Just looking for some general feedback and lip syncing question?

Hello,

I'm working on my first game I have the basic proof of concept working and a general story idea worked out. I'm at the point now were I have to decide on an art style and what technical bells and whistles I need to include or can dump to save time. Being my first game I was hoping I could get some feedback.


My basic concept: The game play is mostly centered around the melee combat system. The idea was to bring fighting game(Street fighter like) combat to 3d Skyrim like environment/Setting. I've actually ended up with something very similar to the Bushido blade series mixed with Mount and Blade if you are familiar. So its very complex and has an very high learning curve. I'm trying to decide if I need to dump the first person option and go with an over the shoulder type camera but I'm worried that would make seeing what your opponent is doing to difficult. I'm considering going the same way Bushido blade did, rendering just the player character side view with a seconed camera and over laying it in the corner of the first person view. Or would it be better to just use a set of Gui icons for character stance/movements.


I'm going with a more story based small setting since I can't create enough content for a large open world on my own. I don't have time to manually create my lip syncing but I have the budget to go with a full Faceshift system. That would pretty much be the end of my budget so I would like to avoid that if possible.


I was wondering if First person games can still get away with text speech or do I need lip syncing?


I would be grateful for any suggestions.

Thanks,

Travis

 

OK, these questions are very much matters of opinions, but I'll share mine in case you find anything useful in it.

First, the basic idea seems sound; I think Bushido Blade has a core group of fans that you may be able to tap into, and Mount & Blade has some adherents too. There is certainly a core of fighting-game players who like realism and complexity.

The danger here is that the complexity will completely alienate casual or new players. So, see if you can introduce that complexity gradually, perhaps by introducing skills one at a time as the character advances through the story. That will make it much more approachable.

As for the camera, I would advise against doing a second picture-in-picture (PIP) view. Better to find one camera view that works, and focus on that. Or at least, one camera view at a time; possibly you could switch between first-person view (good for exploring) and 3rd-person side view (good for fighting) when combat begins, interpolating smoothly between the two views. This still feels like a patch, but less obvious than the PIP patch. Maybe someone else will have a better idea.

As for dialog, I think you can still get away with displaying text. Actually, if you do prerecorded voice, I would expect your budget to be blown on voice acting rather than lip syncing. But either way, recorded speech greatly limits the amount of dialog you can have in your game, and how flexible it can be (there is no way to fill in the player's name, for example, or numbers that relate to current prices, etc.) I think this gives you a good excuse to stick with text and focus your resources on improving the depth of your game, rather than in making read-alongs for the player. It does limit your audience to players who are literate, though.

 

Recent games like Pillars of Eternity and Wasteland 2 have revived text-only dialogue as an option for commercial games. But it really depends on your audience. The hardcore role-players that kickstarted those games want lots of dialogue, and they're willing to read text to get it. If your game is more action-oriented, your players might find it jarring to switch between an arcade action mindset and a reading mindset.

If you decide to go with lipsync, the bigger cost will be paying voice actors, like Joe said. But you might be able to swing deals with theater students at your local university. If you have a lot of dialogue, workflow can be an issue for indie developers. I like SALSA because you can drop in any audio clip without having to do any preprocessing.

 

What can you dump? If your solo, and the goal is shipping a product, you might consider dumping most everything. Dialog scripts, voice-dialog, extra camera modes, open-world environment, lip-syncing. None of those add to the core element of fun in your game, so they can all go.

The shortest path to maximize your skills is to work toward a minimally viable product (MVP). One room/map, one character, and wonderfully joyful combat mechanics. It's a time-tested recipe.

PS - The real red flags are 'very complex' and 'very high learning curve'. Young developers confuse 'niche' with 'complex' and 'high learning curve'. If I knew you in person, I would shake my head, offering one suggestion, "No!". You can lip-sync that if you like.

Gigi

 

It's shameless plug time again! Text speech is great, and I personally love games that expect me to read. But for whatever reason, Joe gamer on the street seems to prefer spoken dialog in their games. And effective lip-syncing always makes spoken dialog go down easier.

If you feel that recording dialog and producing lip-syncing is simply too expensive in terms of time and resources, ditch it in favor of text. This is especially true for the first-pass/beta version of your game. It's the kind of extra feature that can be implemented later after the core game is more complete.

But if you are interested/insistent on getting lip-syncing into your game, allow me to direct your attention here...
https://www.assetstore.unity3d.com/en/#!/content/18746
It's a little free Unity plug-in that I published last year to make it faster/easier to implement lip-synced animations in your Unity game. I'm currently hard at work on a revised version that is optimized for Unity 5.0, so it should be getting a major upgrade inside of the next month. Lip-synced animations can be very costly in terms of time. It just takes a long time to make them, and most of the commercial solutions are quite pricey. I made this plug-in to help small-time developers have a cheap and fast solution.

 

Just finished a lip sync using mic output volume and a little algorithm that measured the delta between samples that gave me a good fake on most visemes and not just an open and closing mouth. The delta gave me clues as to how to get M V F P B phonemes from the live mic and whether to use Oh Ay Uh type visemes depending on the samples delta being similar over several samples..It took me a few hours to code and tweak from scratch. Good enough for a giant screen output at a large music festival in Santo Domingo. Gigi is like a broken record and assumes that all intellects are sub optimal when starting to use Unity. I dove in the deep end because simple was never my gig and never learned S***e from doing the beginner tuts or making a simple game/app/project no matter the app or field of endeavor. I consider it to be a time waster myself.. I have done maybe 40 games/apps for others since then released a few of my own and have three in development I do between pro gigs. I would have used Richard Kain's plugin but this had to be a realtime fake. Also ..most Unity solutions use morphs and I am using a 40 point face control rig and saving the morph targets for future total face changes like say from a pretty woman to a granite jawed drill sergeant.

 

^ Sorry, I'm getting completely off topic, but although I'm a follower of Gigi's philosophy of paring down to the raw essence, I have to agree with ippdev on one thing: never underestimate anyone. I participated in a short game jam last weekend. I was teamed with an artist and another programmer. The programmer had never used Unity before and didn't even have a laptop, so he borrowed a slow Macbook of mine. In two 4-hour sessions, he learned how to get around in Unity, built a graybox scene, set up a navmesh and NavMeshAgents, wrote steering behaviors and higher-level navigation control AI, and implemented finite state machine AI scripts for zombies and civilians. I was pretty impressed.

 

I'm finding I like text-only better than VO, since it enables one to author more dialogue paths, but if I do ever want to add some VO (like a greeting) this is my solution:

In my world its rude to leave your mouth uncovered.