Content

Post new topic Reply to topic

RWBY: From Dust...

Author Message
Shenku RiO Incarnate View user's profile Send private message

Reply with quote Monday, January 22, 2018

Video Link

I've been tinkering with this Ruby Rose model on and off for a few years now, but now I finally decided what the Home for Infinite Losers, might as well make something out of it....

Outside of the player character model itself, and some pre-made marketplace assets/recycled ZEQ2-Lite textures (Namek Grass texture on the ground...), the rest I slapped together mostly over the course of the weekend... (Including the terrible looking untextured/poorly rigged/glitchily animated Grimm Wolf monster, as well as Ruby's rushed re-rig to use the UE4 skeleton to make use of Marketplace animations, cause I'm lazy and terrible at animating......)

Note: Yes, the AI currently does not attack, it only wanders, or chases the player right now, and occasionally gets stuck in their spawner, which I'm still trying to sort out the cause of... I intend to fix that eventually... Maybe...

Honestly though, I'm not sure if I should continue working on this as a more long term ZOPU type project or not, what do you all think...? (Assuming anyone even really checks the forums anymore... I'm too busy for Discord lately.... Sad )

Eagle The Purpose View user's profile Send private message

Reply with quote Monday, January 22, 2018

I don't think it's about what we think, it's about what you want to do. Do what drives your passion; that's not something others decide.

As for this project, I just saw it being mentioned on Discord, and I have to say, I'm liking the way it goes, like the others. You're always showing interesting concepts and gameplay sessions. As always, you still have my moral support, even if it's not much. Razz

Maybe you could kickstart a project from a solid proof-of-concept build? That would greatly help getting the project on track for further development, as well as giving you a fixed goal (I know how it is to go between a lot of projects like you do Razz ).

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Monday, January 22, 2018

Eagle wrote : I don't think it's about what we think, it's about what you want to do. Do what drives your passion; that's not something others decide.


True, but there's no point in making something that no one wants to play. For the most part right now, this is just something that I wanted to play, and the official game (RWBY: Grimm Eclipse) didn't hold my attention very well. Razz

As for this project, I just saw it being mentioned on Discord,


Really? I don't think I posted anything about it prior to the video last night, outside of some screenshots of the WIP model for Ruby a few years ago... Word travels fast, I guess?

and I have to say, I'm liking the way it goes, like the others. You're always showing interesting concepts and gameplay sessions. As always, you still have my moral support, even if it's not much. Razz


Thanks, it means a lot to hear that, considering that lately I've been questioning if I even should be doing these kinds of things, at least solo. I know my limitations, and know I can't make a full game at AAA standards by myself, but the kid in me still wants to try...

Maybe you could kickstart a project from a solid proof-of-concept build? That would greatly help getting the project on track for further development, as well as giving you a fixed goal (I know how it is to go between a lot of projects like you do Razz ).


While normally I would agree with you, in this case, similar to ZOPU, or even ZEQ2/Lite, I can't, because I don't own the rights to make any kind of money off of it, so it's purely just a for fun fan project at best right now...

I do have this small little corner in the back of my head telling me "Maybe you'll get lucky, and Rooster Teeth will see it and try to recruit you like what happened for RWBY: Grimm Eclipse?", but that's such a long shot, I'd sooner expect to get hit by a meteor...

Eagle The Purpose View user's profile Send private message

Reply with quote Monday, January 22, 2018

For the most part right now, this is just something that I wanted to play

That's how it goes. If you don't have a drive for what you do, then it's probably not worth it. It will be more of a weight than anything else.

Really? I don't think I posted anything about it prior to the video last night, outside of some screenshots of the WIP model for Ruby a few years ago... Word travels fast, I guess?

I think Naze linked the video around the time you sent that post? Speaking of, ignore the mention I did there, I wasn't aware you posted here.

Thanks, it means a lot to hear that, considering that lately I've been questioning if I even should be doing these kinds of things, at least solo. I know my limitations, and know I can't make a full game at AAA standards by myself, but the kid in me still wants to try...

I know that well... I'm confronted to this reality every single time I even think about the projects I'm working on. I'm still forcing progress in, though. I'm pretty sure I will someday be able to produce a decent amount, with enough experience. And who knows, maybe I will be able to settle certain things that will permit me to work with a team? I already sorta teamed with a few guys around, but I can't say there's much progress, for reasons I didn't clearly identify yet, but I do have a part in the problem for sure.

I would gladly listen to Brad's wisdom, if he can grant this wish!

Either way, there's no point feeling down on this, it's not what will make you go forward. As long as you are able to do the things you like, it should be okay. And you could expand to more from it as well. Game development isn't an easy path, but it's not something we can't manage for sure. We handled ZEQ2-Lite for quite some time after all!

While normally I would agree with you, in this case, similar to ZOPU, or even ZEQ2/Lite, I can't, because I don't own the rights to make any kind of money off of it, so it's purely just a for fun fan project at best right now...

I do have this small little corner in the back of my head telling me "Maybe you'll get lucky, and Rooster Teeth will see it and try to recruit you like what happened for RWBY: Grimm Eclipse?", but that's such a long shot, I'd sooner expect to get hit by a meteor...

Ah, I didn't realise. The name does ring a bell, but I don't recognize the franchise. That's a pretty difficult thing to manage, and I can't speak for anyone, but the companies/right owners must see a real benefit to shoot the recrut procedure. Complicated matters... In any case, whatever you do is always a pleasure to discuss about, and it is motivational too. Don't look at how active the forum is, it's ZEQ2-Lite (the whole website needs a face lift anyway). You know that at least the usual crew will look into what you're showing. Wink

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Monday, January 22, 2018

Update: https://youtu.be/5-JNSDXNnlU

Added in some more AI functions for attacking, and target selection, added an attack animation for the Grimm Wolves (Just 1 attack... I said I was terrible at animation...).

Also added functionality for happy little NPC Allies, because everyone needs a friend.

/BobRoss


[quote="Eagle"]

I know that well... I'm confronted to this reality every single time I even think about the projects I'm working on. I'm still forcing progress in, though. I'm pretty sure I will someday be able to produce a decent amount, with enough experience. And who knows, maybe I will be able to settle certain things that will permit me to work with a team? I already sorta teamed with a few guys around, but I can't say there's much progress, for reasons I didn't clearly identify yet, but I do have a part in the problem for sure.

I would gladly listen to Brad's wisdom, if he can grant this wish!

Either way, there's no point feeling down on this, it's not what will make you go forward. As long as you are able to do the things you like, it should be okay. And you could expand to more from it as well. Game development isn't an easy path, but it's not something we can't manage for sure. We handled ZEQ2-Lite for quite some time after all!


I can definitely relate. I'd like to think I could work in a team environment on video games professionally, especially with all the experience gained by collaboration here on ZEQ2-Lite, but the industry is very unkind towards people like myself who do a little of everything but are not heavily specialized in any one area. I'd say that Character modeling is my strongest spot, but even that's not up to AAA standards cause I spend all my time working on everything else that goes into my projects, leaving me with very little time to focus on the one area I should be focusing on to get good enough to get a job...

Ah, I didn't realise. The name does ring a bell, but I don't recognize the franchise. That's a pretty difficult thing to manage, and I can't speak for anyone, but the companies/right owners must see a real benefit to shoot the recrut procedure. Complicated matters... In any case, whatever you do is always a pleasure to discuss about, and it is motivational too. Don't look at how active the forum is, it's ZEQ2-Lite (the whole website needs a face lift anyway). You know that at least the usual crew will look into what you're showing. Wink



That's one of the reasons I still use the forums. Sure, they're not exactly modern, but they work perfectly for what they are for the most part. Plus, I prefer posting stuff here over Discord for a few reasons, but namely it's easier to read any feedback given via the forums since it's less likely to be drowned out in unrelated conversations on Discord, and it's also more likely to get responses from the people in our little community who would actually give solid feedback instead of generic "It's so awesome! Download link please? okay, Thanks, Bai..."

Anyways, while I don't know the specifics of what happened exactly, the gist of the story I heard was the guy who was making Grimm Eclipse (Who I think was even in our community at one point quite a few years back...?) was making his game as just a fan game in Unity, and I guess Rooster Teeth saw his videos of it somehow, liked what he was making, and offered him a job to make it as an official game instead of just a fan one. I know the chances of the same happening with my project are remote, but I just feel like more could be done, videogame wise, with RWBY than what Grimm Eclipse offered. Honestly, RWBY's entire concept felt very reminiscent of an action RPG, but Grimm Eclipse was just a straight Action game, so it felt like it was missing the story element that would have let it really shine...

Eagle The Purpose View user's profile Send private message

Reply with quote Tuesday, January 23, 2018

I can definitely relate. I'd like to think I could work in a team environment on video games professionally, especially with all the experience gained by collaboration here on ZEQ2-Lite, but the industry is very unkind towards people like myself who do a little of everything but are not heavily specialized in any one area. I'd say that Character modeling is my strongest spot, but even that's not up to AAA standards cause I spend all my time working on everything else that goes into my projects, leaving me with very little time to focus on the one area I should be focusing on to get good enough to get a job...

Same! I've been thinking about integrating a company for a little while now, but talking with a few professionals and also assuming a few things, I don't think I'm fit for this kind of development context (and I also lack the skills; same problem as you have). That might even kill my passion for it. The best direction I can think of that's officially valid would be making the sort of startup like in the old days of IT. Young people, with a part of professionalism and a part of fun. That's a pretty naive view, however. Not sure how long that thing would last.


That's one of the reasons I still use the forums. Sure, they're not exactly modern, but they work perfectly for what they are for the most part. Plus, I prefer posting stuff here over Discord for a few reasons, but namely it's easier to read any feedback given via the forums since it's less likely to be drowned out in unrelated conversations on Discord, and it's also more likely to get responses from the people in our little community who would actually give solid feedback instead of generic "It's so awesome! Download link please? okay, Thanks, Bai..."

I understand. The downside to this is pace though. You will get a much slower exchange from the forums than a real-time messaging application. The best way you could use Discord for your projects would be to have server(s) dedicated to them.

Anyways, while I don't know the specifics of what happened exactly, the gist of the story I heard was the guy who was making Grimm Eclipse (Who I think was even in our community at one point quite a few years back...?) was making his game as just a fan game in Unity, and I guess Rooster Teeth saw his videos of it somehow, liked what he was making, and offered him a job to make it as an official game instead of just a fan one. I know the chances of the same happening with my project are remote, but I just feel like more could be done, videogame wise, with RWBY than what Grimm Eclipse offered. Honestly, RWBY's entire concept felt very reminiscent of an action RPG, but Grimm Eclipse was just a straight Action game, so it felt like it was missing the story element that would have let it really shine...

He got really lucky if that's true! I guess it's worth a try? As long as you're taking pleasure doing that, it should be okay. The safety rule would be to make things as legaly and functionally portable as possible in case the project wouldn't make it. You would be able to use elements you worked on for others.

Skatter #*&@%! View user's profile Send private message

Reply with quote Tuesday, January 23, 2018

Shenku wrote : Anyways, while I don't know the specifics of what happened exactly, the gist of the story I heard was the guy who was making Grimm Eclipse (Who I think was even in our community at one point quite a few years back...?) was making his game as just a fan game in Unity



I don't know if he bothers with the forums anymore, but you might reach out to Xatoku in a pm or something. He probably has other methods of contact as well.

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Wednesday, January 24, 2018

New Update: https://youtu.be/xtraWLl1SZ8

Adjusted the attacks and damage slightly, Heavy Attacks now deal additional effects to enemies based on Ammo Type instead of just doubling damage like before.

Standard Ammo: Deals 25% more damage on hit
Fire Ammo: Deals 5 damage every .1 second for 3 seconds
Ice Ammo: Slows enemy movement speed by half (To no lower than 50 units) for 10 seconds (Can Stack)
Lightning Ammo: Stuns the enemy and deals 1 damage every .1 second for 3 seconds
Earth Ammo: Immobilizes the enemy for 5 seconds

Also removed auto ammo restoring in favor of a manual reloading system with separate ammo reserves for each type of ammo.

Also some minor bug fixes with the combat mechanics fixing a bug where the player would sometimes be unable to attack for no reason. Additionally, I've updated the HUD slightly to reflect the new ammo types. (Ammo Icons are not final)



Skatter wrote :
I don't know if he bothers with the forums anymore, but you might reach out to Xatoku in a pm or something. He probably has other methods of contact as well.



Yeah, I think that was him. Not sure what he's up to now, and haven't really heard much since he used to post for his Z Warriors Chronicles project way back when. Heck, I doubt he'd even remember me, since we didn't really talk directly much, if at all. I heard Rooster Teeth picked him up to make Grimm Eclipse in an official capacity, but that was the last I heard.

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Thursday, January 25, 2018

Adding in 2 new areas for the story focused stuff. The first is Signal Academy Exterior, and the second being it's Interior... So really, it's just one new area for now...?

Note: I literally have no references to work from for what Signal Academy is supposed to look like, because I don't think it's ever been shown in the show so far. So, with that in mind, I'm just sorta winging it and using some generic Marketplace assets to fill in the city and Academy Interior with stuff that's kind of close to what it could look like (at least until I feel like remaking them to better fit RWBY's art style)... But anyways, choosing this setting cause it gives me more freedom to make things up for whatever overall story I end up going with, at least so long as it doesn't conflict or overlap with the show's storyline...

Anyways, going to do one more story area (Ruby's home) soonish, and then start setting up the functions to load between the different areas, and once that's done, I'll start working on the quest/mission system for actually progressing the story (As well as starting work on a story...)...

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Friday, January 26, 2018

Some more updates: https://youtu.be/omvMAgu_xkU

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Saturday, February 03, 2018

https://youtu.be/M9pHqhXvVns

Added some new icons for the different ammo types with a border from a pre-made icon pack (The icons themselves were made by myself...) Also added two new ammo types, Shadow and Holy...

Updated the mission system slightly, it sorta works now... kind of... But if you leave the area and it's tracking enemies, it auto-completes, so gotta fix that still...

Added a new map area, Ruby's Home, along with travel zones to/from it; layout of the area and house mesh are not final...

Moved the travel zones for traveling to/from the "Grimm Forest" (Name of the forest will probably change...)

Eagle The Purpose View user's profile Send private message

Reply with quote Saturday, February 03, 2018

How do you handle the UI? Is there a new code each time you make an element or are you using a sort of framework? Might want to design it with the necessary assets, to avoid scrapping stuff if it looks bad.

Also, physics-based animations is not something easy to manage. Wonder if it isn't possible to bake keyframes from physics animation, so it will permit manual clipping fixes and prevent unintended visual behaviour. Of course, it will look more repetitive, but probably cleaner.

There's definitely something wrong with the woods map as well. You're not using much visual effects yet it has pretty terrible lags. Now, whether it comes from the AI or the actual rendering is unknown to me, but it's something you might want to address soon-ish, for convenience.

Not sure I asked this already, but is there a test build? I have two different machines which I can test different stuff on now.

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Saturday, February 03, 2018

Eagle wrote : How do you handle the UI? Is there a new code each time you make an element or are you using a sort of framework? Might want to design it with the necessary assets, to avoid scrapping stuff if it looks bad.


I'm using UMG, which is a visual UI widget editor, so I can drag and drop, scale, rotate, exc. each HUD element individually, and bind different behaviors to them, or even have whole separate widgets as children to the main HUD widget if they have really elaborate behaviors that don't directly rely on the other HUD elements. If needed, I can also display fully separate widgets at the same time, I.e. The in-game HUD versus a pause menu.


Also, physics-based animations is not something easy to manage. Wonder if it isn't possible to bake keyframes from physics animation, so it will permit manual clipping fixes and prevent unintended visual behaviour. Of course, it will look more repetitive, but probably cleaner.


Yeah, the cape's been really bothering me with its constant clipping through the torso, still trying to figure out how to fix that with what the engine allows...


There's definitely something wrong with the woods map as well. You're not using much visual effects yet it has pretty terrible lags. Now, whether it comes from the AI or the actual rendering is unknown to me, but it's something you might want to address soon-ish, for convenience.


It did seem like the performance drop was more notable this time, and the only thing I really changed that could have affected it to be worse was some of the particle systems relating to the elemental attacks, so it's possible that may have tanked the performance a bit. It was particularly noticeable when hitting multiple enemies, so I'm almost positive that this is the case. going to go back in and tone down the particles and/or swap them to be GPU based instead of CPU, and that should help. Also going to add in distance culling for the trees and enemies, and that should help as well.

Although to be honest, I'm surprised it performs as well as it does otherwise considering the enemy spawners are spawning up to a max of 50 enemies to run around on the screen at any given time, respawning one somewhere every second, so I'm kind of impressed it doesn't just choke up the performance from that alone.


Not sure I asked this already, but is there a test build? I have two different machines which I can test different stuff on now.


Not yet, still need to add in quite a few more things before I get to that stage, but I'll keep you posted.

Eagle The Purpose View user's profile Send private message

Reply with quote Saturday, February 03, 2018

50 AIs shouldn't be an issue if they only chase the player. As for particles, maybe Skatter or Bradimus can give you tips to optimize them? Sounds surprising they would cause so much lag as well. Nowadays computers can handle quite a lot of particles with even physics based effects if they're well done.

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Saturday, February 03, 2018

Eagle wrote : 50 AIs shouldn't be an issue if they only chase the player. As for particles, maybe Skatter or Bradimus can give you tips to optimize them? Sounds surprising they would cause so much lag as well. Nowadays computers can handle quite a lot of particles with even physics based effects if they're well done.



Generally true, but the problem is partly with how UE4 handles particles in that there are two types of particle emitters, CPU particles, and GPU particles. As implied by their types, CPU particles are processed through the CPU, where as GPU particles are processed through the GPU.

For small simple particles with not a lot going on, rendering via the CPU is fine, but anything with a lot going off all at once or over an extended period of time is better suited for being set to use the GPU. Especially since the CPU is already dealing with processing most of the other game dependent functions, and the added load can definitely lead to frame rate drops if used excessively.

That I believe is where the current bottle neck is that's causing the frame rate dips in this video, since I set the particle systems to the attacks, but didn't adjust them to be GPU based, which really messes things up due to the high number of sprite particles being spawned. Also, the transparency of the sprites can also clog up the rendering thread too, since transparency is a bit costly to render, so any particle with a large number of particles going off will cause a performance hit regardless of CPU or GPU settings.

Long story short, I need to set the particles to GPU, and adjust the amounts being spawned, and that should help a lot, at least for a start, and general LOD/distance culling settings will also help a bit as well.

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Sunday, February 11, 2018

https://youtu.be/CJFclqVykZA

Added in Yang Xaio Long (Ruby's sister, for those who haven't watched the show...) as an AI controlled ally.

Eagle The Purpose View user's profile Send private message

Reply with quote Sunday, February 11, 2018

Generally true, but the problem is partly with how UE4 handles particles in that there are two types of particle emitters, CPU particles, and GPU particles. As implied by their types, CPU particles are processed through the CPU, where as GPU particles are processed through the GPU.

For small simple particles with not a lot going on, rendering via the CPU is fine, but anything with a lot going off all at once or over an extended period of time is better suited for being set to use the GPU. Especially since the CPU is already dealing with processing most of the other game dependent functions, and the added load can definitely lead to frame rate drops if used excessively.

That I believe is where the current bottle neck is that's causing the frame rate dips in this video, since I set the particle systems to the attacks, but didn't adjust them to be GPU based, which really messes things up due to the high number of sprite particles being spawned. Also, the transparency of the sprites can also clog up the rendering thread too, since transparency is a bit costly to render, so any particle with a large number of particles going off will cause a performance hit regardless of CPU or GPU settings.

Long story short, I need to set the particles to GPU, and adjust the amounts being spawned, and that should help a lot, at least for a start, and general LOD/distance culling settings will also help a bit as well.

Could you showcase the said particle system? Maybe there is an chunk of detail you can take off from it without hindering the wanted effect to be.

By the way, what outline type are you using?

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Sunday, February 11, 2018

Eagle wrote : Could you showcase the said particle system? Maybe there is an chunk of detail you can take off from it without hindering the wanted effect to be.



Working on the performance optimization today, kind of skipped it in the video I posted last night... So far, I've improved the performance slightly, but there's still some frame rate issues at higher graphic settings or in full screen/maximized windowed mode. I'm partly wondering if it may just be my laptop, because the Integrated Intel Graphics card often conflicts with the dedicated NVidia GPU, and my laptop's not that powerful to begin with...

By the way, what outline type are you using?


Not really sure what type it would be considered, but it's a post processing effect done via screen depth checks and stuff... I don't really understand it enough to explain it any better than that (It's the same Cel/Outline system as the one in ZOPU), but here's a screenshot of it if you want to try and make sense of it:

Zeth The Admin View user's profile Send private message

Reply with quote Monday, February 12, 2018

Not really sure what type it would be considered, but it's a post processing effect done via screen depth checks and stuff... I don't really understand it enough to explain it any better than that


If it's post-process, it sounds like sobel-edge-detection (image-screen-space outlines). It's the other popular approach next to object-space fatboy/vertex push/hull outlines. As I've mentioned many times, both are techniques roughly from 2001-2003 that each have their own share of limitations/expenses.

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Sunday, February 18, 2018

Zeth wrote :
If it's post-process, it sounds like sobel-edge-detection (image-screen-space outlines). It's the other popular approach next to object-space fatboy/vertex push/hull outlines. As I've mentioned many times, both are techniques roughly from 2001-2003 that each have their own share of limitations/expenses.



I think that's what it was, yes.

Anyways, new video update: https://youtu.be/U4NkGJrVbHM

I've been closely monitoring my performance, and I've been able to marginally increase it to around 25-30ish FPS on average in the less sophisticated maps, 30-40ish occasionally if I'm lucky, but it drops to about 20ish in the forest, bottoming out to 11-15ish when fighting a lot of enemies at once... My biggest problem so far is I'm still not sure how much is from the way I've set things up in the project, versus how much is from my laptop not being able to keep up. I've done as much as I could already to decrease the complexity of things being put on the screen, and swapped out the trees from static meshes to instanced foliage meshes which has fewer draw calls, but the FPS is still an issue...

*Shrug* Maybe it's time to stop putting it off, and finally upgrade and build a better PC instead of working on mediocre laptops...

Shenku RiO Incarnate View user's profile Send private message

Reply with quote Wednesday, February 28, 2018

Update: https://youtu.be/REgj7sBWiXQ

Added Blake Belladonna as a playable character

Updated Yang, Weiss, and Blake to have partial Semblance abilities, Yang deals more damage based on how hard she's been hit, Weiss can double jump for extra mobility, and Blake can leave behind decoy clones of herself to temporarily distract enemies for 10 seconds or until the decoy is destroyed (whichever happens first)

Added death and revive mechanics, as well as game over mechanics. If a party member goes down, a random AI controlled ally will be chosen to go revive them, if everyone goes down, game over condition happens and the player is sent back to the load in point for the level and all status' are reset

Added placeholder for new Grimm enemy type

Post new topic Reply to topic

Actions

Online [ 0 / 6125]