No, Just my suggestion. Because I want to show the hiding ending in Contra(J), It doesn't mind if TASVideos don't use it.
I don't know the old FCEU has bug in it, and I used it because it is chinese version.
On the other hand, I created this file in two years ago, I fix it with FCEU .15 at nearest time. So I continue to use this emulator.
You - instead of us - should decide whether to replace the submitted movie with one beginning from emulator power-on. However, keep in mind that we don't accept a movie beginning from a savestate unless there is a good reason.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
What do you mean exactly? To replace the submission with this longer movie?
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
That was amazing! Nice tricks used. I got a question. Will your run use the "CALL" option to avoid fights? Or is it too slow?
Thanks. I would probably use the CALL command twice later in the game to avoid unnecessay fights. The great thing about it is that the free control given by it is limited by time (630 frames) rather than steps, meaning that with Pulley activated I can slip through quite a few rooms.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
I am not surprised with this result because the main point of this movie was to perform lag reduction on a very laggy game. (And now, has the movie lost its point?)
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
I saved a whole 32 seconds from my old WIP!
Note: According to my strategy, Asuka will be the final team leader who reaches Level 16 first, hence she has to receive EXP from every single battle (except when everyone escapes). Emi and Kazuo may have less EXP.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
One more, thing, since ARMOURs don't randomly appear, wouldn't it be faster in the longer run to fight MAN every 10 steps then to reload the room every 2 fights?
It is a nightmare to manipulate every encounter step number to be a small one (10-15) AFAIK. The game seems to incline to "balance" encounters by rolling out more great numbers (70-74) after you have got several small numbers in a row. Does it reset this behavior if you exit and reenter the same room? I haven't tested this. Anyway, I planned to spread the encounters to many areas to relieve the stress of luck manipulation as well as the audience's feeling of repetition.
EDIT: The Lua script that I used (and revised for presentation :p):
-- Sweet Home TAS GUI --
-- updated in April, 2011 --
-- by klmz --
-- Note: This is a whole mess. Don't use it as a demonstration of Lua scripting! --
--require("auxlib");
-- please ignore this thing and use the built-in alternative instead
local function IPairs(t)
local i = 0;
local n = table.getn(t);
return function ()
i = i + 1;
if i <= n then
return i, t[i];
end
end
end
local function PosToString(pos)
return string.format("%02x%02x.%02x", pos.high, pos.mid, pos.low);
end
local function ReadI8(byte_address)
return memory.readbytesigned(byte_address);
end
local function ReadI16(low_byte_address)
return memory.readbyteunsigned(low_byte_address) + memory.readbytesigned(low_byte_address + 5) * 256;
end
local function ReadPosHML(high_byte_address)
local pos = {};
pos.high = memory.readbyteunsigned(high_byte_address);
pos.mid = memory.readbyteunsigned(high_byte_address + 0x001B);
pos.low = memory.readbyteunsigned(high_byte_address + 0x0036);
return pos;
end
local function ReadPosIF(high_byte_address)
local high = memory.readbyteunsigned(high_byte_address);
local mid = memory.readbyteunsigned(high_byte_address + 0x001B);
local low = memory.readbyteunsigned(high_byte_address + 0x0036);
return { I = high * 256 + mid, F = low };
end
local emu = FCEU;
local keys = {};
local last_keys = {};
local function KeyPressed(key)
return keys[key] and not last_keys[key];
end
local text_color = "white"
local info_color = "yellow"
local warn_color = "#FF3F3F"
local fg_color = "white";
local bg_color = "black";
local osd_inventory_on = false;
local osd_details_on = false;
local offsets = {0, 3, 4};
local team = {{}, {}, {}};
local details = {{}, {}, {}};
local pos = {{}, {}, {}};
local inventory = {{}, {}, {}};
local colors = {"#DFDF3F", "#4F4FFF", "#FF3F3F"};
local directions = {
"-", "L", "D", "U", "R"
};
local stati = {
"normal", "poisoned", "deadly poisoned", "cursed", "paralyzed", "scared"
};
local item_list = {
-- special items
"----", "Camera", "Vaccum", "Key", "Medicine", "Lighter",
-- replacements
"Polaroid", "Broom", "Wire", "Cookies", "Matches",
-- 0x0B
"Shabby Board", "Sturdy Board", "Rope", "Torch", "Light",
-- 0x10, 4 in each row from now on
"Hammer", "Safety Footwear", "Bowgun", "Bucket",
"Pickaxe", "Log", "Iron Bar", "Shovel",
"Crystal Ball", "Jade Ring", "Two Keys", "Golden Key",
"Basement Key", "Knife", "Magnet", "Pulley",
"Empty Box", "Clay Figurine", "Iron Key", "Spear From Armor",
"Slide", "Gasoline", "Candle", "Rope Ladder",
"Diary Key", "Gloves", "Axe", "Child's Coffin",
"Extinguisher", "Photo", "Keepsake Dress", "Blue Light",
"Ruby Ring", "Blue Candle", "Potion Bottle", "Diary",
-- 0x34, weapons
-- knives
"Fruit Knife", "Flick Knife", "Shiny Knife", "Silver Knife",
-- spears
"Spear", "Worn Spear", "Shining pear", "Amulet Spear",
-- axes
"Amulet Axe", "Heavy Axe", "Shiny Axe", "Silver Axe",
-- swords
"Sword", "Long Sword", "Silver Sword", "Holy Sword",
-- other weapons
"Wooden Sword", "Club", "Fork", "Trident"
};
local enemy_list = {
"Spiritual", "Cursed Doll", "Bats", "Grubs",
"Skull Snakes", "Mudman", "Half-Body Man", "Mirror",
"Wall Face", "Mad Dog", "Man", "Copper Armor",
"Machete Man", "Zombie", "Reaper", "Inverted Zombie",
"Skeleton", "Man Turned Around", "Fiery Ball", "Skull Ghost",
"Decomposed Corpse", "Wolf", "Silver Armor", "Maniac",
"Kazuo's Doppelganger", "Akiko's Doppelganger",
"Taguchi's Doppelganger", "Asuka's Doppelganger",
"Emi's Doppelganger", "Madam Mamiya", "Madam Mamiya 2nd Form",
"Ultimate Glitch Monster"
};
gui.transparency(1);
local function Everything()
keys = input.read();
if KeyPressed("I") then
osd_inventory_on = not osd_inventory_on;
end
if KeyPressed("U") then
osd_details_on = not osd_details_on;
end
for i, u in IPairs(offsets) do
pos[i].X = ReadPosIF(0x0C51 + u);
pos[i].Y = ReadPosIF(0x0CA2 + u);
pos[i].FACE = ReadI8(0x0C55 + u);
team[i].HP = ReadI16(0x6194 + u) + 1;
team[i].PP = ReadI16(0x61C6 + u) + 1;
team[i].EXP = ReadI16(0x61DA + u);
if osd_inventory_on then
inventory[i].ITEM_S = ReadI8(0x6255 + u);
inventory[i].ITEM_1 = ReadI8(0x625A + u);
inventory[i].ITEM_2 = ReadI8(0x625F + u);
inventory[i].WEAPON = ReadI8(0x6264 + u);
inventory[i].USED_S = ReadI8(0x6269 + u);
inventory[i].USED_1 = ReadI8(0x626E + u);
inventory[i].USED_2 = ReadI8(0x6273 + u);
end
if osd_details_on then
details[i].LV = ReadI8(0x61A8 + u) + 1;
details[i].UP_EXP = ReadI16(0x61D0 + u);
--details[i].ATK = ReadI16(0x61BC + u) + 1;
--details[i].MAX_HP = ReadI16(0x619E + u) + 1;
--details[i].MAX_PP = ReadI16(0x61B2 + u) + 1;
details[i].STATUS = ReadI8(0x6180 + u);
details[i].WHERE = ReadI8(0x61FD + u);
--details[i].IN_PIT = ReadI8(0x6296 + u);
end
end
for i, v in IPairs(team) do
local last_len = 0;
local gui_string = string.format("X: %03X.%02X", pos[i].X.I, pos[i].X.F);
gui.text(last_len, (i - 1) * 7, gui_string, colors[i], bg_color);
last_len = last_len + string.len(gui_string) * 6;
gui_string = string.format("Y: %03X.%02X", pos[i].Y.I, pos[i].Y.F);
gui.text(last_len, (i - 1) * 7, gui_string, colors[i], bg_color);
last_len = last_len + string.len(gui_string) * 6;
for j, w in pairs(v) do
gui_string = string.format("%s: %3d", j, w);
gui.text(last_len, (i - 1) * 7, gui_string, colors[i], bg_color);
last_len = last_len + string.len(gui_string) * 6 + 4;
end
end
timer = memory.readbyteunsigned(0x000A);
gui.text(200, 201, string.format("timer: %d", timer), text_color, bg_color);
stage = memory.readbytesigned(0x617A);
battle_screen = memory.readbytesigned(0x030F);
-- details
if osd_details_on then
for i = 1, 3 do
local lv = details[i].LV;
local up_exp = details[i].UP_EXP;
local status = details[i].STATUS + 1;
local where = details[i].WHERE;
gui.text(i * 84 - 72, 32, string.format("LV: %2d", lv), colors[i], bg_color);
gui.text(i * 84 - 72, 40, string.format("UP EXP: %4d", up_exp), colors[i], bg_color);
gui.text(i * 84 - 72, 48, stati[status], colors[i], bg_color);
gui.text(i * 84 - 72, 56, string.format("Room: %2d", where), colors[i], bg_color);
end
end
-- inventory
if osd_inventory_on then
for i = 1, 3 do
local item_s = inventory[i].ITEM_S + 1;
local item_1 = inventory[i].ITEM_1 + 1;
local item_2 = inventory[i].ITEM_2 + 1;
local weapon = inventory[i].WEAPON + 1;
gui.text(i * 84 - 72, 120, item_list[item_s], colors[i], bg_color);
gui.text(i * 84 - 72, 128, item_list[item_1], colors[i], bg_color);
gui.text(i * 84 - 72, 136, item_list[item_2], colors[i], bg_color);
gui.text(i * 84 - 72, 144, item_list[weapon], colors[i], bg_color);
end
end
-- battle info
if battle_screen == 1 then
local enemy_id = memory.readbyteunsigned(0x624E) + 1;
local enemy_name = enemy_list[enemy_id];
if (stage == 11 or stage == 6 or stage == 7) then
gui.text(130 - string.len(enemy_name) * 2.6, 48, enemy_name, info_color, bg_color);
foe_hp = memory.readbytesigned(0x6250) * 256 + memory.readbyteunsigned(0x6252) + 1;
gui.text(110, 60, string.format("HP: %3d", foe_hp), text_color, bg_color);
else
gui.text(107, 72, "WARNING!", warn_color, bg_color);
gui.text(130 - string.len(enemy_name) * 2.6, 80, enemy_name, warn_color, bg_color);
end
end
battle_members = memory.readbyteunsigned(0x622A);
if battle_members > 0 and (stage == 11 or stage == 6 or stage == 7) then
call_timer = memory.readbytesigned(0x0EE3) * 256 + memory.readbyteunsigned(0x0EE2);
if call_timer > 0 then
gui.text(200, 209, string.format("call: %d", call_timer), warn_color, bg_color);
else
gui.text(222, 209, "fight!", warn_color, bg_color);
end
else
if battle_screen == 0 then
steps = memory.readbytesigned(0x617D);
gui.text(200, 209, string.format("steps: %d", steps), info_color, bg_color);
end
end
last_keys = keys;
end
while true do
gui.register(Everything);
emu.frameadvance();
end
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Btw, what's the best enemy to train on? In my test run, I trained on *"MAN", "MANIAC", "SKELETON, and "GHOUL(2)"*. What do you think?
They are good choices as long as you can finish them off quickly enough (that is to say, in one round). My original plan was to grind mainly on Copper Armors (easy encounter, use Rope), Decomposed Corpses (use Iron Bar), Skeletons (they are just cheap), Silver Knights (easy encounter) and Inverted Zombies (use Blue Candle, but maybe still too tough) though.
jlun2 wrote:
Lastly, there are some items that I didn't take but could be useful. They are: "Dress", "Pulley", "Broom", and maybe some more. I was worried about inventory space at the end, so I left them. Since you took the "Broom", do you think the other 2 items will be useful?
Pulley requires re-activating if you have activated a different team-wise item like Pickaxe or unite the character carrying Pulley as one without it. But I think it is still worth getting unless you figure out an overall superior inventory strategy that has to take up all six slots. With regards to Dress, it depends on when you are going to spend plenty of PP.
However, jlun2 wrote:
Can we work together on the run?
Sorry, but no, I'm afraid not. I've spent 2 years yearning for and planning on this run, and been believing that this is one challenge that I want to take by myself. I guess I may have also been tired of ending up staying behind the screen, assisting others to make theirs own and then asking for their thanks. I appeciate your offer, nevertheless.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Please write my nickname as klmz instead.
jlun2 wrote:
Why? Because in the good dump, I get killed off for absolutely no reason what so ever; my team members just vanish off the map, and I get a gameover screen in like less than a minute. =p
It appears to me that you were describing what happens with either the [b1] or [b2] dump. Were you using an out-dated iNes auditing tool? I think GoodNES 3.14 is able to give the correct result.
jlun2 wrote:
Edit: Are you sure you used the good rom? Your WIP synced fine in the [B1] version.
Could it be the moving white dots in the battle scenes that made some auditing tool authors to think that it was due to bad dumping? But I believe they also presented in the original images. Well, I don't recall all of the details on this matter.
jlun2 wrote:
Edit 2: Can you explain the Wip I just watched please? Like your enemy choice?
The principle was to minimize the time spent in fighting enemies, as well as that in back-tracking - provided that I could have enough EXP whenever I would be at a puzzle requiring using Mind/Pray Power (abbr. PP :p). Since fighting extra turns would consume a considerable time (1-2 seconds for every team member's individual turn, and over 4.3 seconds for every extra whole round), and all enemies available before the room with two Copper Armors (my own translation to their original Japanese names, might be inaccurate though) all gave too few EXP except for Mad Dogs, I chose to fight the fastest ones getting help from using the effective Hammer on Mad Dogs. However, I shouldn't have bothered with accessing the small room where I acquired the first Log and Rope....
And FYI, the items effective on enemies (remarkable ones in bold):
* Ax: Man, Man Turned Around
* Blue Candle (massive damage!): Zombie, Inveted Zombie
* Blue Light (not sure whether this is useful since there's no enough room for it): Zombie, Inverted Zombie (I think there is a hint about this at the lake), Reaper, Skull Ghost
* Bucket/Extinguisher: Spiritual (artesy name), Fiery Fireball (straight name)
* Camera/Polaroid: Bats, Reaper, Skull Ghost
* Candle: Grubs
* Clay Figurine/Crystal Ball: EVERYTHING!
* Gunbow: Mad Dog, Wolf
* Hammer (weak but early): Mirror, Wall Face, Mad Dog, Wolf, Skeleton
* Iron Bar: Half-Body Man, Mudman, Decomposited Corpse
* Lighter/Matches (very weak but early): Grubs, Skull Snakes
* Log: Half-Body Man
* Rope/Rope Ladder: Copper Armor, Silver Armor (Wondering how? Just watch Star Wars Episode V!)
* Shabby Board/Sturdy Board: Cursed Doll, Mudman, Decomposited Corpse
The only two enemies with no weakness to items besides Clay Figurine and Crystal Ball are the two machete freaks.
These items are mostly useful in the early stages. They become too weak to be worth the precious inventory spaces as the enemies get tougher later on and you can use more PP against them.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Is it possible to avoid using the reset glitch? From the viewpoint of the game, you hadn't beaten all of the first 7 levels before you used the reset glitch.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
The ROM checksum of your movies seems to refer to the good dump, not [b1] that has broken graphics.
You seem to be missing a lot of tricks in this run. I had a WIP that was faster although it had some major flaws at the beginning.
And I think now I have gained the motivation to complete the run again.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Theres no improvements working, and yeah i wanted to get 1796M ... that aint be long anymore now.
The point to delay was more about avoiding somebody accidentally publish it.
You may try to ask a publisher to do that, maybe by putting a significant note at the end of your submission text.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
I haven't heard from the author about anything of this run recently.
There has to be a deadline for this "improvement". The author actually also wanted to delay this submission for a "good" publication number, but it would be a better solution for the author to cancel this submission.
I'll set the deadline soon. That should have been done earlier.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
By the way, I tried both with VBA21 and a newer version, but attempting to encode the .vbm always resulted in a crash at the first Thieves Guild corridor. Some sort of glitch in VBA? So, that section was omitted from the encoding here.
It didn't crash for me, but I didn't try to encode from the start of the vbm. If the crash only occurs that way, the cause of the crash might be the inconsistence in frame timing when reset is involved. A workaround for now would be to split the encodes at the reset points...
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
That doesn't grants a new category given that we have got already 3 categories for this not-so-popular game. While it provides different watch-experience to the audience, it is just alike to the obsoleted movie by JXQ.
More specifically, we traditionally add a category only when the desired criteria cannot be made compatible with the traditional categories like "all bosses". For this movie, it falls under exactly the same category as the old movie by JXQ was under, the category which we decided to obsolete, as we consider an "all bosses" or "100% relics (& spell books)" category should be capable of providing most of that the obsoleted category be capable of and even more appreciated by the audience.
However, I do have realized that we have made a few controvertible precedents. Take the case of movies of Castlevania: Aria of Sorrow ever published for example and comparison:
* The last "out-of-bounds glitch" movie that didn't make use of the death-warp glitch differentiated much from but was obsoleted by the first death-warped movie both under the same "glitched" category;
* On the contrary, not only did the oldest "less-glitched" movie by Atma remain published along with the "100% souls" movie that provided richer contents, but also it was improved with a movie under exactly the same still-standing "any%" category whose naming is inaccurate and misleading.
Another precedent: the published "maxim no warps" run controverted with actually used glitch-warps and wall-zipping has merely a merit: it defeats "all bosses".
Actually, I consider these controvertible precedents as the misdirectors in this case.
We have this:
Judge Guidelines wrote:
Be fair
A judge has the greatest control over the content of this website in the long run. All judges must act towards the goal of having an encouraging and rewarding atmosphere for both the players and the audience. You must be fair towards both.
It is fairness towards the audience when judges disqualify worse submissions and qualify the better ones.
* Too many bad publications turn the audience away.
* Too few publications turn the audience away (and possibly the players too).
The players must have chances of getting their movie published.
We'll final decision on this submission see in a few days.
EDIT: Better format.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
It could have been made easier to judge if the author had chosen a "100% relics" or "all bosses" goal instead. However, there have been several other "in bounds" movies (mostly of Castlevania games) published (on the "main" branches, though) in the same vein. So this category is not unacceptable. While a few technical flaws, mostly during boss fights, can be spotted in the movie, the overall quality is good for a new-comer. I'll wait a few more days to see if there's more interest in this movie.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
I think "no warps" would be inaccurate as this movie used both ingame and glitched warps. Maybe "in-bounds" or "in bounds" to be concise?
EDIT: oops.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
I would like to listen to the opinions (especially on the goal choices) of the audience as well as the author. I would consider it "premature" if the decision were to be made with so little information.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
I voted Yes because this movie is faster than the previous one. I'd have voted Meh had it been slower.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
The statistics in the Wiki: JudgeCounts module are inaccurate. If a judge's decision is made with empty Edit summary, it is not counted.
EDIT: Alright, it's a different bug.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Yes vote for your decision of taking damage to save time, your polished movie and your remembering to thank me.
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
I will be happy to do a no death version.But I still want to make a comparation for you at that time.
I hope you can help me make a decision.
Your opinions are uncertain before the comparation appear.That is what I want to show.
No way. Just adjust the settings at the beginning of the movie to have 4 lives per coin and die 3 times to skip the three trains without having to feed coins mid-game!
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
Yeah
I changed my tactics in the last stage. Because I ignore that piking up S must wait it turn to the pistal. So it can get the 30 S and keep it for the last boss. For most people who like no death,I take it in this one.Although level 5 was slower. It's faster than my past version.
I am going to vote No because you forgot to thank me for my strategical suggestions on the last two missions! [/unseriously]
<klmz> it reminds me of that people used to keep quoting adelikat's IRC statements in the old good days
<adelikat> no doubt
<adelikat> klmz, they still do
