1 2
7 8
Post subject: Shining Force 2
M.I
Joined: 1/28/2005
Posts: 51
Location: Florida
* Aims for the Fastest Time * Manipulates Luck * Plans to Manipulate A Glitch or Two * Uses a Cheat (Special Turbo) I've decided to take up this project alongside the Wiz 'n' Liz project i'm doing right now. There's a good chance that there're alot of people out there (besides all the people on this site) that would enjoy seeing a Shining Force 2 run. I've been in the planning stages for a week now, experimenting with the randomness of the system. It took me a while to figure it out, but if I save state BEFORE the cursor lands on a member of the Force, I can load state until I get the desired results. Misses? A thing of the past. On Test A (below), I didn't use the save states until /AFTER/ the cursor had landed on said member of the Force, and their stats (HP/MP) were pulled up. You'll see one or two misses, but nothing huge on a casual run. In Test B, I used Special Turbo (you'll notice the differences immediately). Special Turbo was a feature implemented by the game-developers to mess around with if you knew the code, in this case, Up Down Up Down Left Right Left Right Up Right Down Left Up and holding down Start until you selected your specific game. This gave you four options: Special Turbo is one of them, which is the only one selected. When Special Turbo is activated, Dialogue passes by a few frames quicker than normal. The cursor (in battle) also moves MUCH quicker than normal. While these seem to be subtle changes, in a game as big as this one, every second I can prevent the viewer from passing out is a second i'm willing to take. :) Test B differs greatly from Test A, simply because I use them both to experiment: while they both go from power-on to the screen blacking after the first battle, there's an obvious difference in time from Test A and Test B. But the real test here comes from you guys. I ask that you watch Test A, then Test B. Feel free to critique (you guys should know by now that I love critique!), and point out things/ask questions. Also, in your explinations, PLEASE include what you'd rather see in the final project: the game at normal speed, or with Special Turbo. However, since it's the keys, and the game auto-saves it at the beginning of the game, where the player can make the first movements, once Test A is over, you'll actually need to go into Shining Force 2 and manually delete the save for !. Then, you can proceed to Test B. If you don't delete the save file in Shining Force 2 manually, the game can and DOES desync, and you'll wind up wondering what the hell is going on. Anyways, once i've collected all the data I need for this from you guys, i'll proceed to begin this as my side-project. I suspect it'll take anywhere from a month to a month and a half to complete perfectly.. and that's if i'm really lucky. :) Thanks, guys! Here's the Links: Shining Force 2 Test A Shining Force 2 Test B Edit: Or, you can just watch Test B. Picky. e_e
Former player
Joined: 5/31/2004
Posts: 375
Test A desynced for me, but test B worked. You take entirely too long choosing your settings, entering your name, and selecting a difficulty. Remember; Frame advance is your friend. Also, just because you don't see a cursor doesn't mean you can't do things with it. For example, this is the first frame you can move the cursor: From here, you can hit left one frame, up the next frame, and C the following frame to select the default name (faster than "A," in fact, and more aesthetically pleasing to see a real name there). Also, you can choose the difficulty in six frames: right, nothing, right, nothing, right, C. Note the cursor looks like it's highlighting "hard" when you do this, but you really do select Ouch! In your vid you take at least five times as long to do these things. As for the first fight, I think it's possible do it without healing, by not spreading out the Force so much that anyone gets surrounded.
M.I
Joined: 1/28/2005
Posts: 51
Location: Florida
Ah. Not sure why Test A desyncs: it works just fine for me. Oh well. Anyways, all of it was done at 33% speed, and I wasn't planning on using either of those movies when I actually start the final project. I mostly just rushed the first battle: I didn't use any tactic outside of using Chester (whom has the largest movement frame at the time, and the second highest defense of the three, making him the perfect choice to lure enemies instead of Sarah). True, I was careless, but when I actually throw myself into the actual project (which'll either start tonight or tommorow) i'll be playing every battle no less than six or seven times. However, here's what i've come up with from experience: Enemies don't move a whole lot (you'll see that in either movie) unless your character is within moving distance. Therefore, making it all the more possible to move the Force in a group with the ones with the highest defense in the front. Later on, when enemies start being able to cast magic, i'll have to spread the Force out more, and then with the later bosses (those packing such powerful spells as Bolt 2/3, or Blaze 3) and enemies have to try to single out magic-casters. The basic bait-and-lure tactic works most of the time, though.. atleast until the later levels, when the enemies begin to wise up a little. At which point, i'll have to switch tactics and move the Force as a single entity. However, since this is a speed-run we're talking about, i'm mostly going to invest my money/time seeking out the most important of items/equipment: Running Pimentos and Running Rings, both which permanantly (or, for the ring, temporarily as long as it's equipped) raise the movement of characters, meaning I can keep the force close, and move further, each turn. Anyways, that's just a basic test to gather information. You weren't missing out on much with Test A: I was more concerned with Test B anyways. Thanks for the info, though. Edit/Update: Also, the reason why I use Sarah to heal is because even when she heals, she gains EXP. The same thing goes for Medical Herbs: if she uses the Herbs, she gains the same EXP she would healing, even if there's no difference in Max/Current HP. So alot of my money early on will go to Medical Herbs for Sarah (since her MP is limited, anyways, so it's a good idea to keep some on her at all times), because this is a quick way to keep her on track, since she won't be getting any sort of real battling outside the first three battles, since her role goes from Part Healer/Part Fighter to Full Healer. Bowie, Chester, and Sarah are wise choices to keep in ANY line-up, so I try to keep them even, instead of hording levels onto the Hero. After the first battle, of course, Jaha joins the group, which is a welcome addition: his attack is rediculous.
Post subject: Shining Force 2
Former player
Joined: 8/17/2004
Posts: 377
I used to play this game years ago, and I really loved it. However I haven't touched it in ages, so I really don't know of any tricks which could be used for speed. But this is a game I would really love to see a TAS made for. Anyone want to give it a shot?
Editor, Reviewer, Experienced player (969)
Joined: 4/17/2004
Posts: 3107
Location: Sweden
Someone tried but appeares to have given up: http://tasvideos.org/forum/viewtopic.php?t=1759
Joined: 8/27/2006
Posts: 883
I typed too fast, I'll try to find how luck manipulation works in Shining Force 2 because it seems to be hard.
Joined: 3/7/2006
Posts: 720
Location: UK
If it's anywhere similar to Shining Force, it depends on what keys you press while you have a cursor selecting who to target. I was able to manipulate a critical, 2 different damages, or a miss, this way. However, I couldn't seem to change spell damage.
Voted NO for NO reason
Twisted_Eye
He/Him
Active player (332)
Joined: 10/17/2005
Posts: 629
Location: Seattle, WA
I tried once already, and after half an hour got absolutely nowhere with the manipulation. All I could do is change whether there was a hit or not. Damage was the same, couldn't get a critical or whatever. That was a couple months ago, and if button presses affect it, well then the game would sure be worth a second shot
Joined: 8/27/2006
Posts: 883
Yeah, well I know when the damage change, and it's when you are about to select attack, there's an RNG running only when that menu is open, and this will change how it works, but anyway, if i have to go trial and error, I might try to modify gens to be able to have a frame number to press the button to make a double hit. That would be faster that way.
Joined: 3/16/2006
Posts: 289
Shining Force 1 is the same with maninuplation. Even on the GBA version! It's like they use a pure math solution with a +/- a little for "randomness." Even maniuplulation that randomness only will get you a little extra damage at least, or if you get lucky a crit or double hit. Crit would be the fastest. The only way I can see getting this to work would be like Catasrophe's work on FF6: take the memory apart and find out how these games tick.
I am just a silhouette, a silhouette of a memory of a solitary night .. nothing more.
Joined: 8/27/2006
Posts: 883
Yeah, well I plan on doing some research about this, I might read the assembly code, or maybe I'll modify the emulator to make it do test with the memory value, like testing for 10000 each time the character do a double-hit, and write the memory value. It's easy to see if there's a double with memory value. I just need to find the monster life, then look for two change in the same testing. For a crit it's harder because I don't know if there's a value change for a crit. Anyway l'll look at that. If you have any idea about memory value or something, It would be really apreciated.
Joined: 3/16/2006
Posts: 289
I wish I did. Then Shining Force 1 would have been done.
I am just a silhouette, a silhouette of a memory of a solitary night .. nothing more.
Joined: 8/27/2006
Posts: 883
Well I got some RAM value for Shining Force 2. Maybe someone who have done some luck manipulation with ram value could help. I have 3 value, one random number, one number that count from 0 to 255. And one other number that is random but only when you are about to select Attack. I think that if I find out how to read assembly code because a hundred frame, I might be able to find out what is the condition to make a double-hit.
Joined: 3/16/2006
Posts: 289
Double is good. Any luck finding crit yet?
I am just a silhouette, a silhouette of a memory of a solitary night .. nothing more.
Joined: 8/27/2006
Posts: 883
Well crit seems not to boost a lot the attack, but I didn't find anything, for the moment, I don't have the time to modify the emulator to be able to do trial and error each time. But I'm taking as much information as I can about the fastest team and such.
Joined: 8/23/2006
Posts: 46
Any updates on this? I am one of "those people" who would really like to see a good run of this.
Joined: 8/27/2006
Posts: 883
Well I don't have much time, and I would like to see this, I found a couple of memory address, but I was not able to manipulate luck. It seems to have to RNG but I'm not sure. I should take a look at the assembly code generated by the genesis to know what cause a Double-Hit or a critical hit.
Former player
Joined: 9/1/2005
Posts: 803
ZeXr0 wrote:
I should take a look at the assembly code generated by the genesis to know what cause a Double-Hit or a critical hit.
While i can't give you an answer as to what causes these, I can tell you that critical hits are damnwell useless, at least from what i tested early on. They did equal damage to a normal hit, or at most a single extra point. A double hit on the other hand, is, well, two hits which may be enough to kill an enemy early on even if it has 9 hp (depends on who you use among other things).
Joined: 8/27/2006
Posts: 883
Well I think that the critical hit will be usefull later in the game, because It might deal greater damage if you are stronger. But it I can get what goes in the assembly code, I could be able to determine what value are used to determine a double hit.
Joined: 8/27/2006
Posts: 883
I don't know if it would be a great tas after all, no ones gonna sit and watch this, I was playing the game for another time, and I can't estimate the time it will take me, but I doubt it would be less than 10 hours. There's 40 battles, and If I don't manipulate enemies, it's going to take forever. I'll do a test run soon on the emulator, and I'll be controlling the enemy. It will be a good start to look for memory address. But I think i'll need a way to see assembly code.
Skilled player (1638)
Joined: 11/15/2004
Posts: 2202
Location: Killjoy
Both Shining Forces could make for very interesting time attacks. I am, or soon will be, working on a time attack for Shining Force 1. Upthorn mentioned via IRC that he will soon be releasing a version of GENS that includes a partial dissasembler, allowing us to crack the codes of critical hits. In shining force 1, I am sure it relies on both controller state and the current frame number. (I've found the frame counter in memory) In addition to critical hits, it appears that forcing the CPU to always miss is a possibility as well. I'd love to see a "Hero takes no damage" SF1 run. The problem will not be critical hits (that should be easiest). The biggest problem will be manipulating Attack queue. It appears to be infinitely modifiable, however, locating the code to determine the attack order will be the biggest hurdle. Additionally, I also determined that the Double attack is determined as the enemy is selected, and not as the attack begins. Thus, it should be possible to coordinate both doubles and criticals, making for a very powerful move.
Sage advice from a friend of Jim: So put your tinfoil hat back in the closet, open your eyes to the truth, and realize that the government is in fact causing austismal cancer with it's 9/11 fluoride vaccinations of your water supply.
Joined: 8/27/2006
Posts: 883
Do you have an estimated time to the total run time?
Skilled player (1638)
Joined: 11/15/2004
Posts: 2202
Location: Killjoy
No idea. Impossible to estimate at this time.
Sage advice from a friend of Jim: So put your tinfoil hat back in the closet, open your eyes to the truth, and realize that the government is in fact causing austismal cancer with it's 9/11 fluoride vaccinations of your water supply.
Joined: 10/3/2005
Posts: 1332
I just did some testing on this, and it doesn't look as though the luck manipulation method DarkKobold found in SF1 works in this game. Second, I don't know if it has been mentioned, but the Japanese version of this game seems to have a time-based RNG, meaning that you could manipulate the hell out of it. Lastly, those two links in post #1 are broken.
Joined: 8/27/2006
Posts: 883
Dromiceius, you seems to have the same conclusion as me, there seems to be two RNG number, and one frame counter. The first RNG number changes when you are not in a menu, and the second one changes when you are in a menu.
1 2
7 8