So are you discussing Donkey Kong Country 2 here or that concept you call Open source TASing? If it's DKC2 you're talking about here, the discussion should be moved here instead of creating a new topic. If it's the open source TASing concept your discussing here, I think you should create a thread in the General or Lab section.
If I had to press left for 100 frames then press B, what difference will make if I copy-and-paste or enter the input manualy?
No Difference, despite the time you spend on it. This isn't like the date or any other hardcoded infos were written, while typing the input.
Dooty wrote:
I know the TAS is kinda open source already, but I don't want to be sued for using someone else's work on my TAS.
Just to clear the point, from the discussion of this topic, it sound like there no way to create a new licensing system(by either copyleft/right way) for a published input file, from this website. So, I think that you can submit anything to TASvideos with your full name there, without problem... just deal with the community after.
Must time, as long as there improvement you can put your name on the input file, otherwise this should be debated in specific workbench submission, in my opinion.
So, if you figured something to save some frames, just copy what you need to copy and go for it. Even if the new movie isn't fully perfect/maxed, your movie should have some good chance to be accepted... but I'll say make sure to show a suitable effort in your submission, sometime this is the hardest part.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
I wish I could explain my idea in Portuguese...
my proficiency in English is like my TASing skills = Zero...
But anyway, I'm very excited about this idea of reusable code.
I'll try to be more specific: I'm not improving the existing TAS,
I'm just reusing the progress that has been done, to show the
possibilities to create (or improve) TASes in weeks, not years.
In two days, I managed to reach Kleever's Kiln, only fixing desyncs.
Remember "I'm not improving", even though I'm 32 frames ahead,
when I have to finish a stage, I always end up losing 10 + frames.
I am old enough to know better, but not enough to do it.
I think I see what he's saying. If you watch, for example, the Mario 64 contest TASes, you'll see that, although the overall routes are the same, even the losers have sections where they're faster than the best overall time. What Dooty wants to do is get, say, 10 TASes and combine all the best parts into one really good TAS. This would achieve the same (or better) results as doing it himself, but be done quicker.
After become open source, the Dolphin emulator had a great improvement.
Similar to OSS projects, Adelikat puts his WIPs in a googlecode SVN. I *think* that's what you're basically trying to get at in this thread, but nobody has brought it up yet. Maybe you'll find the concept useful?
I think I see what he's saying. If you watch, for example, the Mario 64 contest TASes, you'll see that, although the overall routes are the same, even the losers have sections where they're faster than the best overall time. What Dooty wants to do is get, say, 10 TASes and combine all the best parts into one really good TAS. This would achieve the same (or better) results as doing it himself, but be done quicker.
well, this is it
very very interesting.. cuz u can do tas very quicker and more people can be able to do it n also help you with it
but probably as u r not redoing the work u will always get lower or equal speed than someone who tases the entire game (like comicalflop)
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
Let me use the Dolphin example again:
Mario Kart is playable at full speed, but the intro video doesn’t work.
I’m a newbie coder, I don’t know how to create an emulator from scratch.
Fiddling with the code, I found a fix for the video problem, and I fixed it.
But fixing the video broke the controls, and the games can’t be completed.
I sent the code to the guy that make the emulator and he fixed the controls.
That’s what I’m doing, implementing new features on the published TAS.
But I’m not experienced enough to build a TAS like that from scratch.
I’ll not steal it either, but encourage them to reuse their own code.
If I manage to reach 102% at least in the same amount of time,
I’ll send a copy of my idea to the authors, to see if they approve it.
So, I will not post any wip before I make sure it can be completed.
I am old enough to know better, but not enough to do it.
What I'm getting out of this is that he's using the published 102% as a template and just hex-editing improvements in, much as Upthrorn and marzojr have done with Sonic 3 and Knuckles. As long as permission from the author is granted, I don't see a problem.
There doesn't seem to be a proper hard mode in this game, just one that removes DK-barrels. In my opinion it shouldn't be used, because the run uses a lot of tricks which involve sacrificing the other character.
Also, it would be nice if you posted a WIP of your work so far.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
It's some seconds slower, but it's completed! You can get it here.
There's many things I want to implement, but TAS Movie Editor is useless after reset.
holy crap, or I learn how to hex-edit or it's goodbye reusable TASes for me, oh well...
I am old enough to know better, but not enough to do it.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
Thanks guys, but actually I did just a small part of the TAS, the rest is just copy and paste.
those are the only levels I changed or redid:
Krockhead Klamber, I have to enter twice.
Klobber Karnage, I have only one kong to sacrifice, and I need Diddy to do the bonus.
Castle Crush, I used Rambi. (no barrels, remember?)
Animal Antics and the final battle with K. Rool, there's no way to reset the game with TAS Movie Editor.
(now those two levels are also in glorious stereo sound)
The rest was just positioning the kongs to match the NxCy and Comicalflop's TAS,
copy the rest of their movie, paste on my movie and see if it sync, simple as that.
I am old enough to know better, but not enough to do it.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
Almost that, I wish i could hex edit...
Here's a step by step of my work with this TAS:
Open two emulators, one with the source TAS and one with your own project.
Start a new TAS and play until your character is aligned with the "source" TAS.
Stop your TAS, open the source TAS on TAS Movie Editor, look for the matching button press,
(don't worry about the different frames, just the button presses)
copy a small section of the source TAS, open your own TAS paste it at the end of your TAS.
Save your project with the "Save as" function. (the "Save" took forever to complete when the movie is getting bigger)
Now open your project on the emulator and see if it sync, if it not sync look for mismatched actions and try fix it cutting or adding some frames with TAS Movie Editor, saving and testing again.
I am old enough to know better, but not enough to do it.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
Using this concept, I managed to reach Krow's Nest 50 frames ahead of the any%.
I spent more than 60 frames trying to manipulate egg placement, but no luck.
So, I got another of my stupid... I mean, brilliant ideas: Unlucky Manipulation.
I'll post a wip as soon as I manage to finish the battle with Krow, hope it works.
I am old enough to know better, but not enough to do it.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
Thanks for your advice, I was about to change it to Donkey Kong Country 2: Unlucky Manipulation...
I'll edit it just one more time to return to the original name, sorry about that and thanks again.
edit: I'm still reusing the currently published 102% run,
but now I'm going for an improvement! (don't laugh, please)
I know I don't have the necessary knowledge yet,
but I'll compensate it with my passion for this game.
I am old enough to know better, but not enough to do it.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
if you want to do an actual 102%, then I have a few suggestions:
-Go with hard mode. There's only a few stages where you need to sacrifice more than once. NxCy and I actually timed everything out, and we figured that the use of hard mode, in the end, might actually end up saving a little bit of time (Pirate Panic for instance can save a lot of time with hard mode), or come out equal in the end. The difference was so close we were actually testing it out for the any% (where we then realized hard mode means no invisible barrel for castle crush.)
-If you're going to try the 102%, do it completely from scratch. Don't hex anything. I can tell you you will have more fun TASing everything rather than hexing everything (although it is more time consuming). I'd rather much see NxCy and my work improved on every level, rather than just copied and adding us as co-authors. Especially since our optmization really kicked later in the run. I also really want to see the bosses fought differently.
-Check with us for time savers. we know of plenty of new tricks and places where time can be saved for the 102%, as well as possible different routes we wanted to test. In fact, there was a recent 102% speedrun by a guy named LegendOfEvil. His 102% run contains a lot of possible route changes; especially in the first two worlds abusing the warp/start+select variety. Of course, his run was aiming for fastest in-game time, but his routes may lead to something new.
-Another possibility when I saw your very bizarre Krow fight is to make a playaround/glitchfest movie just like this one. Lord knows DKC2 is chock full of weird glitches, and it'd be interesting to compile them all in a playaround movie. I think DKC2 lends itself to some interesting plyaround as well.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
Thank you Comicalflop, I'll follow your tips.
It will be a playaround and my objectives will be as follows:
* Play on Normal Mode.
* Follow the any% route.
* Collect all K O N G letters.
* Collect all DK koins.
* Show all known glitches.
Progress: Krow's Nest.
I am old enough to know better, but not enough to do it.
Joined: 6/2/2009
Posts: 1182
Location: Teresópolis - Rio de Janeiro - Brazil
Hi Paused, sorry for the lack of wips, but I'm using some tricks
that will be present in Comicalflop's any% improvement and he
asked me to not spoil his run.
I'm still doing the Krow fight, but I'm also testing a new glitch in Slime Climbe.
If it work, I may put this playaround on hold and start an 102% improvement.
edit:
The new glitch worked like a charm, but I must go back to my Reusable TAS Concept...
(here's an example of this concept)
edit 2: the new glitch is already used on the any%, so I'll stick with my playaround, oh well...
I am old enough to know better, but not enough to do it.
