Feedback and design discussion

[Lime_x]

So I have not been very active in this community for a while and thought that the developers might appreciate some feedback from an "outsider". While I have followed the alpha videos, I have not really played any alphas after a209, until a226. Within this post I also mention some thoughts about the design direction, which could be up for discussion and I'm interested in hear other people's thoughts(the developers' thoughts would also be interesting to hear).

With that being said, here are my thoughts on alpha 226:

Visuals/Graphics
The graphics looks fairly good overall. There are some improvements that could be made here and there, most of which is probably already mentioned in the "nitpicks"-thread. However here are some of the things that I was thinking about when playing the game:

1. The lighting in some levels work quite well, such as the Waterfall-level and the "Stucco Courtyard". (The reason why I think it works better in those levels is because they are set in daytime where water could reflect the light, making the light environment feel more natural and the pavement in Stucco is light colour and thus could reflect more easily)

Whereas in the other levels "Magma Arena" and "Rock Cave", both feel a bit off, in my opinion. I understand that it's important for the player to have a clear vision of the arena. However, a good lighting could also drastically change the mood of a scene.In the Magma and Rock levels, a suggestion could be to add torches on the walls and thus making most of the places still a bit lit. Also make some places darker so that it feels more "real". (the darker and lighter areas could also be used as a strategical placement for creating interesting combat scenarios, where people could hide in the shadows)
Below are some images of different areas to showcase what I mean:


Above are two example images of the magma Level. I think that the second picture shows more what I was referring to.



Above are two example images of the RockCave level.

In conclusion to the lighting, I think that it could probably be fixed by recalculating the AO(Ambient occlusion) and possibly increasing the contrast. (although I'm just speculating here)

2. The skybox in most of the levels are a bit low resolution and some of them don't quite fit in with the general artstyle. For example the skybox in the Lava-level looks quite like a low-resolution version of a real photo, whereas the rest of the models and textures are not photorealistic looking. (and I'm not saying that you should aim for photorealism, but maybe instead change the skybox to fit more with the current artstyle?)
This also applies to the other levels aswell. However the Stucco courtyard works fairly well with its skybox, even though it is mostly blue. (probably has something to do with the colour scheme being used in that level)
My guess is that the skybox in Magma level also has something to do with the lighting feeling slightly off, as mentioned previously.

This image is just an example, the other levels' skyboxes should also be looked into.
While the skybox itself looks quite good, I think that the "photorealism" in it might cause a slight dissonance in aesthetics in comparison to the rest of the in-game assets.

3. In some levels I have noticed Stretched textures. I found it in the Waterfall arena looking out from the waterfalls.

One interesting thing in this picture besides the stretched textures on the rocks in the background, is that the enemy AI seemed to get a glitch with its head when I took the photo. However I did not notice this while playing the game, so I don't think that it should be in the highest priority. (see further down below for my aniamtion analysis)

4. The Animation in the game is for the most part really well done. It is to no surprise that it is well recognised by other developers and for anyone interested, I highly recommend David Rosen's talk on it on GDC.
Personally, I think that it works really well. It feels very responsive and with the slight changes in rotation of the rabbit's ears, the characters feel more alive. I also really like the dynamic system that is being used in the game.

A very good example of the animation system. Notice how the right foot is in such position and rotation so that it is placed upon the rock. I'm guessiont that IK(Inverse Kinematics was used to do this kind of effect). However, I'm always intrigued to hear more from David Rosen on how he did the animation.

For those interested in the GDC talk, here's the link: http://www.gdcvault.com/play/1020583/An ... e-Approach

5. The particle effects looks in general quite good.
I like that you can see the the blood drips from wounds. However, unless you are actively looking for it, it might be quite difficult to notice and some wounds feel like they should generate more blood from a logical standpoint. However, I understand that not everyone likes blood and I tend to turn it off if I can in games, so I like that you can do that in this game. However, blood could be a cool mechanic of using to track down people, or for the AI to track down the player.

The water in the waterfall level looks quite nice aswell and I think that it doesn't necessarily need much more work for the moment, unless it has for performance issues.
However the waterfall particle effects seem to have some pointy edges and sometimes the textures seem to stack ontop of each other/clipping through each other. While the clipping might also be present in the regular water, it could possibly just be ripples in the regular water.




The first picture shows the "pointy edges". The third picture shows the clipping textures, aswell as the 15 fps in the bottom left corner.

Another thing to note regarding Performance is that I only seemed to dip below 60 fps in the waterfall lever. Which makes me beleive that it is probably due to the waterfall particle effect. (while inside the waterfall, the fps could go as low as 20 or lower)
One idea to solve this could be to have the waterfall be a model with animated(moving) texture and only have some particle effects at the bottom where the waterfall meet the water.
(Though I've only worked with this kind of method within the Unity engine so I'm not completely sure how it would be done within the engine this game uses.)

Audio
1. The sound effects for the most part work really well within the game. The sounds of swords hitting different material, the footsteps of the characters, the foilage in bushes all make the world feel more immersive.
However, here are some effects that I thought about while playing:
- When near a fire, I can't hear any flames. (or rather the sparks of the fire)
- The near a tree, there are no wind blowing in the leaves. This could be intentional in the tutorial as it is not in a windy area. However I have not tested if it is the same in all areas.
- The sound of the bunnies fighting sounds a lot like dogs fighting? (I'm not sure as I don't have a big referrence on dogfighting and this might probably also be just a very small nitpick)

2. The Music is very good. It gives a good feeling of tone of the game and the fighting music has a good rythm to it.
The transition when going to the menu however, is a bit slow and in the beginning I thought it was a bug, but it seems as if it's just a bit slow to transition.

3. The Audio menu is working quite well. Although I think there should probably be a separate slider for the sound effects so that they could be changed separately. (instead of indirectly changing it by changing the master volume and music volume.

Gameplay
1. The Artificial Intelligence(AI) works quite well, although it can sometimes bug out with its pathfinding. Some cases it jumps back and forth two spots. Other times it could stand on the bridge while you are underneath it.

For a further analysis of the AI, see below when I talk about the combat.

2. The character movement works fairly well, especially with the very fluid dynamic animation.
However I'm wondering if maybe the speed and jump-speed is a bit too high. With higher speed the player might get better response from moving the character, but the trade off is that the player moves very quickly through the terrain and the character could also be more difficult to control, and cause less precise movement. With a slower movement speed the player is more encouraged to stay around and look at the environment and possibly also be able to talk to NPCs. (if NPCs will be in the game)
As for the jump-speed, I think it also might be a bit too high as the player can cover a lot of ground with that.
After showing the game to a friend, he mentioned that the jumping felt a bit off and that he suspected the jump speed to be constant until the character reaches the peak when it goes back downwards.

I noticed that it's quite difficult to do wallruns, since the character usually bounces of the wall. I can understand that it is trying simulate how it would work, however I think that it is important that the gameplay is enjoyable for the player aswell. (an example of a fairly good wallrunning is in Mirror's Edge)

The climbing works fairly well, however the I noticed that when holding a weapon while grabbing the ledge, the weapon flies away. I understand that maybe it is meant to make the player sheath their weapon before jumping onto ledges, but I think that to make it seem less buggy, the weapon could simply drop on top of the ledge that the player has grabbed. That way it's less punishing but at the same time it tells the player that they could save a bit of time if they sheath their weapon before hand.
Also I think it would be cool if the player could turn around while holding the ledge and then jump from there to another ledge. (kind of like in Mirror's Edge). That and with the combination of a slower movement speed could generate more precision to the movement, leaving more room for stealth gameplay.

One thing I noticed while Dragging bodies is that it's quite difficult to get it to a certain spot without going there yourself. So a suggestion would be to maybe add a throwing animation for throwing the body. For example throwing a body into the lava.

3. The combat could definitely be discussed about a lot.
I think that it is very fast paced and while there are a lot of different moves which the character can do, a lot of the times I just end up repeatedly mash the attack button. (and mostly win the match)
When the combat is very fast paced it is difficult to start thinking about tactics.
Aside from that, the different combinations, such as different range does not necessarily make it easier to handle them. When the combat combinations are not completely reliable, it is difficult to create tactics around them.
An example could be in Streetfighter where it requires a certain input to make a move and since that same move will always be executed the same way with the same input, it is very reliable to create a combination of different moves as a tactic. (though I don't think that the same input combination from street fighter would necessarily work well within a 3D environment such as in Overgrowth. But maybe soemthing like in super smash brother could work... [just throwing out suggestions])

When discussing the combat with a friend of mine, we compared it to Legend of zelda where most AIs are very different from the player, giving a different combat scenario with each enemy.
Since most of the enemies are very similar to the main character in this game, it could more easily become repetitive.
While having the same opponent as the player character could be argued is for encouraging skilled players, I've felt that the combat so far has been pretty easy.
Along with the fast combat has also come the difficculty to respond to feedback. Usually when I've played and won, I've not always felt like I got a lot of satisfaction from it, or that it was a challange. When I die, I don't always know what made me die and I think that these things might be since there maybe isn't enough of feedback for the player to know what they did good and what they did bad. Knowing what works and what doesn't in an opponent, is good for the player to improve in the game.

Also the restriction of the player not being able to attack unless within range also limits the player from creating cetrain tactics. I've noticed on the waterfall level for instance, that if I stand on the stone and wait for the AI to jump to it, I can't attack before it has come close enough, which then makes it difficult to attack in the expected direction. (What I mean is that I wanted to be able to attack in the place where I expected the enemy the land, so that my swing would hit the enemy when it landed, but due to the range it was not possible to do so)

Story/Level design
While there isn't much of a story in the game at the moment, I thought I might include this to discuss some topics which I think could be considered while making the game.

1. The tutorial level starts out quite well, however the instructions doesn't change depending if you're using a controller or not.
Another bug is that the enemy once continued being red even after I had finished the grabbing tutorial.
Aside from that regarding level design, is that while it works quite well for the combat and introduction to movement, it's a bit harder to do the wallrun in that area considering that it's a closed room, with very short walls.
I think that having a "corridor" could possibly give better opportunity for the player to practice´the different moves and also give the player a sense of progression.
As for the colour scheme, I think that it works quite well. It gives off a friendly vibe and feels like a good tutorial level. Hopefully there could be NPCs around the to make the place feel more alive and give the player more story than just a text message telling the player how to play the game.

2. The arena levels:
- The waterfall and the magma levels are both pretty good designed. They give a lot of room for the characters to move around.
- The Stucco courtyard and the RockCave level also have open spaces, however they don't leave as much room for environmental tactics.

3. I really hope that there will be a NPCs in the final game, that the player can talk to, and that there will be an engaging story in the game.

Final words
While I have mentioned a lot of things, I think that for now, it is very important that the developers focus on things like level design and story. Make more levels so that we can test them.
It is also important not to get too stuck in the feature creep.
Hopefully this post has been informative and helpfull for you and I would love to hear what people(including the developers) think about it.
Thank you for taking the time reading it.

[Constance]

Damn, this thread should be really helpful for the devs in general.
One thing I noticed about the wallrunning and slope physics is that some objects like the Cat Patio Box have walls that are somewhat sloped, but are way too steep to realistically be climbed or be stood on, even though the character has some troubles trying to wallrun and walljump on it.

Hopefully the species will get different combat styles to break up any repetition you get from fighting normal rabbits over and over, not to mention weapons and specific characters.
The water is still very rough, and I agree that the edges are a bit noticeably sharp. To be honest, I personally didn't expect official water being in the game in the first place.

Wind breezing through trees would absolutely give forested scenes a whole lot more atmosphere, and I'd definitely appreciate it. The fire is actually pretty early, but sound effects would probably help sell the effect.
As for the comparison to dog sounds, I never noticed that, but now I can't un-notice it.
I'd say more music and/or variations of the same songs would absolutely go a long way to help each level feel unique.

Sometimes, when the AI can't figure out how to use a jump node (for example, they don't know how to ledge climb), they'll just wiggle on it indefinitely.

And maybe, when you're holding a weapon and you try to grab a ledge, you should grab the ledge with your free hand and leave your hand holding the weapon free. Then you can either choose to sheath the weapon, or drop it below. I also always thought it'd be cool if you could hold SHIFT while you're on a ledge to try and keep yourself hidden from where you're climbing, so shimmying around and below the area enemies are patrolling would be a viable tactic.
Being able to turn around and jump away while on a ledge would actually give a lot more options to freerunning and parkour courses; as someone who loves to create maps, I would have a lot of fun experimenting with that option.

As for carrying bodies, I still say that a Metal Gear Solid style system would be perfect for Overgrowth.

Once you're experienced enough in the combat to be able to land each attack on a whim, the problem of spamming the attack button and not thinking about tactics completely vanishes. It gets a whole lot better a whole lot faster once you are completely certain of each control scheme.
That being said, I bet more moves would actually refresh combat as a whole: but how would you fit more moves onto a simplistic control system?
I'd say:

• - There's only one crouch attack right now, so why not give a few more? A different crouch attack for when you're moving, another for when you're close to the enemy, and another crouch attack for when you're close and moving at the same time.
  
  Either that, or:
  
  - Lock on to enemies Dark Souls style and you can input new moves by stepping backwards or forwards, just like how pressing A and D will trigger a moving attack.

The first option would fit into Overgrowth's mechanics already, but the second option would have to take a lot more work and refining to get right. Since Overgrowth has a focus on using no HUD, getting group fights to work without a visible circle around the enemy you're locked on to would be a bit messy without any HUD at all.

I also agree that the tutorial could teach a lot more about the movement to the game: it's good at teaching you the combat, but maybe there should be another segment where you move into a small parkour course?


This thread is a good reminder that Overgrowth is a game in development, and not a developer kit. I seem to forget that all the time.

[Lime_x]

You have some very intersting points.
Your post reminded me of some points that I forgot to bring up my first post.

I agree that the Metal gear solid style of moving bodies would probably fit really well with Overgrowth and it was kind of that style I was thinking of before.

I also agree that the wallrun could probably need some changes to it to be more easily accessible to the player and to make it easier to perform.

In terms of combat, I see what you mean that once the player has enough experience they will be able to think more about tactics and I do have some sense of how the combat works. However even if I know the moves, I feel like it's difficult to sort get a moment in the combat that feels like it's caused by the player. The roll is one move which I think add such element. It has a very distinct reaction to player input and shows very clearly what happens. On the other hand, while the dodge moves are in fact controlled by the player, since it is such a suddle input choice and a suddle animation, it could be hard for the player to connect the dots and understand the correlation between the input and the dodge. (Side note: The dodge move is not taught within the tutorial, if I remembered correctly)
Edit: After playing some more, I did notice that the dodge animation is fairly telegraphic, however I still think that it sometimes is difficult to get the right "feeling" from using the dodge. Since the animation for it is pretty good, it might be due to the fact that it uses the the analog stick for dodging. (It's the same keys for dodging and for aiming, which might be a bit confusing for the player, though I'm just speculating)

I like the idea of being able to choose whether or not you want to target lock on an enemy. It might also fix the issue with the camera control issue that I mention later on in this post. However, I agree that it might not be according to the usual style of Overgrowth to have a big circle pointing at which it is enemy is targeted. But maybe that information could be conveyed in other ways. such as changing the shader on the targeted enemy to be lighter in colour and the other enemies have darker colours? That way the player would easily know which one is in the focus right now and it would still feel pretty immersive.
There could probably be several ways of doing the target lock.
1. Letting the player choose with a specific button á la dark souls. (and then letting the player change targets with the right analog stick, or keyboard and mouse equivalent)
2. Having the game autoset the target lock when the player gets close to an enemy. (like the current version of Overgrowth)
3. Have the target lock only occur if the player engage in combat while close to or approaching the enemy. For example if the player runs towards the enemy and attacks it, then the enemy is target locked. However, if the player just runs by without attacking or blocking, then the camera doesn't target lock and lets the player continue without interruption.

I agree that there could probably be a few different crouch. The current way that Overgrowth's combat system works is kind of like Mirror's Edge Catalyst and Dark Messiah of Might and magic, in which the player can change the direction of the attack by moving in the direction. I think that when not using weapons, Overgrowth works quite well with that system. However, one thing which I've still found that it's difficult to both manage the distance and the direction of the attack and since the attack changes depending on the distance that also means that it's a bit difficult to prepare with different attacks. Also I noticed that sometimes when I attack I do a one hit kill round house kick and other times the kick just hit the enemy in the side.

However, while the no-weapon combat has quite well telegraphic attacks, I have had a lot more difficulty to see the different attacks when the character is using weapons. One thing which Dark souls usually do is have a lot of visible moves. Even if it sometimes is a bit exaggerated, the animations in dark souls are usually made to let the player have time to react and leaves room for tactics.
One thing that could also give more tactics in sword combat is if the player show slightly which side that the enemy is trying to block. (kind of like in Legend of Zelda Skyward sword)

This is just an example of how the enemy could show it. It's probably a bit too exaggerated, but I thought it would be to show what I mean. Video is of skyward sword skeleton boss.

Here's an example from dark souls. Again, the animation is maybe a bit too exaggerated, but it shows how the player is allowed to react to the enemy's attack.


Here is an example of when I play with the sword. It seems that sword fights end very quickly. Which might seem a bit anticlimactic/chaotic.

Camera control
Another issue I've found within the combat, is that when playing with a controller, the camera control could be somewhat difficult to navigate properly.(this problem is not prominent with keyboard and mouse, since the player then has better control over camera)
1.I did notice that there is an automatic camera options under input menu. However when using it in combat, there were times when the camera would "jump around" and I assume that it would be when the target is changed. This causes a very confused feeling and makes it difficult to track what is going on.
I've also noticed that the camera starts working strangly when in tight spaces while enemies are around, such as on the lower level of the Magma arena. It seems as if the camera tries to adjust to the enemy being there, but can't quite do it since there is a wall in the way.

This Gif shows the part where the camera jump cuts between targets during a fight. It is a bit low in framerate, but I think it is noticable how the camera drastically changes places a few times. Also the camera seems to start look at the players front instead of the back, when using the automatic camera.


2. When the No automatic camera option is being used, the player has full control over the camera. However it can take long time to adjust the camera to the right position with the controller and once that is done you usually have to readjust the position, since the combat and action is so fast. And not adjusting it at all could lead to weird camera angles. (though since the automatic camera option is there, I think it would be recommended for players who use the controller to use the automatic camera option)

This is just an example of using the no automatic camera option.

3. Regarding the automatic camera, one thing I noticed was that the player was usually centered. An idea could be to have the camera slightly move to the side to give the player a better view of the enemy's moves. ()


Settings Menu
While I for the most part think that it is fairly okay, here are a few thoughts that came to my mind when testing the game:

1. Currently it is not possible to rebind input key for the gamepad(at least not when I used an xbox360 controller).
2. The Graphics settings could possibly have a few more options such as different Antialiasing, shadow settings, Field of view slider and being able to adjust the distance of the camera to the character.

This image is the current version of the settings menu for graphics and input.

Another thing which would be good, is if the tutorial could change to show controller inputs if the player is using a controller and not keyboard and mouse.

Tiling glitches in the Tutorial level
While testing the turial again, I noticed quite a few seams in the tiles, which can be viewed in the images below.






The last bit about seams is probably a bit nitpicking, but I thought I would mention it in case the developers sees this thread.

But aside from that, I also agree that a small parkour course could probably be a good way of introducing the player to the movement system.
The game seems to have been like a developer kit for a long time and I think that it looks quite modder friendly at the moment. So I think it would be good if the developers started focusing on the story and gameplay. (which they have been doing a bit more lately)

P.S. Also Timbles, I really appreciate that you took the time to read through my post and replied to it.

[Lime_x]

In hindsight I realize that my second post might be a bit long and convoluted for people to read, so I thought I might add a brief summary of it.

TL/DR;

• While controller support is okay for the moment, it might still need some more work on it to work better in combat. At the moment, keyboard and mouse might be the recommended choice.
• It is sometimes a bit difficult to distinguish the different moves in sword combat at the moment, so possibly make the animations of the moves more easily distinguishable from each other? (both player and enemy moves)

I appologise for any inconvenience I may have caused.
If anyone has any thoughts on it, I would love hear them.