I was quite surprised it worked so well myself. Was expecting severe latency issues and even that didn't really happen. At this point I really just need to reimplement Unity's NetworkTransform component to add some prediction and server-sided lag compensation. Considering this test went so well and took me 2 days to implement we'll probably see a public multiplayer alpha soon.
3 Likes
keep up the hard work. you have gotten a lot further a lot faster than i ever expected. looking forward to the alpha and beta launch keep your dream alive.
3 Likes
Another player model revision. This one is a lot higher detail and took me a couple hours. Has a bigger overall profile as well.
Tried to do some fancy armor plating with the dark gray and black accents.
2 Likes
Very hastily made video showing some further multiplayer testing.
Other than that I've just been doing some housekeeping as far as the code and assets. Some glitches have been fixed such as getting shot by your own projectiles and multiple UI instances spawning for each client.
3 Likes
Things have been a bit slow over the last few days but are still moving forward. I've been reading up on networking a lot and have been trying to add basic prediction for other players and projectiles.
Also randomly made a railgun model in Blender:
This will probably be the first precision weapon in the game. It will either act like a semi-automatic rifle with moderate range and rate of fire or a type of long-range sniper style weapon with a short charge time.
I'm thinking of making precision weapons a lot more advanced with armor, ricochets, and deflections taken into account. We are dealing with vehicles here so hitting an armor plate at a shallow angle and having your shot deflect off with minimal damage makes sense. Likewise striking an unarmored section might result in significant damage. Any thoughts?
1 Like
The polygon mountains look dope. Totally fits in.
2 Likes
Well I've pretty much stalled completely on networking. Prediction is needed at the very least and implementing it in UNET just seems impossible. I have to completely throw out the NetworkTransform component. At this point I'm just going to shop around the asset store for a better and more mature networking solution.
All the tutorials and examples for UNET involve simple hitscan/raycast based FPS games with generic human characters. I have yet to find anything dealing with a game like mine.
I don't really have the funds to spend on any yet but if that's the only option I'll just have to find a way. Worst case scenario is having to build my own solution which may extend development time well past Summer.
I'm not really sure what to do at this point.
1 Like
Kickstarter
Make tons of money just patch in one thing, profit
Lol jk
Well
Or
Should be good
To be honest building a very hacky prototype that just shows off the game to back some kind of funding is probably a viable option.
I'm certain I could build my own solution but I just don't have the time to do it effectively. The game could be delayed till late 2017 / early 2018.
I might honestly start looking at Unreal Engine 4 again. I know C++ and their networking system is complete and does everything I need out of the box. It might be that taking the time to rebuild Genobyte in Unreal will be more efficient than stumbling through Unity.
I'm not giving up but I'm certainly frustrated at this point.
4 Likes
You do not know how excited I am to play this when it comes out. Sign me up for the Beta. The Low Poly Art is just gorgeous, and I love your videos. Keep up the good work and stay strong.
5 Likes
I'm going to dedicate the next week or so to working in Unreal. I need a break from Unity and Unreal is arguably the best engine for this kind of game. I'm going to just start porting code from Genobyte over to Unreal C++ and Blueprint. I have less overall experience with Unreal but catching up shouldn't take very long.
I'll post updates here as usual.
2 Likes
I'd like to voice my approval of Unreal. I've heard nothing but good things coming out of Unreal. You need to pick an engine that solves the hardest problems for you and leaves the easier ones for you to solve yourself. At least, that's the way I'd do it.
1 Like
I second the motion I was going to start making Unreal Comments, I want to make movies in Unreal, and make a game but my computer is waaaaaaayyyyy to crappy. Unreal could be used as a Linux benchmarking system of sorts if you don't want to buy windows for small computer building youtube channels.
1 Like
It's definitely superior in a lot of ways. Unity is great for smaller projects and is excellent for rapid prototyping. Unfortunately it simply lacks built-in functionality in some critical areas and while 3rd party assets make up for it, they also end up costing quite a lot of money.
I'm honestly pretty set on Unreal. It might set me back a month or so while I try to reach parity with the Unity project but a month is a lot better than a decent amount of money and the better part of a season. Plus I know for a fact that this project can be done in Unreal without any major complications just based on the industry.
Unity helped me turn the abstract concept into something concrete. Now Unreal can pick it up and get it to production.
1 Like
Last update for the day before I head to bed. In just one afternoon I've managed to port the following to Unreal Engine 4 from the Unity project:
- Hover Physics w/ Stabilization
- Mouse Aiming & Steering
- Player Model
- Low Poly Canyon
Started today knowing nothing about Blueprint and after some reading and messing around it all clicked and I had a huge amount ported over before I knew it. I might not have to write any C++ code at this rate. Did not expect things to take off on an unfamilar platform this quickly.
3 Likes
Another progress update. I've rebuilt a lot of the Blueprint code and behavior is now almost 1:1 with the Unity version. I still need to add some compensation logic for slopes but otherwise I'd say player physics are about 95% ported.
Here's a video showing the current build in action:
Honestly feels smoother than the Unity version.
Blueprint definitely has a faster overall workflow than C# and I'm loving it. Far easier to debug, modify, and read.
7 Likes
I find UE4 to be better with Hardware snuff even on older ones. Better Physics implementation too,
1 Like
Well I'm definitely liking the engine more than Unity so far. The last few days have been spent learning the C++ API and working on little things here and there. Someone has a project for networked PhysX on GitHub that I might fork and implement in the project. Need to brush up on C++ a bit more before doing that though.
Just the fact that Unreal gives you source level access to the engine makes it possible to do things that Unity simply can't.
Randomly made this model in Blender as well. Might end up being a player model or I'll give it to my friend for his RTS project.
3 Likes
Maybe you can make a tutorial on how you make the maps in blender. I would love to make my own map with some interesting features. Just Cause? shrug Also how you make the player models, it would be fun for me to make some in my spare time.
2 Likes
Yeah sure, I'll try to record the next time I work on something in Blender and maybe add a voice over.
2 Likes