More Radioplays?

Moderator: Milldawg

Is a new Radioplay project worth pursuing?

Yes
4
57%
No
0
No votes
Continue to Pursue Game Projects Instead
3
43%
 
Total votes: 7

User avatar
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8328
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘ జ్ఞ‌ా

Re: More Radioplays?

Post by IskatuMesk »

I made the thread because I couldn't come to a conclusion on what path to take. In such a case feedback can give me external perspectives. I happen to agree with most of them. For now, my time is best invested into LP's.
User avatar
Krazy
Zerg Creep Colony Landscaper
Zerg Creep Colony Landscaper
Posts: 446
Joined: Sun Nov 30, 2008 1:49 am

Re: More Radioplays?

Post by Krazy »

You should make a radioplay-themed videogame?
User avatar
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8328
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘ జ్ఞ‌ా

Re: More Radioplays?

Post by IskatuMesk »

Honestly if I can figure out a solution to characters in Matinee I'm pretty set on going for the CGI solution. That's what I really want to do. There's a few things I need to figure out, like what I'm going to do about Starcraft models. I can port some from sc2, but they'd need rigs for animations because the cinematic models don't have general animations. I can't rig them myself, much less animate them. If I had something simple, say the siege tank or a goliath, with a functional rig, maybe I could learn from there. I'm going to ask someone I used to know about that, but I'm not expecting anything since he's all about the monetary compensation these days.

The biggest problem is definitely Matinee and characters, though. The characters I port from any given game have separate skeletons and animations for bodies, hair, and heads. I can animate the hair by just turning it into cloth physics. It's not a perfect solution, but nothing about this will be perfect. Hypothetically if I knew how Nvidia Hairworks functions I could just use that and ignore the ludicrous and pointless performance cost of having 8 polygons per pixel because lolnvidia. But I'd settle for my tried and true method currently.

I hypothesize I can fuse the skeletons of the heads and bodies and then export the animations individually. However, I don't know what will happen once I build those into composite animations in Unreal. Probably the only reliable way to do it would be to manually fuse the animations and export them one by one. I can do this... but the manual labor time cost would be truly colossal.

Alternatively I can use older character meshes, like the ones from Lineage 2 or WoW, which have no such complications involved. However, I would need to figure out a way how to bring them up to the level of quality expected from the rest of the assets, and I don't really know if I can do that.

I feel like everything else is just a matter of time. I know particles and static meshes and I can model any geometry I can't port. Some meshes I don't have any means to acquire or make just yet, but I can address them when necessary.

If I started with Apex RN I would have a fairly manageable foundation. From there I could go into Apex Prelude and its four segments. I couldn't translate the two latter and any future segments into a CGI due to the asset requirements.

It feels like every step of the road from 2001 to today has been wrought with disappointment and failure. From the disaster of Retribution to the defeat of Armageddon Onslaught, nothing ever quite seems to go my way. Sometimes I don't even really know what it is I want out of a project. But I do know I want to see my worlds in motion. I can't hope to do it in a game because I will never possess the skills necessary to do that kind of work. But this... this I could probably do if I can figure out the pipeline. It is something I want to try, at least. Then I could put the skills of radioplay I learned to the real test and in an environment I know how to work with on other fronts.

Sometimes I feel like posting streams of thought is a rather selfish venture.
User avatar
Krazy
Zerg Creep Colony Landscaper
Zerg Creep Colony Landscaper
Posts: 446
Joined: Sun Nov 30, 2008 1:49 am

Re: More Radioplays?

Post by Krazy »

IskatuMesk wrote:However, I would need to figure out a way how to bring them up to the level of quality expected from the rest of the assets, and I don't really know if I can do that.

As far as expectations go, most of that is what you set the audience up for. I think it's fine to go with older-looking graphics so long as you can keep them looking somewhat consistent in terms of aesthetic. If using current generation animations is a massive time-sink, why bother? Look at something like and it's just made with RPGMaker. I mean, maybe the 3D graphics are the problem, not the solution.

Also I don't know what Matinee is.
User avatar
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8328
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘ జ్ఞ‌ా

Re: More Radioplays?

Post by IskatuMesk »

3d is kind of a double-edged sword. It really only helps you if you make use of it. What I'm looking for is something styled similar to an anime in pacing and presentation. There's a lot of things I won't be able to do, like talking animations or expressions or whatnot, even though a lot of that is critical to visual storytelling (and anime especially). So, it will definitely come down to defining a style and building around that. I've considered many different avenues, but ultimately the most painless is Unreal 4 due to how easy it is to translate content into it and how familiar I already am with many systems critical to any project.

Consider that a project like the one you linked requires exceptionally high skill in Ruby (RPGMaker is very heavily reliant on Ruby for anything beyond the absolute basics, and is why I didn't stick with it even though I worked with it for years in early 2000), and that presents a considerable roadblock.

As for matinee and the involved problems, I fear this is going to become a bit wordy.

Matinee is basically the cinematics editor of Unreal. The proper way of making a cinematic is to motion capture and then tune the resulting animations in 3ds max or whatnot, then bring them into Unreal as a composite asset that basically is your scene minus flash. Unreal adds the flash. Matinee is the track editor that handles everything from cameras to triggering sounds and particles and you can do all post processing in it, like color grading or slow-motion. Think of it as like the cutscene editor in sc2 but a lot more powerful. Unfortunately, it hasn't seen much update since Unreal 3, and it doesn't really interface with Blueprints (Unreal 4's replacement for scripting) which handle all the things like animation blending and fusing my various components together and ensuring they are synced. My cinematics will basically be a bunch of general gameplay animations stitched together in sequence with camera angles giving them context.

Taking a Tera asset as an example, I have to do a bit more than just attach body parts as a component and ensure they are synced. I have to do the equivalent of weighting bones to pieces from opposing meshes.

Image
Image

This spaghetti ensures that no seams appear in the neck as a result of differing weights between pieces of the head and body. It's the same as aligning and linking the bones in 3ds max. Without this, there are very large seams. This is a process specific to Tera assets. I haven't looked too heavily into assets from other games, but the theme of independent face animations is common and, without them, the head usually clips heavily.

Matinee cannot interface with animation blueprints, send commands to them, or even access skeletal mesh child components of an actor. I found mention of an update for Matinee being worked on in mid 2014, but like many things with Unreal 4, it is probably a long ways from being finished. I need to change my content at the source level to suit Matinee if I am to go this route.

Like I said, I have a few possible avenues I would need to test to verify if they would work. If they don't work it is very plausible I can fall back to Lineage 2 or WoW assets. The problem with that is less what I set the audience for, and more that I want consistency. My static meshes and particles are very high detail. I can update Lineage 2 monsters but characters are very dated looking. WoW is... not at all comparable to the asian games. I did experiment with changing WoW model proportions and bone weighting to sort of shift them into a more realistic style, but I didn't have a great deal of success. I could keep trying, I suppose. Though I need to get a CASC viewer and such.

I'd also probably have to reverse sc2 cinematic models and learn how to rig and animate them. I think that would have to be a long-term goal. Rigging and animation are two virtually inscalable walls... but learning either one of them would make this work so much less painful. I could, for example, weight the problematic Tera meshes to the Unreal base skeleton and dodge all of the issues related to components. Or I could weight them to WoW skeletons and alter the animations as necessary. I would probably focus on getting a simple animation setup down first before thinking about trying such difficult things, though. Unfortunately, Tera has no animations for single wielding a 1handed sword... which is something I really would need.

If I could get that working out I feel like I would be in free range territory. I have created materials from scratch that effectively translate texture information and builds higher detail and highly customizable variants out of them, and I have a super high quality skin material.

Image

I did have success in sc2 with everything but the computer AI. I'm not currently considering a project for it, but I am going to ask quantum_menace some technical questions next I get the chance and see what he thinks. I can put the same assets in sc2 without too much trouble.

Image
User avatar
KnoutOut
Terran Goliath Dome Polisher
Terran Goliath Dome Polisher
Posts: 112
Joined: Fri May 11, 2012 4:25 am

Re: More Radioplays?

Post by KnoutOut »

You mentioned that you have trouble with rigging: Did you check out this blender-tutorial http://www.blenderguru.com/tutorials/in ... o-rigging/ for rigging by Andrew Price? He explains it very well and even if you are not using blender, you can gain some good knowledge and insight into rigging by it.

€dit: I made a little mistake, it's not made by Andrew Price but a guest of his. Nevertheless the videos are still very good.
Last edited by KnoutOut on Thu Jul 09, 2015 8:17 am, edited 1 time in total.
User avatar
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8328
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘ జ్ఞ‌ా

Re: More Radioplays?

Post by IskatuMesk »

I know as much about rigging and animation as I do about socializing. I'll watch these sometime.

/e

Also lol post 69
User avatar
ArcCain
Terran Battlecruiser Janitor
Terran Battlecruiser Janitor
Posts: 189
Joined: Wed Aug 01, 2012 4:10 pm

Re: More Radioplays?

Post by ArcCain »

I'd like to see some more radioplay. You could do larger battles in more of an "overview" in terms of description.
User avatar
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8328
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘ జ్ఞ‌ా

Re: More Radioplays?

Post by IskatuMesk »

I devised some new strategies for handling audio design thanks to GCCX, JoJo's Bizzare Adventure and other anime I watched recently. It's hard to put into words but I have a better idea of how I can manage first and third person transitions more smoothly. Rather those ideas were always there but they stood out a lot more clearly to me recently.

I am not sure how I would write Apex H for an audiobook but depending how badly things go in the next few months as far as getting any projects going is concerned I might try to write some prototypes.

Honestly I kind of want to try an AI hybrid theory I had for sc2 first. But it will be time consuming so it's not a priority.
Post Reply