Oh lord the Alliance Skip. Talk about eye candy a game-breaking bug - though you have to be crazy, awesome, or crazy awesome to make it work anyway.
I can look into the PS2 version but somehow I get the feeling that it's going to fix said bugs.
E: The PS2 version looks more like a PSX game than a PS2 one, which makes sense given MM8's original release timing. The impression I get is that it was originally developed to be ported to the PSX but then was changed to a PS2 port once they saw when the release date for that would be. Without updating the assets significantly - the music and sounds are pretty much identical and I know the intro smk is the same. It also looks slow to navigate menus but that's more because I'm used to mouse speed than any d-pad imitations. If the pointer can even be used in walking mode; if it can't then there's no alliance skip.
EE: I've found a copy on ebay for roughly $40. Should get here by the time SGDQ is over.
Hmm... the impression I get is that there's a code interaction here that I can't see because it relies on code I... uhh, can't see. Try some rubber duck debugging when you get a chance.
Hmm... you're testing isSingleMovieLoaded for the false block but you're testing movie.isLoaded for the true block. Why the difference? Still a dumb, you're making sure it's just one movie before you start trying to play it back.
Does BizHawk support fonts installed to the system at all? I want to say there are some outlined fonts that come with Windows (haven't had a Mac ever and I don't have a Linux station at the moment).
Solarplex: Why is this two while blocks instead of a nested if true do while? As written if it comes up false you're getting an infinite loop. Oh you're probably iterating through until it does become true. But couldn't that be done using an if-else statement nested inside a big while loop instead of two separate while loops? Nope I'm still a dumb. When do you call the playback for the movie file?
Add it to http://tasvideos.org/Bizhawk/Features.html - reads ISO, Bin/Cue and ECM/Cue files without outside mounting. I don't like other formats (and didn't know ECM was a thing until very recently) so I don't know what all's supported.
Ah crap, you're right. How I wrote it would break without iterating if it ever comes up false, which isn't the desired behavior. This is what I get for writing code without testing it.
Samurai Goroh: That first one's not an invalid test - if any of the entries are a 2 (except for entry 7), then it can't succeed, and at least one entry that isn't entry 2 has to be a 7 for it to succeed.
jlun2: That second one is actually one if easier than the first one since you're not jumping around in the array - it can be done with just one function.
--this code doesn't actually work, by the way, or at the very least doesn't work as intended.
function test_is_present(array array, int target) as Boolean
for each i in array,
if (readbyte(array[i]) == target) then
return true
else
return false
end if
end for
end function
if ((test_is_present(lift, 1) == true) and (test_is_present(lift, 6) == true)) then
magic()
end if
and you just flip a true to a false in the if around magic if you need the tested value to not be there.
You probably should section those tests off into their own functions and test iteratively since you're going through 1-6 in order. Pseudocode below
E: Looking back you'll also need to embed another if statement into the loops so that you're skipping over lift[2] when you test for 7, so that'll look like
function test_for_not_2() as boolean
for i, 1, 6
if (readbyte(lift[i]) ~=2)
return true
else
return false
end if
end for
end function
function test_for_7() as boolean
for i, 1, 7
if i ~= 2
if (readbyte(lift[i]) == 7)
return true
else
return false
end if
end if
end for
end function
if ((test_for_not2() == true) and (test_for_7() == true) then
do a thing
end if
This should stop each test as soon as it finds a true, but from what I'm seeing you just need one of each anyway.
E: does Lua do type declarations for functions? I forget.
I'm working on converting Xkeeper's Link's Awakening lua script (and by extension the X-functions resource script) for use with BizHawk, but my ignorance is showing - the version of X-functions I'm forking off of has readword in it, what size does/should that correspond with?
E: Never mind; what's in X-functions is just a 2-byte read from what I'm seeing, it's easy enough to replace that out.
Okay new problem - I'm getting an exception when I try to run the script that points to line 29 of the Link's Awakening script, but I don't see how evaluating a boolean in an if statement corresponds to "invalid arguments to method call" - does it think nohotkeys is a method?
(It's line 30 in what I'm working off of since I inserted memory.usememorydomain("System Bus") as line 2)
Those won't work, they're too close to the community to be sufficiently objective for Wikipedia's purposes.
Ready Steady Yeti, this is more of a question for the Wikipedia community than it is for our own. They'll be able to talk you through the process of finding and vetting acceptable references. Lord knows TVTropes has been around forever and their Wikipedia page got deleted more than once for lack of acceptable references.
I'm okay with that interpretation, but as Aqfaq filled the bounty it's her decision as to how the money gets applied.
As for how the page is used, that's up to the editors. I just wanted the page to finally be a reality after five years of disorganized and lost-in-the-thread bounties that still need to be sifted through.
$20 to the creation of a wiki page for outstanding bounties, to be paid as a donation to tasvideos.org (or to a reasonably acceptable substitute as defined by the fulfilling users) within 48 hours implementation.
Memory addresses in BizHawk are relative to the start of the domain (unlike basically every other emulator out there).
Say your address in Visual Boy Advance is 0x9234. Bizhawk sees that as being GB WRAM, which has a root offset of 0x8000 - to get the final address that BizHawk can manipulate, you have to remove 0x8000 from the VBA address. That would give you GB WRAM 0x1234. If the address you're trying to work with isn't in the default domain, you'll need to tell the code to change domains. My earlier RNG Strobe has an example of doing that.
It looks like the address you're after is going to be HRAM 0x18, so you'll need to do this:
Language: lua
local HRAM = "HRAM"
function gameInfo()
memory.usememorydomain(HRAM)
xcord = memory.readbyte(0x18)
gui.drawText(5, 5, "X: " .. xcord)
end
while true do
gameInfo()
emu.frameadvance()
end
Out of curiosity, why Pearl instead of Platinum?
E: Masterjun, that's a bit of a stretch if it's deliberate. I don't really care one way or the other, though maybe call it "Speed/Entertainment Tradeoff" since it's for the pun?
The Void is unwalkable from Platinum onwards.
I are good at saying what I mean. (I recently reacquired Platinum and have it on the brain)
Why Pearl instead of Diamond?
Out of curiosity, why Pearl instead of Platinum?
E: Masterjun, that's a bit of a stretch if it's deliberate. I don't really care one way or the other, though maybe call it "Speed/Entertainment Tradeoff" since it's for the pun?
I generally only use the while true do loop when I have reason to put images on the canvas; for anything other than that (text and simple vector draws) I stick with calling my main function via onframestart. As an example, my strobe function above could be rewritten as
Language: lua
-- "Megaman Battle Network 2"
-- This just generates a strobe effect powered by the RNG in the upper left corner.
local rngvalue = 0x9080
local EWRAM = "EWRAM"
function strobe()
memory.usememorydomain(EWRAM)
local color = memory.read_s32_le(rngvalue)
local invert = color * -1
gui.drawBox(0,0,3,3,color,invert)
end
event.onframestart(strobe)
and it still works the same. The onframestart/onframeend methods do not play well with drawIcon and drawImage though (this is true as of 1.9.4, haven't tried breaking this with the new version yet)
So, if I'm reading this correctly, the game's random seed is generated at startup and doesn't change at all? Is this the case on the actual Apple II hardware as well, or is it an emulation issue?
Probably the case on the Apple II as well; this was before Random Number Generation became an art form. It would be nice to find out if this is the case; how does one induce a memory dump from the middle of the game on that old thing?
