Devlog

[Quentin]

Project Glaurune (*working title)

The goal of this project is to create a spiritual successor to Nox. A remake is not possible because EA owns the rights to the Nox IP and i don't want my ass sued or the project taken down. I'm trying to keep all the good gameplay mechanics, the fast paced and creative combat and expand upon it.

Version 0.3: https://www.youtube.com/watch?v=T2Sd-2iclC4
Version 0.4: https://www.youtube.com/watch?v=Ex96nNByzoI

Trello: https://trello.com/b/YFyZZrg4/project-glaurune

Begin: 31. Oct 2018

[Quentin]

Still not dead, starting to work on it now.

[Quentin]

Got something on the screen:

https://www.youtube.com/watch?v=X1YVhTTFZzY

Basically everything is a placeholder or unfinished. Mostly still learning the ins and outs of UE4 at this point.

[Quentin]

Worked a lot on it but almost nothing new to show. Just to log what i'm currently doing:

- Made the walls, doors and windows modular with LOS implemented, so switching out wall styles will be very easy and works out of the box for new maps. Also fixed some stuff in the LOS system, not entirely done here though, it's not very drag-and-drop yet. There are a few edge cases with with the LOS system.

- Started to learn UE4 animations and the integration of an animation system. It supports proper turn-in-place which looks pretty nice.

- Played a lot with the perspective. Perspective is mostly an issue because of visibility of players behind walls. Noticed my map was only 30 degrees rotated instead of 45 degrees which is the standard for an isometric look. True isometric doesn't work in UE4 though, the orthographic camera is all kinds of broken. You can fake parallel projection by reducing FoV of the perspective camera and zooming out, that kind of works nicely until a certain degree because if you zoom the camera out too far other UE systems start to break (lightning, shadows, LODs...). The lost depth also doesn't help in making the game look modern. The same goes for the camera set too far top-down.

- I implemented the usual "character highlighting behind walls", but Nox also makes the walls transparent, which looks even in 2D kind of weird at a closer look. Transparent walls testing is one of the things i'll do next though.

- More mana stones testing, still not entirely satisfied with them.


Todo:
- Test transparent walls
- Implement ability system (There is an ability system in UE4 that was used for Fortnite and Paragon that i'll use, it's kind of complex and not well documented but supports proper multiplayer)
- Implement HUD
- Implement abilities & classes
- Fix, refactor and polish inventory & item system
- Implement modular chars (male and female)
- Implement ragdolling, blood, removeable body parts (last part is still up to testing)
- Implement menu system
- Implement a basic character creator where you can customize your char and start gear
- Learn proper lightning techniques
- Setup a master server for server discovery (still need to do research about the Steam master server, maybe i'll code a custom one)
- Login system, probably based on Steam

[Quentin]

Transparent walls are in. They look way better than character outlining behind walls. Windows are quite important in Nox so you'll find yourself behind walls pretty often compared to traditional RPGs. And being outlined all the time sucks big time.

Im over HUD, ability system and ability protoypting now.

The ability system prototype will be:

- 8 Elements (QWERASDF like Magicka) but no element mixing, everything is a unique spell
- Probably a very basic elemental-counter system (fire - water e.g.) i feel like that's really expected for the obvious counter elements.
- Every Element has a basic spellset of 7 or 8 spells
- (1 spellbutton for moving to a new spellset and potentially allow for unlimited spells per element)

I will test 3 options to navigate the spellsets
1) combo (auto reset back to main selection after spell is cast)
2) manual reset (stay in the element spellset as long as you'd like until you press an button which brings you back to the elemental selection)
3) 1-8 to select the spellsets directly while within another spellset (or bind your own keys)

I really need to test it first. Maybe there will be an option to allow for more than one style.

[Quentin]

Got a basic spell system working and HUD implemented. Spells are just crappy placehodlers though, i make a new video once i have a few core spells running smoothly.

Also did a looooot of research on networking recently. Physics is going to be complicated to make work with clientside prediction. Clientside prediction + physics requires modification of the prediction code and ability to rewind and replay the entire physics state, it's especially problematic because physx is not deterministic and continous resync of the state will increase the bandwidth by a lot.

Nox didn't do clientside prediction but it's pretty much expected in 2019. I also looked into lag compensation but i'm not sure if it's a good fit for top-down games. I hope i can get a good enough system running.

Plan B is no clientside prediction like Nox did it. Actually i even prefer it over a bad clientside prediction implementation like the one wizard wars had.

[Quentin]

Even though estate can't make it in the final game and it's kind of a waste of time i can't help rebuilding it:

https://i.imgur.com/7ZgmPmE.png

Currently working on spells, i have a few core spells prototyped and i'm going to implement a proper spell system now.

Also there is good news, apparently Epic is replacing Physx with their own Chaos engine in the next few versions. Hopefully it improves determinism and physics networking. That means i won't go beyond simple networking tests with physx until there's more info about Chaos.

[superjack]

Why can't estate be in the final game???

[Quentin]

Unfortunately map layouts themselves are copyrighted and i don't want to give EA any ideas.

I won't use any copyrighted material which is basically everything but gameplay mechanics which thankfully can't be copyrighted. 1:1 use/copy of any art, music, sound, story and map layouts won't be possible.

I'll eventually make a map that will feel similiar though. For now i'm mainly testing on estate to adjust the gameplay and visibility mechanics because i already know how playing on estate should feel.

Well, if the game becomes successful a map editor would be nice to have anyway and the copyright problem solves itself i'm pretty sure. Map editor is currently not on my roadmap but wouldn't be that hard to implement as i keep everything as modular as possible.

[Quentin]

Project Glaurung 0.02 - what is the best game mode and why is it flagball?

https://www.youtube.com/watch?v=_OxBa2hVIWY

Ok, the ball might not make it into the game, at least not in that form, that was mainly me getting sidetracked while testing things.