Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Spider-Waffle wrote:
I had some crazy ideas for the wily stages but some how I doubt they'll work :\
If it depends on room adjacency, small tunnels without ladders or involves close proximity with spikes, it probably doesn't work.
Otherwise, please explain :)
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Hmm.... I didn't know this topic existed.
(And now that I do, I combined some existing discussion about the subject into this thread.)
Well, now that I do, I might as well mention this movie:
https://files.tasvideos.org/bisqwit/smb3j-frog-test.fcm
It plays the first few levels of the world 1 using the frog suit.
It requires using the master cheat code.
(Address $A9C5, value 204, compare 53, type "read substitute")
(In Game Genie syntax, KKKZSPIU)
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Spider-Waffle wrote:
Can somone explain what happens in the quickman level when they do that long zip through the top of the map and it ends with the tunnel turning down, how do they get to the next screen from inside the wall?
In Mega Man 2, there are not really concepts of "up, down, left and right" in the map. There's only next and previous.
However, the next room can be arrived into even by going to the wrong exit.
When a wrong exit is applied to get to the next room, the screen doesn't scroll properly, and a difference will form between actual and displayed room. At some point, the player must return to the actual displayed room.
What Finalfighter did was that he went to the "next" room not from the bottom exit, but from the right side exit. He then played a few screens further, and at some point of time later on, he returned to the actual displayed room, which as always is the previous screen from the point where Mega Man is.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Hoe wrote:
It apears that you're just OR'ing the last X frames together.
Nope, that would create overimposed colours (and generally hide black pixels). My algorithm doesn't invent any new colours.
(Ok, the GIF was reduced to 8 colours for better compressability so that's not entirely true, but it's not the point.)
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Blublu wrote:
Whenever a pixel changes, it's given a set number of frames to live, and when it expires, it's changed to the original image's equalent pixel. If it is changed again before it expires, the "frames to live" is reset.
Seems to work like that to me...
Your theory doesn't work: When a magnet beam part disappears from the image, it also counts as a "change", and your algorithm would keep the change - the sky - visible for the set number of frames, instead of the beam.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
adelikat wrote:
This algorithm looks really promising for future runs. Perhaps you can compile this into some sort of program to make available for time-attackers. It look extremely helpful for comparing various routes.
To avoid confusion, the algorithm I referred in my last post was about the GIF linked in the same post, not the different-routes-side-by-side image that adelikat seems to talk about...
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
For fun, I've been trying to create a version of the above-shown animation where the magnet beams are contiguous instead of flashing.
Among many algorithms I tried, here's the most promising so far:
http://bisqwit.iki.fi/kala/snap/rm1bombdemo2b.gif
Now, who can guess how exactly the algorithm works?
Hint: The animation generator has no access to the emulator's knowledge of sprites. It operates only on images.
Hint 2: Except for the background (array of most commonly appeared pixels during the movie clip), the animation generator has no knowledge of future.
Can someone suggest a better algorithm?
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Spider-Waffle wrote:
nice if you could place the first 16block MB such that you could place the other 2 with only 1 jump, maybe adjust the placement of the very first MB.
Of the 9 possible positions where the first beam can be put without having to jump twice, I tried last two. The last didn't allow proper placement at all and the one that worked, only allowed proper positioning if the placement jump is done immediately after landing.
I didn't test more positions...
Edit: Oh, there's the thing that
- The beam has to reach far enough (to be able to jump on the second zipping beam)
- The higher the beam is, the less far it needs to reach (because a jump reaches longer).
So I placed it as high as possible.
If I place it lower, I have to delay the release and as result, the first zip would be shorter.
But be sure I'll try to optimize it when I play the part for the actual movie.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Spider-Waffle wrote:
Why is two refils with only two switches unlikely?
Because the cannons are far from each others (I recall they activate on proximity) and because the big weapon powerups are rare (tedious to manipulate the randomness).
Perhaps it's possible to take one from cannon and other from sinewave-flying bomb, but without testing it's impossible to say whether it really works.
> Also I think you could do F with only 1 extra beam if you used the technique I brought up with the iceman jpeg.
The underlying beam can only be 256 pixels wide. And it must end just before the next floor starts (in order to be able to jump on the next beam). This means that it must begin very late. Add there the time needed to construct the actual F-zipping beam -- the F-zipping would be very short. (Note that the F-zip must end just before the hole in the floor.)
Edit: You mean like this? It seems to work nicely enough. Saves 1 beam.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
There's one particular thing that fascinates me in this game.
In the end of my Iceman battle, you can see that there appear extra Rockman characters for a few frames. I don't know the exact cause of it, but I think it's a buffer overflow glitch of some kind.
There's one particularly interesting item in the game: item number 0x43. It's a teleportation item.
The item can be found only in the last level, Wily 4. Each teleport contains one of those items, and they are invisible.
If the item is activated in some other level, weird things can happen such as teleportation forward in the level. I'd like to find a way to trick the game into cloning such item.
Of course, there's also item number 0x44, which is the level-end trigger...
Other items of curiousness include 0x2F, which invokes a level-specific behavior (found in Gutsman level (lifts) and in Wily3 (water tunnel)),
0xFF (water tunnel along many more functions), 0x4A (random powerup) and 0x3B (boss activator)...
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Walker Boh wrote:
Is it somehow possible to zip through the wall at the first ladder you are supposed to climb upwards? That would be awesome since the boss room is on the other side of the wall.
No, it's not possible.
The level does not really have a 2-D layout. The only relation between rooms is "next", "previous".
Going through the right side of the room will take you to the next room (and only partially - you can't really progress there).
For ease of visualizing (and to keep the image in sane size), I did not make the map 1-D.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
New Bombman level plan.
https://files.tasvideos.org/bisqwit/bombmanmap2.png
These choices were made:
- A instead of B (B is almost nothing)
- R=1 (number of recharges)
- E instead of F (no zipping)
- D instead of C (in Elecman level it was a surprisingly big gain)
As concluded in previous posts, the end of the level (after the 1up room) will not feature any special tricks.
Doing two separate recharges is a bad idea, because it takes a lot of time.
If I can somehow manage to do two recharges with just one switch2 (which I think is unlikely), I'll change the plan: I'll either do F (+2 beams, 3 excess) or the 4-beam ending (1 excess), depending on which works out faster.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
It desynced at me in 4-3.
Using Windows version (fceu-0.98.12-blip.win.rar).
I'll check at home using some other version.
For what I saw until that point, I have one question:
Often it seemed like you didn't do perfect corner cuts.
When you had to fall a floor down to the right, you didn't press against the wall while falling. Instead, you approached the corner from a few pixels of distance. What is the reason of that?
Other than that, it looks perfect.
Basically this.
