Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
I just realized that if you simply want a fair die with 1 (whats the point), 2 or 3 values.... then you could just use a regular 6-sided one and re-purpose the values :) (e.g. 1&4=1, 2&5=2, 3&6=3)
then you'd have equal probability for all values AND for all faces :) it's just that face-count isnt same as value-count
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
the article arflech linked has an explanation why dice with unequally shaped faces will never be fair... basically, it'll be influenced by the elasticity of the die and surface it lands on (e.g. a flexible surface might slightly favor one face while a less elastic surface might favor another face)
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
Hmmm I think it'd be possible for any N>2 to make fair rolling-log dice with equally shaped&sized *landable* faces? (each sides of the log would be N triangles resulting in non-landable faces that form a point side)
for even N, the top side could be indicated with the result
for odd N, numbers could be written on/over edges, or the botton/landing face could be used as the result
edit: oh yeah, it's be a N-sided prism with N+1 sided pyramids attached
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
I cannot really figure out a failsafe method of finding the best strategy... so far I'm simply using lots of alternatives to find a fast one
1.1 in 2197 (30 frame improvement)
edit:
final screen of 1.2 goes black at 6976
edit2:
found some nice tricks in 1.3, trying to figure out the fastest way of maneuvering the remote-control bomb to a target, and whether it pays to run part of the way rather than stay back as far as possible
oh and a perfect example why this game is a **** to optimize... When riding the elevator to the bottom level of 1.3 (and when you go left to collect RC bombs), there's several possibilities of where/when to perform jumps (which is faster).... I found 3 options that all lead to an equal time used to cross those screens, differences start playing up 5 screens later however... in this particular case with a 30-frame difference between the worst and best of those 3. So, who is to say whether waiting 1 frame might gain me another 30...
edit3:
Hmm, found some new insights... and a new strategy for 1.3 :P gotta redo a fair bit, and still need to work out how to use the wasp fastest :X (although it got simpler than before)
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
Hello, nice work on your project! To avoid memory limitations you could use a different approach... but it's hard to give any advice without know what you currently have :) (quite possibly you don't want to share it yet ofc)
for example a classic depth-first tree search has very low RAM cost
I used several Sokoban (similar game) solvers for my Young Merlin TAS and found quite some differences in their outcome (too lazy to make my own ;)
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
One friend is *slightly* interested in the concept of TASes but doesn't watch any (only maybe of zelda OoT... his favorite all time game)
My brother knows I create and watch TASes but is in no way involved or interested
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
I scanned through a review of 'The Path' this week.... seemed like that game would be useless to TAS
from what i understand, the fastest way to complete the game is what most players will do the very first time.... follow 'the path' as little red riding hood, arrive at grandma's house and that's it, game over and the player is informed that maybe the game would be more fun if he goes off the path
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
Oh my bad... I uploaded the same file twice because I copied the wrong file by accident :X (so I no longer have the cool one myself either)
anyway its approach was:
skip 1st enemy and climb the ridges (instead of the stairs)
wait at the 'warp' a bit to manip the 2nd enemy
use your 3rd hover-bomb to destroy 3rd gate and 2nd enemy simultaneously (and collect its drop for your 4th bomb)
so didn't have to use my gun for that approach :P couldn't manip the enemy towards the door fast enough though
edit:
Found another way cool trick (in 1.2).... that costs ~100 frames :( crap
edit2:
lol there was a typo in my spreadsheet.... it saves ~100 frames instead of cost :D
edit3:
Ohw, just found another way to save some time in some places... including the very first screen. That gives me 2 opportunities (ok lets try to see it from a positive side :)) 1st I can see how well this game hex-edits, I'm guess not very very poorly and it'll desynch the same screen as i try to edit it. 2nd I can have a new go at the no-need-to-use-my-gun-trick in 1.1 (since I'll arrive at different times)...
Received the awesome suggestion to make a LUA script that'll perform the various moves for me, hadn't thought of that so I'll see if I can make that work as well.
all in all lots of work to be done! :P Back to the drawing boards :)
edit4:
oh dear... movements in the game are quite irregular, up to the point that delaying an action can result in different (possibly faster) actions being available :( E.g. when kyle starts running at frame N kyle will be able to start a jump at tiles T+1 and T+3... but if he starts running at frame N+1 he can jump from tiles T+2 and T+4
not sure yet how exactly this works but it seems optimizing will be hell :/ first thing is to figure out how far consequences can/will reach
edit5:
just found 3 different approaches that have 3 different frames (1227, 1225, 1224) for picking up an item, yet they all lead to the same states (exactly same RAM) at frame 1609.... definitely convinced there's a frame-rule involved, but dunno what it stems from :P
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
Alright first WIP post...
tried some variations... unfortunately the coolest one was 13 frames slower :(
I timed the first black frame of the last transition in lvl 1 (so when going up the elevator)
Aqfaqs: 2257
fastest: 2227 (here!)
coolest: 2240 (accidentilly thrown away :P)
comments & synch-checks welcome!
ps. ROM = Blackthorne (32X) (U) [!].32x
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
I may have a go a this game (I'm assuming nobody is still working on it since 2007 :P)... first a playthrough on Gens to see if the speed is as varying as XYZ said in 2007
searched on TASvids, SDA, google, gameFAQs and youtube... couldn't find much about any TAS or speedrun or misc info, any info appreciated
edit: oh I found the ghost LUA script ofc! :) didn't try to run it tho :P also, it didn't mention for what platform the script is :D
edit2: oh I found maps :) awesome! (they're not the correct chronological names for Gens because the snow/ice world is missing, but it's still nice to have :))
http://www.vgmaps.com/Atlas/SuperNES/index.htm#BlackThorne1.1, 1.2, 1.3, 1.42.1, 2.2, 2.3, 2.43.1, 3.2, 3.3, 3.44.1, 4.2, 4.3, 4.4Sarlac
edit3: Started getting to know the game from a TAS POV (input moments/effects, RAM addresses, etc) and currently some test runs of the menus and first level (Figuring out why I'm losing some frames on Aqfaq :D)
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
Yeah I'm getting some latin language too....
"wanna know the temperature of the sun? on google you'll find the answer!"
"wanna know who invented paper? on google you'll find the answer!"
never seen anything english/dutch there
thought I got this because my PC's been abroad :P
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
none of my emulators can play your movie (unsupported movie file) :/ what emulator version are you using?
note: your previous movie (hhssom.smv) played with the 1st entry without problems
I tried (copy-paste from "Help" menu entry):
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
the lvl 5 whipe charge strikes in all 4 directions, when this move is performed near a whip-pole while facing the 'wrong' way, the characters will actually jump the wrong way... combined with the super-charge glitch, this may save some time in a few places for a TAS
source = http://www.flyingomelette.com/oddities/oddities25.html
edit: oh... *old* news
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
Oh I noticed the value FFFF while quickly scanning through the topic :P sorry for erroneously assuming it was overflow related :)
hmm from your explanation I don't understand why Wall is needed... iirc you can directly target party-members in SoM? or is there some direct check in the game that checks whether target is partymember? :P
Experienced Forum User, Published Author, Active player
(354)
Joined: 1/16/2008
Posts: 358
Location: The Netherlands
Interesting to see a new go at SoM :)
did you try getting crits instead of regular hits? pretty sure I read/seen something about it on TASvids but I'm too lazy to look it up... sorry
What's your plan for leveling/magic/weapon-upgrades/etc?
GL either way