Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
c-square wrote:
1) It looks like you slow down on the last letter of TWINS and LEFT, even though it's possible to pick a letter at full speed. You don't slow down on SWAN or COMEDY. Why the difference?
I'm guessing on this answer based on my own work on the game.
Due to the frame rule from the end of a puzzle to the next die roll; delaying a final letter's input allows for a 1st frame die roll of 6 without sacrificing total frames. So it can appear 'sloppy' but it has a purpose.
On an unrelated note. Spikestuff's April Fool's submission was rejected on the grounds of this game being a board game and thus not vault eligible. Even in the judgement the statement was made that the arcade sequences wouldn't change this perspective because they can be skipped.
I think any submission of this game is going to be fighting a major uphill battle to be considered for publication on the site.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
feos wrote:
When I first suggested the term "warp glitch" I used this logic:
- We're getting more and more runs that skip directly to the game end using some glitch. Let's call it "game end glitch".
- What about runs that use glitches to skip very far ahead, but not straight to the game end? Let's calls it "warp glitch".
I don't think it matters how you skip to there precisely. You arrive to a point you weren't supposed to visit that soon, and it's not a game end. Even if it happens consecutively instead of instantly, you still do skip to there. So you effectively warp. The key here is not whether physics engine is used or ignored, but the result of arriving to some completely unrelated area. Does it make sense?
Makes sense to me. Also removes the necessity of differentiating in the branch whether or not the physics engine is used to effect the warp. The submission text should be valid enough to differentiate.
Truly, using 'warps' vs 'warp glitch' in the branch should be enough to differentiate between whether or not the physics engine was used (as an intended warp isn't really a glitch).
'Game End Glitch' is a pretty straightforward description.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
What ROM version are you using?
EDIT: In the Japanese version I have, the UFO's Horizontal position has RAM values
0x072B and 0x072F for the left and right sides respectively.
I can't figure out how to identify individual aliens, but here are some RAM values regarding them.
0x0032 is enemies remaining
0x0033 and 0x005c have to do with their movement
0x0029 is lives remaining (if you care)
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
feos wrote:
Most importantly, if I get hit by a truck, how do new staff members that need to handle this labeling know if it's moot or not? What do I say as I record my valediction to them (before I die on that road), teaching them how to resolve this problem?
Interesting choice of arbitrary way to disappear from the site....I'm a bit surprised you didn't come up with a more obscure method.
For what it's worth, I'm hoping you don't leave anytime soon.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
Yes vote for adventure games! I know i'm in the minority, but I think it's interesting to watch them being beaten as fast as possible.
I'm not super familiar with this particular game, but it looks to be a nice run. Too bad 30ish% of the run is waiting for loading screens.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
This is one of those games that looks like it would be a fun challenge to play through from a casual perspective. It's unfortunately also one of those games that's just not very fun to watch. 'Meh' vote for entertainment, but well done otherwise.
A 100% run of all levels and saving all the bombs would probably be quite a challenge even from a TAS perspective. While it would obviously be much longer, I think it would be more impressive than this run from a technical standpoint. I don't know if it would be any more or less likely to published though.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
Interesting.
Update for WIP using the 'CUP' sequence: I manipulated all the die rolls to be 6 on the first possible frame. I saved a whopping 3 frames total. Current run-time is 6099 frames.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
Thank you both for your perspectives.
And thank you Spikestuff for the more specific comparison to the previous run. The 15 second estimate was simply an estimate with which I was mentally including the resetting of the world X position to = 0 after the boss battles as well. Sorry if that wasn't clear.
I respect both of your perspectives regarding the value of the entertainment choices made in this TAS. I don't find this game very entertaining to watch myself. I personally doubt there's much of anything that can be done to bring this game up to moon level. But that's been my opinion since my last submission (one of my earliest) was rejected. That rejection actually helped me better understand how runs are perceived by the community.
EZGames69 was the one primarily interested in doing this game. He believed it may be able to be made entertaining enough, and approached me about helping with the run due to my previous work on the game. While I hadn't originally planned on revisiting this game, I agreed to help.
I personally don't anticipate many 'yes' votes for entertainment on this game. But I believe that this game is indeed vaultable due to the fact that there are significant optimization opportunities. A 4.5 minute difference over the human RTA record is significant, especially considering the pseudo-autoscroller (or 'glorified autoscroller') aspect of a large portion of the run.
I've never been one to shy away from vault runs. I believe they are important for archival/posterity sake (which shouldn't be a new concept to anyone who's read my opinions before).
DwangoAC: for what it's worth, I think if both this and the sequel game can be done with identical (or one set of) inputs; a side-by-side TAS would make a slightly more interesting viewing. I didn't know there was a sequel until about half way through this prjoect when EZGames69 pointed it out to me. I even mentioned to him at that time how incredibly similar they appeared. Other than the sprites, the sequel even has an identical end cutscene with the code for hard-mode.
EDIT: Not to be picky, but....regarding the rejection of the two previous submissions of this game:
The first was rejected for sub-optimal play. The judge in that case also incorrectly stated that the autoscroller portions couldn't be sped up or slowed down (they can).
The second (my original) was rejected more for goal choice (collecting all stars possible) than for having autoscroller components. It was even suggested by the rejection note that the game may still be vaultable if the time wasted collecting stars was saved.
EDIT 2: Ok, that might be a bit picky.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
I have a HUGE amount of respect for the work that went into this run...unfortunately this is simply a boring game to watch. Meh vote for entertainment.
Something else to factor in is the time required for die rolls.
I don't think this is going to have a big enough impact to make us choose another word sequence. In my fun run I think I had to wait like 20 frames for one roll, but otherwise, a 1 in 6 roll is extremely quick to manipulate. The way I see it, all possible runs are equally affected by die roll manipulation, and I doubt we'll be unlucky enough to see an outlier. But you never know...
In my WIP waiting just two frames before selecting the game mode changed the die rolls enough to save over 30 frames on the final TAS. I still enter the team name on the same frame to get the 'CUP' word sequence. 30 frames is a lot considering the die roll RNG changes every frame; and to me, half a second is a lot over a 1 minute 41 second run.
Please don't take my comments wrongly. I'm very impressed with your work on the scripts (partially because I recognize that I'm currently incapable of producing what you have). I just feel that there are too many factors to automate everything in determining the fastest potential. At some point it's going to take some 'brute force' testing of the handful of options that show the best potential.
EDIT: I have no idea why the font size changed in this post. Thanks Spikestuff.
Testing frame 529...
TWINS (124)
SWAN (108)
COMEDY (168)
LEFT (160)
LIP (112)
STOP (116)
26 characters
788 frames to type + 44 frames of waiting = 832
!!BEST SO FAR!!
My original sequence takes 1092 frames. Yours takes 1148 frames. None of these numbers account for the 128 frame rule. It also assumes you type each word by either holding left, or holding right; it can be optimal to stop and turn around, but my script doesn't account for that.
Great....but if your script doesn't account for changing directions, what's the point in calculating how many frames it takes with the shoes just going one direction? I wouldn't be surprised if that factor alone significantly changes the outcome.
To truly determine what run may be fastest, we must consider both the time waiting to get a later starting frame as well as any frames saved by changing direction. Not to mention the frame rule (which I, admittedly, need to look into more myself.)
My WIP using the 'CUP' word sequence starting at frame 481 has its final input on frame 6102.
I may TAS a few other word sequences that appear as though they may be close and see if there is a better one than 'CUP.'
EDIT Something you could consider doing is calculating how many spaces away in the alphabet one letter has to be from another to make changing directions more efficient than wrapping around the alphabet. (If you decide to do this, don't forget to calculate the blank spaces between Z and A.)
EDIT 2: Something else to factor in is the time required for die rolls. Sometimes the RNG manipulation to guarantee a 6 roll can take quite a few frames. This can be affected by changing when the 'Start' button press happens to select "Regular Game."
So when the game mode is selected , the die roll RNG is affected. When the team name is selected determines word sequence.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
Memory wrote:
ruadath wrote:
Even I voted Meh on my own submission here.
At least you're honest...?
There's nothing wrong with finding your own work unentertaining.
Regardless of how much a TASer may try to inject entertainment through the TASing process, some games simply don't offer much in terms of entertainment value.
Some of us also feel that the archival of fastest-known runs regardless of entertainment value is a worthwhile endeavor and just as important as archival of entertaining runs.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
It's still hard to say. Your changes sound like they alter both timing and button presses, so either or both could still be contributing to what's altering the RNG.
I'm not familiar enough with the mechanics of the official MegaMan series to know how their RNG is affected, so the best I can do is speculate on a hacked ROM.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
Ultimatesaber27 wrote:
Here, you can see that it appears in the top screen. Well, for me, it appears in the Top-Right corner of the screen. Even losing some frames before entering the boss gate does nothing.
The RNG that determines his position may not be time related...meaning waiting any number of frames may not have an effect.
For example, the RNG instead may be based on any (or all) of the button presses up to that point in the game.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
warmCabin wrote:
Holy shit!!! Have you heard of the elseif keyword? I think you might like it :p
Nope....never knew of that until now. You're lightyears beyond me in terms of lua scripting (and probably programming in general). Thanks for the mini-lesson.
On a more positive note, my experimental WIP of the TAS currently uses the sequence starting with "CUP." Which means, I just need to clean it up a bit and I'll be done!
Curiosity question. Your script generated multiple sequences labeled "Best so far." Does it take into consideration how many frames you have to wait to initiate the other sequences besides the one starting with "CUP"?
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
Spikestuff wrote:
DrD2k9 wrote:
I can simply press start to not input a guess at a word and move onto the next word in the sequence.
Works in Alternative Mode. Not Regular.
Regular Mode will force you to get an answer right.
You're correct for progressing through the game...but I was referring to building a spreadsheet of answer sequences. If you give up trying to guess a word, the game will move on to the next word in the sequence.
I'm doing this as much for the RTA community as I am for TASing.
EDIT: For anyone interested, here is my (in-progress) spreadsheet of word sequences. I've listed the first 10 words in each sequence. https://docs.google.com/spreadsheets/d/12_ealGeGx_VdxQLL0J2xylnTWGoEjA7itqnwEeJF7wc/edit?usp=sharing
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
Wow...I just realized how much of an idiot I am. I can simply press start to not input a guess at a word and move onto the next word in the sequence.
Well that was a bunch of unnecessary time wasted.
Editor, Experienced Forum User, Judge, Published Author, Expert player
(2094)
Joined: 8/21/2016
Posts: 1029
Location: US
CUP and TEARDROP are the starting words for the two short sequences I've found thus far.
I'm not good enough at lua to completely automate the testing.
I've just been using one to display the answer on the screen and one to limit token movement to 1 space at a time to speed up the time between puzzles. I've been manually inputting the letters.
Here are my lua scripts:
--Display Answer
local x = 0
local y = 10
local i = 0
local l = 0
local lvalue = 0
function letter()
if lvalue==10 then
return ("A")
else
if lvalue==11 then
return ("B")
else
if lvalue==12 then
return ("C")
else
if lvalue==13 then
return ("D")
else
if lvalue==14 then
return ("E")
else
if lvalue==15 then
return ("F")
else
if lvalue==16 then
return ("G")
else
if lvalue==17 then
return ("H")
else
if lvalue==18 then
return ("I")
else
if lvalue==19 then
return ("J")
else
if lvalue==20 then
return ("K")
else
if lvalue==21 then
return ("L")
else
if lvalue==22 then
return ("M")
else
if lvalue==23 then
return ("N")
else
if lvalue==24 then
return ("O")
else
if lvalue==25 then
return ("P")
else
if lvalue==26 then
return ("Q")
else
if lvalue==27 then
return ("R")
else
if lvalue==28 then
return ("S")
else
if lvalue==29 then
return ("T")
else
if lvalue==30 then
return ("U")
else
if lvalue==31 then
return ("V")
else
if lvalue==32 then
return ("W")
else
if lvalue==33 then
return ("X")
else
if lvalue==34 then
return ("Y")
else
if lvalue==35 then
return ("Z")
else
return ("")
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
end
while true do
while i<16 do
i=i+1
l=l+1
x=x+15
lvalue= memory.readbyte (0x447+l)
gui.text(x,y,letter())
--gui.text(x,y,str);
end
emu.frameadvance();
x = 0
i = 0
l = 0
lvalue = 0
end
--limit token movement to 1 space
while true do
x=memory.readbyte(0x89)
if x<255>0 then
memory.writebyte(0x89,0)
else end
end
emu.frameadvance();
end
There may be much more efficient ways to do these, but my programming knowledge is so limited that this is the best I can come up with.
