Amaraticando, what are you working on and does it still go slowly at a simple 1x or 2x size?
I'm working on a Super Mario World script. In fact, if the size is 1,the slowdown is small. However, there's no way to show all the info in such a small resolution. In higher sizes,the slowdown is very noticeable.
Therefore, I'll have to accept that and include less functionality than the lsnes script... :(
It's dreadfully resource-intensive. I was averaging 55 fps at 4x size, so machine power is going to be a factor with this one.
It looks like it'd be very useful for shenanigans though, just not as much for general play.
EDIT: I think I fixed it but I'm not sure why this fixes it - for whatever reason it just fails at drawing images when you use event.onframestart(), but not when you use the old while true do loop method.
I don't know anything about this, having no real knowledge of Lua, so this may seem like a stupid comment, but...
OnFrameStart() would imply to me something that happens before the frame is drawn, such that if you draw the frame is going to draw over that afterwards. So you'll get the exact problem you're describing.
Not sure if there's any reality to that, but if I were to draw against someone's API I would expect to do it after they do, so they don't draw over me.
I tried it with OnFrameEnd() just now and it gives the same problems as OnFrameStart(). What's interesting is that it only happens to drawImage (and maybe drawIcon? I don't know if I tested that or not).
CrazyTerabyte wrote:
As I said, I only tried it once, discovered what happened, and never tried again. I don't even know what side-effects it has.
Adding gui.DrawNew("native") as the first line of main() in my zelda 2 script leads to this:
Amusing, but not expected. And not particularly helpful for replacement ui shenanigans, but probably good for popup messages and the like.
EDIT: And the duplication turns out to be because I didn't clear the lua session properly. So bluh there. Also using Native and not having the client window match the game resolution/aspect ratio is a good workaround if you need more space than you've got.
Not sure why that makes a difference, but it does. With that said, I think I'll just stick to making a generic lua folder outside of the bizhawk path since I don't like disabling my security software.
Isn't Mupen ridiculously outdated at this point?
Bizhawk's N64 core is an adaptation of Mupen, anyway, so switching over shouldn't be a hassle unless you're either lazy, stubborn, an idiot, or some combination of the three.
Do you even know why people don't bother to change to bizhawk?
TAS input is more clunky, savestates are slower, memory addresses are different and most games aren't more stable than on mupen (glitches still occur on both emulators). So idk why most people would bother.
Looks like most people that wishes N64 TASers change to Bizhawk haven't tried using the emulator lol.
I actually use BizHawk for basically anything that's supported. Outside of some graphical glitching that's probably been fixed by now, I haven't had any major problems with BizHawk running N64 games.
With all that said, the main reason people aren't bothering to change to bizhawk is because of a combination of how it has the same bad rep as Byuu's things as far as system requirements, and the fact that people are stubborn and the older emulators just have better exposure. Hell, I watched a tas the other day that was made in ZSNES of a SMW level that was made within the last 12 months, and the author of the run admitted it's not optimized.
Update time, Zelda II flavored.
edit: updated code at bottom
Right now the problem I'm having is that it won't draw the key graphic at all, regardless of if I'm trying to drawImage or drawIcon. The image file is in the same folder as the script, so I'm probably missing something obvious that I forgot.
There should be two keys being drawn, and I can't see it if it is.
EDIT: I think I fixed it but I'm not sure why this fixes it - for whatever reason it just fails at drawing images when you use event.onframestart(), but not when you use the old while true do loop method.
Language: lua
-- "Zelda II: The Adventure of Link"
local bagCounter = 0
local currentMP = 0
local currentHP = 0
local maxMP = 0
local maxHP = 0
local hearts = 0
local jars = 0
local armor = 0
local intel = 0
local skill = 0
local xp = 0
local toLevel = 1
local percentToLevel = 1
local keys = 0
local selector = 0
local magic = {[0]="No Magic","Shield","Jump","Life","Fairy","Fire","Reflect","Spell","Thunder"}
function main()
bagCounter = memory.readbyte(0x05DF);
currentMP = memory.readbyte(0x0773);
currentHP = memory.readbyte(0x0774);
hearts = memory.readbyte(0x0784);
jars = memory.readbyte(0x0783);
armor = memory.readbyte(0x0779);
intel = memory.readbyte(0x0778);
skill = memory.readbyte(0x0777);
xp = (memory.readbyte(0x0775)*255)+memory.readbyte(0x0776);
toLevel = (memory.readbyte(0x0770)*255)+memory.readbyte(0x0771);
keys = memory.readbyte(0x0793);
selector = memory.readbyte(0x0749) + 1;
percentToLevel = xp / toLevel;
maxMP = (jars*32)-1;
maxHP = (hearts*32)-1;
gui.drawBox(0,0,255,41,"White","Black");
gui.drawText(0,8,armor.."|HP: "..currentHP.."/"..maxHP,"Red",11);
gui.drawText(0,18,intel.."|MP: "..currentMP.."/"..maxMP,"White",11);
gui.drawText(0,28,skill.."|"..bagCounter.." enemies since last drop","Yellow",11);
for i = 1, keys, 1 do
gui.drawImage("key.png",95+(i*8),8);
end
gui.drawText(95,18,"Spell: "..magic[selector],"White",11);
gui.drawRectangle(245,8,10,(32*percentToLevel),"Green","Green");
gui.drawBox(245,8,255,40,"White");
end
while true do
main()
emu.frameadvance()
end
editwo: For drawText, how do I fiddle with the font family - am I restricted to certain fonts or can I use any installed font?
Okay, I've given the game a fair shake and the design is more focused around air dashing than the TAS shows (the drawbacks of going ultrasuperfast in a mod that's already been cranked up from the original), so I am officially rescinding my argument that the design didn't account for it.
What makes Hacks different is because we would get swamped with bad hack submissions if we didn't have a Notability Rule, which would crowd out both good hack submissions as well as submissions for official games.
Or... you can just accept good hacks? The amount of fallacies and unmentioned arguments used here is amazing.
Notable does not equate to good, not by a long shot.
I myself voted yes as far as the run; my main concern is that the hack itself is suspect and that my yes vote will get vetoed by the judge when the time comes.
So votes don't actually matter?
One for the judges, that. My understanding was that the voting was primarily used to separate moons from vaults.
It does not help that Air Sliding appears to have been a late addition to the hack.
Because we don't have runs that prove Super Mario Bros. is beatable without running, or Mega Man X without dashing, nope...
Apples and oranges. This TAS extensively uses the Air Sliding mechanic, which as I've said in... basically every post I've made on this run does not appear to have been planned for in the design. This scenario is basically the exact opposite of the ones you cited.
Garrison, you keep bringing up that most of us haven't played the game and are condemning it because it's a hack that doesn't bring much to the table on the surface.
That's because the people who have played this are in the minority. A significant minority. Most of us won't be exposed to the gameplay of this hack outside of encodes, which is true for almost all submissions to this site anyway. What makes Hacks different is because we would get swamped with bad hack submissions if we didn't have a Notability Rule, which would crowd out both good hack submissions as well as submissions for official games.
Do I think this hack is publishable? Yes. Do I think it's publishable alongside R4MI and Rockman no Constancy? I don't know - I haven't played any of the three, and R4MI is the only one I've seen played outside of a TAS. I think that for most of the community, this will be the case, so arguing from the standpoint of someone who has a vested interest in seeing this hack published isn't going to do this game any favors. We need impartial voices voting on hacks, not just those few of us who want to see more movies of mods on the site. I'd love to see A Super Mario Thing on the site, but that's impossibru as ASMT uses ASM that breaks actual hardware, let alone decent emulators, and I doubt even a fifth of the TAS community has played it given - surprise - it's a niche hack.
As for your concerns that this won't be published... worst case scenario, either flat rejection or gruefood delight. If it's rejected because of play issues, make your own and submit that. If it's rejected because of bad hack choice, petition to have it added to Gruefood Delight. I know I will; this is a blast to watch.
Which brings me back to the poll results. Just looking at the Poll Results, this is getting a strong positive as far as entertainment. I myself voted yes as far as the run; my main concern is that the hack itself is suspect and that my yes vote will get vetoed by the judge when the time comes. It does not help that Air Sliding appears to have been a late addition to the hack.
In that imgur set, I counted exactly three spots where Air Sliding is possible and Rush Jet is not, necessitating it.
Garrison wrote:
Good music, good level design, good mechanics, good weapon changes, good robot master changes, good tricks etc.
R4MI also brings these to the table, and as it's considerably far more well known there would be no issue of whether or not it's worth publishing were no movie already published. Rockman no Constancy is in the same boat. This hack is not well known, in fact I had heard nothing of this nor the predecessor hacks for 3 and 5 until I saw this thread. As notability is one of the acceptance factors pertaining to hacks that doesn't apply to raw games, that needs to be taken into account, and it's going to be the saving grace when this one goes to judging as the Air Sliding does not look like it was deliberately designed for.
Instead of being vitriolic, perhaps you could answer the questions here.
1) Was everything sped up in the hack overall?
2) How many places in the game absolutely require the Air Dash, as opposed to simply it being a more convenient answer to the problem?
2a)Can the game be completed without using the Air Dash even once?
2ab)If not, where is the Air Dash required for progress (screencap please)
3) Why this hack specifically?
The reasons for the questions:
1) Watching the YT encode, it looked like everything was sped up, not just Rockman.
2) As I stated in my previous post, the stages do not appear to have taken the Air Dash into consideration when they were designed, which leads me to believe that the Air Dash was added late in "development".
3) This is the question everybody wants an answer to but is afraid to ask.
Because of a specific fanfictions called Those Lacking Spines, in which the canon universe is literally under assault by bad writing, prompting Vexen, Xaldin, and Lexaeus to go save their comrades. :)
Caught myself cheering for Vexen before I remembered that this is the Canon Universe :p
Question: Is Omnislash really the fastest available method for taking down Riku III? It seems a bit... slow, I guess.
For the poll option, I'm voting yes on the grounds that "yes, I found this run to be entertaining." Especially with the playback speed controls on the YT encode. Ultra Super Fast Mode 2 Turbo, The Game: The Movie: The Game.
However, I do have some reservations.
1) As has been stated previously, the Air Sliding feels like an unnecessary addition to the game. Yes, that it is in the game allows you to refill RM weapons during Cossack 2 with almost depressing ease (I'm sure Wily would agree there), BUT I can count on one hand the number of cases where the Air Slide was even considered when the stage was designed, and none of those were in the RM stages.
2) Was it just me, or was everything unnaturally fast in the TAS as opposed to "just" Rock? Was this a documented increase in speed all around?
3) This is not going to be an easy sell to laypersons for the simple fact that it's not R4MI. I like it because it is accessible to Laypersons, but R4MI is established and - let's face it - more well known. This feels more like a niche hack showing off the hacker's ability to hybridize two other hacks' gimmicks, and as mentioned in point 1 is shows in the stage design.
I'm just messing around with causal plays of BN2 and 3 at the moment, but does anyone have the memory locations (or at least values to look for) for tracking style change progress?
E: I've got a lead in Combined Wram 0x2d60 through 0x2d70, but I got a rewind-induced crash before I could discern if I was barking up the wrong tree or not. Got a bizarrely-lucky-for-me virus battle out of it that I can use to really pin down Guts and Custom points in battle and rank growth post-battle. Too bad a new game is really hard to get Team and Shield points/ranks for.
Anti-depressants have different effects for different people. They helped me but in an unexpected way because of an underlying physiological problem that responds to them, but I have a friend who doesn't respond to them at all.
The bullying may or may not be as much of a red herring as you'd think in this day and age. Doxxing is a real threat that leads to real-world consequences, and it's not too hard for someone intent on finding the information to get and use it even in the absence of a dox. If the bullying spilled out of cyberspace and into her daily reality...
The main argument for the "no hacks" rule is that it's the same performance in the same old way. Most hacks don't fundamentally screw with the underlying engine enough to stand as a separate game (see EMB), and many hacks are what can generously be called "YouTube Levels," designed specifically to provoke an antagonistic reaction out of the player in exchange for cheap Youtube hits.
This hack, while it doesn't fundamentally screw with the underlying game, does things that are interesting from a casual point of view, and that's where the judgment call has to be made. I think it's worth keeping at this and improving it specifically because it doesn't require such a technical mastery that's demanded by YouTube Levels to enjoy casually. In short, non-TASers can enjoy a run of this game just as much as a TASer can.
E: With all that said, does the death shortcut in 8-4 truly justify a second loop through 8-5 to get Fireflower? I think that's where this run is losing a fair chunk of entertainment and is throwing our evaluation of the run from a technical standpoint into confusion.
It very much is. I also got the impression that they wanted to do "This is why you have Applejack and Big Mac and Granny Smith" but were told by the execs "No heavy stuff" so they compromised with this.
The game seriously had a required stage where the intended route was blocked? That's a really dumb bug / lack of playtesting, like that one stage in Battletoads that Player 2 can't finish.
That's highly debatable. It's not like the alternate exit he uses is some glitch. It's a valid exit coded into the game. The 'normal' exit requires traversing a gigantic maze, collecting a bunch of key cards, and then when all the doors are opened, the door itself is blocked by spikes. So if you could get to that area, there's no way to get through the door.
Here's a map.
http://www.shikadi.net/keenwiki/File:Ck3lv15.png
What's blocking the area itself is the 1 block high gap on the bottom left, but the spikes past the doors make it impossible as well. It's very likely that the 1 block gap is a design oversight, but it's hard to call the spikes an oversight. It's more likely just supposed to be frustrating (there are also next to no bonus items in the maze while the rest of the game is full of them) and force the player to think outside the box and go around the maze instead of through it. You could also see that it was impossible to get through the exit from the alcove above it, which was accessible.
That said, Keen 3 was full of bugs. Half of another stage was fully designed but there was no way to reach it.
Even ignoring the SGA, the point items in the stage are all the "rare" top-tier 5k-apiece items. There's no way this stage wasn't deliberately made like this. Especially with the cache of ammo near the achievable exit.
Maybe I'm missing something here, but are there no mushrooms or fireflowers in the rest of the game? Was it truly necessary to run through the final level twice to grab a flower?
1-1 and 6-2 both have powerups that are revealed but ignored.
The notes don't even touch on the powerup situation aside from "Oh right you need fireflower to kill Bowser in this", so I would second getting an author's explanation as to why the second loop through 8-5 instead of getting Fireflower during the bulk of the run.
As far as the goals go, I am fine with the one use of Arrow Gas. That was clever, actually. It feels cowardly, in a sense, to have Darkwing Duck try to get out of the action and climb over a wall to avoid having to fight an enemy ;-)
I know the Arrow thing came up in the previous pacifist submission and it was decided that since it's a choice between using the arrow or killing an enemy, Pacifism beat out Not Using Gas.
Amusing, but not expected. And not particularly helpful for replacement ui shenanigans, but probably good for popup messages and the like.
EDIT: And the duplication turns out to be because I didn't clear the lua session properly. So bluh there. Also using Native and not having the client window match the game resolution/aspect ratio is a good workaround if you need more space than you've got.
BizHawk 1.9.4 on Windows 8.1, bizhawk directory is C:\Games\BizHawk
