I’ve always felt DoA was way ahead of all the other 3D fighters when it came to animation.
Good luck with your game.
Ninja edit: I just realized that comment might come off as a joke, I was not referring to boob physics :lol:
Have you investigated motion capture? I think that missing element in the animation might be more easily captured using some motion capture. I just did a quick google search there might be some affordable options.
Yes, if I follow SC, I should invest in motion capture. As a consequence, the game speed will be a little bit slower.
I’m setting up a markless mocap system at home, using xbox kinect. So I could improve animation quality, also reduce time and cost.
I also consider to buy
https://www.youtube.com/watch?v=Imm8oL68HNM
or
https://shop.xsens.com/shop/mvn-biomech/
Great yeh I was thinking the Kinect2 would probably be a good investment too.
https://www.youtube.com/watch?v=_POTSkgARHc
You probably don’t need to go to the level of detail the other motion capture device gives you.
DoA has been using motion capture since DoA2 too its still a quick game
https://www.youtube.com/watch?v=qeSjDCh6DPY
I purchased 6 ps3 eyes from Ebay, they will be shiped in next 2 - 3 weeks.
I will use ipi software like like that
https://www.youtube.com/watch?v=1-jmUafH0Gs
I still have to use keyframe animation for crazy moves (about 20 % of total anims).
You shouldn’t put too many frames of animation in. Most normal attacks should only be in the single digits in terms of total frames of animation. For example, your weakest punch should have something like 2-3 frames of startup and be active by the 3rd frame, and then only 1-2 frames of recovery.
I’m not sure mocap is used as much as you think it is. A lot of companies are opting out of it since it tends to look clunky without a lot of touching up. Soul Calibur, for example, stopped using mocap in SC5. They actually outsourced animation to Cyberconnect2 (Naruto, Asura’s Wrath, upcoming FF7 remake). They do everything by hand and it looks better. Even MK only uses it for certain things like fatalities if I’m not mistaken.
Realistically speaking, mocap will save you some time, but again, if you want it to look good, you’re going to have to put some serious time into making sure things flow right which means you’re going to have to heavily modify everything you’re doing.
I don’t know what sort of look you are going for, but if you want a more stylized looking game, mocap is going to ruin that for you completely. If you want complete realism, it might work, but it’s going to be tough making it not look like you mocapped everything.
I hope this works for you though.
All valid points but I don’t think unless I am mistaken was suggesting using 100%, nobody does that. You need a foundation to work from and thats how mo-cap should be used.
Don’t get me wrong, I think mocap is good for some things. I’m not saying you guys want to use it for everything at all. I’m sure you already know that.
All I’m saying is that I’ve working with mocap before. It’s a bit tedious in my opinion. I received a great deal of mocap animation and had to fix it. It ended up being a lot easier to use mocap to grab keyposes and animate the rest on my own. I still had to modify the keyposes heavily, however, to make the poses more dynamic.
So, again, mocap is nice, but, personally, I think it’s better for subtle things that are hard to hand animate (such as character acting).
Really, though, if this is how you want your game to look, go for it. It’s an aesthetic that may be off-putting to some, but it will probably work fine as long as the game play is solid.
Fighting game animation will, by the genre’s nature, require alot of hand tweaking. This is because of how fighting games are tied to their frame data. Once you’ve recorded animations, you’re still going to have to adjust them for the frame data, tweaking the startup, active and recovery frames.
Even just the amount of startup and recovery frames a move has can affect the feel of the animation. For a good example of frame data, you can look up 3rd Strike’s here.
Also, changing startup and recovery isn’t just a matter of increasing the number of frames. Doing so can make moves feel mechanical. There needs to be a snappiness to the moves to give them impact. For example, when giving a kick more recovery, instead of just slowing down the recovery animation, you can keep the part where the leg snap back fast, and then add the extra frames to the later part where the character returns to neutral.
Thanks all you guys for long comments. I have to wait until I can test the markless mocap system (in next 2 weeks), so I can discuss more about animation issues. But animation is the core feature of this project, so I will do my best for it.
At the moment, my main concern is gameplay. I’m working with two first characters, there are about 20 actions for each character in this demo (I will add more later).
Orc just use two blades in hack and slash style, because he is a melee fighter. But he could summon a dragon for projectile skills. The dragon can stomp, breathe fire, scratch… in a long range; but these abilities cost a lot of gauge points.
Dwarf is short, so he couldn’t crouch and jump so high. He would cast ice projectile skills from his long axe easily (they are mid range).
Another important thing is AI, my little brother will create AI rules with UFE Fuzzy AI plugin. I just hope the AI is acceptable.
This is such an important thing to consider if you’re using 3D models and animation for a fighting game. You really need to think of moves as startup -> active -> recovery 1 (I call this follow through) -> recovery 2 in terms of your frame data. So much of the beefiness and impact of your attacks will be due to this balance and where the timing of your animation is defined for those two recovery states.
You can also apply this to startup frames as well, having something like wind up frames and throwing the attack frames, but from my own experience it makes the attack animation look wonky.
Here is a clip about Dragon Summoning Skills of Orgk (I will add hurt boxes for them later)
https://www.youtube.com/watch?v=lI7nBFcgXV8
Hey, does your game not have diagonal inputs?
@liero116 : Yes, the diagonal input feature will be added by the UFE Game Engine soon. At the moment, there are only 4 basic directions.
I want to explain more about the game idea here:
- The fire breathing skill is anti-air move. It is quite slow and unblockable, so it could be used for defending or initiate another combo. For dodging, opponent must crouch, only Titan still get hit because of his large size. Although Dwarf and Halfling couldn’t crouch, this move still wouldn’t hit them (except they jump up). Centaur couldn’t crouch too, but he is in normal height, so he will be hit easily (he has to dash back, so Orgk will have a chance to send a kick right at his face).
- The dash and stomp skills are unblockable too, so opponent must jump away to dodge these moves, opponent’s height doesn’t affect in this case.
- Orgk is strong at close range, for balancing, his far range moves and projectiles consume a lot gauge points.
In this game’s philosophy, size and movement style do matter, each race each character has pros and cons. I don’t want to let a race could counter entirely any specific race.
For example, Angel could fly with wings, so in normal idle state of flying stance (like stances in Tekken 3), she is in mid air and her feet do not land at ground, therefore she would dodge low attacks like crouching sweep kick easily, she will gain advantage over short races like Halfling and Drawf . But, in this stance, she couldn’t crouch immediately, so some character with long range attack could throw a ton of fun to her. For example, Rikky The Halfling would use her gunaxes for shooting Angel like a bird.
Centaur has 4 legs of horse, he can walk, run, jump so fast, but horse couldn’t run backward smoothly or crouch instantly. So he will have a back stance, can attack with his two back legs and run or dash backward more efficiently.
Mermaid is just a walking fish, she don’t have ability to jump like a frog, but she might summon a tide to bring her in the air like a whale (these skills just cost a small amount of gauge points ).
The main purpose when I start this project, not just for entertainment, but also for education. In particular, it’s a language learning system for hardcore gamers.
The are many language learning games, but most of them are for kid and girl. With true gamers, those simply games are so boring.
In my dream, this game could teach people how to write kanji of Japanese.I know it will be very risky, but these are what gamers need from indie game developers (implementing crazy ideas is my duty).
There will be two modes. The first mode is Normal mode, it’s just a normal fighting game, nothing to discuss here. The second mode is learning mode, where we should master Japanese to defeat opponent (your win depends on your playing skills like combo commiting, strategy… and your Japanese vocabulary, grammar… the ratio is 50:50).
Because this game is designed for hardcore gamer, so the system will be quite complicated, then it isn’t suitable for popular audience. It will also help gamer could learn Japansese for playing Japanese fantasy video games, so the vocabulary will start with epic words like: dragon, sword, magic… Modern words like airport, car… will appear later.
There are many quizzes inside a battle. The quiz has form: What does X mean? and three choices. Player choose his choice by attacking opponent with rule: first choice is high parts like head, chest… second choice is middle parts like lower torso, hands… third choice is low parts like thighs and foot. If a player answer right first, he will get bonuses like HP, DEF, ATK, GAUGE, MOVEMENT… If the answer is wrong, he will be punished.
There are also RPG elements here. Guardian Card ( just like Guardian Force in Final Fantasy 8) will boost your character’s status a little. Here are some examples (note: each race has specific Guardian Cards)
The Genius Gnome (Halfling)
- 2 ATK in each strike of combo
The Wild Orc (Orc)
- 10 ATK in first strike of combo
The Blue Lizardman (Reptile)
- 15 ATK in last strike of combo
The Mountain King (Dwarf)
- 5 DEF
The Dark Elf (Elf)
- 20 MAX GAUGE
The Centaur Archer (Centaur)
- 10 MOVEMENT
The Sealed Summoni (Oni)
- 5 DEF of enemy
The Young Mermaid (Mermaid)
- 5 HP when you Parry
The Minotaur (Beastman)
Special skill Stomp the Earth
The Last Angel (Angel)
- 2 HP per 10 second
The Brave Titan (Titan)
- 20 MAX HP
Player could equip several GCs (maximum is 5). Each GC has 10 words. If player answer right a word of a GC in battle, the GC will gain EXPs. The GC could level up, then unlock the new word and increase the status. Max level is 10. If player anwser wrong, the GC will lost EXPs, and the level of GC may be decreased (so player has to remember words). When a GC achieve lv 10, player will gain its passive effect permanently, so player would unequip it and learn another GC.
For GC collecting, he will have to defeat insane bosses, with high HP, DEF and badass skills. The boss could equip GCs too. The bosses will appear follow the storyline and player’s learning progress.
I want to ask you guys a question: DO YOU NEED A FIGHTING GAME FOR LEARNING JAPANESE?
I really want to hear your opinions, listen what you want.
And of course , there will be another languages like Chinese, Spainish…
May be, you just need a game for fun, so I will continue make a normal fighting game. But if you say YES, I will create demo about learning game ideas.
Honestly, I’m not interested in playing a fighting game to learn any language, but I’m sure someone out there will be.
I have another question though. Is this your own original artwork?
@liero116 : I bought them from Unity Asset Store.
Why do I choose Fighting Game?
Because we cram a lot of information into our heads as fighting game players. If you are good at fighting game, there is no doubt that you have a great mind. So, the learning curve could be more suitable.
Yeah, gamers we spend a lot of time in playing game, just want to beat any challenge while gaming. Society says it’s a waste of time. So why we shouldn’t learn a foreign language at the same time? It will help us get a better life. (It’s just what about gamification).
This game even has a real plot for the reason why you have to learn a foreign language:
Humans are haughty
They conquered Myth-earth
The other races
Allied together
Defeated human
After ten decades
The peace is over
New war has begun
Forever again
An Elf scholar discovered an ancient language (it’s the language you want to learn: Japanese, English, Spanish…). It has extreme power, called Rune Force . Everyone who understands it, could gain an unbelievable strength. But Rune Force is very hard to learn. So The Human invented a device called Guardian Card, for storing Rune Force in creature’s form. The Human could build great armies because they would control Rune Force more efficiently than the other races. The Human was so ambitious, they did want to be King of Myth-earth. They conquered anywhere, and no race could stop them. The other races allied together for preventing The Human. The Human was lost, and other races drove them away from Myth-earth. Guardian Cards were banned in Myth-earth. After 100 years, these dangerous weapons appear again. Every race fights each other for possessing Guardian Cards. At the same time, The Human is planning for their revenge.
So you have to learn Rune Force(Japanese) for mastering Guardian Cards.
You could imagine the Rune Force here is something like spell in Harry Potter or Force in Star Wars.
When fighting with bosses, it will be very hard to defeat them, if you don’t remember words. So you can not unlock more Guardian Cards, characters… unless you actually study the language by mastering Guardian Cards.
In PvP mode, you can challenge other friend, for proving that you learned more than them. Your friend could choose a different language from you. For example, you are studying Japanese, but your friend just only knew Spanish as second language, so there will be two quizzes in both two languages at the same time, you and your friend will have separate objectives to target. In each round, there will be about 5 quizzes (15 seconds per quiz).