Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Well, you're doing the GP run, mine is a time-trial attempt, so he can wish us both luck ;)
As a side note, TASEditor I encourage you to post your emulator movie files (.m64) along with your encodes - I'm amazed that with all the work on this game over the years that almost nobody shares the files that would let people view the input required for various strats - it makes it a lot harder having to reverse engineer them from AVI's!
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
I'm happy to announce that after only about 20 hours of fruitless banging my head against a wall (and mupen64) I finally was able to float in Mute City 1, which will hopefully lead to some progress on my long-stalled time-trial effort.
EDIT:
http://dehacked.2y.net/microstorage.php/info/1075350257/Lord%20Tom%20FZeroX.m64
Looking to get some help from MrFixItOnline re: steering in the air, but can't post links until I make 10 posts and 1 week passes...doh!
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Indeed - the zebetites (pillars before mother brain), the red doors in Tourian, and mother brain itself all require missiles as far as anyone's ever been able to determine.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Definitely the most entertainment I've had in an SMB run for a long time. Given its simplicity and the extent of known bugs, a playaround is where this game should really shine.
I agree with many that the goal of "low score" has big negative side-effects and that a shorter run with the same tricks, or an equal length run with a greater store of tricks would be superior. But until such a run appears, I vote yes for this to be published as a playaround, for obsoletion by a more entertaining playaround without reference to the score goal.
A further drawback to having the score goal which I don't think I saw mentioned is that you lose opportunity for entertainment in levels where 0 score is possible - 1-2, 4-2 and 8-4 end up being just like the any %.
My ideal playaround for this game would show all manner of glitches, probably using some (not all) warps for completion time ~15 minutes. Further, I'd use subtitles to advertise arbitrary (but entertaining) goals for certain levels, e.g. walkathon for whatever -3 level it was (5?) that took the community months to find a solution for.
One can't deny, though, that HappyLee knows this engine, and spared no effort with the entertainment, despite the limiting goal!
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
I'd like to see some solution to the "racing game" issue. Having TAS'd Excitebike, it's clear that in some cases it is reasonable to have one published movie for all tracks.
BUT, having started TASing N64 F-Zero X, it is clear to me that in some cases it is NOT reasonable. Doing the first track took me about 40 hours (granted, some of this due to unfamiliarity/difficulty with the emulator). There are 23 other tracks, though, so even for such a popular title, a complete TAS would be a massive effort. If some team were to complete this, obsoleting it would be another massive effort.
The result, which I see as a negative, is we have no F-Zero X presence on the site.
The SNES F-Zero seems about halfway between Excitebike and F-Zero X; it would be reasonable to think a single motivated TASer could complete all tracks, though no one has coming into the site's 7th year.
Another reason I think the track-by-track approach is reasonable in cases where there are many tracks of moderate to high TASing difficulty is most racing games have no "state" that carries from track to track, ie it can reasonably be stated that 30 single track efforts is just as legitimate an accomplishment as a single 30 track effort.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Ok, we give up on trying to get rid of that lag frame in 1-1. There are various ways to eliminate the lag, but all of them seem to require losing at least 1 normal frame. And losing even 1 normal frame wrecks the hammer brother manipulation at the end of 1-1.
I contacted andymac, and he has not done so either. However, Mitjitsu is correct in that were someone to find a way to remove this lag without losing any normal frames, that would be an improvement that would be immune to the now sizable pseudo-frame-rules of World 8.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Update on the 1-1 lag issue. It may not be possible to remove that lag without losing a normal frame (which would ruin the hammer manipulation). It certainly isn't easy, at least! If anyone has actually seen it done, please speak up, otherwise I'll ask the judges to give us until the weekend to succeed or fail.
And thanks for the comments!
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
mklip2001 wrote:
Wow! I have a few questions though:
* Is it possible to hex in a 1-frame improvement in 1-1 without too much trouble?
* When you stomp the boomerang brother in the first stage of World 8, you look like you're drifting down slowly to land on him because of the P-speed. Is it faster to let go of the B-button for a few frames when running to him to avoid getting P-speed? That way, you're in the air for less time. (It might not matter though, if the boomerang still has to leave the screen before the chest appears...)
* Are you going to apply these new improvements to the warpless run again?
In any case, fantastic work once again!
-I'm investigating the 1-1 issue. It will be no problem to hex it in if that lag frame can be eliminated; however the technique that I'm aware of to eliminate that lag involves losing 1-2 pixels, so I have to see if the lag can be eliminated w/o losing time. This wasn't relevant in the warpless b/c we needed a different hammer manipulation, so we were losing time intentionally for that.
-On the boomerang brother, we have all the time in the world to actually kill that guy; the boomerang is by far the limiting factor
-I'm open to doing the warpless again, but I'd prefer to hit some other projects first! This run really only has the BoomBoom y-subpixel trick to add to the warpless, so it'd be a very tiny improvement to bother hexing in.
DarkKobold wrote:
Did you mean faster?
Also, easy yes vote.
Plus, is there any way to test earlier frames, to see how many frames you'd have to save to get 6 hops
No, 240 is slower than 204. It's because for negative/upward vertical speeds, the absolute speed is 256-x; so 240 is the same speed as 16 in the opposite direction; 204 is the same as 52, etc.
One can test earlier frames by hex-editing Mario further ahead than he can really go. We did so, and mention in the last section of the comments (you didn't read to the end?! sniffle) that one would need to get to Bowser 17 frames faster than we could (or 38 frames faster than this submission, which wastes 21 frames).
Etchenba: the 1-up sound glitch occurs when Mario tail-flips five frames after the 1-up sprite appears. I believe there's a great technical explanation in the SMB3 forum.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
I'd vote no; the difference just isn't fundamental enough to switch ROMs. I find it more entertaining to see the bullet glitch where it's used because it's so difficult - PAL just makes the same glitch easier.
Contrast that with Rygar and Blaster Master, where totally different behaviors (jumping over river, air-jumping) and routes (passing through walls, skipping levels) are enabled by the PAL ROM.
None of this is to take away from MUGG's project - it's great to see this new avenue explored, and by all means link to it from the published run.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
I suggest:
-We keep the "swordless" title - that's a familiar term that's been around for years and best conveys the challenge.
-State in the first sentence of the movie description how this movie handles the conflict between the challenge and what is required (at this point) to complete the game. If someone finds a truly swordless technique in the future, everyone understands this movie will be toast.
-As I read things, only a small minority has actually taken issue with publication of the movie as is. This small minority would be unlikely to be satisfied by any of the possible permutations to this movie's ending (which would require redefining "completing the game" as opposed to appyling a caveat to "swordless"). There's been a ton of semantic discussions - very frequent, fun and healthy around here - but nothing has arisen to justify taking away the author's prerogatives to make the reasonable decisions jprofit22 made.
-Let's publish this bad boy! :)
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
I like the comparison to "pacifist" Contra - a challenge, which, while it doesn't fall into the any/low/100% pigeon-holes, is a very natural one given the mechanics of the game and leads to very interesting changes in strategy and tactics.
I'm not familiar with the CCC issue in M64, though it appears that challenge was felt to fall short in terms of the novelty of its gameplay. I don't imagine it helped that the submission text mentioned that it was improvable by over a minute, though the judge's comment doesn't mention it.
I still think the guideline of limiting branches to a game is a good guideline to avoid nonsense like "gets blue ring and magic boomerang and kills every bat with aforementioned boomerang", but I think there's a pretty compelling case that adding this category will enhance the site.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
So glad to see an entry on the workbench for this run, 6 years (!) after the topic was first raised in the forums.
Looks great; a big improvement over your last WIP. I especially like how you enter level 6 with half a heart and keep that throughout the entire level - definitely emphasizes the tool-assisted-ness.
Everyone will have their opinion over the best way to handle the swordless/Ganon fight conflict, but given the extensive discussion in the forum, I think it's silly to argue that this solution is invalid.
Voted Yes, and congratulations to jprofit22 on an ambitious first submission!
EDIT: Cardboard: LOL, I was thinking of making some sarcastic note to that effect, but yours was much wittier! :)
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
That makes sense; I'm glad you got the money-game count down to 4, that makes the tradeoff very reasonable. Be sure to point stuff like that out in the submission text, it'll save a lot of questions!
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
22 hours in 2 days! Yep, those were the good old days! :)
One other thing I noticed is on your prior wip you used the money-making game 5 times; IIRC the published 1st quest uses it only twice and grabs 100 rupees from one of the "It's a secret" dudes. Just something to look at if you haven't already; you're buying an extra bomb and using the new overworld glitching, so that may change the math in favor of the money game.
Sounds like you're making great progress; I look forward to seeing it finished! No rush, though...it's easy to get burned out doing a TAS. I try to finish at a point where I'm getting a bit sick of the game but not yet feeling like I'll throw up if I ever see the title screen again!
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Yeah, it's the "wiggle-spawn" technique described in the published glitched and full-length runs. A roller will spawn each time the player's x-pixel crosses a certain boundary; by wiggling in place (barely visible in realtime) up to 4 rollers spawn within 8 frames, so they're right on top of each other.
Watching on frame-advance you can occasionally see there's more than one as they come slightly out of synch.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Warp: A max of 4 rollers can be onscreen at a time from that spawn location. But an item counts as a roller until it is collected, so a max of 3 rollers can be spawned while an item is onscreen.
I think for the any % I found it was faster to just spawn 3, but I can't say I remember 100%, and it's a different rom...Cardboard?
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Glad to see a full run-through, swordless! I'm sure small amounts could come off most dungeon rooms with further effort, but overall a great looking TAS. A few comments...
1. I'd drop the 'minimalist' concept and just stick with 'swordless.' I think the topic's been kicked around enough and generated enough interest that you don't need any other category. Anyway, it's not really minimalist in any real sense, since you get MORE items than the any %! (which doesn't get the wand, though obviously it gets the sword much earlier)
2. Money, life and start-screen management do stick out as being a bit iffy...not a game-breaker but could definitely be a source of improvement if you redo
3. I was surprised at how similar the battles seemed to the sworded runs; guess one gets used to bombs exploding all over the place and doesn't really notice the sword much...
4. Overall, sweet! I'd encourage you to submit this run after 1 more redo, if you can stand it. (I'd vote for it to be published as is, but it'd clearly be vulnerable to obsoletion which might draw some protests)
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
I have tried the overworld glitching at multiple locations with the U-rom many times and never got it working. I'd be very surprised if it were possible.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Derakon: yeah, that's a lot, especially if one ponders the Metroid 100%, which is several times longer. But I imagine one could process the run in segments, splice them together, then synch with the sound.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Andy: Thanks for the how-to info! Also, I must have missed that you already had all the info about the Metroid run being tool-assisted; please ignore my earlier request.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
If you watch closely in the any %, there's a point where the player drops vertically by several pixels for no apparent reason. I think it has something to do with the game trying to line the player up with the floor. No idea if it's doable on the U-rom, but it saved a fair chunk of frames (maybe 15 or so), so that might account for any difference you found.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
AndyDick, I'd be interested in a brief overview of how you're currently doing this.
Also, to avoid misleading or irritating anyone, can you mention somewhere in your youtube post that the metroid run was tool-assisted?
Following with interest!
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
On screen 2, isn't it slower to drop straight down to the door, instead of dropping off the earlier ledge and pressing right while descending?
Otherwise, looks pretty good. Nothing too exciting, but much faster than last time, I'm in strong-meh to weak-yes territory...yes it is.
Experienced Forum User, Published Author, Expert player
(3276)
Joined: 5/25/2007
Posts: 399
Location: New England
Watched it; totally amazing! I'd say it should be automatable in some fashion with some hackery:
1. dump a file with camera x,y changes & corresponding frames
2. dump screenshots for every frame
3. write a program to animate it
4. synch with sound
I'm 100% unsophisticated with video stuff, but the above would definitely be possible; perhaps others know a better approach.
I'll add the NES Metroid 100% to my wishlist!
