COMMUNITY - FORUMS - GENERAL DISCUSSION
[Sticky] Pre-Alpha Advanced Traversal Mechanics Gameplay Video

Greetings friends,

The game team has been working on perfecting the advanced traversal mechanics lately. We are getting preeeeeetty close to being able to get the playable version of this into the hands of our Alpha 1 backers. Below is a sneak peak of some of that gameplay.

This is currently being developed in our Prelyria art style, which is a low-poly style that is a lot like greyboxing--if you're familiar with the term--only prettier. Because it's not just boxes and global lighting, it's allowing our art team to explore mood and ambiance, as well as experiment with lighting for organic, indoor spaces. Once we've tested the mechanics and are happy with them, the team can worry about amping it up to the full-fidelity Chronicles of Elyria art style. It's a "Measure Twice, Cut Once" kind of process!

This environment is part of a cave system that The Waerd have carved out and turned into a testing ground. You may have read about it recently!

Stay shiny!

Vye


... ...
3/7/2020 7:59:55 PM #61

Posted By Rovert at 05:42 AM - Sat Mar 07 2020

It took 3 years to make the pax demo a playable demo to test?

i think it took 3 years to complete all the milestones that were prerequisite to releasing a transversal alpha1 test platform.


3/7/2020 8:51:31 PM #62

Posted By Avastar at 7:59 PM - Sat Mar 07 2020

Posted By Rovert at 05:42 AM - Sat Mar 07 2020

It took 3 years to make the pax demo a playable demo to test?

i think it took 3 years to complete all the milestones that were prerequisite to releasing a transversal alpha1 test platform.

But... that demo doesn't touch any of those milestones unless I am missing something. This is the first time since PAX where we actually get to see gameplay from COE and it doesn't seem to be online (or at least there is no one there to show that), it doesn't feature NPCs, it doesn't feature...really anything but some very generic and quite dated platformer mechanics.

So what milestones exactly did it need to be released?


Looking to join some fun community on the EU server ;). Friend code: C2D13A

3/7/2020 9:17:26 PM #63

Posted By Puciek at 1:51 PM - Sat Mar 07 2020

So what milestones exactly did it need to be released?

it consists of ~2500 players all sending potentially launch quality data to a server to be analyzed, so all the milestones that make all that possible. this stuff has the potential to be used at launch. i think it also has basic character creation and tracking, so also all the database stuff to retain individual character data.

i think pax stuff was only a proof of concept that didn't have any of the above.

imo, movement is the most basic thing that makes this type of game possible. without moving around, coe couldn't function. focusing on movement, bringing it up to launch quality first seems efficient. then they can add in the other alpha1 (adventurer) features. First movement skills, then combat skills, social skills, crafting skills, etc. each one is unique and can be implemented separately.


3/7/2020 10:49:05 PM #64

Posted By Avastar at 9:17 PM - Sat Mar 07 2020

Posted By Puciek at 1:51 PM - Sat Mar 07 2020

So what milestones exactly did it need to be released?

it consists of ~2500 players all sending potentially launch quality data to a server to be analyzed, so all the milestones that make all that possible. this stuff has the potential to be used at launch. i think it also has basic character creation and tracking, so also all the database stuff to retain individual character data.

i think pax stuff was only a proof of concept that didn't have any of the above.

imo, movement is the most basic thing that makes this type of game possible. without moving around, coe couldn't function. focusing on movement, bringing it up to launch quality first seems efficient. then they can add in the other alpha1 (adventurer) features. First movement skills, then combat skills, social skills, crafting skills, etc. each one is unique and can be implemented separately.

Yea... Sorry to burst a bubble here by gathering diagnostics and this type of gameplay is something available off the shelves, including having stats to affect the movement system. Granted, you still need to put work in to make it work for your use case, but there is just nothing revolutionary here. And this is a game that's been over 5 years in development now. Hell, I'll give you 4 for those who didn't read Kickstarter page, and that's all there is to show for it.

Seriously. Can we at least get the single-shot of the trial run, without those suspicious cuts?


Looking to join some fun community on the EU server ;). Friend code: C2D13A

3/7/2020 11:40:44 PM #65

Posted By Puciek at 3:49 PM - Sat Mar 07 2020

Yea... Sorry to burst a bubble here

No worries. I'm sure that it is very likely that there are far more milestones that needed to be fulfilled aside from whatever was needed, even if nothing was needed, to specifically gather diagnostics. Are all of these transversal mechanics, terrain elements, models, animations, etc really available off the shelf ("this type of gameplay" integrated into ue4 by default)?


3/7/2020 11:56:16 PM #66

Posted By Avastar at 11:40 PM - Sat Mar 07 2020

Posted By Puciek at 3:49 PM - Sat Mar 07 2020

Yea... Sorry to burst a bubble here

Are all of these transversal mechanics, terrain elements, models, animations, etc really available off the shelf ("this type of gameplay" integrated into ue4 by default)?

For pretty much all the modern game engines, Unity store is easiest to naviage, so here's an example - feel free to follow related for much more (and this is one of the cheapest, disclaimer I have zero relationships with the author, nor derive any profit from it in any way shape or form): https://assetstore.unity.com/packages/templates/systems/all-in-one-game-kit-elc-character-system-55598


Looking to join some fun community on the EU server ;). Friend code: C2D13A

3/8/2020 1:57:17 AM #67

wonderful nice to see some good progress.


3/8/2020 2:11:02 AM #68

Posted By Puciek at 4:56 PM - Sat Mar 07 2020

For pretty much all the modern game engines

thank you. that does show that many character movements can be bought for cheap. a few character movements i didn't see were those related to ropes and wall running. i don't think it comes with a procedural dungeon generator either. so, while some things could be bought and integrated, they still would of had to code some of their own, and if they are going to code some, might as well code the remainder as well so that the coding is all uniform and unique to CoE.

while it may not come across as revolutionary, some of the milestones that they have accomplished before now are revolutionary. the maps from the procedurally generated worlds, for example. just because it has taken this long to release movement and procedural dungeon testing, doesn't mean that they have been slacking off or not accomplishing things that we are somewhat unaware of. the fact that huge processing power, ram, and time was needed to generate each world/server is impressive to me.


3/8/2020 2:43:50 AM #69

Posted By Puciek at 3:56 PM - Sat Mar 07 2020

Posted By Avastar at 11:40 PM - Sat Mar 07 2020

Are all of these transversal mechanics, terrain elements, models, animations, etc really available off the shelf ("this type of gameplay" integrated into ue4 by default)?

For pretty much all the modern game engines, Unity store is easiest to naviage, so here's an example - feel free to follow related for much more (and this is one of the cheapest, disclaimer I have zero relationships with the author, nor derive any profit from it in any way shape or form): https://assetstore.unity.com/packages/templates/systems/all-in-one-game-kit-elc-character-system-55598

The answer is yes... but no. I'll explain.

Puciek is correct that the two most commonly used game engines today (Unity & Unreal) both have marketplaces that provide a ton of readily available content and plugins for adding functionality to your game.

For example, there's a few parkour plugins in the Unreal Marketplace, there's dozens of inventory system plugins, and there's a ton of asset packs for medieval assets that aren't too unlike what we're using for CoE. Heck, there's even some low fidelity asset packs that we've used in the prelyria client. So if we told you none of the functionality we're working on was available via off-the-shelf solutions, we'd be lying. In fact, we used many of those off-the-shelf solutions ourselves during the first iterations of CoE gameplay when we were doing a lot of prototyping.

The problem with the content/asset packs is they're built separate from one another, by different individuals, using different materials & shaders. That means in a small, single-player game it's perfectly acceptable to download a few asset packs, throw them in your game, and they'll more or less get the job done. However, in a large game like ours, where we expect to have hundreds or thousands of entities on the screen at the same time, we need as many of the models as possible using the same materials/shaders in order to minimize the amount of video RAM being used. We found early on that as we started throwing more and more asset packs into the game with different materials/shaders attached to them that performance started to tank pretty quickly.

Similarly, there's plenty of asset packs on the store that provide their own models, and quite a few that provide animations. Unfortunately, many of the models do not have proper triangle topology, so when you attempt to animate them with skeletal animation you end up with some pretty nasty artifacts.

At the same time, most of the animation packs on the market are designed to be used with the UE4 base pawn and rig. But our rig and skeleton must incorporate body dynamics, and as such, is far more complicated. Retargeting animations from store bought to our rig/skeleton doesn't always look right.

All the above said, we've found over time that store purchased models, materials, animations, etc. work great for smaller single player games, but not so great with a game that's as large and dynamic as ours.

Now let's talk about the plugins. To Puciek's credit, he's not at all wrong. There's a ton of RPG, Adventure, and other genre-specific plugins on the Unreal marketplace that can be purchased and used to add functionality to your game. However, virtually all of them are built using the UE4 Blueprint system, or in some cases, C++.

As with before, if you're making a single-player game, or a small-party adventure game using UE4's built-in replication system, then in many cases you can purchase the off-the shelf solutions, drop them in your game, and they'll work out-of-the box. You might have to tune a few things, but they're a great starting point.

Unfortunately, as with before, CoE isn't a single-player game, or even a small-party adventure game. In an MMOG like CoE, virtually all of the gameplay mechanics must be implemented on the server, not on the client, and unless we plan to run hundreds or thousands of headless UE4 clients on the server (which we don't), it's not possible to use any Blueprint solutions on the server.

Furthermore, even the ones that are built in C++ for "multiplayer" assume you're using the built in replication system and do not have your own client/server network stack used for communicating between game clients.

So much like with the game assets, you download the plugins and quickly find that their Blueprint-exclusive solutions, their multiplayer stack, or simply the fact that all game data is stored and updated using UE4's Actor model and not your/our proprietary ECS game system, renders the majority of what you downloaded useless.

And to underscore that final point, we used many of those plugins in our earlier prototyping to "find the fun" with features such as parkour/traversal problems, combat, archery, and more. And then it was time to get serious, and the last couple of years has been about creating the necessary functionality NOT using off-the-shelf solutions that won't work in an MMO, and won't work with our ECS, but implementing versions that will.

Over the last year we've been focusing on combat, NPC interaction, questing, identities, dynamic story-telling & dialog, NPC AI, world interaction, procedural map generation, dungeon generation, and yes... locomotion/traversal mechanics.

This upcoming pre-Alpha test for Early Access is just one-such set of mechanics we'll be testing this year, and is just one-such set of features we've been working on. So I ask that when you look at the video in the original post, don't judge it too harshly or ask yourself "is this it?" The answer is no. There's way more good stuff to come.


3/8/2020 3:06:20 AM #70

Thankyou for that reply Caspian. That really explains things perfectly for me.


Divide et Impera

3/8/2020 3:08:55 AM #71

Excellent post Caspian, thanks for sharing the lessons learned during COE's development journey to date, good stuff.

It really is harder to do than it first looks.


You must all be quick to listen, slow to speak and slow to get angry. JAMES 1:19 NLT

3/8/2020 4:12:55 AM #72

Awesome post Caspian.


3/8/2020 5:26:50 AM #73

Thanks for the context, Caspian. I know the dev process isn't exactly linear, and this explains a lot.


3/8/2020 6:32:49 AM #74

Caspian, thank you for posting that.

That is exactly the kind of information I love to see.

Seriously, awesome.


Imgur

3/8/2020 7:14:19 AM #75

Ok that post by Caspian forced me to come out from under the bridge a little. I had my own reservations about the shiny that I was gonna keep to myself till I watched it again this morning... go to post and I see his reply. One of the first times in a long time I feel like he has been open and honest about why things are as they are. Looking forward to more of that in the future.


Upvoting and downvoting indiscriminately

... ...