Hey there. So I wanted to know where in this forum I could ask for feedback on a movie, but I'm unsure about where I could do it... in the game thread, in the "newbie corner" or something else... but if this is the right place, then I'd like to know if this run is optimized enough (or if it isn't, what could be improved)
http://tasvideos.org/userfiles/info/39911649688522623
(BizHawk 1.13.0, SMA2.gba)
Currently focused on PS1 games Looney Tunes: Sheep Raider / Sheep Dog 'N Wolf, and Toy Story 2: Buzz Lightyear to the Rescue.
well you either go in the game thread and ask for feedback
or you go right here : http://tasvideos.org/forum/viewtopic.php?t=10152
also you´ll get more feedback if you make a youtube encode, most of the time ppl are too lazy to watch the movie files.
well i took a quick look at it:
can´t you do p-jump switches in the first castle?
donut plains 4 also looks improvable
i noticed that you didn´t use 1/1 swimming in the lake stage
2-do:
Smurfs Nightmare, The (EU) GBC 10%
fin :
Mega Man: Dr. Wily's Revenge improvement: submitted
Mega Man II Improvement: submitted
Mega Man IV Improvement: submitted
Mega Man V Improvement: submitted
future plan:
-n-
well you either go in the game thread and ask for feedback
or you go right here : http://tasvideos.org/forum/viewtopic.php?t=10152
also you´ll get more feedback if you make a youtube encode, most of the time ppl are too lazy to watch the movie files.
well i took a quick look at it:
can´t you do p-jump switches in the first castle?
donut plains 4 also looks improvable
i noticed that you didn´t use 1/1 swimming in the lake stage
the p-switch jump glitch is fixed in the gba version afaik
Currently focused on PS1 games Looney Tunes: Sheep Raider / Sheep Dog 'N Wolf, and Toy Story 2: Buzz Lightyear to the Rescue.
Ok. A quick bump here (if thats even possible) just to say
I have been pointed towards the userfiles. Used it. That didn't quite work. I got little to no feedback. And honestly it seems like this forum is very inactive.
I'm trying to improve. But how can I improve further, if there's no feedback and the forum is dead most of the time?
Not to mention my last submission attempt (LEGO Island 2) had optimizations. There's no other way around the dialogue skipping - that was frame perfect. The movement, yes, could be improved, but still, there was no set standard for a LI2 run since that was the literal first run of it ever.
I've also seen many people who don't give a crap to the "Watch before voting!" advice in generated submission topics, as well as people who assume a run is bad simply because of the methods used (in my case, rewinds which are literally just save states but they don't add to the re-record count... until I used a manual rewind counter.) so there's that.
tl;dr: I can't improve my runs further if the forum is inactive and no one gives feedback so I don't know what to do now.
Currently focused on PS1 games Looney Tunes: Sheep Raider / Sheep Dog 'N Wolf, and Toy Story 2: Buzz Lightyear to the Rescue.
To salvage my border-flame post here, what you pointed out here is in my opinion the biggest weakness of this site; often the forums are inactive. This is especially damaging for newcomers, because TASing takes a lot of effort, and often you won't have people jumping at your TAS and ponting out the problems that it has, especially if it's a more obscure game. (I'm not counting comments in game threads saying "Great work!!" as helping here, they help for encouragement but not so much for the quality of your TAS..).
There are some places that can help you with this, such as going on the IRC (see the top of this forum, "We also have an IRC channel") which is way more active than this forum and has a bunch of experienced TASers lurking, or looking by the Wiki (which has a solid grasp on methodolgy and emulator features), see http://tasvideos.org/TASTutorial.html.
To criticise the method you use a bit, rewinds are very bad for making a TAS, because instead of having a much more broad plan of what you want to do, you only can optimise the immediate victinity, and making changes to a level (or even going back much earlier) is near impossible. I'd suggest using TAStudio instead, which has a different and more convenient workflow. As well, learn how to use tools like RAM search, and some basic lua scripting if you have the profiency to, which gets you a long way.
Finally, don't take this all too seriously (in a job-like way). As TASers, we spend months, sometimes years competiting against ourselves for shaving frames off some video games; it's a wierd hobby, that we take strangely seriously, but it doens't mean we're not having fun doing it. We're certainly the least qualified people to pretend what we're doing is important, nor should we dress ourselves in grandeur or intolerance.
You couldn't possibly mean that one instance 3 years ago where Spikestuff voted once without watching the movie.Which he then corrected by still watching the movie.
I'm still not seeing that "many people".
Did you know that making an encode of your run on YouTube can increase your odds of getting feedback?
Did you know you can improve your TASing skills by comparing yourself to already existing runs? We have an existing GBA SMW run, did you try at least practicing on that?
You still have not a single reply in any game thread.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Did you know that making an encode of your run on YouTube can increase your odds of getting feedback?
Did that already. In the WIP thread.
Masterjun wrote:
Did you know you can improve your TASing skills by comparing yourself to already existing runs? We have an existing GBA SMW run, did you try at least practicing on that?
the existing run is for Mario only (and it is the any% route). Luigi is slower, naturally, and has different physics.
Masterjun wrote:
You still have not a single reply in any game thread.
I don't because as I said before the forums are very inactive and messages from that thread are like, many years old.
I'll take xy2_'s reply
though I still believe rewinds could be used efficiently (but also with regular save states, of course)
anyways, thanks.
Currently focused on PS1 games Looney Tunes: Sheep Raider / Sheep Dog 'N Wolf, and Toy Story 2: Buzz Lightyear to the Rescue.
Joined: 12/28/2013
Posts: 396
Location: Rio de Janeiro, Brasil
Lu9, I'll be honest - you picked a bad game to start with. SMW is a very hard game to optimize, you would certainly need to spend more than a year only learning about the game before starting your run. SMA2 is a bit easier because some complex glitches are fixed, but it is still not simple.
If you do want to get into SMW, I suggest reading the SMW game resources page, the submission text of the published TASes, and watching on youtube the 40+ videos Let's Glitch Super Mario World series by nathanisbored. With all that information, you will be able to watch closely the lsmv of the published movies and understand what is going on. Some tricks aren't described anywhere, unfortunately, but they are in the movies and following these steps you will eventually learn them by yourself. You may also join the SMW Discord server, where you can always ask more experienced TASers when you have questions.
I don't because as I said before the forums are very inactive and messages from that thread are like, many years old.
What I meant by "the forums are inactive" is that there are not many people on the game forums, not that they're a dead mess. Unlike many forums, this one strongly encourages necrobumping, because any information, even decades old (as has happened) will still be relevant because the game stays the same.
Ultimately, it's up to you to make the threads, and ask for feedback (I laid some ideas for this in my previous post). Of course, making encodes of your run, and explaining what's going on, as well as doing this frequently takes more effort than just blindly submitting entire runs, but it's also what will make your TAS better.
If you react so badly to criticism, then it's probably a good idea to avoid TASing altogether. We are quite brutal here, because you do not understand this: effort is a two-way street. If you feel more the need to defend your run, than to try to improve it, of course nobody will bother to help you further if you retract at the first reply of "here's why what you are doing is bad."
I know it looks rather demotivating that the forums seems rather dead, but I recommend you at least still post your findings and WIPs on it. More than once I ended up reinventing the wheel, because I forgot to post some tricks I found, then completely forgot they existed until I went back to the run much later.
Even if you get no feedback, it would still help anyone in the future looking back at the thread, if they were ever interested in the run.
You conveniently ignored the points I made against which you had no defense. Your other points make no sense.
Good luck with the site. I'm done here.
the only other point you made was about the guy who voted without watching the movie, but still kept the vote after watching.
But that's not the only time - in other submission threads I've seen votes in that poll appear in just a few seconds after the thread was even posted.
Also is there really a need to sound aggressive like that?
xy2_ wrote:
If you react so badly to criticism, then it's probably a good idea to avoid TASing altogether.
I didn't react badly to criticism. I just thought the criticism I got was very vague and didn't explain anything I could do to improve.
And I don't think "the dialogue skipping is shoddy" counts as criticism.
One thing in this thread - P-Switch jump, I discovered it was indeed possible in SMA2, so I did it, along with 1/1 swimming.
xy2_ wrote:
f you feel more the need to defend your run, than to try to improve it, of course nobody will bother to help you further if you retract at the first reply of "here's why what you are doing is bad."
To me it looks like I'm getting more "it's bad, period" than "here's why it's bad".
But I will keep posting, hoping that I will get actual feedback sometime...
EDIT: Also just to note that I looked at Masterjun's criticism of my last SMA2 submission and already improved it, using a lua script, frame advance, rewind, save state, etc.
I just didn't make an encode of it yet
Currently focused on PS1 games Looney Tunes: Sheep Raider / Sheep Dog 'N Wolf, and Toy Story 2: Buzz Lightyear to the Rescue.
And I don't think "the dialogue skipping is shoddy" counts as criticism.
First off - how is it not a criticism of your run? Do you think hand-waving it as "it's not criticism" will suddently make it not true?
To further my point, I looked at your TAS, and it's indeed true; the dialogue skipping is shoddy, probably because you TAS using rewind. Indeed, first off you don't skip dialogue at the first possible frame, and sometimes you even forget to skip it properly; instead of A > (blank) > A you do A > (b) > (b) > A.
Your movement is bad; you don't cut corners, you pick up speed after talking with NPCs in an unoptimal way; there's even a point in the run where you bump into terrain for no reason and slowly have to pick up speed afterwards.
And by the way, the splashes (with company name, etc) at the start of the game can be skipped, saving a bunch of frames. So is it a far stretch, not to tell of your run that "the dialogue skipping is shoddy", but rather "it's an unoptimised pile of garbage"? When you say in the submission text, "There is not much that can be saved TASing this game as it is mostly running around skipping dialog.", perhaps you ought to take a look at what you submit first before justifying yourself.
See, I was almost ready to argue in good faith, but in particular these two parts of your post made clear that you have no intention to do so:
lu9 wrote:
Also is there really a need to sound aggressive like that?
This is just straight up baiting. We are being agressive here because you attack us instead of improving yourself.
lu9 wrote:
EDIT: Also just to note that I looked at Masterjun's criticism of my last SMA2 submission and already improved it, using a lua script, frame advance, rewind, save state, etc.
I just didn't make an encode of it yet
Are you popular in Dungeons & Dragon circles? Because this sounds like a lot of roleplaying. You defend yourself with words, rather than actions.I would have almost believed you if you didn't include the "rewind" part, which easily points out you are lying, because no TASer uses rewind. Also, it takes time to make a lua script, and you casually crossed over my suggestion in my first post of using TAStudio mentioning frame advance and savestates. A bad attempt at saving face, really; nobody cares how you TAS.
To summarise, please stop baiting people here, and trying so hard to defend yourself. Masterjun gave up a while ago on this thread, and I am doing so as well because you don't seem to want to change your attitude.
First off - how is it not a criticism of your run? Do you think hand-waving it as "it's not criticism" will suddently make it not true?
if it is criticism then it's definitely not constructive and is still vague.
xy2_ wrote:
To further my point, I looked at your TAS, and it's indeed true; the dialogue skipping is shoddy, probably because you TAS using rewind. Indeed, first off you don't skip dialogue at the first possible frame, and sometimes you even forget to skip it properly; instead of A > (blank) > A you do A > (b) > (b) > A.
rewind is useful if used together with frame advance, to return to a previous frame.
it wasn't lack of testing, because i did try the earliest frame in a dialogue box, and the first frame that i could skip the dialogue was that one. I should look into that A > (blank) > A sequence, though
xy2_ wrote:
Your movement is bad; you don't cut corners, you pick up speed after talking with NPCs in an unoptimal way; there's even a point in the run where you bump into terrain for no reason and slowly have to pick up speed afterwards.
the movement was pretty rushed, indeed, if I try to tas this game again I'll improve it
xy2_ wrote:
And by the way, the splashes (with company name, etc) at the start of the game can be skipped, saving a bunch of frames.
I didn't know, and I actually tried. But from this I'd guess I didn't try enough which again, it's ok, I was lazy, I know, no need for reminding me that, I will improve and test things more.
xy2_ wrote:
So is it a far stretch, not to tell of your run that "the dialogue skipping is shoddy", but rather "it's an unoptimised pile of garbage"?
there's quite a difference between saying that, and actually giving constructive criticism, which you are actually doing right now, which by the way, thanks for it.
xy2_ wrote:
This is just straight up baiting. We are being agressive here because you attack us instead of improving yourself.
And I was just "attacking" because of previous aggressiveness, so it's an endless cycle. I didn't want to argue. If I did, then sorry. All I wanted was constructive criticism, and I got that now. I'm not a native english speaker so maybe some things I've said made no sense, even though I'm pretty confident about my english knowledge, so I apologize.
xy2_ wrote:
Are you popular in Dungeons & Dragon circles? Because this sounds like a lot of roleplaying. You defend yourself with words, rather than actions.I would have almost believed you if you didn't include the "rewind" part, which easily points out you are lying, because no TASer uses rewind.
I still don't get why can't rewind be used effectively in TASes. It is literally the same thing as save states, but easier. If you have to return to a previous part, like a level, then yes, use regular save states. But for small segments in one level, for example, I don't see why rewind can't be used.
xy2_ wrote:
Also, it takes time to make a lua script
I probably didn't make it clear that I used an existing lua script made for SMA2...
xy2_ wrote:
you casually crossed over my suggestion in my first post of using TAStudio
Well, to quote on what you just said:
"nobody cares how you TAS."
So if a run is good, who cares if I used rewind, save states, TAStudio or whatever? And I know my submitted TASes are below average, but I'm still improving. And the actual criticism that I got will help.
If TAStudio is easier, then probably I just like doing it the old-fashioned (but probably harder) way more. But I will give TAStudio a try.
--
So for short, I got the constructive feedback I wanted, tips on where to look for more people willing to give feedback as well, so this thread probably has no further use, unless you want to add more to it, or argue against my points (which I encourage! really, I'm not looking forward to being "the only person right" here, I just wanted to defend my points, which could still be wrong, so I'd actually like to know if they're good or not)
Currently focused on PS1 games Looney Tunes: Sheep Raider / Sheep Dog 'N Wolf, and Toy Story 2: Buzz Lightyear to the Rescue.
I agree on the point that calling it "unoptimised garbage" is uncalled for. And a lot of the rest of my previous post is baseless attacks on you. That's what I get when I write posts at 2 AM without thinking about it; I apologise for the attacks and the rest of my rude behavior, which is especially uncalled given the context here.
To answer the rewind point:
I still don't get why can't rewind be used effectively in TASes. It is literally the same thing as save states, but easier. If you have to return to a previous part, like a level, then yes, use regular save states. But for small segments in one level, for example, I don't see why rewind can't be used.
This is mostly a matter of workflow. First, not every BizHawk core has rewind, and other emulators do not have it all. But the main reason I push TASeditor/TAStudio is because it allows a much better workflow for TASing; you are not limited to 10 savestates - instead every frame is a savestate, and you can jump to it very easily, unlike in traditional TASing; when you go back to a frame, the rest of the movie is not truncated unlike with savestates where they do not store the input after the savestate; you have the capacity to undo; and you are not forced to TASing a small subset of the movie at a time, but can make modifcations anywhere at anytime by simply clicking on a frame, as well as seeing the complete input log-- this is far better than the small step back of a rewind, or the limited segment of traditional savestate TASing.
