infiniteImage

Hey all, Been doing some research for meter gain properties in P4AU and just wanted to post my results, both to proliferate the information and for others to proof read the work. I'm going to put the TL;DR version first, so you can just get the raw facts, then after I'll go into detail about how I found those things. So here goes! Meter Gain Facts: Meter gain is based on the base damage of the move. Each ~267 damage will provide 1 meter (e.g. 800 damage moves provide 3 meter). Moves provide a static amount of meter no how much their damage is reduced or increased by a combo. This means no meter proration and no bonus from counter-hits. The first hit in a combo will always provide 2 additional meter. Shadow characters' moves will generate 1.5x meter (not including the 2 meter first hit bonus). Meter gain is halved for blocked attacks. Example: If Shadow Narukami counter-hits with his 5C, he will do 792 damage ( Base damage * Shadow penalty * CH bonus = 900 * 0.8 * 1.1 ) and gain ~7.06 meter ( [base damage / Meter gain constant * Shadow bonus] + First hit bonus = [900 / 267 * 1.5] + 2 ) Implications: Because there is no meter gain proration, doing high meter gain moves at the end of combos is insanely efficient (ex. All Out Rush will provide ~12 meter for a normal character and ~18 to a shadow character even as the very last hit!). A shadow character will reach full bar after having done almost HALF (8/15ths) of the damage than what it takes a normal character to reach full bar. Because meter gain is still 50% on block, you can gain significant amounts of meter if you have high damage, safe blockstring attacks. Now for the details-- Meter based on Damage After extensive testing with Chie's moveset, I was able to discover that there is a linear assocation with base damage (that is, how much damage a move does as the first hit) and meter gain. I went into training mode and did the same attack on the training dummy and recorded how much total meter I had. Thus, [Total meter / number of hits] gave an approximate number for each move's meter gain. After getting data on about 20 or so moves, I graphed the results and found an extremely linear curve. Tracing it back to its origin, it seemed even the lowest moves would provide at least 2 meter (more on this later). Doing some quick math to get "damage per meter" I found that all moves hovered around 265-270. And as such, providing me with the approximate "Meter gain constant" used above. Meter gain does not prorate Once I had an estimate amount for how much each move would give, I set about finding how much they deteriorated in combos. I took a sample of three moves (400, 800, and 2000 damage) and performed them all as the second move of a combo. I noticed that all of the moves seemed to be providing 2 less meter than the original, regardless of total meter provided. This lead me to test the theory that all "first hits" provide 2 bonus meter. Sure enough, after placing all of my attacks in various parts of a combo, they all provided the same amount meter. To stress test my theory, I set the training dummy to not tech and repeated the same attack 40+ times in a row. It consistently provided the same amount of meter throughout the entire 'combo'. First hits provide bonus meter As previously mentioned, while testing the proration of meter, I found that all moves generated 2 less meter than what I had recorded when used after the first hit. This also made sense when considered against my linear graph of meter generation starting at 2. Shadow Characters get 1.5x meter Once all the other data was in place, this was fairly simple. I took another sample of three moves as a shadow character and tested the generated meter. I found that these numbers also gained the 2 meter bonus and were not affected by the Shadow characters' damage penalty. Meter gain on block Similar to the Shadow characters tests, once I had the other data this test was fairly straight forward. I took another sample of a few attacks and measured their gains on block. The results were all consistently half of my previously recorded values. If you made it this far, thanks for reading! If you have any questions, feel free to ask.

Prepare for math-- Following the formulas for combo proration found here I've set up a quick excel sheet to predict combo damage/meter gain/etc. Take the following Chie Combo: Move Damage Prorate SMP EffDmg TotDmg TotPrt EffPrt 2A 200 600 0 200 200 0 100% 5B 550 100 0 250 450 600 76% 2B 400 100 0 172 622 700 72% 5C 800 100 0 326 948 800 68% Based on the standard prorate formula that 2B should do 172. But in game, it scores a 192. If I replace the 2B with a different move, it experiences a similar boost in damage, even supers. For instance, doing God's Hand C instead of 2B nets 1440 (when it should be doing 1296). Then, after the 2B, the 5C goes back to being predicted correctly. Interestingly, I can switch the 2nd and 3rd moves to other moves of similar strength (5B, 2B, 5C, etc.) and the 3rd move will still receive bonus damage after starting with 2A. But doing three A moves does not cause the third move to have bonus damage. *hair pull* Using the same prediction method, if I just set the CPU to not tech and do a dead heat combo with the same move over and over, I get erratic results: Repeated Chie 2As should do: 200, 91, 81, 76, 67, 62, [57], 55, 50 etc. In game they do: 200, 91, 81, 76, 67, 62, [67], 55, 50 etc. I can't predict where they will happen or what it's based on (I've tried tying it to many variables with no clear results), but random hits in my combos are just getting these weird boosts. Note: this happens with both normal and shadow Chie and with some cursory checks this also happens on other characters (Yu, Naoto, etc.)

So I've been messing around with other ways to convert AoFC into a ground combo. The most consistent one I've come across is: AoA->C > IAD > j.8D > AC (turn around) > j.236B (input as 214B) > 236B > damage finishers It's at least a pretty easy way to get them back down to do dragon kick/god hand finishers. But still no knockdown setups that I can find so far. EDIT: The best knockdown setup I've found is still just j.5D. But I did find a way to do a safejump out of it: AoA->C > j.5D > whiff j.A > 9jump > j.2B

So has anyone else noticed the combo damage being erratic? I can't find any threads on changes to the prorate system, and yet I'm getting a bunch of anomalies when testing for new P1/P2 values. Quick example: go in training with a CPU that doesn't tech and dead-heat combo with the same move (2A). Eventually one of them will do the same or MORE damage than the previous one. Did they add some sort of combo bonus prorate system? Or maybe it's a result of how they changed SMP?

Did you mess around with pausing your 6B? Because yes, while it's not ALWAYS the answer, it makes the link much much more consistent. As long as you tag that first 6A low enough, the staggered 6B is pretty much guaranteed. I haven't really ran into a situation where I whiffed the 6B and didn't expect it, most of the time I realize either "did my 6A too high" or "did my 6B too early". I've definitely spent my fair share of time whiffing 6B since my friend plays Litchi/Tsubaki/Valk, it's gotten me used to watching for the weird floats and such. And yeah, the three reps thing was just me messing around with the gatling pause, seeing just how far I could push the limits. It does technically end up doing more damage and building just as much meter, but it's char specific and hard enough that I don't think it's practical at all (especially since the 6C fatal combo is rare as hell anyways).

Just ran into that 2B > 5B > 632146C combo in the corner, soooo good! Also, I wanted to mention something about the 6A > 6B. It WILL link on the entire cast. It's just more difficult on certain characters. The trick is to pause the 6A > 6B gatling. Hit 6A, hold your 6B so that it will hit right before their tech opportunity, and you can link somersault loops on even carl/cyborgs/etc. Of course, catching the 6A lower to the ground helps tons too. Using the method above, I've even been able to link three 6A > 6B reps into the 6C Fatal combo. but it only adds like 100 damage, not really worth the insane amount of precision for that third rep, lol. (maybe if you aren't going to make it to the corner? *shrug*)

Sup Mu boards, Thought I would come in and say hello in the GD thread before I started posting everywhere else chattin' about Mu strat. That way I don't get as many "who the f*** is this guy?" comments. d: So yeah, SoCal Mu, haven't been to too many tournaments yet, but going to start shortly. Got a pretty dedicated group of guys that run in-house ranbats every other weekend; we get some good competition going. Anyways, I'll be around, lurking threads, posting random insight, etc.