Jump to content

Redwall

Member
  • Posts

    1,105
  • Joined

  • Last visited

Posts posted by Redwall

  1. thats not enough, the professors that teach the class i am taking have full labs because these courses are high demand. my friends and i usually email people in other depts including but not limited to those researchers that teach grad courses. and it is waste to read papers because the ones that do reply to emails usually reply with "sorry my lab is full." and im trying to go for efficiency here and i feel that the emails i write based on their website description is pushing it since that requires effort to and i should be sending out form emails to as many professors as i can.

    Would you mind posting an example of the emails you've been writing (with personal info removed)? Maybe I or someone else could help you revise.

    the problem with reading work/lab description is the fact that the last paper that the researcher could have published might hav ebeen years ago and the webpage may not be updated.

    You can use Google Scholar to find more recent papers of theirs.

  2. Have you been attending office hours for the classes you're currently taking? It's easier to find a position if the professor has a good impression of you and knows you're serious. The only reason I had an undergrad research position was that I got the top grade in my instructor's class (I was not the smartest, I just worked harder than everyone else). Before speaking to them, read at least a few of their papers, and don't just stop at the research description on the website. There are plenty of professors who could use a bit more help but don't broadcast.

  3. I like it. The small-ish screen doesn't bother me since the only game I've played on it is FE14, and I don't miss the 3D feature at all. I hear battery life is marginally better than that of a regular 3DS (~30 min extra). If you're mainly using it for Pokemon, it should be fine.

    edit: According to the thread I linked, you'll have to call Nintendo to transfer the NNID, assuming the guy who bought your 3DS hasn't deleted its system data. Unlinking =/= deleting, so if he's deleted it, then it should not be possible to regain your old games.

  4. I'm not bothered by the enemies having exclusive weapons since we players have access to an even stronger weapon, our ability to think critically. I do dislike Lunge Ninja, but more because they highlight the need for high-ish stats (or *breaker skills) in efficiently clearing levels, which I think is a more fundamental design problem; I'm otherwise fine with enemy-only things.

  5. If you're using a full team, it will be easier to simply break your own wall and have Niles use the Ballista on the Spear Fighters during PP1, allowing you to KO them with a single unit during EP1; if you have someone capable of ORKOing the axe users, you can move him/her to the right during PP1 and kill two or three of the axe users during the enemy phase. A calculated offense during the first few turns should give you enough breathing room to handle the final part of the map with ease.

    Killing Takumi isn't particularly hard if you break your own wall, but I don't like to go for him because of the low hit rates.

  6. Sorry, made another mistake. I don't think Competitive should get modified by Dragonskin. So assuming DS halves DF,

    75 = 2*((Atk-26) *0.75) + 0.25Atk - 2

    is what we've got?

    EDIT: neglecting things like Rose's Thorns, of course

  7. Are you sure thats correct? I believe with the the Shadow Yato you do 75% damage because of dragonskin. And dragonskin reduces Dragon Fang's effectiveness as well.

    Sorry, my mistake.

    75 = 2 * 0.75 * (Atk - Def) + Dragon_Fang_modifier * Atk

    Anyone know how Dragonskin modifies Dragon Fang's bonus when using the Shadow Yato?

  8. Yay got to Conquest Lunatic Endgame. Time to start reseting for that Double Dragon Fang proc.

    On a side note, does anyone know the minimum damage MU needs to do against Takumi so that he can kill with a single Dragon Fang proc? I'm thinking it's around 30.

    75 = Atk - Def (or Res) + 0.5 * Atk

    PhAtk must be at least 78 if Def = 26 + 1, MAtk must be at least 77 if Res = 22 + 2 + 1 assuming you don't double.

    If you do double,

    75 = 2 * (Atk - Def [Res]) + 0.5 Atk

    PhAtk must be at least 52, MAtk must be at least 50.

  9. +1 Str on Jakob is more to kill the boss turn 5 with Gunter attack and then Jakob attack+Gunter tagteam. Corrin seizes that turn. Sometimes Corrin doesn’t have the Str to 2hko the boss and you can try the Jakob strat or Gunter might have to get the kill (as I think he can’t survive EP4 vs the boss+archers). Though I think sometimes Corrin+Pocket Gunter+Jakob aura can survive PP4/EP4 against the boss, which gives you 3 chances for ideally Corrin+Corrin 2hko, but Corrin+Gunter 2hko is a backup, or Corrinx3 for the 3hko.

    It might also work if Corrin doesn’t have the Def to split the bridge enemies since it’s Corrin + Jakob (in pocket maybe, not sure if astance works) survives on the DV turn 3, clean up/split turn 4? Do the bosskill and seize PP5?

    For the bridge it might also help in another approach, though depending on how C2 exp worked out, he might not have a level yet (may need to feed him the north myrms in C3).

    For C4, I seem to remember the benchmarks to one-round the higher level faceless are pretty high?

    Nice, thorough C5 strat. I guess using Ryoma means Kaze doesn’t need a bunch of Str procs and dodges.

    Re: final boss and talents: I think Samurai Talent can oneround without procs which makes it pretty alluring, especially for resetless or such. But enough Str to oneround with at least 1 proc isn’t too bad efficiency-wise (like ~50%ish?), given how short the chapter and C27 is. I forget the exact numbers because Dragonskin does weird things but recalcing quickly…Wyvern!Gunter, Rally Str, Elbow Room, Defender is like +15 (16 S) Str, Shadow Yato with A swords and triangle is 20 att. 31+2 Str in Nohr Noble with tonic, -Luck, is 68 att. –26 Def and dragonskin doing 3/ 4th and floor is 31 damage/44 with proc which is just 75.

    Also, if Paladin!Jakob is used, Corrin doesn’t reclass until C10, at which point Camilla has arrived. So maybe Corrin not having Cav early doesn’t even matter too much in full teams? (can still Friendship/Partner into Paladin/Dark Knight later)

    So tbh Samurai seems pretty solid indeed. Also apparently there’s cheesy stuff to 2-3 turn C17 and C19 is easy nowadays with Beastkiller wyverns (usually Camilla), so I personally don’t think Ninja is that appealing over Levin Sword builds anymore.

    Yeah, I probably should have been more clear about the ridiculous benchmarks for ORKOing the Sealing Faceless. I suppose the best reason to choose +Spd over +Def or +HP, even though +durability is necessary for the absolute lowest TC, is to avoid going crazy on these sorts of chapters.

    Ryoma's penchant for dying before putting the C5 boss in KO range makes the five-turn clear dicey, but the nice thing is that it can transition into a six-turn clear pretty easily.

  10. Has anyone done a Lunatic LTC yet, at least for Chapters 1-6? Can't find a sign of one anywhere. There's a Hard Mode LTC, but the Chapter 2 strategy, if even possible on Lunatic, requires extreme luck with a lot of dodges against 90% hit rates and whatnot. So, does anyone have a real Lunatic LTC strategy?

    Gwimpage's breakdown of his first serious try is here, and he noted various spots in his reddit thread where he could improve.

    The C2 strategy XeKr mentioned is pretty much it--I think you can reset for post-EP1 enemy formations that can get you more EXP, if you have the patience.

    Assuming no dodges, C3 will require either a) 8 Def on Corrin, so that the lower-right Archer won't target him, or b) +1 Str on Jakob, presumably to KO the Samurai with a Dual Attack from Gunter (I haven't explored b) at all). The idea is to kill one of the three enemies beyond the bridge and manipulate the other two into a formation that doesn't block your way. You can get a four-turn clear if you crit the boss with Gunter, or a five-turn clear with Corrin taking the kill if you choose not to crit. Given the low hit rates, it's not the most reliable strategy either way.

    C4 is pretty ugly. I expect you can still get a five-turn clear by triggering the middle Dragon Vein on turn three, as mentioned in your reference (if you don't have a +Spd boon, you'll probably need more crits on Lunatic though). Previously I got a six-turn clear, triggering the middle DV on turn four instead, using a Corrin strong enough to ORKO every non-boss enemy and relying on her baiting the Sakura-bound Faceless by triggering the final DV with a Dual-Guard-loaded Corrin in range of death. This wasn't reliable since a) if an enemy missed prior to that, Corrin wasn't in range of death, and the Faceless would target Sakura instead; b) if Corrin scored a crit at any point, the DG-meter building changed from what I expected, leaving her without a fully-stocked Dual Guard for that particular Faceless.

    Without ninja dodges, C5 in five turns requires +1 Res on Kaze, and either +1 Spd (to double the Mercenary) or +1 Str (to enable Azura to finish off a boxed-in Dark Mage during EP3). Assuming the +1 Res is fulfilled, if you have both +1 Spd and +1 Str, you must not attack the Mercenary straight away because the DG meter will not fill up in time to block the Dark Mage's attack; if you have +1 Spd and +0 Str, you must attack the Mercenary; if you have +0 Spd and +1 Str, walk one step further than the periphery of the initial range of the enemies. The idea is to finish the bottom cluster of enemies during PP3 and EP3; and to next have Kaze take out the upper two Dark Mages, and, if the boss is not within KO range by Corrin at the start of PP3 PP4, the bottom-right Dark Mage. Ryoma must survive long enough to get saved by Corrin so that he can take out the bottom-left Dark Mage.

    C6 in two turns is easy and not unusually unreliable once you notice the Dragon Vein near Xander.

  11. Never really seen the appeal to Paladin Jakob either; just go let Corrin take some abuse and Silas turns into a Paladin.

    I'm only at C10 right now, but Jakob's been very good so far as a Paladin.

    C7: if you're playing with Ranking awards, Jakob may have to be a Paladin to get the seven-turn clear. Not sure though since I haven't played online.

    C8: with +1 Str and +1 Spd from base, and a C-level Pair Up with Corrin, Jakob ORKOs the boss on PP4 with 100% displayed Hit given the appropriate tonics. iirc the odds of getting at least +1 Str by Lvl 4 are about 80% if you do the binomial calculation, while getting at least +1 Spd happens about 87% of the time.

    C9: Jakob can ferry Corrin to ORKO the lower-right Archer on PP3, then switch to Jakob to break the wall and receive a Song from Azura to ferry Corrin to OHKO the Enfeeble user on PP4. The defensive support Jakob provides (Pair Up + personal skill) is much appreciated for tanking the EP3/4 onslaught, and even allows Corrin to tank a hit from the boss without healing if you're going for the unreliable five-turn clear.

    Long term I can see him falling off, but the combination of high Movement, decent damage output thanks to Elbow Room, and good EXP gain is too much to pass up early on.

  12. Good idea with the EP1 Dual Guard on C5. I previously thought that the most reliable six-turn clear of C5--namely, one without a WTD attack from Rinkah--required +1 of either HP, Def, or Res from Kaze's C4 level-up to tank C5's RHS enemies on EP1, but it looks like +1 Spd will also do. Corrin needs bulk that cannot be too far below average (I think yours is enough), but fewer than 1% of Kaze (Ninja) level-ups will leave all four of these stats unchanged.

    edit: ack, thought about it a bit more carefully and realized that the EP1 DG alone won't do for an axe-less six-turn clear, and that reaching the stat benchmarks for an axe-less six-turn clear is probably less likely than connecting one or two axe hits from Rinkah. Regardless the aggro DG-meter build looks like it should be a useful tactic in general.

  13. How much is known about this game's AI? I've been playing the earlygame and feeling a bit confused about what the AI prioritizes in its decision-making.

    Is there a team in "I"?

    After hearing from XeKr that a five-turn clear of Lunatic C3 could be done with an 8-Def Corrin, I rigged myself a strong Corrin and tried placing Jakob (7 Def) and Gunter on the top half of the bottom bridge to KO an Archer, and found that the Samurai targeted Corrin, and not Jakob like I expected (which would have ruined that five-turn clear since the Archer could have moved to initiate a near-fatal Dual Attack). This is consistent with the idea that the AI does not search for fatal combinations of individual attacks, let alone fatal combos involving Dual Attacks, during the enemy phase.

    I have wondered if the order in which enemies attack (which arguably matters more than ever because of Dual Attacks) can be determined in a way similar to what we saw in FE12--highlighting an enemy and cycling through all the others with the L button. However, I've seen at least one case in Fates where the AI broke this L-sequence to deliver a DA that wasn't fatal, so I'm not quite sure what to think.

    Passing up big damage

    On PP1 of Lunatic C5, I placed Corrin and the Kaze/Rinkah pair so that each was in range of both the Wyrmslayer Mercenary and the Dark Mage, with Corrin in clear danger of getting killed during EP1 from the two enemies ganging up on her (Kaze had enough Res to survive).

    You might have expected EP1 to feature the Mercenary attacking Corrin, who would've eaten 20 damage at 77% displayed hit, but he instead attacked Kaze/Rinkah for 10 damage at 74% displayed Hit, something that wouldn't have happened in, for example, FE12. I assume the Mercenary did this because Corrin was capable of OHKOing him back during EP1, and Kaze wasn't.

  14. Good stuff, this should be fun to watch.

    I will leave it to somebody a lot more patient to go for a sub-7 turn of chapter 4.

    I've been struggling with this myself on Lunatic. I was able to trigger the DV during PP 6 and get the Sakura-bound enemy to attack a Dual-Guard-loaded Corrin/Kaze pair during EP 6 because Corrin was within KO range (obviously the enemy attacks Sakura otherwise), letting me end the rout within six turns. I don't know how to reliably time the DG-meter-building alongside Corrin's HP depletion properly for this level, but this may be the key.

  15. Good stuff, looking forward to this.

    I believe this [four turns] is the lowest that is possible [for C2] without doing something extreme like fishing for criticals (which can be hard to do).

    If +Spd Corrin goes two-for-two on her first two level-ups (0.7^2 = 49% odds), she can double Rinkah for a three-turn clear. I didn't explore it very much, though, so I don't know how much it depends on enemy positioning. With a male Corrin, because Felicia is so weak, the only three-turn clear I found required the axe guy to move on turn 1 six squares away from where Gunter would end on turn 2, but there might be more reliable approaches.

    I do not think a 5 Turn is possible [in C3]

    Shephen and Gwimpage on reddit have both found five-turn clears using Felicia and Jakob respectively. I'm not sure how they did it, though; the only five-turn clear I could get on Lunatic (Jakob) required a crit on turn 5.

  16. Marth has 10 base Speed. 6 Speedwings gives +12 Speed, so Marth would end with 22 Speed (+2 with Emblem). That leaves 24 Speed. Medeus of course has 30. Even after a Rainbow Potion - 26. You can escape it with a Speed Bond and Rainbow Potion I believe, but even so that just shows how demanding of resources Marth really is.

    Are you sure that Bonds stack with the RP? I'd test this myself if I still had the game. Not that this is the biggest obstacle to a 0% H3 run, of course...

  17. As others have mentioned, it's necessary to eventually share your suffering with someone you trust. If not yet prepared for this, just break up the problem into smaller, easier problems and approach them day by day. Escaping/blunting consciousness through napping or video games may refresh you enough to tackle your problems with a clearer head, but doesn't directly solve anything.

  18. I imagine some of the first complaints that come to mind about Smash not being a "real" fighting game are that the directional-influence mechanic precludes long flashy combos, and that the mobility options trivialize the game in some (vague, undefined) way. But seeing as fighting games are, at their core, strategy games with dexterity requirements, DI arguably enriches the game by allowing for player-player interaction in the midst of a combo (unlike in a lot of other fighting games where the defender is stuck in any properly executed combo if no Burst is available). Added mobility options allow for, among other things, more attack angles and such, which are nice.

    The execution barrier is also lower in Smash than in Street Fighter, but I mean, who cares?

  19. although technically the mathematics required for both mechanics and em are pretty complex, intro courses require only strong comprehension of single-variable calc.

    AP Physics C is slightly easier than the typical freshman mechanics+E/M sequence; students who learn what an integral is (which happens very early on in AP Calc BC) are all set for AP Phys C. I took Phys C concurrently with Calc BC and had no difficulties with the math, and I say this as someone who is pretty bad at math; all the problems I had in that course were due to me not paying attention and misunderstanding the physics.

  20. AP Physics C is much less time-consuming than AP Chem and may be a better fit for you this year than AP Chem. The first term of Physics C revolves around F = ma, and the second term revolves around the four Maxwell equations; while it demands slightly more problem-solving skill than AP Chem, AP Phys C should be an easy class for someone like you provided you pay attention.

    Doing really well in AP Chem requires a bit more work in that the scope of the class feels broader (forcing you to memorize a bit more information, some bits of which can feel disparate given the light treatment of the material). And at least for me, homework (read: lab reports) were/was much more time-consuming in AP Chem than in AP Phys C.

  21. Robes are somewhat less desired across the cast than speedwings I thought. With a robe I'm pretty sure Draug can scrape a level somehow in chapter 1 and the start of chapter 2 to get to 12 speed as a hunter, then with RP he can double a low rolling 10 speed dracoknight in chapter 2, which is a ORKO with just Irons and base str + RP. That is some very immediate cost saving, although obviously I don't really know how that plays out in the long run. Anyway I think the robe arguably has the same positive feedback loop it has on Linde, the level he'll get in c2 resulting in speed again means he can ORKO all the c3 Dracos with Iron + RP too so you still don't need a bow forge and so on and so forth. Since it's CR that's a lot of EXP and chance to build bow rank.

    What level is Luke coming out of Prologue normally? I think he needs 4 over his base to be able to do the same thing, that seems unrealistic without a speedwing.

    I don't disagree that Speedwings are more valuable than Robes or Arms Scrolls, but since the four growth units currently populating the middle of High Tier each ask for a fair amount of boosters, I wouldn't say that Ryan is noticeably worse than the other three, as -Cynthia- has argued. In the post-Prologue chapters, I like Draug a bit more than Ryan for the former's ridiculous Spd enabling things like what you mention.

    Fair point about Luke's Spd, but since PKL has shown you can eschew the bow forge without loss of turns, it's not much of a stretch to feed one of the Alteans an early booster.

×
×
  • Create New...