You are right, it’s better like that.
I finally played the demo today and it is absolutely phenomenal! Everything is so beautifully polished! Right from your custom intro and the animated menu to the camera system and super-cool “power up” tablets that pop out at you, I’m really impressed. If I can make my demo even half as polished as yours I’ll be happy.
Most importantly perhaps, the gameplay is so much fun too! I played the whole thing almost 2 times, great job with the combos, fancy sword particle effects and destructibles too. In games like these smashing barrels just to see how they fall apart is half the fun 
If I could offer any suggestion it would be to add more variety to the combo system. The dodge-hit-hit-hit-hit-JumpSpecial-dodge routine is fun but I instinctively find myself trying to create new combos by varying the timing, number of hits, etc. All those hours of my life spent playing DMC probably calling out to me!
VSZ, thank you for trying out the demo and thanks for the positive response. I really appreciate it the fact that you spent some time playing it.
With regards to combos… yeah you have no idea how badly I want to make a much more complex system but I just don’t know if I’ll be able to work it into this project. It’s a lot of work just to get everything in there and a lot is still not in there, just generally speaking. As far as combo goes, I still have to fully animate the existing system. What you get in the demo currently is actually re-used animations to make a complete combo, where it should be all unique in the full system and will feel different once it’s fully unique. It’s a lot of work to create it all and I can spend the length of the project just working on combos. I pretty much deliberately kept attacking on only one key so there’s no to try and find different combos. The only exception is the “finisher” where you use the Jump key to engage.
So, I try to strike a balance in making this game, I keep everything simple and manageable. If I find at the end of the project it makes sense to add to it, regardless of what it might be, but say, adding a more complex combo hierarchy, yeah, it’s a possibility, just not a planned one. I think ultimately if it can survive it’s gonna have to be on it’s simplicity and if I get a to build on that for a possible second game, then that’s the idea… you know.
Edit:-
I’ll mention that you can press “B” to unlock all the moves. Basically it’ll add the jump attacks, which are the most powerful attacks. Jump straight up and then attack, or jump towards an enemy and attack. So that opens up the attack choices a little more. Those are not officially in the demo because you unlock them later in the game, but using “B” is a bit of a “secret” to unlock them… 
Can you can make some basic combo solution allowing you to create different combos by simply adding any moves and entering different parameters ?
http://forum.unity3d.com/threads/released-universal-fighting-engine.218123/
It could be interesting instead of customizing all combos in Blueprints.
The real problem isn’t really creating a combo system, it is to animate it all and make sure it all works just perfect. The animation needs to all work together and feel right. Even if you look at my current animations, they are about as simple as can be where it still feels ok as a motion but feels good as feedback from your input. It can still improve on various levels, it just needs to feel important enough for me to spend time there. So, that doesn’t just happen, it takes a lot of iterations to get that. Animation of any fighting type game play is key. Anything a little more elaborate and it’s just way harder to do. I’ve attempted to add in some kicking moves as a combo choice and failed miserably. I might give that another shot some time. I’m not an animator by any means, I mean I can get by, but it’s hard for me. I have set realistic goals for this project. They’re certainly not static goals, otherwise my game would be done already. My original idea was much simpler. But still realistic goals where I’ll flex where I feel I need to. Going beyond my current fighting system means a pretty big commitment on a few levels. It’s not impossible, just, not practical at this point. I’m very aware of methods to make different types of systems, but I have chosen what I did for a good reason, it keeps the project in the realms of possibility.
So, currently I have basically 2 animations that represent the combo or the loop of the combo. They fit together, right, left, right, left… etc. I have 5 animations total that represent start, loop, finish. Although as finisher moves I have 3 animations for 3 different strengths finisher, but that I need to replace with a sectioned system so ultimately less animations with more end possibilities. The current solution was a quick and dirty one. The start is represented by 2 animation, to get the combo going. Those moves you can do while not hitting someone and they basically loop until you hit someone to get the combo going. So this is pretty simple, yet, took me months to get it just right.
My plan at this point is to have, for the loop section of the combo, 4 or 6 animations that can possibly be used in different orders. So it’ll be either 2 left and 2 right, or 3 left and 3 right. In this case you get a combo loop that can vary in it’s look and feel but still simple by just tapping the key. Or if I feel it’s the right thing, bring in a second key for choice. This whole system is based on old school arcade brawlers like Final Fight or Ninja Turtles. Those games work because they are fun, even with only one string of something like 3 or 4 moves from start to end. Sure, we don’t live in the 80’s anymore, but this is the game I’m making and I’m putting a lot of effort on everything, not just combo animations. If I can make the kicks work, that will instantly create a possible hierarchy choice or split in which case a second key comes into play anyway. But that’s just really unlikely at this point. I do hope to take another shot at that and see if I can make it work, by which I mean, make it feel right. Setting up the blueprint for that isn’t a big deal, but the rest of it… well… is…
Totally understand. I commend you for setting realistic goals for your project, I could take a leaf out of that book myself too.
I think this game will do well, as a player everything felt polished, smooth and I guess that is what matters in the end! I’ll try out the “B” cheat code some time, I just knew there had to be a special move where you jump straight up and then attack!
WOW this game is AWESOME!!! Please look at the game Im working on! I would love your feedback! https://forums.unrealengine.com/showthread.php?99873-Coreupt-(action-beat-em-up)
VSZ - Cool man, thank you, yeah try out the jump moves, once you have them it’s pretty cool to deal with bigger groups of enemies. Later on in the game when things get more difficult, it’s much needed… But yeah in the demo you can get a good feel of their power and especially the forward jump move, slicing through groups of enemies, it’s pretty cool.
JWright68 - Thank you!.. I’ll check out your post.
Thank you SO MUCH! Your game is a lot of fun man! VERY VERY AWESOME STUFF!!!
I’ll say thanks again… I’m glad you liked it, and appreciate that you took the time to check it out.
I know this is lot of work and not easy to add new moves, but there is movements packages sold cheap or some ways to make some motion capture today for cheap. I agree keeping a the project possible to make is better than targetting unrealistic goals.
For sure, there are various options, I’m aware of that. I still choose to make my own animations.
Great to see that there is still progress!
Quick question - About blending to rag-doll and back to standing…Did you ever get it working again in a newer version of the engine? Or did you figure out another way of doing it?
No man, sorry, I was not able to make it work again, but I have not tried again since maybe 4.8. Really the reason is because I don’t intend to use it anymore. Maybe if it wasn’t such a big problem I would have kept it but to try make it work again with it being actually a really small feature in my game, it’s a waste of time for me to keep picking at it and get no results.
What you could do, which is what I did back then as well, just check out UT4 and see if their Feign Death feature works correctly. I think it was on 4.8 when I checked it and their ragdoll blends were also broken, in which case I assumed it can’t work. To be honest with all the testing I did back then, I was pretty convinced it couldn’t work, but seeing it broken in UT4, that convinced me completely. I don’t have a recent version of UT4 so haven’t been able to see if they ever fixed it. Once their system works then you can be sure it works and then it’s all about figuring out the method. But trying to figure out a broken system is counter productive.
Ok thanks Ill check UT4 out. Although…I was messing around some more and I think I got it…Like it works basically perfectly. The problem was, in the Phat, there was nothing on the Root joint (the one on the ground at 0,0). I believe thats whats causing the spinning when changing the blend weights. So I put a rigid body on the root joint, disabled its collision, and then simply set all the joints above the Root to simulate physics , and then use a timeline to blend back the blend weights. Also, when the character is simulating physics, I disabled the movement-component and set the capsules location to the pelvis joint of the mesh, and added half the capsules height to the Z, so that the capsule stays with the mesh wherever it falls. The Root joint stays with the capsule at the bottom since it isn’t simulating physics, so the pelvis and root are in the same location. I still have to test out a getting up animation but I think it should work.
Oh and one thing to note: When you add a rigid body to the root, The Pelvis joint will get locked. So you need to open the constraint properties and set everything to Free, including the linear movement.
Cool, I’m glad you got it working!..
I’m fairly certain they must have done something to fix it unless you can confirm your method works in 4.7 or 4.8 because I’ve tried this among other things. I mean to add a body for the root… etc. and I haven’t been able to get good results. But, again, I haven’t tried this since 4.7 or 4.8.
If I remember correctly the problem I always had was with the root not staying… at the root, so when I tried to basically lock it down using this method you describe, that was ok, the root stayed there, but it still didn’t work out properly. I think it was with aligning the capsule, there was a reaction on the ragdoll. If I remember correctly what it did was pull the ragdoll around, even though it was technically “disconnected”, not constraints, or anything. So I couldn’t align the capsule, no matter what I tried. I actually could get good blending, if I left the capsule alone. There were some definite problems with it. I think, that was it.
If you got it working, I can try mine as well and see what’s the result, if I get something different from before. I’m sure once I start looking at it again, it’ll jog my memory on things. But from the sounds of it, and from what I can remember about my earlier attempts, it sounds like, they may have changed things to make that function correctly.
But anyway, that’s good news, it gives me a reason to check it out again…
Haha ok. Yeah I did get some strange behavior, but I messed around, tryed out a few combinations, and now what I have seems to be working real well 
I added a get up animation and it looks really good! What do you think? It still needs some tweaking but Id say its a great start.
Cool, yeah that looks good. Nice work…
Does it still work if the ragdoll gets thrown further away?.. or fall on his stomach? or some other position other than directly where he stands falling on his back?. I realize the animation isn’t made for anything else, but I’m asking for the sake of the blend itself, because that is the best case scenario for the blend to work and I wonder if you test a worse case scenario and it still works?.. meaning, really fling the ragdoll away from the falling location, having it also land in other positions, other than on his back.
Yeah it does! The capsule will follow the ragdoll anywhere. And the blend will work no matter what. It doesnt matter what position the character is in, or what state of movement really. I would just need another get up animation (from stomach) to make it look good.