Posts for Nitrodon

Posted: 4/18/2008 3:26 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

Amazing run. Yes vote. I noticed a few times (starting in stage 6) that you pressed the jump button before you were clear of the ceiling above you. This increases your Y coordinate by 1 when you are actually able to jump, thus delaying your landing on the block. You lost a few frames that way whenever you were in that situation. (As a side note, that also allows you to jump across gaps 2 spaces wide, but I haven't found any places where that would be both possible and useful that way.)

Posted: 4/16/2008 3:33 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

Looks good, but as you predicted, I found a small optimization. On turn 3 of chapter 15, instead of moving up to Caellach and attacking him, I think it would be faster to move up to Caellach and not attack him. That way, you don't have to watch the base heal him.

Posted: 4/12/2008 12:49 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

cheetah 7071 wrote:

I suppose I didn't even think of that option because that would require suffering an attack from the shaman. I guess that takes about the same amount of time as killing the shaman outright, though (unless he goes for Saleh, which is of course possible).

He does go for Saleh. AI-controlled units just love attacking people with less than 10 HP, even if they won't deal any damage.

Posted: 4/11/2008 11:06 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

The units you have blocking the door could be one space closer to the enemy. Because of scrolling, I'm not sure if this would save or lose time. You can win in 8 turns simply by moving Seth one space further on turn 3, then moving him as far as possible on the next 3 turns. Two enemies will be killed by counters rather than normal attacks (one on turn 3, and one on turn 5), but the middle area will probably be faster. If you want to break the silver lance under the same circumstances as the current WIP, you should equip it on turn 3 for this strategy. I should probably say something about Vanessa's options. If/when she promotes, she'll need 85 additional lance exp if she plans to use Vidofnir (and get that ever-important 5% critical hit bonus). This may or may not require giving her a steel lance. If she doesn't reach S lances, you'll lose about 12 seconds by letting Fomortiis counter, and you would also need more manipulation for that battle. If you don't promote her at all (the animation takes about 20 seconds in FE7), killing Lyon in the final chapter will require a double critical from Seth with 2 more strength than he has now, and you'll lose 22 seconds because you would need to attack Fomortiis twice. Also, chapter 17 could be annoying without a promoted Vanessa, but I suppose you could use Myrrh under those circumstances. P.S.: You don't really need to make your entire post a link.

Posted: 4/10/2008 2:48 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

I checked Molotov's run to see how many javelin uses might be required. Chapter 15: The TAS strategy will differ greatly from the console strategy, so I can't say anything useful. Chapter 16: No ranged counters are necessary throughout the entire chapter. Gerik attacked a couple times with a bow during the player phase though. Chapter 17: Myrrh was the only one to get into a battle. Enough said. Chapter 18: I suspect that the number of javelins will be more relevant than the number of javelin uses here. They'll mainly be used to reach gorgon eggs more quickly, instead of the usual counter massacres. Chapter 19: Molotov's run uses a javelin on turn 1, then doesn't bother to unequip it, but there aren't any ranged attackers to counter. If someone other than Myrrh has the swiftsoles, that unit can follow exactly the same strategy without a javelin. Chapters 20, 21-1, 21-2: No javelins necessary. In conclusion, I expect that after you pass chapter 15, you'll have enough javelins for the rest of the game. I would therefore suggest that you find a strategy for 15, and then determine how many javelins you need for it. A general route for 18 might also be useful.

Posted: 4/7/2008 9:25 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

Chapters 16 and 18? Personally, I'd be more worried about having enough javelin uses for chapter 15. With the amount of javelins/spears you currently have, you'll probably have to rely on Saleh a lot in that chapter (which is probably a good idea anyway, since he and Vanessa are the only good units you have who aren't slowed down).

Posted: 4/2/2008 8:23 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

Rounds 2, 4, and 5 are improveable. I tried this game, and rounds 10 and 23 are not reused in any other Lolo game that I have played. Some other rounds are different enough that you can't copy the TAS of the corresponding room in one of the US Lolo games. I reached round 28 before getting bored, so there are probably a few other rounds unique to this game, but not very many.

Posted: 3/27/2008 7:43 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

I should be able to write a more coherent post in the morning, but for now I'll just type a few things. The only entrance to the throne area is the door in the top right corner, so the only reasonable path is the one on the right. I suspect the fastest way to deal with the cavalier reinforcements may be to put an unarmed unit in the way. Kyle and Dozla have the highest defense (11) out of all of your units except Seth, and either of them would easily survive in that position. The cavalier with a javelin is the third one to move, so he should be blocked by his own allies and unable to attack you. EDIT: I knew I should have gotten some sleep before posting. Of course, this is a TAS, so anyone can dodge one attack per turn easily. The other easy possibility I thought of later is to have 3 units there to kill them, but it's probably faster to just block them instead. There's no general way for Seth to waste RNs faster from that position. There may be a sequence of cursor movements that will waste that particular set of RNs more quickly, but I'm too tired to try to find one. Molotov's run beats this chapter in 8 turns, compared to 9 in your run. I haven't checked to see what he did differently, but I have noticed that it's possible to win in 8 turns if Seth is one space ahead of your run after opening the door.

Posted: 3/26/2008 10:31 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

When I tried the one-turn victory for fun, I made Seth kill the first knight with his silver lance, which didn't need a critical hit. That would explain why I didn't need much manipulation.

Posted: 3/26/2008 9:57 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

Why did you do all that switching in the formation screen? The one-turn victory works just as well if you only switch Vanessa and Saleh. Also, this is mainly a curiosity, but I happened to notice that you can win in one turn with very little manipulation. Saleh gains HP+def+res that way, so the manipulation you did might make it faster.

Posted: 3/26/2008 7:18 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

You should be more careful. Frames 127846-127870 (while Tana is manipulating the RNG) have no input, wasting 25 frames for no reason. Other than that, it looks good.

Posted: 3/25/2008 7:52 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

The end of that looked a bit ugly, but I suppose that's because any one of those gargoyles could have easily killed Vanessa in one hit at that point, and/or the boss was too hard. I have some solutions for that. Vanessa with an axereaver can kill the boss in one critical and one normal hit. If you do exactly the same until this boss kill on turn 6, then the enemy phase will merely consist of dodging all 4 gargoyles with three hit rates at 52% and the other at either 62% or 47% depending on how the AI deals with one-hit kills (and depending on Vanessa's position, which may depend on whether the bonewalker moves). All counters will be fatal without manipulation, except that one of them (which may not even be possible to counter) will need an extra point of speed to double with the axereaver. The gargoyles are actually triggered by stepping into a certain area. (The most succinct and unambiguous way to put it is "use any command in the trigger area, including 'Wait'"; being dropped/warped/rescued doesn't count.) I've mapped out the trigger area, but it doesn't seem to be avoidable without losing time.

Posted: 3/24/2008 10:02 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

cheetah 7071 wrote:

Even if Vanessa doesn't level up at all, there's a heavy lance floating around. I shouldn't have any issues.

You need a ranged weapon to reach that knight on turn 1. Tethys would be a possibility, but she's already needed so that Saleh (or Seth) can reach Aias on turn 1.

Posted: 3/24/2008 9:32 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

Those calculations seem right, until you notice that Demon Light gives +10 defense (among other stats). Myrrh would need 12 more strength for a OHKO. If you plan to attack twice, Seth+Myrrh will certainly be able to do it at her base stats, as long as Seth gains 3 points of speed and 1 point of strength between now and then (which should happen even if you don't manipulate his stats). I calculated earlier, and a double pierce with one critical would kill Fomortiis. You'd need 20 strength, and the probability would be extremely low without a high level. The triangle attack would give you a free critical, which certainly won't hurt, but the probability of success would still be fairly low. EDIT: I checked the 1-turn victory in Hamill Canyon. I'll assume that double criticals are impossible to manipulate, since the RNG usually isn't that cooperative. Vanessa needs about 2 additional points of strength to kill the armor knight, depending on stats. If necessary, Vanessa can use Ewan's energy ring, but the kills from chapter 12 should probably be enough. Seth would need 3 additional points of strength to kill Aias with a javelin, but it turns out that Saleh can kill him with Thunder at base stats.

Posted: 3/24/2008 5:57 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

Honestly, I'm not sure how feasible Vanessa with Vidofnir is. Vanessa has 73 lance exp at the beginning of this chapter, gains 40 on promotion, and needs a total of 251 for an S rank. A double pierce with one critical would still do it (but would be hell without leveling up Vanessa). Molotov's run used Myrrh, which is still a possibility if you don't mind watching her level up several times.

Posted: 3/24/2008 5:32 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

I'm not sure how useful it is to recruit Dozla for that chapter. Based on stats, Vanessa is perfectly capable of dealing with the last two enemies with enough luck manipulation. Then again, that recruitment did get rid of one "Other Phase" message. Based on this chapter, I'm assuming you don't plan to train Vanessa. As long as she is powerful enough to kill an armor knight in Hamill Canyon with a javelin (she probably is) and you have a different plan to kill Fomortiis quickly, that seems to be fine.

Posted: 3/9/2008 4:22 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

My old WIP of this game reached the end of stage 29 before I gave up on it. I compared the times: Stage 1: Randil, 3 frames Stage 2: Randil, 8 frames Stage 3: Randil, 6 frames Stage 4: Randil, 33 frames Stage 6: Randil, 2 frames Stage 7: Randil, 1 frame Stage 9: Nitrodon, 22 frames Stage 10: Nitrodon, 150 frames Stage 11: Randil, 7 frames Stage 12: Randil, 39 frames Stage 14: Randil, 15 frames Stage 15: Nitrodon, 3 frames Stage 16: Randil, 12 frames Stage 17: Nitrodon, 59 frames Stage 18: Nitrodon, 192 frames Stage 19: Nitrodon, 39 frames Stage 20: Nitrodon, 353 frames Stage 21: Nitrodon, 127 frames Stage 22: Nitrodon, 35 frames Stage 23: Randil, 57 frames Stage 24: Nitrodon, 20 frames Stage 25: Randil, 38 frames Stage 26: Randil, 110 frames Stage 27: Nitrodon, 15 frames Stage 28: Nitrodon, 115 frames Stage 29: Randil, 130 frames Total: My WIP is 359 frames ahead. Still, I'll probably vote yes after watching the rest of the run. EDIT: See also my unassisted speedrun of stage 33, 3 time units faster than this submission.

Posted: 2/28/2008 6:14 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

The Change Rod (and the Lilith Rod for that matter) increases wisdom by 5, which makes black magic deal more damage. With both the Change Rod (wisdom +5) and Tiara (wisdom +10) equipped, Rydia's damage output is increased by about 66%.

Posted: 2/22/2008 11:15 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

I just watched the test run, and it was much better than I thought would be possible. I especially liked the fact that you never needed to heal in battle. Saturn: Having Robo in the active party would add a few seconds of dialogue when Ayla joins.

Posted: 2/21/2008 9:51 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

The gameplay of the two Eggerland games is similar, but the maps and puzzles are completely new in the second game. A run of one game will not obsolete a run of the other.

Posted: 1/29/2008 1:44 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

There are two useful facts about the solutions to that problem: 1) The number of digits in n increases exponentially with respect to the number of digits in the highest value of s seen so far. If n has 5 digits or more, there is no way for s to have enough digits to make the product large enough. Therefore, n has at most 4 digits, and s is at most 36. 2) s = n (mod 9) because it is the sum of the digits, and s^2 = n (mod 9) because the reversal has the same digits as s (and is thus the same mod 9). Hence s*(s-1) is a multiple of 9, and thus s = 0 or 1 (mod 9). From here, we can just enumerate the possibilities: s=1: n=1*1=1 (works) s=9: n=9*9=81 (works) s=10: n=10*01=1 (doesn't work) s=18: n=18*81=1458 (works) s=19: n=19*91=1729 (works) s=27: n=27*72=1944 (doesn't work) s=28: n=28*82=2296 (doesn't work) s=36: n=36*63=2268 (doesn't work) Therefore, the (only) solutions are 1, 81, 1458, and 1729.

Posted: 1/11/2008 9:23 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

I use Geiger's debugging tools anyway, so the ability to use a debugger without needing to open another emulator would be great. However, I've noticed some occasional repeatable bugs with Geiger's debugging tools which would affect synch, and thus adding them as-is to a recording emulator would cause problems.

Posted: 1/6/2008 1:28 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

jdurg: You need the "record reset" version of Snes9x to watch this run. http://files.filefront.com/record+reset+snes9xzip/;9106812;/fileinfo.html

Posted: 12/24/2007 9:31 PM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

All current emulators used here store the movie data with the savestate, so loading the wrong state won't cause a desynch unless the movie is in read-only mode when loading the state and you either save state or start rerecording later.

Posted: 12/12/2007 9:45 AM

Quote

Nitrodon

Experienced Forum User, Published Author, Active player (283)

Joined: 3/4/2006
Posts: 341

The first time a character joins, it's always at the same level. When a character is temporarily out of the party, he or she gains the same amount of experience as your living party members at the end of every battle, so the level when rejoining can be said to depend on your level in a way.