Third Person Controller - Third Person, AI, Multiplayer, Mobile Framework

@scissorman1 - I have a fair bit of experience using all three of the top third person controllers on the Asset Store (Opsive's, Invector's, and ootii's), so perhaps I can provide some additional insight beyond the answers you've already received.

It's the animations. Opsive's TPC uses Animator.Crossfade() when switching states, but the movement is all handled using standard blend trees. Replacing the stock animations with higher end ones, such as the packs by Kubold that @Teila mentioned, make a huge difference. I have most of Kubold's packs now, and I've set them up for all three controllers. You can get smooth and responsive movement out of any of the three.

One thing I did notice: I was prototyping a concept using each of the three controllers. I used the same character model (one I whipped up on Fuse), the same animations (Movement Animset Pro), and the same environment (the demo scene from Urban Underground). No other characters in the scene.

Opsive TPC gave me the best performance, at a very consistent 58-60 fps. ootii Motion Controller was pretty close, but not quite as consistent a frame rate. Invector TPC (Shooter) was averaging slightly lower (53-55 fps), but it would get these random drops to half that frame rate for no apparent reason (just walking around). I've noticed similar results in another environment I tested all three in (Medieval Village).

You do have some control over the parameters of the jump in TPC (Force, Air Speed, Air Dampening). As Justin said, swapping out the animation used might help. I've found that swapping out the jump animations is rather tricky to get right, across the board (all three controllers). You can, of course, build your own custom Jump and Fall abilities if you find that you need to do more than change a few physics parameters or swap out animations. Honestly, I find the jumping to be one of the weak points of all three controllers.

But on the topic of customizing the code... that's one area where Opsive's TPC stands head and shoulders above Invector's. While it has come a long way from where it was a year ago, I still find Invector's code base to be a bit of a mess. When you take a peek under the hood, it's immediately obvious that Opsive's code is written by an experienced professional software engineer. The code is clean and well-organized, and there is a cohesive design and structure across all of the parts of the controller. Most of the time, when you want to customize something, you can inherit from an existing class and build your customization there. If you have to modify the source code, it is unlikely that you would need to modify the "core" files (such as the RigidBodyCharacterController), thus reducing the ripple effect of making changes. (All of what I just said regarding Opsive's code and design applies equally to ootii's as well)

It's more difficult to customize Invector's controller without having to modify the core source files. For example, they have a Rewired integration on their forums, which gives you instructions on replacing all of the Unity Input functions with Rewired versions. With Opsive (and ootii), you just grab the Rewired integration script and use the Rewired Input component in place of the controller's Unity Input component. Invector uses a fairly deep inheritance chain for the primary components of the controller (input, animation control, movement, etc), whereas Opsive tends to follow the "favor composition over inheritance" principle. This makes it a lot easier to make changes that don't require editing the main source files. Not that it is difficult to make those edits, but it can be a supreme hassle to maintain regular updates from the publisher when you have to merge in your own changes every time they release an update.

That being said, I can't deny that Invector easily has the most polished and impressive demo. And while I have issues with how their code base is structured, I've read a number of comments from people about how much easier it is to understand Invector's code than Opsive's. That's a potential downside of taking an approach that favors loose coupling, programming to abstract interfaces, using event aggregation, etc. The layers of indirection keep the code more flexible and easier to extend/modify, but the flow of logic can be harder to follow if you're not used to using those design patterns.

It's easy to reposition the weapons if they are off. There is an option to apply a bit of recoil force when firing, but that mainly seems to affect the gun and the character's forearms. If your shooting animation contains a recoil movement, you wouldn't really need to do anything else.

If you haven't already tried it, check out the demo for the Deathmatch AI Kit. There is a free integration package for Behavior Designer that add support for TPC, but the Deathmatch AI Kit shows what you can really do with the TPC + BD combination. (As an aside, I have zero interest in that type of game, but I picked it up so that I could have a concrete example of an advanced behavior tree to study)

The tight integration with Behavior Designer is an advantage that shouldn't be overlooked. Sure, you could use a behavior tree with Invector's TPC if you want, but you'd have to code your own set of tasks for doing so. BD with its add-ons (Movement, Formations, etc) will give you a far more robust AI that what you'll get with Invector's built-in AI. Of course, it will be substantially more work to get it all set up the first time.

While I understand your point, I would much prefer that Justin focus on getting TPC v2.0 finished and released before spending his time making tutorial videos or nicer demos. Now, TPC v2.0 should absolutely have a polished demo that wows people in time for its official release.

Put it this way: Opsive TPC has 323 ratings with an average of 5 stars (Behavior Designer with 493/5-stars). That speaks volumes about the quality of these assets. Requiring that he "wow" you with a flashy demo comes across as kind of silly to me, if I'm being totally honest.

I'll see if I can build the TPC version of the prototype that I mentioned above. There's no way that I'm putting a public link to that up as an example of my work, as it is nowhere near polished and plenty of stuff doesn't work. But if you want to see how TPC feels with a good set of animations, drop me a PM and I'll see what I can do.

 

Thank you for being understanding. Some people respond negatively to criticism so I was initially worried about posting my concerns.

I definitely understand your issue with not wanting to spread yourself thin on animations, since you are right about people replacing them. However, they are also a big part of the first impression the demo scenes make, so the strange way the Shooter demo character holds his rifle really taints the way the controller comes across. I'm sure that's why those animations get such significant criticism, and I think you're doing yourself a disservice to rationalize that issue away. I don't know if you could use some free animation asset or some from the Standard Assets, but I feel like putting some polish on the things that are supposed to sell us on the product would be very helpful. If you could take a look at the current demos and perhaps fixing a couple of things to make the presentation a bit better, I think that could go a long way.

The Platformer demo definitely feels better, so it might be that the root motion when the character switches from his upwards jump to the idle animation causes him to be pulled down. But that's the kind of thing I was talking about- a small omission like not having a fall animation makes it seem as if your controller's entire jump ability is broken. Its like if you were shopping for a car, and one was sort of beat up and the engine sputtered while doing a test drive, and then another car is all polished and clean and runs smoothly- you'd probably go with that second car.

I also totally get wanting to just work on making what you feel is a good product, but those of us who might want to buy your product can't really see the cool stuff you're doing under the hood unless you show us through demos or videos. For instance, there is a flight action documented on your site, but there are no videos or screenshots that show it, nor is it in the demo as far as I'm aware. Or the integrations- I'd love to see some videos or demos showing how those work. If somebody owns Inventory Pro and was able to see your asset using it, that might push them to buy it!

I guess what it comes down to is even if you have a good product, you've also got to market it so that customers will want to buy it. If the craftsmanship of your end product is sloppy it covers up all your hard work.

Thank you again for your time, and I hope you take this into consideration.

If you have or want to make a Youtube video your character using the TPC with those animations, I would love to see it (and I'd wager other people would too.) Seeing people build stuff like that might do a lot to shake the stigma that the built-in animations.

Thanks for replying. I definitely understand the performance issues between the two controllers- Invector's poor performance on weak machines is largely what drew me to Opsive, but glaring and apparent flaws in the presentation materials are also what keep me on the fence.

As for demos, I completely understand current owners of the asset wanting the developers to go full steam ahead on creating V2, but that also means those of us who would be purchasing the asset won't be getting anything until around the end of this year. Ignoring the current product on the market in favor of a potential future product leaves consumers in the dark until this new asset is done.

I also think it's a little unfair to characterize wanting to see the capabilities of a product before a purchase as "silly." I understand that you may have used the asset inside and out, but you've got to understand that those of us looking at the product from the outside only have a few old Youtube videos and some somewhat sloppy and incomplete demos to go on. To return to my car analogy, if one car is able to show you a supercharger, WiFi, and a big audio system during a test drive, while the other's air conditioner doesn't work but the salesman insists that it will if you buy it... It's clear who makes the better impression.

If you are willing to build a simple demo just to show people what the controller can really do with fresh animations, I would absolutely love to see it. I would however urge to reconsider making it private. I am sure I am not the first or only user to have misgivings about the asset based on the existing demo materials, and I'm sure Opsive would appreciate having something new they could point potential buyers to as a demonstration of the product.

 

I'm looking at the rifle aim now and here's what I see:


Is this the same thing that you are?

You can fly in the clean scene with the 'Z' key We have made a lot of progress on v2 and are a couple of months out from starting our major marketing push with that. Once we start this I think that you'll be seeing way too many videos from us - it'll be nice because for the most part any video that I create for TPC will also apply to UFPS so we're getting twice the benefit of the videos. The same goes for the demo - we have a *single* demo scene planned for for both TPC and UFPS with just the core weapons so that scene will be extremely polished. I know that doesn't help now but we are definitely aware of it and plan on making changes with v2

 

Since we do not have anything set up with the animations I am sure you want to see, it will have to wait a bit. We have a deadline to reach first. However, we have tested all of Kubold's melee animations, many of them with UMA, in complex scenes with no issue.

We are not using shooting animations or anything of that sort.

Actually, my programmer reminded me that our goal is a trailer by the end of September so you can wait until we do that and see the animations. Or you can just trust the folks on here who have told you that better animations work better.

 

The stock TPC animations are certainly clunky and more than a little awkward-looking (sorry Justin; fantastic controller but the animations aren't pretty). However, taken as a whole set they are visually consistent. Plus they are designed such that upper body "action" animations can be blended with the basic locomotion animations running on the lower body.

There are no free animations that cover all of the actions that TPC needs to animate. Mixamo has some decent animations, but while those are free to download, you cannot distribute them, which makes them a poor choice to use in your asset's primary demo scene. Mixing and matching free animations from multiple sources will not give you a polished-looking result.

I think that Kubold's "Animset Pro" series are generally considered to be the top-tier animations on the Asset Store. And it's pretty easy to replace the basic walk/run animations with those from Movement Animset Pro and have it look pretty good.

BUT...

If the default jump animation looks/feels weird to you, then this will only be amplified when you substitute the higher quality walk/run animations. Same with the stiff climbing animations (and this leaps out at me in Invector's demo... Most of the movement is fluid looking and then you get that super rigid looking stock Unity ladder climbing animation).

So while TPC's animations may be clunky, what looks truly sloppy and unprofessional is a mishmash of various free animations, or an incomplete implementation of Animset Pro animations. If you look at Kubold's packs, you'll see that all of the melee weapon sets have the character in a realistic fighting stance, the hips rotated such that the upper body isn't square with the lower body but the character is standing at about a 45 degree angle. The attack, defend, etc animations can't just be blended with the normal walking animations; you have to use the full body stances or it looks weird.

So it takes a significant amount of time to replace everything so that it looks consistent. And this is why I don't want to simply post a public link to a demo, as I have not spent the time replacing and tweaking all of the necessary animations. My demo looks sloppy and unprofessional because there are a lot of rough edges and a lack of consistency. You've said it yourself that you need to see a more polished example before you feel comfortable making a purchase; I foresee getting into a long discussion about which aspects of my demo look bad because I didn't finish the work or didn't spend time polishing it. And it's the polishing that can be truly time consuming.

TPC's Clean Scene demo looks and feels like an early game prototype should. I've worked on a couple of AAA titles and this is pretty much exactly the sort of thing we started with.

Kubold's animation packs are generally about $65 per set (each set covering one weapon type), which is almost the cost of TPC itself. Invector uses a lot of free Unity animations and some custom animations to fill in the gaps. But there's only one set of melee animations (one handed with shield). TPC is a lot more comprehensive and covers more actions and weapon types. I think you're seriously underestimating the time and cost that would be involved in improving TPC's included animations.

It does have a fall animation, and the jump is controlled by physics, not root motion. Might just be the animations that you have a problem with. But as I said before, I've found that swapping out the jump animations rarely seems to be quick and easy (with any controller). Usually you need to chop up a longer animation into multiple pieces (as the code controls moving from one state to the next), and it can be very time-consuming to get this looking right.

The end product is the character controller, not the demo. Just wanted to stress that point.

Also, Opsive TPC is the top-selling third person character controller on the Asset Store, by what appears to be a considerable margin. I don't think that he has had much of a problem with customers not wanting to buy it.

He's not ignoring the current product on the market; he's got someone who is updating and maintaining TPC 1.x. He's just not investing the time and money to product a flashier demo for a product that will be replaced by the end of the year.

Very, very few assets on the store have demos that even approach being polished. A lot of them are just plain broken. But the majority of these assets are also created by either small teams or individual developers, the vast majority of whom don't make enough money on the asset store for it to be anything other than a hobby. In general, you've got to use your imagination a bit as to what you can accomplish with a given asset. Ideal? No, but I'm just being realistic.

When you're developing a game, you have to be able to extrapolate what the finished product might look like from rough, unpolished raw materials. Like when we do greyboxing.

You mean exactly the same things I had to go on before I first used the asset?

TPC's "Clean Scene" demo shows off multiple controls types and camera perspectives, all/most of the different Abilities (climb, vault, cover, swim, ride horse, pushing objects interact, crouching, ledge strafing, balancing, moving platforms, etc), and lets you try using a variety of different weapons. It's not incomplete.

Invector's demo scenes are outliers as far as what you'll find on the Asset Store goes. Don't use those as your benchmark, or you'll find almost everything coming up short.

And that's why you then do your own research and read reviews of the cars you are considering and completely ignore what the salesman told you. The flashy demo speaks to your emotional reasoning, not your logical reasoning. Taking a car for a test drive only lets you experience the car on a superficial level. You can't experience the long-term reliability of the car first-hand on that test drive, nor can you really experience its safety features (at least I hope not!). You generally have to rely on independent reviewers, the experiences of other owners, and the opinions of people who know how cars work.

Also, we're talking about $75 game assets, not $25,000 cars.

 

You are considering buying a 3d controller, not animations. Opsive is not an animator. The wonderful thing about the asset store is that you can buy a great third person controller that is made by someone who knows exactly what he is doing. You can then either make your animations (a very good way for you see how difficult they are) or you can buy some decent ones made by an animator who knows exactly what he is doing.

You are buying the product described, not the animations. What you are implying is that anyone who builds animations should also include a really great character model to show off those animations. Anyone building a terrain tool should also make their own amazing textures to go with it.

That is not how it works. Assets are tools and you are supposed to add the polish. I understand that you are hesitating on buying this because you have not seen the tool with better animations. You have other options. Some of us have tried more than one option for that very reason. Choose the one that looks best to you and use that. Hopefully, your first pick will be the best.

Or wait until after Unite in Austin and we will show off our use of TPC with Kubold's animations. I am not sure what else can be done to make you happy.

 

I totally get what you guys are saying, but after weighing everything I'm just going to have to go with Invector. There are just too many issues with your controller at the current stage for me to feel comfortable building a project around it.

Thank you for your time, and I wish you luck in the future.

 

No problem - while I do think that you'd be better off with our controller I do hope your project goes well. We will be working on a polished scene for version 2 so make sure you keep an eye on this thread or Twitter.

 

Hey squared, is the documentation on Opsive’s website right on how to update the project with the source code? I don’t see a ThirdPersonController/scripts folder

There is a lot in wrappers, and many under demo scenes but no scripts folder. And when I import TPS source code it goes into assets\TPS but my project has it originally set up as \assets\ThirdPersonController

Any help would be great!

 

The documentation is correct - if you haven't changed anything then there will be a Assets/Third Person Controller/Scripts folder right after import. Just to be sure you have the latest try removing the cached package from your Asset Store download folder and then download the asset again.

 

Hi! I what to know if somebody could help me with it issue with the crouch ability, the player have some parts under the floor. Could it resolve with some propierty in the Third Person Controller? O I need change something in the model?, When the player walking or stand up, all is perfect. Thank you!

 

Hmm, it looks like the proportions of your character is causing the crawl animations to move lower than normal. If you click on the animation you can adjust the y offset - this should allow you to move the character up.

 

Thank you! it work!

 

I wish there were crafting animations, as my game will be mostly about crafting. I would have liked to see such animations as: sawing, coocking & tasting, hoeing, planting, working at a worktable, reading, chopping, watering, Hammering, etc.

 

This asset is geared toward being a character controller that you can use with any genre and if you have crafting animations you can use those animations without modifying the core code. Take a look at these videos for how to use the genetic ability and also creating your own ability:

http://opsive.com/assets/ThirdPersonController/videos.php?id=9
http://opsive.com/assets/ThirdPersonController/videos.php?id=7

 

Thank you for your kind advice. I was just hoping that someday someone would make a character controller that looked a little beyond combat manoeuvres and parcour as standard.

In your video you said that people should make requests, so may a make this as a hopeful request.

Thanks for listening. I very much appreciate it.

 

Noted We are currently focused on version 2 which will basically combine the featuresets of TPC and UFPS so it won't contain any crafting abilities but after that we should be able to iterate on updates pretty quickly.

 

oh thats greate, I would be so gratefull, I've been looking for a long while.

 

I want to perform melee attack in direction player is facing.Character movement is adventure type and camera is in third person, character can walk and run in direction which it is facing but when i perform melee attack it rotates in direction camera is facing. Ex- character is running backwards facing camera but when i perform attack it automatically turn around in direction camera is looking.

 

Do your attack animations rotate the character at all? I just tried melee attacking within the adventure scene and the character did not turn to face the direction of the camera.

 

Hi there! im trying to setup a networked Scene, but i continue having this error:

Error: A NetworkIdentity component was found but the ENABLE_MULTIPLAYER symbol is not defined. Please define it within the RigidbodyCharacterController inspector. The source code must be imported for the symbol to be detected.

I Already had "pushed the button" on the inspector Add multiplayer symbol, as the documentation says, but im still having the same error, all the time...
I tried on the sample scene, pressed the add symbol, and still error.
on the same demo scene, created a new character, and on blank scene created a new character, always with networking checked, and now the "Add symbol" not even shows on the rigidbody inspector.

Any hints on this would be really appreciated.

 

Make sure you've followed the steps from this page. Step 2 is especially important. You'll also want to make sure you aren't using the WebPlayer build target since the WebPlayer doesn't support uNet.

 

Hi! I have a issue with the push, when the character push a box backfoward and going to the wall, it don't stop and the character cross the wall. I try to make a playmaker event to stop that, It work but I want go to idle state again (Stop the ability of push) so the player can press F to restore the action. I see in the Adventure demo have a little trick and the crate only move where the collider is available. But in my game I need move the box around the scene. Thank you!

 

Are you running the latest version? The Push ability has been renamed to MoveObject, and I just tried it in the clean scene which doesn't have a track and the box stopped moving when it hit a wall.

 

Thanks for you quickly answer! Yes I have the last version, in the demo Adventure work fine, I can't find why don't work for me, my collider is small (0.4)? all setting is very very similar to Adventure Demo. I copy the same values in the crate , naomi and the move object.

Thanks!

 

Does the clean scene work correctly for you? This scene doesn't have a track so is more similar to your use case.

The collision is handled by the rigidbody component. Depending on how quickly you are moving you may need to set the collision detection on the crate to continuous, but your object is large enough that I wouldn't have thought that you would need to do this. Continuous collision detection is normally used for small fast objects such as projectiles.

 

Hi! I made some test, new project in Unity 5.6.1p4 and only Third Person Controller the version 1.3.7 and the problem sitll there



And I use the demos scene too without the lab_Extra tube



Thank you!

 

I have been trying to reproduce this but haven't been able to:



Did you make any changes to the adventure scene in order to get the crate to move through the wall? The Third Person Controller doesn't do the actual collision detection, it instead relies on the rigidbody in order to determine if it is colliding with another object or not. This is a pretty simple use case so I'm surprised the collision detection isn't working properly.

 

Hi! Try to pull (not push) to the wall (backfoward) and remove the lab_Extra tube or use a simple cube with boxcollider as a Wall. The only modification I made is in this values. The player will go to inside the wall but not the cube

 

Ah ha! That did the trick. The issue relates to the MoveObject.UpdateMovement method - it is setting the position of the character directly which bypasses the Rigidbody collision checks. I'm thinking of a proper solution for this and I'll let you know!

 

Try Mixamo.com. You will need to create an adobe account if you do not have one already, but the animations are free. A lot of our crafting animations have come from there and some we have simply had to do ourselves, usually by modifying others. Nice thing about Mixamo is that they also have characters you can use for prototyping and you can upload and rig a character using their autorigging system.

 

New integration video for HAP 3 and Opsive TCP... enjoy!!!! (HAP 3 ... 3 weeks on pending review )
Still working on the combat while riding integration

 

It looks too amazing!
I held off on using the current version because I can't wait to use the new features in this version.

Any word on the upgrade price?

 

yes is 15$ for all who own HAP2

 

Can anyone tell me if there is any integration with UFPS already? I would like to buy this Asset, but I also need a First Person integration;

 

There's no integration with UFPS yet - we're making a lot of progress but version 2 won't be released until later this year. In the meantime one of our community members did a first person mod.

 

Is there a date for updating HAP?

 

Nice, tks

 

Hi! I want to be secure about the Integration of - Adventure Creator with The Third Person Controller. Is it work good ? Because I want to buy this asset.

I will wait for this! please and thanks!

 

The Adventure Creator integration allows your character and camera to be recognized by and work with Adventure Creator. It's a relatively simple integration and I don't get too many comments on it so that's a good sign

 

Hi, where can i get the script for this fps mod as seen in the video ?
Thank you.

Edit: Ah, found it. Problem solved.

 

Hi Opsive,

I'm having a series of errors as I try to follow along with the video tutorials. I'm following along with the videos to a T and whenever I try to build a weapon as in

, it attaches the weapon to the character arm and then throws a null reference error:

NullReferenceException: Object reference not set to an instance of an object
Opsive.ThirdPersonController.ItemBuilder.BuildShootableItem () (at Assets/Third Person Controller/Scripts/Utility/ItemBuilder.cs:125)
Opsive.ThirdPersonController.ItemBuilder.BuildItem () (at Assets/Third Person Controller/Scripts/Utility/ItemBuilder.cs:38)
Opsive.ThirdPersonController.ItemBuilder.BuildItem (UnityEngine.GameObject item, Opsive.ThirdPersonController.ItemType itemType, System.String itemName, ItemTypes baseObjectType, HandAssignment handAssignment) (at Assets/Third Person Controller/Scripts/Utility/ItemBuilder.cs:27)
Opsive.ThirdPersonController.Editor.ItemBuilder.BuildOrAssignItem () (at Assets/Third Person Controller/Editor/Builders/ItemBuilder.cs:194)
Opsive.ThirdPersonController.Editor.ItemBuilder.OnGUI () (at Assets/Third Person Controller/Editor/Builders/ItemBuilder.cs:58)
System.Reflection.MonoMethod.Invoke (System.Object obj, BindingFlags invokeAttr, System.Reflection.Binder binder, System.Object[] parameters, System.Globalization.CultureInfo culture) (at /Users/builduser/buildslave/mono/build/mcs/class/corlib/System.Reflection/MonoMethod.cs:222)
Rethrow as TargetInvocationException: Exception has been thrown by the target of an invocation.
System.Reflection.MonoMethod.Invoke (System.Object obj, BindingFlags invokeAttr, System.Reflection.Binder binder, System.Object[] parameters, System.Globalization.CultureInfo culture) (at /Users/builduser/buildslave/mono/build/mcs/class/corlib/System.Reflection/MonoMethod.cs:232)
System.Reflection.MethodBase.Invoke (System.Object obj, System.Object[] parameters) (at /Users/builduser/buildslave/mono/build/mcs/class/corlib/System.Reflection/MethodBase.cs:115)
UnityEditor.HostView.Invoke (System.String methodName, System.Object obj) (at C:/buildslave/unity/build/Editor/Mono/HostView.cs:249)
UnityEditor.HostView.Invoke (System.String methodName) (at C:/buildslave/unity/build/Editor/Mono/HostView.cs:242)
UnityEditor.HostView.OnGUI () (at C:/buildslave/unity/build/Editor/Mono/HostView.cs:76)

I should note that the included demos seem to work fine. Any help would be greatly appreciated, I am at a complete loss on how to get off the ground. I'm using Unity 5.5.1f1

 

Ok I fixed the issue above, I wasn't using the right asset type for a weapon. However, now I have another problem where I cannot run and hold a weapon at the same time. No errors are being thrown, but I can run after putting away a weapon with T but not while it's equipped.


Edit: I deleted my project and created a new one and this time there's no such issue. Don't know exactly what I did to the first project to make it not work, but it affected every scene. Working now though so ignore me haha.

 

Simple question, what are these 2 parameters for?

 

They are used if you want to have a starting and stopping animation for when the ability starts. The HeightChange ability in the Adventure scene has an example of this for crawling -



In this case the State Prefix is left empty but the same results could be achieved by placing "Crawl" in the State Prefix field and not having "Crawl." in the other four fields.

 

Thanks, it helps.

I have another question. I want to use the mouse wheel to zoom to the character. Invector is doing it natively and it quite standard in RPG. I cannot find an option for doing it with your TPC. Could you point me to the solution?

Detail: If the camera has a min of 4 and max of 30 I want to zoom in and out from 4 to 30 with the mouse wheel.

 

On the camera state if you enable "Step Zoom" this will allow you to do it

 

Thanks.

A couple of feedback that you may find useful for the next releases, it is my first contact with your asset:

1. it would be great to have a way to apply the settings from one character to another. Then your demo character could be used as a template. Right now it seems to be complicated to adapt the Clean Scene characters to your own character.
2. the documentation is dated, most of the pictures do not correspond to the current release. Same with the video. It is confusing because if you never used your asset, then you just try to follow the video (abilities for example) and you get a different results than the video.
3. having the source code and the abilities in different places is a mess. Many other assets are checking if the assets are here and support them natively without asking their customers to manually do it. I personally find like a turn off to have to spend hours understanding what needs to be done to have a complete product. Selling your asset in kit is disappointing for the end user who paid to get something without a hassle. For instance, the source code is available in one place, the instructions to install it in another place. Natural for you maybe, but a headache for the customer who is not aware.
4. Climbing is not working on terrain. Why not having it in the default abilities. Like if the slope is <45 (variable number) you walk and run, if it is >=45 (variable number) then you climb. Try to make your TPC climb a cliff on a terrain. Using boxes as climbing trigger won't work on terrain. Actually your character can only climb a ladder right?
5. Swimming is not going inside the water. If there is something at the bottom of the pool, you cannot get it.
6. Climbing short, is not working out of the box. The character could evaluate the height of an object and jump it automatically without tagging it.

Thanks

 

Thanks for the feedback @jimburn!

Making it easy to copy abilities from one character to another is high on my list. The issue is that when you add an ability it also adds a hidden component (abilities are just MonoBehaviours) so you can't just copy the RigidbodyCharacterController. In v2 abilities are derived from System.object and a serialized separately so that'll enable you to just copy the main controller component.

Climbing works with ladders, vertical walls, and pipes. It does not work on steep terrain though and the climb ability is one of the things that I plan on doing a major revamp for.

If you add the Dive ability you can go under water

Hmm, specifying the object layer should be all that it takes (using layers instead of tags since string comparison is slow). In the clean scene you can use the short climb ability on these platforms:

 

Thanks for your feedback, I have not noticed the dive ability! It is good to have the 2 options tho.

On the positive side, I love your camera. The easy option to take in account obstruction for example, and in general from all the TPC I have tested it is the one that works smoothly and is not doing CPU spikes