Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
Alyosha wrote:
Nice work! one long unexplained bug figured out!
I wouldn't say that. It's still a very rare bug, I haven't been able to do it without lua yet.
$FFE9 does normally update even during lagframes. Only in 0.1% of cases, it doesn't and we need to figure out why.
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
I reproduced scroll glitch by using lua, to freeze enemies on the screen to create lag.
Due to the lag, the scroll glitch happens (at a rare chance).
The game skips a vertical row.
This is what it looks like normally:
But due to the glitch it is like this (notice the tree is gone):
As a side-effect, the game still scrolls as far as it normally would, thus wrapping around the level.
The bug happens because $FFE9 doesn't update after a lagframe occured.
So in the end, it's a very interesting glitch but probably not so useful for TAS purposes. However, I secretly hope lagframes could lead to other more severe tricks.
Language: Lua
memory.usememorydomain("System Bus")
local tbl={}
while true do
--freeze mario at the sky
memory.write_u8(0xE201,0)
--freeze enemies on the screen
if emu.framecount()%3==0 then
for i=1,10,1 do
val= memory.read_u8(0xD100+16*i-16)
if val ~= 255 and val~=0x0A and val~=0x0B and val~=0x49 then
if tbl[i]==nil then
tbl[i]=true
end
end
end
elseif emu.framecount()%3==1 then
count=0
for p=1,table.getn(tbl),1 do
if tbl[p]==true then
memory.write_u8(0xD100+16*p-16+3,40) -- x
memory.write_u8(0xD100+16*p-16+2,20+p*12) -- y
end
count=count+1
if count>5 then break end
end
tbl={}
end
emu.frameadvance()
end
I will try to record a movie file without lua/cheat.
The game uses map patterns and reserves certain tiles for hidden content.
Certain "reserved" tiles are unused, in 3-3. --> put this on TCRF
In the two autoscrollers there is a wall of hidden blocks at the boss, respectively. --> put this on TCRF
Whoever made the map for VGMaps didn't know about certain blocks... (This is also the case for some of the Mario & Luigi Superstar Saga maps at VGMaps.)
There are no more invisible ground tiles aside from those in the middle of 2-2. And there is no lift in 4-2, so
foundit=false
hiddentile = 0x5F -- invisible block that reveals a lift/1up/powerup
invistile = 0x62 -- invisible ground in 2-2
address="script started"
while true do
if emu.framecount()%90==0 then
address=""
foundit=false
for i=0x9800,0x9AFF,0x01 do
if memory.read_u8(i) == hiddentile then
foundit=true
address=address .. "\n$"..string.format("%x",i)
end
end
if not foundit then
address="tile not found"
end
end
gui.text(50,50,address, 0xFFFFF070)
emu.frameadvance()
end
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
I tried the latest interim (commit ffbb1f339).
- The GB menu lacks the core picker (Unlike the GBA menu)
- GBHawk and Gambatte use separate gamepad config (Unlike the two GBA cores that share their config).
- Can GBHawk not run GBC games?
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
Looks like you're doing some pretty good work. I read that your core will even surpass Gambatte in terms of accuracy!
So I will look forward to testing a fewcertainbugs once it's ready.
Maybe I will be motivated to do some SML1/2 TASing, too.
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
Note: I fixed the problem I'm describing here, see the edit below.
I'm using lua to press keys on the keyboard, in order to tilt GBA Yoshi Topsy-Turvy.
The tilt is delayed one frame which is fine, except when I want to load a state...
For example,
I want to keep tilting right:
frame 1: tiltRight
frame 2: tiltRight
frame 3: tiltRight -- I save a state here
frame 4: whatever
I load the state and do
frame 4: tiltRight
but at this point, it behaves as if I didn't tiltRight on frame 3.
A way to work around this is to hold the tiltRight key and load the state.
But it's very unintuitive and would leads to mistakes if I had to remember this every time I need to loadstate.
I was wondering if it's possible to fix this, maybe by using userdata functions (to save the info that I did tiltRight on frame 3, so after loadstating, it does the tiltRight automatically). I'm not familiar with those functions, so I would appreciate your help.
My current script that has the behavior as explained above is here line 523 in the pastebin.
Edit:
I found a solution.
Userdata wasn't of any help because the code is always executed at the end of a frame. So I put the tilt function like this:
Language: Lua
event.onframestart(function()
if Displays[4][2] then -- "tilt" is active
doTilt()
end
end)
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
It doesn't matter where Yoshi is standing. The line gets drawn depending on the angle the game is tilted. In both screenshots above, the angle is 0. Please note I'm not very good with math things. I achieve much of what I want through trial and error.
Language: Lua
-- draw line from yoshi to this destination (*80 factor is how far the line goes)
des1 = math.sin(AngleDegree*RPD)*-80
des2 = math.cos(-AngleDegree*RPD)*80
des3 = math.sin((AngleDegree+90)*RPD)*-80
des4 = math.cos(-(AngleDegree+90)*RPD)*80
gui.drawLine(Xscreen,Yscreen,Xscreen+des1,Yscreen+des2,0xFFFFFFFF) -- below yoshi
gui.drawLine(Xscreen,Yscreen,Xscreen-des1,Yscreen-des2,0xFFFFFFFF) -- above yoshi
gui.drawLine(Xscreen,Yscreen,Xscreen+des3,Yscreen+des4,0xFFFFFFFF) -- left of yoshi
gui.drawLine(Xscreen,Yscreen,Xscreen-des3,Yscreen-des4+1,0xFFFFFFFF) -- right of yoshi (this is the line that looks off, but by adding +1 it looks as before)
I added +1 and it looks like before. This is good enough for my purposes.
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
converted the lua to Bizhawk
https://pastebin.com/4j9b1C0C v0.1https://pastebin.com/VujaUZ7s v0.2 (tilting is not delayed 1 frame anymore)
Same as before, except it now finds and updates the dynamic addresses (xspeed, yspeed, angle) automatically.
Timer and objectives are removed.
You can edit your position/speed/current level number.
Press E on the keyboard to tilt left, R to tilt right.
EDIT:
I wonder if game can glitch if there are too many musical notes...
https://www.youtube.com/watch?v=uaml44qUmZU
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
I'm in the middle of converting and re-writing my Yoshi Topsy-Turvy script, from VBA to Bizhawk.
I stumbled on a problem, where I want to draw an axis on top of Yoshi, depending on the current angle the game is tilted in.
But after I finished the re-write it looks different. The x-axis looks off by a small margin.
I noticed using this gives a different result in both emulators:
Language: Lua
angle = 0 -- as read from memory when not tilting
AngleDegree = angle * 1.40625 -- converting from 0-255 to 0°-360°
RPD = 0.0174532925 -- radians per degree
print(math.cos(-(AngleDegree+90)*RPD))
VBA prints: 1.79489673697e-009
Bizhawk prints: 1.79489673696339E-09
Is this a bug?
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
I feel like it will require someone more proficient (Masterjun LordTom etc.) to look into the memory and find out where the ACE is hidden in this one. It's basicly going to be a repeat of SMB3 and Zelda 1.
Question:
How is this done? I would like to try this in other places.
https://www.youtube.com/watch?v=3UpXJDN8CCs
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
Hello Foulco,
I managed to do it :-O
http://dehacked.2y.net/microstorage.php/info/1970539497/StarTropics.bk2
It would be good to break down and understand the glitch.
So if you can record an input file of the credit warp or the debug menu, I would be grateful. I'm sure other users will be happy to investigate, too.
Maybe the bug can be done at the Snake boss, too.
You can upload your input file here: http://dehacked.2y.net/microstorage.php
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
whaaaaaat
Well this is interesting. I will play around with this tomorrow.
Maybe will run it too, since up until the Kraken was as far as I managed in the past.
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
There is a small graphical glitch I encountered while TASing on Bizhawk-Gambatte (ver 2.2).
I load a state A (this is the starting point).
I advance to a certain frame*.
I load state A again.
The next frame, it shows the glitched sprites. The frame after that, it is normal again.
If I advance to a different frame, and then load state A, it doesn't happen.
I think other effects that I saw was that my character and enemies turned invisible.
I'm noticing this problem especially in the current level I'm in (The Smurfs GB, level 11 "The Cliff").
I turned off the luascript and it still happens. No memory freezing or lua memory editing is going on. Seems to be graphical only.
___
*This frame will always produce the problem. The chance of landing on such a frame is really high in The Smurfs GB, level 11.
Here's another emulation glitch [...]
As a side-effect, the dragon's head will be gone.
I haven't checked it just now, but on the real console you can't run out of present boxes nor does the dragon's head vanish [...]
In GB The Smurfs, in the dragon boss, VBA has this bug: if you pick up too many presents, dragon's head despawns and you run out of presents [...]
But on real console and Gambatte, it works correctly: You can pick up as many presents as you want [...]
We need to talk about this glitch.
The glitch mentioned above always happened to me when I was playing/testing on VBA but it never happened on my real GBA. As a result, I've always considered it to be an emulation glitch.
But right now I'm making a new TAS in Bizhawk-Gambatte to improve my old VBA run and the Dragon's head despawned while I was playing around!
I was really confused how this bug, which I thought of as an old VBA bug, slipped into Gambatte. After some further testing, the Dragon's head despawning actually turned out to be a version-exclusive glitch. It only works in version 1.0. It never was an emulation glitch.
I always used the 1.0 ROM, but I have a 1.1 cartridge. And due to VBA having many other (confirmed) emulation glitches at the time, I quickly came to a false conclusion. Sorry about this.
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
I did lowscore because some guy requested it.
http://dehacked.2y.net/microstorage.php/info/1267827518/lowscore%20sml%20testrun%2010110%20points.bk2
(Bizhawk 2.2, Super Mario Land v1.0)
10110 points:
2-3
10 points from score tallying. Goal pedestral can't be touched at 0 seconds left.
3-3
100 points from unavoidable coin near the beginning.
4-3
5000 poins from killing the cloud boss.
5000 points from killing Tatanga.
Editor, Experienced Forum User, Published Author, Expert player
(2315)
Joined: 5/15/2007
Posts: 3856
Location: Germany
I just watched the MLP movie. It was decent but...
https://pastebin.com/RXbm4Y5z
+
For a movie that was in the works for 2~3 years, it was lacklustre. I can get more out of the Equestria Girls movies than this, personally. The lack of known characters is a big killer for me. The villains never seemed so world threatening. It was overall a nice watch but not something that will make me come back to watch certain parts or listen to certain songs.
But due to the glitch it is like this (notice the tree is gone):
As a side-effect, the game still scrolls as far as it normally would, thus wrapping around the level.
The bug happens because $FFE9 doesn't update after a lagframe occured.
So in the end, it's a very interesting glitch but probably not so useful for TAS purposes. However, I secretly hope lagframes could lead to other more severe tricks.
Same as before, except it now finds and updates the dynamic addresses (xspeed, yspeed, angle) automatically.
Timer and objectives are removed.
You can edit your position/speed/current level number.
Press E on the keyboard to tilt left, R to tilt right.
EDIT:
I wonder if game can glitch if there are too many musical notes...
https://www.youtube.com/watch?v=uaml44qUmZU
I noticed using this gives a different result in both emulators:
I load a state A (this is the starting point).
I advance to a certain frame*.
I load state A again.
The next frame, it shows the glitched sprites. The frame after that, it is normal again.
If I advance to a different frame, and then load state A, it doesn't happen.
I think other effects that I saw was that my character and enemies turned invisible.
I'm noticing this problem especially in the current level I'm in (The Smurfs GB, level 11 "The Cliff").
I turned off the luascript and it still happens. No memory freezing or lua memory editing is going on. Seems to be graphical only.
___
*This frame will always produce the problem. The chance of landing on such a frame is really high in The Smurfs GB, level 11.In GB The Smurfs, in the dragon boss, VBA has this bug: if you pick up too many presents, dragon's head despawns and you run out of presents [...] But on real console and Gambatte, it works correctly: You can pick up as many presents as you want [...]