Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
Very nice to see someone working on this again!
Some parts of that last level looked a bit dodgy though. Also, not moving at the end of the first big boss would have saved at least 5 frames due to lag reduction.
Anyway, keep it up :)
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
Not really... if you have read all the posts, you would see that his story was indeed very unlikely. He had been talking about having a wip, and wanting to post it as soon as possible for over a week. Posting a wip can be done within seconds. I even wrote a detailed explanation how to do it. Then all of a sudden his WIP got mysteriously deleted, and he decided not to remake it (even though he had only done a little, and didn't lose THAT much progress). So yeah, I considered this story to be very unlikely, and I generally don't believe very unlikely things (especially when coming from a relatively new user on an internet forum) unless I see some kind of evidence.
When he actually posted his suggestions, I was the first to try them out and admit he was right.
Edit: Nice Stoneman wip btw Glitchman, it seems like using less superarrow to avoid weapon changes more really pays off.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
In Spezzafer's submission thread, bullethead2202 posted two movies showing walljumps in Yoshi's Island. A walljump could for instance save time at the end of 2-1, where there is a small wait before a rock that's falling down is low enough to jump on.
NxCy and I don't have a consistent way of performing these walljumps, and bullethead2202 stated in a PM that it was more trial and error. On the (E) version, we can get walljumps quite easily by randomly trying, we haven't been able to do even a single one on the (U) ROM. Also, even on the (E) ROM we haven't been able to do a walljump on a moving wall/rock.
As it's only a result of randomly trying, we aren't 100% positive that it's not possible on the (U) ROM, or on moving walls... but considering the ease of doing them on the (E) ROM on stationary walls, we decided that we're sure enough that it's not possible to move on.
(One of our theories is that Yoshi's speed seems to be in the 900s in the (E) version, which is a bit higher than in the (U) version, this might also influence it... the (E) version also doesn't run nearly as stable as the (U) version it seems.)
Either way, if someone were able to do walljumps either on the (U) version, or moving rocks, it would be nice to let us know, otherwise we'll continue :).
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
Sorry to go slightly off-topic, this is a bit general, but related so this movie (as it has relatively few rerecords).
Althos wrote:
*The number of rerecords is not important, why would you rerecord a lot if you do it perfectly the first time.
How do you know something is perfect without rerecords? The most basic thing, pressing a button on the first frame possible, requires quite a few rerecords to perfect... you only know approximately what is the right frame. Even if it looks fast, it's still trial and error to find the absolutely earliest frame possible.
This is a 50+ minute TAS, and has probably lots of situations like this... even if no things like lag reduction, luck manipulation, strategy testing, route planning etc is required for this TAS, and only pressing a button on the right frame is needed, the rerecord count is very low.
What also generally true (although you might be able to find exceptions), is that if a TAS does require things such as lag reduction, luck manipulation, strategy testing, route planning etc, (which require lots of rerecords), the game is generally more interesting to see TASed.
I haven't watched this TAS yet, and I know the number of rerecords displayed can be easily edited etc, but the number of rerecords used to make the TAS is not completely unimportant I'd say, as it can be an indication of some things.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
Í assume the glitch can only be done off the ground, and not off another super arrow?
If it could be done off another super arrow, it could for instance save at the third room of napalmmans stage... to glitch through the upper wall in the section where those instant kill blades come at you in the narrow corridor...
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
I was going to reply that this was nothing new. Zipping through the wall has been done quite a while ago.
This however always required taking damage, and there is no spot in the game where it actually saves time :S. (In the TAS of the hack that woabclf just linked, it apparently does save time at least somewhere, here at 3:30.
Anyway, what you just showed is certainly something new (to me at least). Although I think Megaman still has to take damage if he wants to go through walls that are thicker than that. Either way, nice find! :)... I hope it can be used in the TAS somewhere.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
I probably went in watching this having too high expectations, given the positive feedback. You described the movie quite well, so it's not your fault... but I ended up being disappointed.
Maybe the reason for this however is that the idea of for a TAS like this isn't new. I realized that doing everything with UFO or WHEEL fails to show off one of the most entertaining (imo) parts of the game. Since the SWORD and MAGIC ROD are required to beat the game anyway, I figured I'd do a run using only those powerups, in order to show off this aspect of the game, and still have consistent goals. You can read the response to it in the linked thread... but looking at the WIP I posted in that thread (this is an fm2 version btw, in the thread it's fcm... also PRG0), I think a full optimized TAS with these restrictions would be very nice.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
You talk about 3 improvements, but only list 1-2 and 1-4. I know level 1-4 very well, and I don't see a very obvious place where wall jumps would save time.
Either way... maybe I should have learned not being so sceptical about claims of improvements without any evidence, after Glitchman (his claims about Megaman TASing turned out to be true)... but unfortunately I still am.
Could you post an smv movie file showing these improvements? I think one is only justified to make statements about improvements if one can provide that (or the improvements are too obvious, which isn't the case here).
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
No, I can't/won't encode them. Maybe you can convince one of the people that usually make encodes for TASvideos to encode them... but otherwise you'll just have to watch them on the emulator.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
Bezman wrote:
With the next playaround, would it be possible to have multiple patterns/ pictures? Vertical stripes would be cool, as would the suggested sprites or 'TAS' lettershapes. I seem to remember Tetris DX's blocks falling faster, so maybe the pictures could be wiped and new ones created?
Hmm, one of the aspects that I like about this TAS is that you can somewhat follow the puzzling that is going on. I think that acpect might be lost in a Tetris DX playarond. I also then would rather see a regular GB Tetris playaround, as that will be roughly equally fast (though be it without real colors), and isn't represented on the site yet.
However! I am not planning on doing either of these. Somewhere in the (maybe far) future I want to do a Tetris DS TAS of (almost) all modes. Getting highscores in some modes requires unlocking level 20 first in a seperate run. This would be very well suited for doing a playaround.
Tetris DS would be a different kind of playaround as individual pieces can't be manipulated, and one has to make best of what one gets. This means a lot more lines need to be cleared for getting something particular (and maybe the hardest patterns / pictures (like the one in this TAS)) aren't even possible. However, that more lines are needed for them doesn't really matter as the blocks can be places so much faster, and the shape can probably still be formed quite fast (if not faster).
Bezman wrote:
so maybe the pictures could be wiped and new ones created?
Particularly mission mode lends itself perfectly for such a playaround. Clearing a mission in mission marathon mode drops down the column a couple of lines, so the entire field can be cleared within seconds.
Tub wrote:
Baxter wrote:
I did consider the picture idea, but there are some difficulties. The field is only 10 blocks wide, which is not enough to draw even the most basic Mario or Link sprites. You are also limited because every line needs to contain at least one hole. Either way, in the ends, I couldn't think of a nice, recognizable image.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
TASVideos Grue wrote:
om, nom, nom... blech, bitter!
If you don't want it, I'll have it back then.
I uncancelled this as it was more popular to have this seperately instead of grouped together.
I probably acted a bit too quickly before. Sorry!
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
The lack of even trying to do anything remotely interesting during this boss battle is almost enough reason for me to vote no.
The next boss battle made up a little for it.
Either way, there has been some discussion on doing a glitchless run for NES Mega Man 1. Rules for that would seem very arbitrary probably. I think I'd rather watch something like this. This however containing the first three games might be a missed opportunity to also show the upgraded graphics for Mega Man 2 and 3... I bet those runs would also be quite different from the NES ones too, that bonus game might also be interesting.
Overall, it felt this run was a bit lacking (especially that boss battle I mentioned)... if polished, and with the other games added, I'd see me voting yes.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
MUGG wrote:
In my opinion this feels like throwing the any% run with a 100% run of a game together. While I think that in this special case mode A and mode B in one movie wouldn't hurt, the playaround should stand alone. The playarounds have been strongly distinguished from the real speedruns in the past anyway, right? Isn't there even a category or tag for those?
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
Paused wrote:
NxCy wrote:
Today we finally finished world 1...
This means that by the time we finish world 6, we can release a WIP...
Grats, but...
Go on, finish teasing us and tell us how many frames you have saved thus far, please?
Hehe, it's not that bad, every time we complete a world, we will release a wip with 4 levels. Because we already released a wip of the first 4 levels right when we completed them, the wait is a bit longer. We however didn't want to have shown everything when we complete the TAS (eventually :P), so we decided we'd stop releasing wips after world 3. The wait for new content between that wip and the submission would even be a lot longer than this wait probably, so we hope people understand.
We will release the next wip (which completes world 1) when we have completed world 2 (we should work a bit faster than we did sofar, but keep in mind that the levels in world 2 are significantly longer than the levels in world 1).
Not to let this post be completely without content: in the second half of world 1, we saved 30 seconds! This means our wip of world 1 is currently about 45 seconds faster than Spezzafer's wip! (Do note that we will lose a LOT (probably all that we've gained sofar) in 2-1, since we won't be taking the warp Spezzafer used (our run is warpless).)
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
JXQ wrote:
Two things come to mind.
1) What happens when someone later obsoletes one of these three? Do they have to do all three before submitting because the precedent was set here?
2) If the GB run was not accepted because it's similar to the NES version, I don't think those circumstances will change if this run is published and obsoletes the other NES runs. It's still the same amount of Tetris, just a different amount of separated movie files that contain this.
1) It's probably the same as if someone would be faster in one level on any other game. That doesn't mean he/she won't have to TAS the rest. I agree that this makes it less appealing to even try... if each mode were to be published individually the bar would be lower. I wouldn't be against that, but I can see the point of view from (other) judges.
2) I think the amount of published movies was the problem here. If it were just about the amount of content, then there wouldn't be anyone against these modes all being published seperately. I think the main issue here was that one might be able to come up with a dozen interesting concepts, that would all warrant their own published movie then. If one keeps it at one published Tetris movie per game, and just adds all concepts per game into one movie, then this problem is avoided.
Experienced Forum User, Published Author, Skilled player
(1408)
Joined: 5/31/2004
Posts: 1821
I did consider the picture idea, but there are some difficulties. The field is only 10 blocks wide, which is not enough to draw even the most basic Mario or Link sprites. You are also limited because every line needs to contain at least one hole. Either way, in the ends, I couldn't think of a nice, recognizable image.
Things like spelling TAS are obviously possible, but I plan to make a Tetris DS (where each block shape has a different color) TAS in the future which will also include modes where one can make playarounds like this, so there might be something like this in the future.
I picked this pattern because every block really has to be handpicked to get this to work. A pattern like this would be pretty much impossible on the DS as the pieces can't be manipulated. The strength of this TAS I think is that a difficult puzzle, which would take hours to solve normally is being solved in a fast (but still watchable) pace. An image which might have relatively large areas of the same color wouldn't display that puzzle solving element that well (would even be a bit easy when one is able to manipulate each block), and might be better suited for a DS TAS, where the speed is too fast to keep up with what's going on anyway.