Hahaha, thank you.
Here are the annotated system changes - get reading, gentlemen!
SYSTEM CHANGES:
There are really three kinds of system changes here:
1) The removal of glitches or excessively poor game mechanics.
2) Gameplay enhancers (which we define as encouraging thoughtful play).
3) Flow enhancement
The first is easy to defend; no one wants poor game mechanics, such as the current TAC system to remain in place. For the second, it is important to distinguish between two kinds of good in a fighting game. There is the good which is strong, yet thoughtful, and when players use it they feel good about themselves for making a good choice (Bionic Lancer, for example), and then there is the good which requires very little thought, is useful in many situations, and has negligible consequences (mashing s.L with Magneto for an anti-air). This second type is detrimental to the general enjoyment of the game. The players who use it do not feel any strong sense of enjoyment, and the opponent does not particularly respect his loss. We would like for patch balance to move the game more toward thoughtful play, and this is a recurring theme for changes we have made throughout the list.
The third point is flow enhancement, and this is difficult to define precisely. It may be enough to give examples of poor flow in this game:
1) Long periods where only one player is playing, whether because a combo is excessively long (Zero combos, TAC infinites) or because they are given no choice but to block for an extended period of time (Vergil with Hidden Missiles and a large meter stock).
2) The general winning makes you stronger ethos of Marvel. Whoever gets the first hit in UMvC3 tends to win, because the pressure gained from the first hit generally leads to more mix-ups and then a dead character. Then, the incoming character must survive a series of difficult mix-ups and pressure without any assists, creating a 2v1 environment. X-Factor was designed to combat these problems, but it is simultaneously too much and not enough, as it puts all of the potential to re-establish yourself on one input per match. Losing a character is more than losing 1/3 of your health you have lost tools to be competitive in the game. It is for the benefit of the game to reduce the landslide win tendency the game has.
We ask that you keep an open mind to our suggestions, and remember that the thought behind the suggestion is more important than the suggestion itself. We have done our best to address the symptoms of problems in this game, but there are always other approaches that are worth looking at.
TAC Changes:
*Costs 1 bar, always a guaranteed exchange in one of three directions.
*Meter gain/loss from directional choice removed.
*Hitstun deterioration returns to 0 after a TAC and builds normally from there; returns to pre-TAC hitstun levels combined with post-TAC hitstun levels when either character touches the ground.
*The TAC user cannot gain meter after a TAC until the opponent enters recovery (exception: Soul Drain, meter gain assists).
This is likely the most controversial change on our list. We would like for TACs to become completely non-random and adjusted based on that alteration. The meter cost, combined with an inability to build meter post-TAC, ensures that they are usable but not abuseable. Hitstun should be reset for the simple reason that players have created a lot of creative combos in the current system. While it is not desirable for all of these combos to remain (infinites, etc.), combo creativity is a major aspect of this game, and some facet of TAC combo fluidity should remain.
Meter gain/loss is removed because, most obviously, a downward TAC would be free in this system. Secondly, meter loss/gain is a poor mechanic because it deemphasizes one of Marvels most important gameplay aspects, which is meter management. Several characters are especially affected by a new TAC system, such as Phoenix and Frank West. We have taken these changes into account when balancing these characters.
It is popular to suggest that TACs simply be removed from the game. We are united in opposition to this suggestion. While TACs are not enjoyable in their current form, they are an intriguing idea. As a team game, adding new and interesting ways for team members to interact is a boon. We hope that with this system, players may TAC to teammates for high-damage follow-ups, such as the combo enders characters like Iron Man and Chris are capable of creating due to their moveset. Knowing that this will happen, we also recognize how important it is that post-TAC combos do not build meter for the player using the TAC, as this might lead to team infinites.
As an alternative to these TAC changes, another option we strongly considered was extending the counter period and having the TAC coloration show up on the same frame the input was entered. This is a distant second place option for us, however, as it leaves several problems intact still: meter management disruption, the TAC XFC exploit, infinites, and the still essentially random nature of TACs. In the words of Viscant, TACs are only a mindgame in the same way Twilight is literature. TACs are a poor and frustrating mechanic when players are left to guess, and unless one is looking at the opponents joystick, an increasingly distasteful ye common practice, TAC counters are simply a guess. There is the additional concern of whether TACs would become useless if the counter window were too large and noticeable. This would render TACs useless, which is not desirable because it is an interesting aspect of team construction.
Throw Changes:
*Throw tech window increased by 3 frames (from 7 to 10).
The current throw tech window is too small, and this is widely agreed upon amongst competitive players. Throw tech periods should be balanced between having to predict the throw and being able to tech a throw on reaction. The former is too little time, and the latter is too much time. We feel as though a 3 frame increase would make throw techs a little more reasonable.
*You cannot be grabbed during tech recovery.
This is a highly situational change for characters such as Wolverine who, when spread up to maximum speed through Berserker Charge and XF3, can air throw characters during their tech recovery.
*Ground throw techs put more distance between the two teching characters (slightly out of most characters s.L ranges).
Throw tech mashes are an eyesore. When an air throw is teched, the characters are pushed back into a neutral position, but in melee range on the ground, characters will often tech throws 2-3 times before entering a neutral position. This is especially irritating when one character has significantly longer grab ranges than another. A slight distance increase after ground throw techs removes what is presently an irritating gameplay experience.
*Throws can only be initiated with f.H and b.H.
This is a change that goes under the thoughtful play category. Option select throws are too good right now, and a character like Wolverine can simultaneously block, dive kick, and throw option select all with one button push. Tridashers typically use j.H even over their better aerial normals to catch chicken blockers with an air throw, which gives them a strong advantage over characters with fewer aerial mobility options. This change still provides powerful option selects available, such as Dantes j.H box dash, without providing three whole options within the single input.
X-Factor Changes:
*X-Factor sum values changed to: XF1 40%; XF2 60%; XF3 80% (from 40/70/100).
Even after being reduced from Vanilla Marvel, X-Factor values are simply too high still. It is still too often that a character uses X-Factor 3 and proceeds to stomp an entire team through 2 successful 50/50 mix-ups after XFCing blockstun to kill the first (or something similar). We believe that X-Factor, like TACs, is a conceptually good mechanic, but it is too strong in its current state. We recognize that when a character dies, a tool is lost, and comebacks in Marvel are far more difficult than in any other fighting game due to this fact.
Two other approaches to comebacks were discussed, and they are both worth mentioning for future iterations of the series:
1) When the last remaining character on a team comes into play, they receive an automatic buff that allows them to be competitive without team assistance. For example, Wolverine could receive a passive speed and health regeneration buff, or Arthur could receive a new level of armor with even stronger projectiles. This change was rejected because it would involve introducing an entirely new system, which is a better fit for a sequel or substantial update than a balance patch.
2) When a character is the last remaining character on a team, the A1 becomes a solo assist button. Each character would have a character or ability that could be called in from his/her respective source material to assist them in a solo setting. This would ensure that every character has a team-esque option when alone. Each characters solo assist would be something especially beneficial to that characters playstyle. For example, Shuma-Gorath could call in a tracking extradimensional tentacle that attempts to swipe at the opponent, and Viewtiful Joe could call in Sylvia to fire her gun from Viewtiful Joe 2 (not Desperado).
*X-Factor now prevents time outs while active.
We were a little uneasy about this change, but with the universal health buff and universal X-Factor damage nerf, there is an increased risk of time-outs. Allowing X-Factor to prevent time-outs during its active period would create some substantially exciting moments and would help prevent time-out situations. This is likely the least essential change we have suggested, but it is one we all agreed on.
*X-Factor buffs now stack additively with install hypers, not multiplicatively.
This is an important change for characters who are regularly anchored and stack their installs with X-Factor values, such as Vergil. The current system provides stat bonuses far out of what is reasonable. This change will ensure install hypers are still beneficial during X-Factor while toning them down somewhat.
Damage Changes:
*All character health increased by 20% (individually changed character health does not factor this in).
To state it again: if a character has a health buff in the patch notes, that number is before the universal 20% increase. Felicia, for example, does not have 900,000 health in our version of the game, but 1,080,000 (900,000 x 1.2).
That being said, there are three considerable reasons for this change:
1) Right now, the first touch on an opponent for many teams still leads to a dead character. The Marvel team recognized this was problematic in Vanilla MvC3, because damage was generally toned down a bit, and X-Factor valued were reduced. Dying from a single hit at the start of the match skewers the game toward characters with strong early games, such as Wolverine.
2) In theory, there are two kinds of rushdown characters in UMvC3: reset-based, and damage-based. In practice, there is only one kind: damage-based. There is absolutely no reason to pick a character with good resets when you can choose to kill your opponents character instead. This is why, despite having a lot of interesting options, Felicia is nigh unpicked in the current tournament scene. It is not worthwhile to have to reset an opponent twice successfully to kill when you can just kill off of the first hit.
The most common criticism we have heard in response to this change is that the high damage characters will still be preferred because they are high damage, while Felicia will be even worse off because now she will need 3 resets to kill. This is untrue because, first of all, requiring a reset on a high damage character strongly impacts their offerings on a team. Second, we have also tweaked individual character damage across the board to reduce the damage disparity somewhat.
3) We strongly feel that a symbiotic team structure system is more appealing than the present Point/Support/Anchor system. That is, presently the vast majority of teams are ordered to have a point character with a strong ability to KO entire teams backed by his/her best assist, and the third character exists solely for comeback potential. The most common example of this team formulation is Zero May Cry (Zero/Dante/Vergil), which places Vergil on a team where his assist (Rapid Slash) is of only marginal usefulness to his teammates. The teams goal is to either KO the entire team with Zero using Jam Session mix-ups, or to use XF3 Vergil with Devil Trigger to win after Zero (and probably Dante, since hes poorly supported) dies.
This current formulation is prominent solely because of the high damage nature of this game. If Zero is killed, then it is entirely necessary to have a strong anchor waiting to make a comeback. And if Zero is hit, he will be killed, necessitating this structure. In the interest of greater team variety, we would like the game to move toward teams which function with any of the three characters on point. While MvC2 still had the unfortunate philosophy of assist tier characters, it was superior in that teams were considered from a wider variety of orders due to the lower damage levels present. As Yipes said early on in MvC3s life, In Marvel 2, if your opponent had one full-health character and you had three characters at 1/3 health, that was a good place for you to be. In this game, its the opposite. In addition, a universal health buff encourages players to use underused game mechanics such as snapbacks and alpha counters (to save a point character). Presently, snapbacks are rarely used because one is in a position to choose between killing the character presently being hit, or having an opportunity to mix up another character. If the choice is between having an opportunity to mix up another character or having the opportunity to mix up the present character, a snapback becomes more appealing.
*Chip damage increased by 5% (to 35%).
The simple motivator behind this change is that we do not want keepaway to suffer too much from the universal health buff. It is already much more work to chip opponents down than it is to do highly damaging combos, so some chip increase is necessary to ease the change for characters like Hawkeye.
Combo Changes:
*Light attacks which chain into themselves may only do so on hit or block.
This is primarily a universal s.L anti-air nerf in the name of thoughtful play. Presently, characters such as Magneto and Nova will simply stand under opponents and mash s.L into an anti-air combo. While their s.L moves are very fair in terms of frame data and hitboxes, anti-airing an opponent should be a thoughtful and well-timed option.
Assist Changes:
*Assists depart with invulnerability frames if the point character is hit before the assist appears on the screen.
This is simply returning assist calls to MvC2 status when the point character is hit. Certain combinations are very strong right now because the current system encourages reckless, thoughtless play with assists covering you. The characters who are the most reckless tend to be the strongest. The most iconic example is Wolverine backed by Plasma Beam. Wolverine can charge forth and do as he pleases when Plasma Beam is called, because it is guaranteed to protect him from nearly any mistake he makes. We feel that a system which encourages intelligent use of assists is more beneficial to the game.
*Assists cannot be called during recovery frames or rolls.
When your opponent finishes a combo and sets up a proper reset opportunity, that is a time for you to be blocking. Allowing assist calls during recovery rolls forces certain teams into a defensive position when they have earned offensive ground. The clearly best example is Double Lariat, which can be called on top of ones opponent while rolling under him/her, creating an invincible cross-up that leaves you with frame advantage and pressure on wake up. Once an opponent has completed the recovery animation (standing), assists may be called as normal.
*Assists can no longer be called after recovering from a hit after having entered a superjump state.
Superjumping is a powerful tool in this game, and it locks out assist calls for a good reason. Right now, it is not always bad to be hit after superjumping, and that should not be the case. For example, a Sentinel player may superjump toward Morrigan backed by Hidden Missiles to get in. If the Sentinel player allows himself to get hit by the missiles, he/she will recover and be able to call an offensive assist to pin Morrigan, who is underneath him in this scenario. Getting hit should not provide a free offensive opportunity.
*When an assist begins recovery from knockdown, it is invincible until it exits the stage.
Punishing and juggling assists is fine and excellent. However, at times an assist will land, enter recovery, and then be hit still. We take this to be an unintended glitch, and feel that it is best changed.
*Crossover counter now always give you full control of the character after the crossover counter is initiated.
Crossover counters are the most underused mechanic in the game presently, and this is the case for two reasons:
1) Crossover counters are generally not strong neutral assists.
2) Crossover counters often do not allow post-counter follow-ups even when they hit (Viewtiful Uppercut).
While the first of these is difficult to change, and is probably best left as-is, the second point can be fixed. The successful expenditure of a bar of meter into a move that hits should allow for post-hit combos in all situations where that move would normally be able to do so.
Other Changes:
*Guard Breaks removed (1 frame untechable/unblockable setups; glitch).
To be clear, what we are not changing is this situation: player 1 blocks an assist on incoming, and, having blocked, is now command grabbed by Super Skrull. This is a legitimate setup that took thought to do, and the incoming player had an option to get out of the situation (getting hit by the assist, or pushblocking to increase guardstun time). What we are talking about is the 1 frame unblockable that is possible in certain situations due to the player simultaneously being hit and thrown. This results in the throw being untechable. While it takes a great deal of skill to do this, and it is rarely done, we feel that untechable, unblockable situations should be removed as much as possible in the game for improved gameplay flow.
*Ground neutral tech recovery is now 5 frames slower and more deliberately visualized.
This is a small change, but presently neutral wake-ups recover significantly faster than roll recoveries, and this leads to mash-happy situations. This is primarily due to a lack of visualization in the wakeup. A slightly greater visual indicator would be beneficial.
*Character deaths do not cause inputs to drop when there are multiple characters being hit simultaneously.
When a character dies, all inputs are negated for a small period of time. While this can be helpful to aid a player in saving unnecessary meter expenditure, it is more problematic than helpful, and it is fine to hold players accountable for their inputs. In situations where two characters are being comboed, it is frustrating to have input drops that lead to combos ending prematurely. This is most notably in high-execution settings or multi-button inputs such as Dantes Volcano into Bee Hive.
*Input reading changed: holding forward/back is no longer a forward/back input in terms of moves. Example: holding back, and then doing qcb, currently gives one a reverse dragon punch motion. In general, directional input buffering is toned down a bit.
Easy mode inputs may be fine for Street Fighter, but they are very harmful in Marvel. Presently, if two qcf.AA inputs are made to DHC, the second qcf input is read as a dragon punch input. This can be solved by making a hcf input instead of a qcf input, but the better solution is to tone down input buffering. Marvel is already a very input lenient game; the present input reading causes more problems than it solves. An example familiar to Dormammu players: 2D1C is charged, and Vergil does s.H x Round Trip. There is a brief gap here where you try to let 2D1C loose as a counter, but because you have gone from holding back to block to a qcb motion, you instead teleport above Vergil. Of all the fighting games we play, Marvel is the only one that has input buffering so lenient that it causes problems.
*Raw tags no longer hit overhead.
Safely tagging out a character on the defensive normally costs 1-2 bars, depending on your options. Raw tags are too likely to succeed, and in many circumstances lead to full combos on success. Allowing players to block raw tags in a crouching position would make them a less appealing choice.
*Attack cancelation does not change when in X-Factor (currently, you cannot cancel Morrigans s.H in certain X-Factor levels).
A simple change. Anything that works in a characters normal state should function properly in X-Factor. There does not seem to be a reason to prevent Morrigan from fly-canceling s.H after the first hit in XF3 when she can do it in XF1.
*All snapbacks universalized to be 2 frames and +1 on block; attack travel speed instantaneous for all characters; active and recovery frames universalized on snapbacks.
Presently, some characters have great defensive options, and others do not. We feel as though snapbacks should provide a universal 1-bar defensive option against opponents to help balance out poor matchups and strong pressure situations. All frame data for snapbacks needs to be universalized to ensure they can all be used as semi-reversals, and we understand that a few snapbacks, such as Ghost Riders, will become exceptionally strong. We find this to be acceptable, as it fits his character archetype.
