Sunday, September 9, 2012

QB Battle 34.8

Mirrors: App + Media + Extra + Fanmade (update)
Sendspace: App + Media + Extra + Fanmade (update)

Corrected a first batch of bugs, and to celebrate 200 characters, added the fanmade Panty & Stocking (in adult form), plus SSF4's Juri (thanks to Rumormonger Baofu):


Plus Masamune Shirow's Juliona and the WIP character Ayra (thanks to Ayrawinla):


Coming next, Persona's Mitsuru Kirijo.

Also, there's a Grimoire section in the official QB site now, but no news for the moment.

56 comments:

  1. Wow thanks a lot again! I was especially hoping for the Panty, and Stocking character!

    ReplyDelete
  2. In regards to an earlier discussion about Claudette's Thunder Clap, would it make more sense for Cattleya to have augmented the sword with lightning gathering capabilities? Since it is proven she got it from her, and that she used the magic gems to make it stronger, i'd say the power to do so is within Thunder Clap itself.

    ReplyDelete
    Replies
    1. I guess that was in dire need for some digging... and everything seems to point to that, actually.
      From what I gather from the guys at the japanese wiki, the magic power of lightning was added to the Thunder Clap Claudette used, and she excelled at using it in her fights.
      In fact, they say the Thunder Clap was the collaboration of Owen's blacksmith skills and Cyan's alchemy to insert magic into it. Apparently, Cattleya was only involved with selling it, and periodically performing maintenance, perhaps.

      Delete
    2. Japanese Wiki?
      that actually makes sense that way, with the fact that Cyan and the couple were all good friends and whatnot.

      Delete
    3. The wikipedia's QB japanese page, I thought there was no more juicy info there, but I guess not... ^^U

      It also seems to be the most reliable (albeit small) place on QB info, they don't have a fancy, exhaustive wiki like ours :D

      Delete
    4. Oops, I mean the most reliable "japanese" place, obviously :p

      Delete
    5. you're right about that one, at least we can sorta say we have the most extensive english wiki out there.

      Delete
    6. I suspect that I'm going to be taken to school here, but my impression was as follows.

      While I don't know much about Thunderclap as it is during QB, I thought that Claudette's QBR profile said that it had been augmented - by wizard stones, which I assumed meant that the augmentation was the work of Ymir - and no longer had to "charge up" to fire lightning.

      Delete
    7. Hello Alzrius! nice to meet you, im NinjaRenegade, one of the Admins over at QBWiki! But in regards to your theory, with the way we have it, it seems that the Thunder Clap already had the power to do so and she can now call upon lightning without using her sword.

      That leads me to believe somewhere along the lines two things may have happened: Ymir found a way to utilize the Wizard Stones for elemental purposes, or Ymir found out how Thunder Clap was able to fire lightning(Reverse Engineering/Forging?) and applied it to Claudette's Wizard Stones, in case someone ever disarmed her.

      Could be wrong though.

      Delete
    8. Well, when Fabio originally translated her Rebellion profile, he did say "without even readying her sword", but it seems the kanji there meant more like "manipulate"/"operate", that is, "without using her sword".

      The new Thunder Clap seems to have been remodelled with magic stones, and Claudette seems to have powered up as well, so in a way, you're both right (we know we can't underestimate Ymir's wisdom in Rebellion ^^U). Still, Claudette's power-up might be more related to what was revealed in the 3rd volume of the stories, though...

      Delete
    9. NinjaRenegade, thanks for the cordial greeting! I've been spending some time reading through the QB Wiki recently, and I'm quite taken with it, to the point of *almost* being read to start making some entries of my own for it.

      Just be warned though, when I do you'll probably want to look through and clean them up, since it's been a while since I've participated in a wiki. ;)

      LB, thanks for clearing that up regarding the translation of Claudette's QBR profile. I have some proficiency with Japanese myself, but not so much that the (written) language comes easily to me.

      I had been wondering for a while (and even asked on here) if Claudette (in the original QB) was able to use lightning without her sword, and the general consensus seemed to lean towards that her lightning came from Thunderclap. It's ironic that she may be able to do so without the sword in QBR.

      Delete
    10. I always check and revise pretty much everything that comes on it, just to make sure it looks alright, so no worries there. and feel free to contribute to it, there's definitely nothing wrong with that.

      I'd also ask you to sign up if you haven't, it'd be easier to tell who you are and give credit for your edits.

      Delete
    11. No worries, I made a wikia account several years ago for a different wiki (d20 NPC), using this same name, and I'm much too narcissistic to remain anonymous.

      Delete
    12. well alrighty then, i'll look forward to having a chat with you whenever you feel the need to drop by.

      Delete
  3. Thanks for putting the least horrible-looking page for Ayra in the preview :) I've been low on QB-dedicable time in the last two days so there's been nothing done on my side, except for reviews of the new characters... And since I'm still low on time I'll cut it short and just put the reviews in:


    Juri: BUG: Violent Stomp goes to Fuhajin High page instead. Art is pretty darn good! Strength-wise, she's certainly one of the best street fighter. She's more durable than the majority of them and she got a good base damages and a great scoreless block (Kasatushi). Her jumps are also pretty decent, on top of her multiple other sources of bonus damage for the following turn. Won the 10 win challenge with 0 losses.

    I do have one complaint however: She feels clumsy to play due to Fuhajin. It's a neat idea and works well, but there's a few issues in my opinion. The first is that you have to answer the promp every single time you do a Fuhajin. It's on the basic Side Swing move, which is one of the most commonly used option in a fight, and since it's also an amazing move for her, you tend to do it more often. Secondly, you know the result before picking: Samus missiles do work the same, but there's two key differences: A) Samus doesn't get the promp when she miss (which makes it a lot less annoying) and B) She has limited resources so it makes sense to sometime not want to expend them, and you get no benefit on block/parry. Right now, for Juri it gives either a +2 guaranteed damage if you hit a non-parry or a guaranteed +2 later otherwise. It's win/win all the time with that move.

    So I think that move would definitively be more balanced if you picked before you know if you hit/what you hit. If you prefer having it stay this way (or it cannot be changed easily), then I'd HEAVILY recommend making it auto-hold if not on a score page: It'd at least reduce the promp annoyance considerably! Beside that, big thumbs up!


    Panty: First, what game are those from? I can't say I ever heard of it. Artwork style is kind of inconsistent, but individual drawings are quite nice. Strengthwise, they're pretty solid. Decently durable (13bp, 2nd best Daze in the game, only drawback is the 1s in unbalanced states), pretty decent base damages (an innate penalityless 4 and a few 3s) and Catfight Club can be extremely useful in the right situations..... But I have to give two ratings to her. Against an AI, I'd say she's pretty good. She's really a slugger-type and she does quite fine at it. She won the 10 wins challenge with 2 losses (for what that means). Against a human player... She's quite possibly the most annoying thing ever. "... of the dead" absolutely cripples some characters. Get hit by that once (even on a block or parry), and you find yourself having to play with a tiny moveset for the rest of the entire match. Against an AI or as the user, it's a great fun innovative move. But if you are the receiver, it's a fun-destroying move especially if the match is long.

    ReplyDelete
  4. I might as well be fair and review my own... Most likely biased!


    Ayra: Artwork is TERRIBLE! Some people are really shameless to put up something like that. It's consistent in it's awfulness, though. Against a human player, strength-wise she should be pretty decent. Great long-range capabilities, acceptable durability and some okay tricks at close range (especially incendiary bomb). Against the AI, she's probably overpowered. Since the AI loves using Charge so much, if you do Quick Sling Low on first turn you are nearly guaranteed a critical damage wound. 2nd turn Load (using 38 preferably), 3rd turn Heartseeker or maybe Incendiary if you are feeling risky. Congratulations! You just beat the majority of AI characters without breaking a sweat... Won in 10 without a loss, usually within 3 rounds. The AI also gets decimated by Incendiary Shot since it cannot understand how it plays. There are an handful of other characters that can chop down AIs as fast or faster though (Alleyne, Alice, Claudette, Annelotte, Nyx, Aldra R, ect), but it's rare... I'd suggest me to make her matrix a bit more risky at close range, since other changes I can think about would require a full character concept remake... An "old" Lost World style matrix should do the trick in making her weaker close range as intended.


    Juliona: BUG: If you do Eternal Optimism and the enemy does Away, it goes to the TNT page instead. Awesome artwork! Beautiful and extremely consistent (obviously). Strength-wise, great long range and very bad long range. Base durability is around average. She does have a 2bp heal, but it's not spammable, and using it leaves her with only limited dangerous options. Great base damages when mounted, further boosted by various conditions (though she cannot force them to happen). Has exceptional abilities at finishing off enemies with I am the Law, Villains cannot Hide and Spur Kick. On the other hand, multiple of her moves do cause her to take more damage so using them can be risky, and a lot of her moves leaves her with few options especially if not mounted. Eternal Optimism is an awesome damage-reducing move against low-hitting characters and is a very handy addition. Luck of the Draw is far too unreliable to be counted on, but it can be used as a scoreless block and hitting the jackpot can be fun. In my personal top 5 most fun character to play. Won the 10 win challenge with 1 loss.

    ReplyDelete
  5. I thought Juri's continuous prompt would be an issue, but since I was using the same "selection box" for both styles and hypnosis (though only Veltigg used hypnosis until now), I couldn't figure out how to have both. Though the display is kinda cramped there, maybe I can squeeze both in and stop asking all the time (or at the very least, not asking when not scoring).

    Panty and Stocking are from an anime series with that same name, that usually has a kiddie look, is highly reminiscent of the Powerpuff Girls (maybe you know that one? You should, it's awesome too) and sports a lot of references to other shows. To add a few references of my own, that's why I made them with 13 points (bad luck), an attack directly called "Climax" and two attacks named after chapters from the series (the two you highlighted). But I digress...
    I didn't think "Of The Dead" would be that bad. Rogue takes both Down Swings and Skills as well with no penalty, so even if you take the enemy's Skills, both parties losing 28 and 30 felt kinda balanced (I thought it couldn't be worse than Elina's critical-causing attack).
    Also, someone should mention that like Dizzy, they recover both weapons at once, gotta bring barely-used restrictions back into fashion :)

    ReplyDelete
  6. I'm glad you like Juri. I'm hoping her Ultra system mechanic can be applied to the other characters that have Ultras (Cammy and Chun Li)

    Basically, the more damage you took, the more Revenge meter you got, and the more meter you consumed for the Ultra, the more damage it dealt. A maximum of +5 for having less than 10% BP makes sense.

    I took the either/or route because Juri has two different Ultras that do two different things (Well, technically Cammy has an Ultra 2 that's basically a counter with the properties of a focus attack. A hard counter that deals massive damage if she is attacked)

    ReplyDelete
  7. Yeah the ultra system does reflect a bit better the base game (more meter = more damage). The either-or for specials is fine since it's not overpowered, and picking at the start would just limit your options (and since the moves require you to be at low bp, there's no need to restrict them even futher).

    Leecherboy: Hmm? Elina's critical attack? You mean Whisper Evil Things? I thought that restriction to be mostly useless in her case! Here's why:
    1) It only applies on Daze
    2) It's on a move number that scores not terribly often (44) and can score on a lot of other things beside Daze
    3) It's on a +3 move that gets boosted by a ton of other moves, so it reaches 4 or more extremely easy and +4 already deals crits.
    4) A +3 score move on a Daze page actually insta-kill a decent amount of characters so the critical restrictions on Daze doesn't ever apply for a lot of characters.

    I play Elina a whole lot (she's also on my top 5 favorites) and I had that restriction applying maybe twice in total. I do agree that it would be pretty overpowering if it were a "apply critical restriction on any page" and that it was placed on side swings or thrusts though :) It'd be nearly as overpowered as Sigui's Bind with Chain move.

    Of the Dead isn't overpowered in my opinion: It is a good move but since it is on a not overpowering character it's fine. It is incredibly annoying to get hit by it however, since it is a huge hard lock on your choices for a whole game. Beside the annoyance factor of a human being on the receiving end of the skill, it's perfectly fine. Rogue doesn't count by the way: She's a marvel :)

    I kind of thought that recovering both weapons at once was the default! But after checking, it turns out that it's not... There's definitively not too many multi-weapon characters!

    I got my new books already! Yay for unusually fast shipping (It seems that it's faster to get mail from Japan than it is for intra-country mail somehow). Anyway, Pieces 9 is pretty nice in my opinion with lots of interesting pictures, and it's also much less "hard" than Pieces 7 and 8 are (ie. The tons of images of Juliona and Shion that I can't possibly use ever even with tons of censoring) which is good in my book. What does this mean? Well, I mentionned last week that I might do a new Shion and that I considered that bad news for you. Now, using the same metric, I have really, really bad news :) Also, there's a mention on the stuff packed with the book that mentions that there MAY be an Hellhound 2 that would come out (so more Shion). For one, I'd be completely okay with that!

    I also got W Tails Cat featuring Cyril, and... It's mostly lost money. It's expensive, it's pretty "hard", and it's completely unappealing in style. It feels incredibly lazy and "soulless" if you will. The worst thing? There's maybe an handful of pictures I could use at best for QBBattle... With Pieces 2 it's very likely enough to make Cyril, but the book crappiness would have stung a bit less if it had been of some use at least...

    You wouldn't have happen to have the 2012 Shirow calendar (Saber Tooth Cats 4) by any chance? There's two awesome Shion pics in there I'd love to use that aren't seen anywhere else, and the only stuff I've found on the internet have a big "Sample" written on them... I don't own that calendar, and while I would to use those images, I'm certainly not going to buy a calendar for a year that's more than halfway over just to use two images :)

    In any event, Kat's going to be finished before anything else of course. I think I got everything thought up, but I might have to ask a few questions depending on bad it craps out when I try to playtest it up!

    ReplyDelete
  8. Right now I'm working on Chie. Persona 4 character, same series different game. She will adhere much more closely to the base game of Persona 4 Arena. Yes, I know, another hand-to-hand type character, but I hope to include her Persona in her attacks more often this time.

    ReplyDelete
  9. Ah, so Elina's critical thingie isn't that useful? That's a waste of a perfectly good restriction, we're going to have to think of someone else to use it properly, lol.

    Looking forward to more pics from Shirow even if it's the same character with a different job :) And nope, I hadn't ever heard of that 2012 calendar, sorry.

    Also, if you were kinda wondering how I'm releasing the chars Baofu gathers seemingly out of thin air, we have a Dropbox thing going on, it's a neat thing to have where lots of fixes and corrections are involved.

    ReplyDelete
  10. Hehe, you spoiled the surprise. A magician, especially one who's a Persona fan never reveals his secrets. :P

    She makes a good point though. Elina's move does enough damage to land a critical hit on its own and most of the time you won't land a Daze on a 44 unless they picked 18. Still, that'll give me an idea for the next character.

    I forgot about Sakura as well for applying the new Ultra system.

    Angel plays nicely. I test characters and the game itself differently by playing challenge mode and gaining 1000 XP on each character. Changes the dynamic really. Good characters become broken and "bad" characters become good.

    ReplyDelete
  11. To be entirely honest, I wasn't too mystified about it: I just assumed you were using e-mails or something like that :) I can definitively see the use of the drop box, but I don't think I'll bother with one at the moment since we don't have that many transfers ongoing... If that changes for some reason I'll look into it though!

    I don't have any short-term plans about using the critical-causing restrictions but I'm definitively interested in seeing where you guys will use it more effectively!

    For the record, I'm never going to take account bonus exp when making or balancing characters, so if any character I made become absolutely broken with bonus exp, I claim not guilty! :)

    I never played any Persona game, but now that I have a Vita and that Persona 4 Golden is comming out soon, I was thinking about giving it a try. Is Persona 4 a good place to start?

    Oh, hypotetically speaking... Would it be possible to make a dual bp character where
    1) if either BP bar runs out you lose,
    2) where the character who takes damage is the one matching the previously used move
    3) where using the same character twice in a row would result in that character losing 1 bp?

    Since it's all hypotetical I haven't studied the help files or anything (I like thinking base character concepts based on images available), but I think it'd be a nice idea that'd make sense thematically and would lead to a character you need to spend a bit of time thinking while playing her... But I don't think it's actually possible to implement right now. I can see condition #1 being doable with an Always and an Eval clause, #3 being partially doable with making a state based on who did an action last but you can't do "splash" damage to your 2nd BP bar, and #2 not being possible due to which character taking damage is already pre-defined in the matrix file. I suppose it might be sort of possible to implement #2 in a roundabout way if it were possible to deal splash damage to the secondary bp: This way you could have all attacks dealing damage to the main BP bar, but if the state is set to secondary, you'd heal up the amount of damage taken and deal the same amount to the second BP bar... I THINK that might work, maybe...

    Oh! Question about Kat I wanted to ask before and forgot to. I want Kat's Gravity Shift effect to only end AFTER her attack is done. Since I wanted her to use her height and everything before it gets reset by the clauses in the Sheet file, I used aftercheck so that it doesn't get reset before making it's use but you said to use Always instead. Would Always still apply after all my regular restrictions are used?

    ReplyDelete
  12. I'm an old-school kind of guy so I'd suggest you try and hunt down Persona 1 and Persona 2 for PSP. The originals on PS1 are better but they are kinda dated. I'd play them in order, myself. Preferably on a PS1 or PS2, but the PSP remakes are good too.

    ReplyDelete
  13. The only thing I know about Persona myself is what I gathered from hiimdaisy's spoilerific strips (the MGS ones are hilarious), and what I see from the new fighting game.

    I guess that dual-BP character could be done, yes. #1 is doable right now as you say, #2 would need a "force choice" restriction (you know, to not let you choose who takes the damage) and two long "move xx-yy-zz-aa-bb" restrictions, and #3 would need states and I guess an improved "lose" restriction that would let you say "lose 0-2" and your second BP losing 2.
    (Also, before I forget, as you can see from Juri, the Eval and Result now accepts decimal numbers, in case you want to do percentages and stuff).

    About Kat, I think I recommended "Always" because of the "Enable move" types, that kinda override the movement restrictions before it. Those should have "Aftercheck", and maybe the rest could have "Always"... unless one depends on the other and you get an entangled mess of restrictions :p
    In which case I guess you'd need to have "Always" and use "counters" like the Ougon guys...

    ReplyDelete
  14. Sorry, kinda messed up there. I meant, use "Always Enable" (or disable), and try "Aftercheck" for the others.

    ReplyDelete
  15. Okay. I'm starting to get pretty frustrated with Kat... I can't get anything Height related to work at all. If I use
    Instant change height +1
    it crashes the game. If I use
    Height < 9 Instant change height +1
    it brings the Height to 8 instantly. It also doesn't work for values greater than 9.

    The ONLY thing I managed to get working is
    Change height 1
    which unfortunately takes 1 turn to activate bonuses, and it's not remotely enough to implement her. If you could give me some restrictions to get started I'd definitively appreciate because right now I'm certainly not having much fun... I need things alongside these lines:

    1) Setting height to be equal to some base value. For example, if knocked down while Gravity Shifted. Example:
    Change height = 4

    2) Setting height to be equal to some evaluated base value. For example, losing half your bonus gained height when pausing on a building to recharge. Example:
    Change height = Eval + 4 / 2

    3) Using your bonus height in damage equations, so something like 1.5 times your bonus height in damage.
    Score Extra Eval height - 4 * 1.5

    4) In the -s file, I want to cap Height to 10 (else damage could get too insane). Since there's some moves with gains greater than 1 point, it'd be easier to put that into the -s file...
    Always Height > 10 Change height = 10

    So yeah. Kat's status is completely stuck right...

    ReplyDelete
    Replies
    1. Sorry about that, the height might not be polished enough, true, not many chars are using it (and I guess change height stays in effect until it reaches the limit, that's why it crashes it there isn't one :p).

      I'll see what I can do, I'll get back to you on that (also, equations? Can't say I didn't expect it sooner or later ^^U)

      Delete
    2. Sorry to disapoint you, but the next two characters won't be using any equations (But they'll have their own nifty stuff) :) After that... who knows!

      I think I'll just use special 2 instead... I'll simply boost her height to 9 when in gravity shift mode for flavor sake (as she does roam higher than dragons or dinosaurs easily), and use special 2 for her bonus altitude. It'll have the advantage of not having to substract 4 in each equation (which seems impossible to have both a - and a / in the same equation without crashing) and also automatically keeps the value between 0 and 6. I'll give that a try and tell you if it ends up working a better...

      Delete
    3. Well, it is working substancially better. There's "only" 3 issues now that I can see.

      1) Resetting height when the meter runs out. This line in the sheet doesn't work
      Aftercheck State 0 Height > 4 instant change height -1
      My other Afterchecks do work, but none have a "dual check" like here.

      2) Taking damage when falling to the ground or when failing to land the Powerbomb move (so using Eval with Lose). When it scores, the move works perfectly fine. When it doesn't score, it crashes the game. Here's the condition:
      Score Extra Eval special2 * 2,Not Score Lose Eval special2 / 2,Disable move all,Change state 0,Change special 0/-6,No restrictions

      3) Reducing a special by a non-fixed value. I'd like to lose half the altitude on refil (So that it's better to refill before crashing, but not easy to reach max altitude) instead of a fixed value. It's not an absolute must, though.

      Also... in the help files it's written that you can have up to 5 clauses in the Sheet... Would it be possible to have 6? I can live with 5 but I'd have to copy/paste one restriction in like 20 pages no matter which one I pick.

      Delete
    4. Yep, that's our Ayra, nice job finding a workaround.

      1) Is it possible that you're aiming for a normal "change height" (not instant, for next turn)? Maybe that'd work just fine, even with an "always".

      2) Oops, I think the "lose" restriction doesn't currently evaluate (kinda sloppy, I know). I guess you can try a "regain" with negative value for now...

      3) Hm, with the special able to have two sets of values, the evaluations would look weird... for the moment, maybe you can make do with repeating the instruction a few times for each value.

      4) You know what, I'll give you 20 clauses effective right now :D (no, seriously, I changed that recently)

      Delete
    5. Kat's been nothing but endless frustration for nearly every move... I was still feeling like doing work on a character but not on Kat so I started doing a bit of work on another character I've had the images ready for about a week. Lo and behold! EVERYTHING worked on the character. There was a tiny amount of debugging, but everything went super smooth even though it's not a terribly plain character either. It's been a very enjoyable process and it did marvels to wipe off the frustration of plugging a ton of time on a character to still have a tons of things broken.

      Everything works great. Everything but the very last condition, that is. It's a pretty core skill, unfortunately... Basically, the special bar varies from 0 to 5, and the skill is only usable when you have 1 or more point (this part works). You get a better bonus with more points spent, but it often make sense not to spend all your points so I offer a dialog box to pick how many to spend. However, I have to make sure the person can't pick a number higher than they actually have...

      So what I came up with was multiple Result starting with 1 point (which has no dialog box) all the way to 5. So if you have 2 points, you get a box with 2 and 1. If you have 5 points, you get a box with 5, 4, 3, 2 and 1. The multiple results actually work: It shows me the appropriate box.

      There's two problems however.
      1) The box never displays more than 3 results, no matter what. If I have 5 points, I'll get a box with 5, 4 and 3.
      2) If you pick anything else than 5, you'll get bad values. For example, if you have 3 points and pick 3, you'll get +10 (instead of +4). If you have 5 points and pick 5, it'll work right (+4 for 3 turns). If you have 5 points and pick 4, it crashes with an Index out of Range.

      A bit of help would be appreciated in getting this working! Beside that, everything's perfect (will require playtesting though)... I'm pretty happy with this character, except for some of the censoring job. So when I send you Kat (probably over the weekend if I can finally manage) you'll get a bonus character too! Lucky you! :)

      Consider that as an apology for having to look at this monstruousity of a restriction... :)


      restriction[31]=No restrictions,Bonusonly move 9 Demon Lure,Result special + 0 < 2 Change special -1,Result special + 0 < 2 Repeat self for 2 Bonus 2,Result special + 0 < 3 Chooseoption "2 Points"|"1 Point":Change special -2/Repeat self for 2 Bonus 3|Change special -1/Repeat self for 2 Bonus 2,Result special + 0 < 4 Chooseoption "3 Points"|"2 Points"|"1 Point":Change special -3/Repeat self for 2 Bonus 4|Change special -2/Repeat self for 2 Bonus 3|Change special -1/Repeat self for 2 Bonus 2,Result special + 0 < 5 Chooseoption "4 Points"|"3 Points"|"2 Points"|"1 Point":Change special -4/Repeat self for 3 Bonus 3|Change special -3/Repeat self for 2 Bonus 4|Change special -2/Repeat self for 2 Bonus 3|Change special -1/Repeat self for 2 Bonus 2,Result special + 0 < 6 Chooseoption "5 Points"|"4 Points"|"3 Points"|"2 Points"|"1 Point":Change special -5/Repeat self for 3 Bonus 4|Change special -4/Repeat self for 3 Bonus 3|Change special -3/Repeat self for 2 Bonus 4|Change special -2/Repeat self for 2 Bonus 3|Change special -1/Repeat self for 2 Bonus 2

      Delete
    6. Sorry to hear that you're still having a hard time with Kat, hope you can figure it out before it gets too overwhelming :S

      And woah, quintuple choice? You don't know the meaning of "holding back", right? :D
      Sure there are only 3 choices maximum now, I couldn't even begin to imagine anyone with more. Since 5 lines could look too weird if you are only showing a number in each, maybe I'll make a new restriction to choose numbers side by side.

      Delete
  16. Great work as always. You know who would make great additions? She-Ra and Teela.

    Maybe I'll try and and work on these on my freetime. Unless someone more talented can take a stab.

    ReplyDelete
    Replies
    1. I kind of know who She-Ra is (I know she's from He-Man and I have a vague idea of what she looks like), but I have no ideas who is Teela. I'll pass on working on them though because

      A) I'm not familiar with the characters so making a fitting moveset is hard

      B) I already have... hum... 7 movesets lined up to do right now in total so I don't think it'd be wise to add two more

      C) I'm most likely not more talented than you are :)

      Delete
    2. She-Ra and Teela, huh? Besides me barely remembering a thing about He-Man, I'd say they aren't at their peak of popularity right now... but DeviantArt does have an impressive amount of fanart of them, surprisingly.

      lonecow, even if you don't quite talented enough to make a book out of them, if you could wade through all those pages and pick the best, varied pics, I can always help and finish the job.

      Delete
    3. Awesome. I might just need advise on which book to try and modify. Teela is a warrior but has some magic abilities as she is supposed to be the most powerful Sorceress one day. So close range fighter type with weak long distance magic.

      I'm a Graphic Artist by trade so I have some confidence in my abilities, it's just finding time to work on it (due to new baby more than work schedule). But I am better at Photoshop manipulation than actually drawing, so we will see. It will be a fun side project and like you both pointed out, I think I've picked some obscure characters that I doubt someone will beat me to the punch lol.

      Delete
    4. You could even use one of the 90+ unused ideas for characters that a LW fan called Peters designed a long time ago:
      http://www.sendspace.com/file/hc8v7o


      We barely mentioned these like a year ago, but now that Ayra has levelled up lots in character designing, maybe it's the time to bring these up again. I just hope no one insists on the really funky things, like the double matrixes or the summons ^^U

      Delete
  17. Kat wins my official "Meh, good enough" seal of approval... Her core is working and she's usable. Her representation is surprisingly close to her game's playstyle and I'm happy with that: Get distance while evading, and charge in for massive damage if you started far enough. It makes her playstyle a bit too clumsy and she's probably the character that'll perform the worst if you aren't careful (Watch your gravity gauge constantly, for one). That makes her incredibly bad when used by the AI (Summoner Shion level), but no miracles to do there...

    However, she has bugs... Bugs that are not consistent and imprevisible. For example, sometime it will take a turn more for her Gravity Shift moves to get reactivated, sometime it will be instant. Sometime, her height will "freeze" at 0, other times it will switch as wanted. Sometime, she'll get something like a +10 on all Oranges, when I don't have any orange boosting move. Sometime, she'll get "stuck" on a move. As in, move 24 has no link to page 25 in the matrix ever, but I saw once that she kept going to page 25 each time I did move 24... But that didn't happen again.

    All of the above are rare: Like one of them will happen about once every 3-4 matches or so. If you can debug that, it'd be nice, but I suspect it'll be really, really hard... She does work currently, just not always as expected! Oh, the "Gravity Kick" (page 25) states that it adds Altitude * 1.5 score, but it crashes when I use it. I thought we could use decimals now? Anyway, I changed it to * 1 and added +1 base damage to it for balancing purposes, but I'd prefer the old way if possible.

    On a more positive note, Demon Tamer Shion is 95% functional. Everything works spot on with no issues I could find. The character is fun, sexy, works well, requires some thought while playing, the AI is sort of competent and I'm very happy with her (not Juliona or Seductress Shion-happy, but very close). I changed the monstrous condition to only offer 3 choices (still 4 triggers though) and cleaned it up and it doesn't crash the game or give wrong values now.

    The only bug I saw was that twice, the Octopus Helper move with 5 octopuses (which is +4 for 3 turns), I didn't have the bonus on turn #2. On both occasions, I had the correct bonuses for turn #1 and #3 and on turn #2 I had nothing (and the enemy didn't do any unusal move). I haven't been able to reproduce it at will unfortunately.

    You may notice that the Spider sequence is not present... That's because it's part of Priestess Shion. I did warn you that I had very bad news for you before :) All the images for her are ready (and censored too but few needed that) and the moves are decided (but not the specifics yet). She might not be next however: She's relatively complex, and after working on Kat and Demon Tamer Shion, I feel that I deserve a break and I might do Farrah next which is going to be very WIMPIL-like (so much easier to get everything working and balanced).

    Oh, one good news is that out of the images sets I have ready for multiple characters (... or variation of characters), Demon Tamer Shion is by far the naughtiest so if this one gets approved as is, all the other characters I have on my list should have no problems getting approved.

    I also included the fix to Juliona's matrix. Comments from anyone regarding the new characters are very welcome from everyone! While I do try my best to be impartial when judging my own characters, as the character creator I might very easily be misevaluating some things or missing some aspects completely.

    http://www.sendspace.com/file/qoan08

    ReplyDelete
    Replies
    1. I knew you'd manage to tailor Kat to our needs, and she doesn't look half bad there either. I'll see what I can do about the bugs, but the char sheet of both of them is missing the black moves? (Sorry if I'm kinda picky, it's a bad habit :p)

      Great work on censoring the naughty bits on Shion all around (I'm assuming, never saw the pics), yay for almost-naked pics ^^U. I'd complain about Fish Guardian and Unbalanced kinda showing a little nipple, but eh, if you don't feel like tweaking those a little, I'll still let them slide.

      So yeah, nice job there, and it goes without saying that you deserve the break. Also, you know that if you have any ideas on strategies the AI should follow to be more challenging, I'm all ears (just because it has the fancy name of "balanced" doesn't mean it can't be improved if needed to).

      Delete
    2. I... probably didn't change the text color when I wrote the black spaces.

      Demon Tamer Shion did indeed need censoring on over half the pages... Censoring quality varies from nearly flawless (like Extended Ready and Octopus Helper) to alright (various water splashes and Demon Charge) to horrible (like Leg Wound or Shuriken Toss). I actually had a different censoring of Shuriken Toss and when I was checking things before sending, I thought that it was so bad that it needed a complete different version. The current one is better, so imagine how bad the censoring for that page was :)

      For the almost naked, it was pretty much done on purpose. I scanned the new book and divided up the images and added the old unusued ones in the appropriate places... So this one got all the "Aquatic" based images, which by nature are less covering and had a few demon pictures in. So I took all the fitting Shion pictures plus an handful of demon pictures (the "Overwhelming Beauty" demon just had to be featured somewhere!) to make the demon tamer. I would have prefered if I had near-naked pages for the entire book for consistency-sake, but unfortunately there's an handful where the wanted image just didn't exist. It's maybe a bit shameful to admit, but one of the other criteria for my image selection for this book was "Nice behind... you go in!" Theme consistence is important! :)

      I was hoping that the pictures would be okay but I'll add a few more dabs of water splashes to the Fish Guardian and Unbalanced pages and send it to you later this week... To be fair, I was far more scared about Octopus Helper since I didn't have any sort of replacement image for it so I'm pretty glad! As mentionned before, I'm fortunately not major expecting issues with other currently planned characters as the spider sequence is the "worst" thing in the pipeline.

      Gameplay-wise, I'm pretty happy with her. She got a core basic moveset that's hers (The effect on each is just because I wanted each page to do a little something), resource system for the other pages that's unusual, a few unique moves (Cleanse All Evil and Set Traps) and she requires some thought if you actually want to use any of the demon stuff. Kat also has some unique aspects (even more so), but she plays just so clumsy that she's not as enjoyable to play even though she's "accurate"...

      In unrelated news, Juliona is now officially the first self-made character (others are Mirim, Echidna and Saber) which I ever managed to get to 1000 Experience points without spending any and without dying! Yay!

      Delete
    3. DAMMIT! I just lost my huge post about the AI and I don't have time to redo it all :(

      In short:

      1) AI do Charge way too often. It makes the Black moves nearly unusable (as charge hits it) and the white moves way too good. Even characters with no long-range attacks beside Charge should do it less often and use their crappy attacks to close range more often.

      2) The most commonly used Jump is Duck. Duck is the worst jump because of
      A) Lowest chance of success
      B) Highest chance of getting really bad damage pages
      C) Worst move limitation with usually small bonuses
      D) It gets hit by the most damaging attacks
      So in short: Dodge a bit more often (far more successes, no move limitation and you don't get hit by worse than unbalanced) and Duck a bit less often.

      3) Do 36 and 24 too often. It's not uncommon to see the AI do things like 24-36-24-36. 24 is a great move for the non "old matrix" characters, but not when overused especially when combined with the 36-using characters. Basically, it makes Duck nearly suicidal to use: I duck maybe 1 time per 5 battles at best (Up is only slightly better) and it makes me use Dodge over and over since it's the safe pick 80% of the time (number approximate). More side swings, less 24 and 36 (Especially for characters who have both 24 and 36).

      That doesn't hold true for all AIs of course: Those who have long range attacks tend to Charge less and are far more deadly for example, some tend to spam less 36 and 24 or have a priorized 48 (which hits most dodges so it makes it far less safe), ect.

      I'm not saying I do better with my AIs though! Ayra was designed to be a great AI-controlled character ever since I started building the moveset... And she's not exactly amazing in AI hands either.

      Resource-based characters are nearly impossible to be effective in AI hands, especially those like Summoner Shion where justicious gain and use of those resources is everything (especially when it costs BP to gain those resources). It's why I was happy that Demon Tamer Shion works kind of well: What saves her is her good amount of resourceless moves, unpredictability at what's comming your way (since nothing is highly priorized beside the 5/5 move which normally don't happen) and the fact that a lot of moves are really variable costs. It doesn't prevent her from occasionally doing Demon Lure when she's at 5/5 demon tamed, though...

      Delete
    4. Oh, I guess I do use Charge and strong attacks too much for the AI, guess I'll have to tone down those and change the jumps as you pointed out.

      But you know, I was kinda more concerned about what you say about AI not knowing how to use resource-based characters:
      -Do you think detailing what to do best with 1 point, then with 2 points, etc, would suffice?
      -Maybe it'd be a good idea to try and have the CPU "analyze" your moves somehow, and learn little by little from what you do (yay for behavioral AI ^^U)
      -Or do you have any other idea for the AI? If you can explain it more or less coherently, I could try and program it that way.
      Those would probably mean a little more work for me, but eh, it's not like I'd need to implement them right now.

      Delete
    5. Behavioural AI that analyzes your move choices? How is that possible?

      Delete
    6. Well, you know how in some fighting games (mostly 3d), the tactics you use in the first matches never work in the last ones, and you have to keep the cpu guessing... so I just thought maybe I could try something like that.
      I guess the idea would be something like: for each char you use, the cpu would observe which maneuvers you choose, your favorite combinations and how you play it safe when you're in a pinch, and try to learn and/or counter you... or something ^^U

      Delete
    7. Well, it's a good concept but I'm a bit wary of the implementation to be honest. It's something that would require fine tweaking and I can easily see that it could go bad and annoying very quickly depending on how it's tuned. If you are confident that you could make it work fine, go for it. But you might want to put in the menus an option to turn it off :)

      If you are willing to change/improve the AI system, in my opinion the main issues and recommendations are the following... Beside #1 they might not be too realisticly doable without too much effort though!


      1) Even with a new and improved system, I think it's important that no move is too heavily emphasised as previously mentionned so I still stand by my previous post content. It's good that the better moves comes out more often, but not to the point of using only the same 2-3 moves all the time, especially when the same counter works for all of them (namely Dodge in the vast majority of them). I already talked about that last post so I'll skip to the following one.


      2) Currently, the AI doesn't take in consideration at all what the opponent can do. The best example is if you are turned around: Your two choices are usually only Dodge and Away. Yet, the AI will still constantly do 36 at that point, even though it's 100% impossible that it will hit.

      Also, a good part of winning a QB match is about landing one attack (or the opponent using a move that restricts them heavily) and keeping the offensive up and only "resetting" if the enemy manage to get back their main offensive options. So if the player has to use only Jumps (or green and yellow), the AI shouldn't waste it's opportunity by doing, say, a 26 or a 4 (Especially since they are one of the extremely rare moves that get beaten by the yellow 40) or doing jumps. The only "sane" option in that situation are thrusts, sideswings and moves that share those properties (like 6 sharing thrust properties mostly for example), except for an small handful of characters who have special effects even miss.

      Currently, that problem gets worse when the AI is in the under 50% bp mark since it dodge and blocks more often. The AI might get you in a bad position, but then it'll an up that does nothing but limit their next turn option and allow you to recuperate. In other words, it should tweak it's set priorities based on the enemy options (No real offensive options available means the AI should be attacking) and favor moves that will currently hit the opponent instead of doing a good move on paper that has no chance to land currently.


      (#3 incomming in a bit)

      Delete
    8. 3) AI inability to use resources correctly. The only way I can see that working is if we can do some basic scripting regarding general move usage alongside with some conditions. The really basic point system like the Street Fighter characters work fine since they are just really "fluff" and not a core thing. MP-based characters tend to do okay-ish since you put some sort of MP regen on all jumps, but this means that they never can end up saving enough to use the bigger spells. If the resource is critical to a character however, then the current system cannot handle it at all.

      I'll use two examples of it, one a middling case and one critical: Both will be my own characters since they are the ones I've tested and tweaked the AI more intensively so that I'm more familiar with the issues.

      Demon Tamer Shion is middly affected by the AI restrictions. Since the minimum cost on most skills are extremely low and that she has a large quantity of "free" skills, she will function mostly like a normal character. However, there's still a few issues. Her most reliable way of regaining resources is by using Demon Lure, which costs her 1 BP and has a -6 attack score. Under the current system, she'll still use the skill when she has 1BP, which is a guaranteed loss but the move still has to be usable under 50% BP and there's no way to tell the AI that.

      The second problem is that what is possibly the best overall usage of her resource is the Octopus Helper move. It gives massive bonus over multiple turns (depending on how many octopuses you used): If you have the bonus active, you want to attack. Throw something at them: Even the basic thrust is going to do good damage. You have no restrictions, and you know that any trade is going to go to your advantage. Yet, you will often see the AI just doing Jumps or Demon Lure and completely wasting it's massive bonuses and resource expendiure away.

      The third is that the AI cannot recognize when not to do the resource gathering move. It will occasionally do Demon Lure when it is at 5/5 tamed demons. The move cannot do damage (except on the very first turn after Octopus Helper usage where you can't be at 5/5 and after a duck) and will cost her 1BP for absolutely no benefit. That situation is sort of rare due to assigned weights, but it is still an issue. Still, the character does function somewhat at least... Most resource-based characters have the same sort of issues as her.

      Delete
  18. Now, let's take the character that's most crippled by AI resource management: Summoner Shion (though some others are close, especially Kat).

    The basic concept of Summoner Shion is that she has extremely high BP (with utterly devastating score pages), and her "ultimate" spells deal damage to both her and her opponent. So in general, the basic strategy is to keep your BP higher than your opponent while you charge up one of the ultimates so that when they are used, they die and you don't.

    Her biggest tools to accomplish this are:
    1) Dweller Army: She lose 10BP (Though it cannot bring her under 0BP) and the opponent lose Dweller Points on score or Dweller Points / 2 on miss. The safest way to use this skill is to gather enough dweller points so that Dweller Points / 2 will bring the opponent to -1 or lower.

    2) Endless Assault: This moves applies 1BP damage per turn for both characters, and the move can be reused as often as you like for more damage each turn (You need at least 8 tentacle points to use the move, but it only costs 2 points). So the safest way to use this move is apply this once or twice, and play safely (you still want to attack of course as any damage you land will make victory come safer and only spamming jump is actually riskier than combining both).

    3) Heal: It's by far the biggest heal in the game, and it's often critical in being able to succesfully use Dweller Army or Endless Assault. It gives you back as many BP as you have tentacle points, you lose all your tentacle points and it forces you to jump. It's an extremely useful skill if you have lots of tentacle points stored up or if you need a bit more BP to survive a turn or two more of Endless Assault.

    ReplyDelete
    Replies
    1. How does the AI manage with all that? It simply doesn't, and it can't possibly do so with the current AI setup without being able to script things.

      Using Dweller Army when at low BP is a smart strategy: If you are at 15 BP, you lose 10 BP. If you are at 10 BP, you lose 10 BP and get KOed. If you are at 2 BP, you lose 2 BP and get KOed... But if KOed, you still win if your opponent is lower than 0, and your effective BP loss is lower when at low BP.

      However, if using the skill wouldn't bring the enemy under 0 even on a score, then it's completely stupid to be using if it is going to KO you. Also, while you always lose 10 BP, the damage goes up depending on how many Dweller points you have stored. If you use Dweller Army as soon as you have enough points (8 points), then you will lose 10 BP and will do either 4 or 8 damage which is not a terribly attractive proposition unless that's enough damage to finish them off. If you have 16 dweller points, then you still lose 10 BP but deal either 8 or 16 damage, which is far more beneficial. The AI can't figure any of that of course.

      If you want to use a "Put Endless Assault and Heal" strategy, then you need to store an absolute minimum of 8 tentacle points (10 is probably better but 8 can be fine depending), do endless assault first and then heal up afterward. If you want to do that strategy but use Heal before endless assault, you reset your tentacle points to 0 and end up not being able to apply your endless assault at all.

      Regarding Heal itself, there's no point in doing Heal when you have only 1 or 2 tentacle points since it forces you to jump afterward, and that 1 or 2 BP won't get you anywhere with Summoner Shion (Her scores are by far the worst in the game so you lose more than 1 BP even on a parry). Since we can't tell the AI that "Using Heal with 1 tentacle point is a complete waste but using it with 10 is pretty darn good)", it's going to use that skill whenever it feels like, no matter how effective or useless it can be.

      Combining everything together, the last main strategy with summoner Shion is the "brawl it out" way and it's usually your best option against high BP characters, or mid-fight if you are finding yourself on the losing side. Basically, it involves pokes that don't damage Shion herself much, gaining both Tentacle and Dweller points to be able to use Dual Summon which is her only high-damage option that doesn't damage her, using your other leftover tentacle points for heal and gaining leftover dweller points for an hopefully final dweller army to close it out.

      The AI will just throw whatever at you, and summoner Shion just can't win that way: Most of her moves are at -1 base damage, she has score pages varying from 1 to 10. All her top skills are only effective when used in the right situations. I've seen AI Summoner Shion beat me a grand total of once, and I faced her plenty of times: Even the worst characters like Nowa, Revy or Leina fare better than that.

      Without being able to tell the AI things like "Never use Dweller Army unless it can at the very least KO the opponent on a score" or "Don't use Heal for under 5 BP unless you'd die from poison in the next two turns", I really don't see how Summoner Shion or other more complex characters can possibly work when in AI hands. An adaptive AI would help for regular characters but it wouldn't be helping much for resource-critical characters or those depending on multiple move interaction (like Seductress Shion with her shielding, reflect damage move and love points).

      Delete
    2. Whew, that was a lot there ^^U
      I guess the "learning" AI isn't the solution we're looking for right now, but maybe I'll try it for fun sometime.

      But now we're talking, see?
      -A "reactive" AI would make a lot of sense, and I guess it'd be one level over the "balanced" one. A CPU that "pays attention" to the moves you can use right now and punishes accordingly should be more challenging that the current "deaf" one indeed, and considering that we know which moves usually beat other moves, maybe it's not that hard to implement.
      -And a "scripted" AI isn't that farfetched, and is close to what I had in mind.
      If I let the CPU read "restrictions" from the AI file too (like "Result 5 - body > 0 Can't use move X", or maybe a new restriction to lower its percentage use), maybe it'd be simpler to do than it looks?

      Delete
    3. Ever consider a Gambit feature? Like in FF 12 you would set Gambits for the AI depending on their health, their status, etc.

      Delete
    4. Wellll... I maybe typed a bit more than I really needed to, but I figured that exposing the whole concept was important so that an optimal AI could potentially handle all that :)

      Anyway, if the learning AI strikes your fancy the most, I'd say go for it! If it's the most fun thing to build, then it's what you should do: The goal of the game is to have fun after all!

      I think that being able to add or substract percentages based on a list of conditions could maybe do the trick indeed...

      A gambit system would be more cool of course.

      One thing to keep in mind is that the AI system needs to be usable: If it's too overkill and that it needs like 10 hours to make a single AI file it's not that much fun for the creators! :)

      Talking about fun... I did start to inventoring Farrah's images, and it's not looking good. I'm just about 1/3rd of the book done so far, but there's bunch of text everywhere, images overlapping so not really usable, the scans are low res and Farrah's pretty much doing nothing all the time. It's like "Okay, that would make a good away, this one would make a good body wound, this one unbalanced", but there's only one single image so far usable as an attack and it's heavily cropped due to images stacked on top of each other. It's definitively not an image set that gives me interesting ideas so far either... So yeah, Farrah might not happen.

      Delete
  19. Just in case it might help a tiny bit (although it's super vague information):

    I played a match against Kat, and I eventually found myself having deeply negative values for all my Oranges. I checked her scores, and she had like +20 something in all oranges and reds. The values kept increasing (for her) and decreasing (for me) each turn. So basically, the height checker seems to bug out occasionally: It keeps stacking bonuses and penalities instead of just handling a temporary +1 as supposed to.

    I've actually seen that happen a few times with Praxides in the past too, although it's never been with such absurd values (Maybe because Kat gets so much height?)

    ReplyDelete
    Replies
    1. I've seen this happen in many other instances, but the changes are only visual, they don't actually increase the damage. It's just a weird visual bug. I never mentioned this because it simply didn't affect the game.

      Delete
    2. Oh, I thought I had fixed that long ago, but if it's only the visual part of the bug that remains, I guess I don't need to hurry.

      Though, with a character that actually has various levels of heights, I wonder...? Gotta check that out.

      Delete
    3. Yeah, basically the reds and oranges will gradually increase to a ridiculously high mod score. Fortunately the actual damage is equal to what it was prior to the numbers inflating.

      Delete