fresh view on the game in question, so it's easier for them to spot some omissions in the old movie.
3) New authors are supposed to have full information from the authors of the old movie, so they don't have to reinvent the wheel, and thus they have more time to expand the existing knowledge base.
With all these advantages, a simple resync is laughable.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
MESHUGGAH wrote:
AnS wrote:
Could you show more details of this Rygar example?
When a level/scene changes, the game flags frames as lag frames and loads the next level/scene. If you press an input before this load, you can be 1 frame faster.
As my uploaded WIP shows, I can jump earlier by pressing an input before the lag frames. The last non-lag frame before the 1st frame of a level. This trick could work on games where you need to press actions 2 frames long or where game logic isn't fully loaded once the game started.
Oh, now it makes sense, I thought you meant something different.
MESHUGGAH wrote:
I don't think that many people know that you can change the PPU clock by doing this. Prove me wrong.
Now I don't understand what do you mean by "changing PPU clock".
MESHUGGAH wrote:
So what I want is to make a database with techniques and tricks detailed with examples in one place.
So yes, this is good idea, but before making a database you'd better collect as much data as possible (in text/html form), because currently your data is too vague, and it's not enough for a database.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
MESHUGGAH wrote:
Pre-input processing: try to start your input before the last non-lag frame (works in NES Rygar)
Huh, I don't understand this. By default, lag frames are those frames where the game doesn't poll input, so you can't change anything by pressing buttons on these frames. Could you show more details of this Rygar example?
MESHUGGAH wrote:
try to press various buttons at any non-lag frames on different controllers
This is the most generic technique, and it works in any game. :D
MESHUGGAH wrote:
Okay, I must admit this wasn't the best example as I can't really think an idea thoroughly in short time and I could write a lot more, but I would really like to get feedback about would you be interested in, who could contribute for it, and how should it be made (I vote for generating content like using filter on movies insted making multiple wiki pages).
For now, I think you should just make your own page at TASVideos and collect everything you come up here. So far your idea doesn't seem to be very clear, so it's too early to think about the format, you need to refine the idea and write more examples and details.
MESHUGGAH wrote:
edit: of course my goal is to make a reference page that when I want to make a TAS in the future, I check the categories I want to optimize so I don't need to track every little things I already tried to test and keeping multiple movies for this solely reason. I think this should be a very good resource for every TASers (even super godlikes through first timers) for both learning and making TASes as highly optimized as possible.
Yes, it would be useful to collect all this stuff in one place. Usually people come up with those mini-tricks naturally, by struggling with Input when trying to squeeze more frames from an already very optimized movie.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
Publication text wrote:
The 2nd Youtube encode has been edited to have ending using proper graphics instead of glitched mess.
Now this text suggests the encoded video was edited in After Effects to include normal ending, because only garbled ending could be achieved after using the glitch.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
Tangent wrote:
It could also very well deter people from sharing their tools, processes, and workflow if those could then be used to trivially wipe out their credit simply from picking a different emulator. They have to at least do something different from the published tricks and tools to improve it and replace (or add) their names to the run. Removing the "do something different" requirement could very well dissuade people from sharing the code or tools they create.
This would be real disaster, actually. The most valuable aspect of TASVideos was the atmosphere of openness, and anything that undermines it can not be justified.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
MESHUGGAH wrote:
edit: " As for garbled graphics, I think I've seen this trick used too much, so I've become apathetic to it." - and I saw TASes too much, so I've become apathetic to it.
Haha, but it's incomparable! TASing is art, while garbled graphics is simply a trope.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
If a run only syncs with an interim revision and doesn't sync with any official version then it's similar to not releasing input file at all (and only providing an encode), because who will be able to replay the movie ~5 years after?
I remember once it was difficult to replay FODA's NES Battletoads because it used very old version of FCEU and didn't sync with any later versions. Fortunately it wasn't completely obscure revision of the emulator. If it was, then noone would be able to re-encode it in HD or use the movie file for comparison/etc, because the source code is not available anywhere now.
RachelB wrote:
3.5 was released 2 years and 900+ commits after 3.0, and we are now almost 1500 commits past 3.5 in the last 7 months, and still quite a ways away from 4.0.
Then why won't you release 3.7 of something? What's so cool with round numbers?
RachelB wrote:
But can you guarantee the archive will be available, say, in 5-10 years? Stable versions are usually spread around the Internet, so it's not a problem.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
Masterjun wrote:
Scepheo wrote:
Masterjun wrote:
However, deprecation can still be a reason for rejection.
turska wrote:
That's not the case.
How is this not clear?
And then where is the explanation to that?
I think it's pretty clear. Deprecation is not a goal in itself. The goal is to make movies more likely to sync with real hardware, and here it's not the case.
The Rockman movie plays all levels and doesn't skip right to the glitched ending, thus it looks solid. The whole movie actually prepares viewer by progressively increasing the level of glitchness.
The Kirby movie starts as normal, but then glitches out without building any dramatic tension to let people know that the result was arranged (that it indeed was a success and not a failure). The current glitched version just doesn't look satisfying enough. As for garbled graphics, I think I've seen this trick used too much, so I've become apathetic to it.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
riking wrote:
I know that other movies have had multiple YouTube and/or archive.org links when there's a slightly different encode that some people might like better (e.g no cutscenes).
Why not edit the current movie with this file?
Here's a 'mockup' I made. alternate link here
That's exactly what was done with Goonies movie made in Famtasia but later resynced to FCEU by the site admin.
Unfortunately, in current case the author of resync spent so much time on it that he wants some recognition/credits, I can understand it, but I wouldn't want other people to do the same (it's just a waste of time, especially when done using an ineffective method).
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
mklip2001 wrote:
The discussion of whether resyncing can be done automatically is just largely academic at this point.
It's not purely academic if we stop relying on absolute statements like "it has to be either full automation or hard manual labour". Some very efficient semi-automatic approach already exists for NES.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
ALAKTORN wrote:
about the submission, I’m all for having runs resynched on better emulators, but after having read the post of that one guy that claimed it should be an automated process and it’ll be possible in future with BizHawk, I agree with him and think we shouldn’t have a new submission every time a better emulator is made, but rather something automatic to fix the current submission to work with the latest emulator
I didn't claim that input resyncing can be fully automatic, just that it could be done with much less effort. Some human involvement will always be necessary, unless it's a bruteforce bot (and even then someone has to setup the bot and verify the end result). But I want to clearly distinguish between these two types of activities, which are so mixed only because of ancient tools we use (namely the linear recording method):
1. When you resync input, you are working within the frame of a single strategy. The process stops right after the strategy successfully shows the expected results. With right tools and a right state of mind it can be done mindlessly.
1002. When you TAS, you are testing different strategies and comparing them to each other in order to find new factors and define the best strategy. The process can go for years (it's only limited by your own patience and the game complexity. Usually the patience runs out much earlier). And the process, of course, includes occasional resyncing of input, but is much more than that.
Masterjun wrote:
I don't just dislike the idea with the automated TAS converting thing, I absolutely hate it!
Seriously guys, creating a TAS is effort, doing thousands of rerecords to be sure that this is the best way one can do. And then just creating a bot where you can't be sure that this TAS is perfect? I think it's not nice to see a TAS you created to be converted to a better emulator and then be given credit for a possibly suboptimal run.
I don't see what is to hate here. You first resync the movie, then you start improving it. Two different things. Yes, in order to ensure the perfectness, you may want to do both things before submitting the new run to the workbench, but it's not required. If you submit a suboptimal run, someone will beat it later.
Example: Phil&Genisto's NES Goonies - the old movie was padded (by Bisqwit I presume) with 3 frames in order to sync with FCEU. Authorship wasn't changed. Then it was improved and obsoleted by other people.
Of course, it's very natural to search for improvements in the middle of resyncing (especially when you use tools that require you to repeat the old input by hand), but then the only thing that matters is the outcome - whether you've actually found those improvements or not. If you didn't find any improvements then your thousands of rerecords are wasted (which is normal even for usual TASing - you may spend many days testing an alternative strategy only to throw it away after finding another one). And it doesn't really matter if you've wasted thousands of rerecords and many months, or you've wasted hundreds of rerecords and a few hours - the result is the same: no improvement.
Masterjun wrote:
Converting movies by hand gives the opportunity to find optimizations or even new strategies to beat the game.
No one takes away those opportunities from you. Once the movie is converted (or e.g. once the first half, which you consider frame-perfect, is converted), you can use it for comparison while making an improved version.
Also, even if you convert the movie by hand, after that you still have to search it for optimizations and test new strategies. You can't claim that you've already tested all possible approaches while resyncing, because resyncing ends right after getting the same result as in the previous run.
Masterjun wrote:
Are you really thinking that a bot can be created that is so reasonable that it can convert the slightest RNG manipulation and lag reduction strategies without writing a bot for each game in its own?
I guess I was misunderstood. I was not talking about bots doing all the work. I was talking about the resync workflow (the one described in that thread is outdated though) which shortens the feedback loop so tight that you can do it using only spinal cord (figuratively speaking). In a way, you become bot yourself. RNG manipulations and lag management don't require changing the original strategy (if they do then it's a huge problem which entirely invalidates the old run and then yes, you have to TAS from scratch). So you leave the old input and make small adjustments only to the vital parts that desynced (inserting/deleting frames and buttonpresses, and seeing the final result immediately). I know it's kinda hard to get from the traditional TASing viewpoint, so I'm not saying this movie should be rejected. I think it should be published after adding the previous movie authors (like in recent Kirby framewar).
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
TASVideoAgent wrote:
And for entertainment,
Glitch the ending graphics
this is input- and PPU clock/CPU cycle dependent. The reason I used this is to destroy the game after tedious TASing sessions.
Is it possible to have a movie that beats the game in the same amount of frames, but shows normal graphics in the ending? I would prefer to see it.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
mklip2001 wrote:
I'm glad to see I'm not the only one in this thread who thought of [2058M|Poko-Nyan] when I watched this. While simplistic, it is quite amusing and also very Japanese.
Except the game and actual franchise has nothing to do with Japan. The game was made by European company and never released in Japan. You people are judging things by cover. This is totally different kind of cuteness!
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
Patashu wrote:
Even from a technical perspective there is ambiguity as to when the game is considered beaten or not, because 'the game is beaten' is a psychological phenomenon.
Well, I agree that movie validity should be evaluated case-by-case.
However, I don't agree that psychological aspects of the game should have as much weight as Warp seems to suggest.
Since it's TASVideos, we're not playing games, we're solving tasks, right? A video game is the source of task, but not the only source (we often create additional challenges like multiple games with one input, pacifist runs/walkathons/no-damage/etc). Thus our goals depends less on the game's traditional goals and more on how interesting will be the task/how entertaining will be the final movie/etc.
And it does not matter if e.g. the game developer agrees with TASer that the game was beaten - as long as the target audience agrees, the TAS should be published. Here I'm talking about Moons/Stars tiers, because I'm not very interested in Vault runs - I guess if a movie can't be entertaining, it really has to follow all those boring conventions in order to gain at least some audience.
So, the technical confirmation of the game completion often serves as good enough task. No additional confirmations are required, unless you really want to cater to some additional audience.
But if I were to choose from the following 2 tasks:
1) beat the game using glitches which will alienate some purists but produce funny video;
2) beat the game while catering for as wide audience as possible, likely producing a shallow letsplay (tool-assisted);
- I would certainly choose the first option, simply because I want to have myself some fun in the process of TASing, since the task is more formal and the result is more personal, etc.
feos wrote:
I have no idea why you keep bringing that question up in each submission glitched like that.
Shh, he didn't notice that Kirby submission also uses heavy memory corruption.
Nach wrote:
So, do I need to wait for a frame-war on this one too?
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
Warp wrote:
The way I see it, there are two different aspects of "completing a game": From a purely technical point of view ("does it reach routine X, after which it proceeds to routine Y?") and from the viewer's point of view.
Isn't it essentially an "objective point" vs. "subjective point of view"? Subjective definitions of game completion will never bring any consensus.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
Yes, TAS Editor expects you to start movie from console power on. Movies starting from savestate are not published on TASVideos anyway, so you can't go with single level movies forever.
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
Whoa, what do you mean quit speedrunning, why is that?!
As for tediousness, yes, it's obvious part of TASing, but nowadays it can be greatly reduced by using TAS Editor (since we're talking about NES game here). Plus, in general, the tediousness becomes less noticeable as you gain experience. Anyway, someone else will probably be able to TAS Batman, but I doubt anyone else will speedrun Bucky O'Hare better than you!
Emulator Coder, Experienced Forum User, Published Author, Experienced player
(724)
Joined: 2/23/2006
Posts: 682
nesrulz wrote:
btw, AnS, perhaps this feature could be added, pls?
OK, it is added in r2962. I don't think the actual checkbox is needed, emulator should just always save the last entered name, like TAS Editor always did (when creating a new FM3 project).
By the way, since you're only beginning to learn, I suggest you to try TASing in TAS Editor instead of recording movies the traditional way.
nesrulz wrote:
And, whether this menu can be seen from full screen? As can be seen in window mode.
Well, this is old request, but I don't think it's important enough to warrant the hassle.
