Lumberyard Game Engine

Denzil

I'm a pretty attached to Unity Game Engine and have been using it for all my projects thus far. But this looks great! https://aws.amazon.com/lumberyard/

It's Amazon's Lumberyard Engine. As far as I can tell, it's 100% completely free and opensource. It's cross platform and has twitch integration.

It's C++ based I think. Normally I go for C# but I'm really considering this strongly for future projects!

What do you guys/gals think?

Elyaradine

I do not like the C++
But I will use it if I must.
I do not like to be the first.
(Early adopter bugs? The worst!)
More competition's always great.
Low prices I appreciate.

Denzil

@Elyaradine Haha, fantastic!

roguecode

C++ is a big blocker for me. I feel less productive in C++ than C# (likely just because my experience is very low there), but beyond that it seems hard to find C++ people locally.

Performance-wise, I like where Unity is going with IL2CPP.

But yeah, the more competition the better, and this looks pretty cool.

rustybroomhandle

This engine is based on CryEngine - so if that's your kind of jam, then you'll be right at home. :)

Tuism

I agree with most above me, also I want to add that I'm not a fan of learning a new language :/

kidult

It does have a system similar to blueprint in UE4 and word on the street is that it's even easier to use... guess that's better for a non coder though.

SkinnyBoy

I use cocos2d-x so it will be cool for me to check it out

SkinnyBoy

I just notices the engine size is 10gb, so I'll pass

Denzil

@SkinnyBoy haha, yeah I noticed it too this morning. BIG.

quintond

There is always a catch ...

Free – with Full Source
Amazon Lumberyard is free, including full source code, so you can deeply customize Lumberyard for your team and your vision for your project today, and for future projects in years to come. There are no seat fees, subscription fees, or requirements to share revenue. You only pay for the AWS services you choose to use.

... yeah AWS services ... make you bankrupt in seconds if you don't pay attention. I would probably choose Unity over this ... ummmm ... obvious promotion for AWS.

* Sorry if there are any AWS employees on these forums, not my intention to offend anyone just my humble opinion regarding AWS.

kidult

@quintond - So bit of a noob here but what would you use the AWS for?

roguecode

The fine print says that if your game uses a cloud service then it has to be AWS. But if you're self hosting your own server then that is fine too.

quintond

Personally ... I would assume that you would use it for shard servers or maybe server hosting or stat tracking systems ... everything about Lumberyard points to AWS integration.

Amazon Lumberyard is a free, cross-platform, 3D game engine for you to create the highest-quality games, connect your games to the vast compute and storage of the AWS Cloud, and engage fans on Twitch.

So if it is a typical MMORPG for instance, the game engine supports AWS from the get go ... I kind of see this as working similar to Hero Engine.

By starting game projects with Lumberyard, you can spend more of your time creating great gameplay and building communities of fans, and less time on the undifferentiated heavy lifting of building a game engine and managing server infrastructure.

So once again it looks like you make your game and then get access to AWS for all the deployment of servers, social media, Twitch streaming ... etc ... but all off AWS.

With Amazon Lumberyard’s visual scripting tool, your designers and engineers with little to no backend experience can add cloud-connected features to a game in as little as minutes (such as a community news feed, daily gifts, or server-side combat resolution) through drag-and-drop visual scripting.

Focus on the game and not worry about the backend services, which in theory sounds very good but AWS can work out extremely expensive if you aren't paying attention. I for one would rather have a tight handle of my systems, be able to build my backends myself and hence be able to debug them or fix them if something goes wrong.

In theory this sounds awesome but my gut says this could end up getting a lot of Indie Developers in to trouble due to lack of understanding of the complexities of AWS.

Using Amazon GameLift, a new AWS service for deploying, operating, and scaling session-based multiplayer games, you can quickly scale high-performance game servers up and down to meet player demand, without any additional engineering effort or upfront costs.

This is very cool but is also something quite easily achievable through companies like Digital Ocean and their API ... and probably for cheaper. The problem here is that systems built for AWS can not be deployed on to non-AWS systems with ease and visa-versa ... so if you pick one you might end up never being able to move.

It is very cool that they are supporting AWS from outset but that site mentions AWS constantly ... which seems to me and maybe I am wrong ... that the system is geared only to AWS and is just a way for them to hook more people on to AWS.

Please note that this is just how I am interpreting the information on the site and applying my experience with AWS for other projects (non-game related) that I had to use AWS on. If you really want the flexibility of AWS type services then stick with Unity or Unreal and rather use something like OpenStack and combine it with things like Vagrant and Chef.

I personally use a home built system called GameFlow (that I wrote) and I have hooked it up to Digital Ocean via their API. Digital Ocean allows me to use standard Linux Virtual Machines that I can either create directly on Digital Ocean or create them on my test environment in Virtual Box and port them across to Digital Ocean when I am ready.

Okay ... this response was longer than I anticipated ... sorry. 8-{

* Sorry if there are any AWS employees on these forums, not my intention to offend anyone just my humble opinion regarding AWS.

garethf

Clever how they're differentiating their engine by targetting the e-sports market with integrated scalable multiplayer and twitch streaming out of the box.

So it's looking like it may shake down to:

Unreal if you want raw graphics horsepower
Unity if you favour easier cross-platform support.
Lumberyard if your focus is scalable multiplayer e-sport like products.

Not that you couldn't make whatever you wanted in any of those engines (or another engine), but it's interesting to see what territory the big players are trying to target.

It's also interesting to see world-class engines being offered for free as an incentive to use some service. What a time to be a game developer. :)

dislekcia

Did you all see the zombie apocalypse clause?


Thanks to Sos for pointing that out ;)

quintond

roguecode said:

The fine print says that if your game uses a cloud service then it has to be AWS. But if you're self hosting your own server then that is fine too.

So in essence that is no difference to Unreal's 5% Royalty. Also, what do they define as a cloud service ... no thanks.

roguecode

quintond said:

roguecode said:

The fine print says that if your game uses a cloud service then it has to be AWS. But if you're self hosting your own server then that is fine too.

So in essence that is no difference to Unreal's 5% Royalty. Also, what do they define as a cloud service ... no thanks.

I also find the Twitch thing a fairly odd sell. Yeah, you get a few cool examples of integration, but is this really something that most game developers and gamers want?
And aren't they attracting slightly the wrong crowd? If I were to market my game engine I'd want to show awesome games with a lot of fancy tech. Said fancy tech would have been made by fairly competent developers. Fairly competent developers would also be able to write their own Twitch integration.

Asbestos

dislekcia said:

Did you all see the zombie apocalypse clause?


Thanks to Sos for pointing that out ;)

My understanding of license law is not authoritative, but I think that in the event of civil collapse, there will be few lawyer-survivors willing to represent game engines from a collapsed super-consumer enterprise. Especially not those engines being used to run the makeshift centrifuge that enriches the nuclear fuel powering their small, but fast-growing, faction.

dislekcia

roguecode said:

I also find the Twitch thing a fairly odd sell. Yeah, you get a few cool examples of integration, but is this really something that most game developers and gamers want?
And aren't they attracting slightly the wrong crowd? If I were to market my game engine I'd want to show awesome games with a lot of fancy tech. Said fancy tech would have been made by fairly competent developers. Fairly competent developers would also be able to write their own Twitch integration.

I'm not sure I follow the logic here... Why would fairly competent developers be who they're going after here? And fairly competent developers use pre-written engines and plugins all the time, it's why they're fairly productive ;)

But something like Goat Simulator with free twitch integration, or whatever the next step in internet hilarity that people can make odd faces while playing ends up being, is clearly what they're after. The marketing isn't designed to catch people on tech bullet points, it's designed to make as many people come up with as many crazy ideas as possible in the hope that some non-zero fraction end up actually making something that goes viral.

roguecode

dislekcia said:

roguecode said:

I also find the Twitch thing a fairly odd sell. Yeah, you get a few cool examples of integration, but is this really something that most game developers and gamers want?
And aren't they attracting slightly the wrong crowd? If I were to market my game engine I'd want to show awesome games with a lot of fancy tech. Said fancy tech would have been made by fairly competent developers. Fairly competent developers would also be able to write their own Twitch integration.

I'm not sure I follow the logic here... Why would fairly competent developers be who they're going after here? And fairly competent developers use pre-written engines and plugins all the time, it's why they're fairly productive ;)

But something like Goat Simulator with free twitch integration, or whatever the next step in internet hilarity that people can make odd faces while playing ends up being, is clearly what they're after. The marketing isn't designed to catch people on tech bullet points, it's designed to make as many people come up with as many crazy ideas as possible in the hope that some non-zero fraction end up actually making something that goes viral.

I'm not meaning that they don't use pre-made engines or plugins. I'm just trying to say that in a venn diagram with "developers who will build the type of games that would be good showcases of the engine" on the one side and "developers who choose an engine because it has a free Twitch plugin" on the other, there is very little overlap. Yes, Goat Simulator could be a good example over being in the overlap - but I just can't see many people deciding on an engine because of that, or the easy AWS stack.

dislekcia

roguecode said:

I'm not meaning that they don't use pre-made engines or plugins. I'm just trying to say that in a venn diagram with "developers who will build the type of games that would be good showcases of the engine" on the one side and "developers who choose an engine because it has a free Twitch plugin" on the other, there is very little overlap. Yes, Goat Simulator could be a good example over being in the overlap - but I just can't see many people deciding on an engine because of that, or the easy AWS stack.

Yeah, it's the "type of games that would be good showcases of the engine" bit that I'm not following. This seems like an untested assumption about the type of games that Lumberyard is built to produce being inherently technically astounding... And a corresponding assumption about the types of people that can produce those games only being good coders. Neither of those things is necessarily true, so I'd argue that they're hoping to push for server-heavy, streamable, community driven games - which points to things like eSports, subscription/F2P/service multiplayer games and stream-bait fumblecore to me.

You're right that the C++ and open source angles don't gel with AWS and twitch integration in terms of who they talk to as users, but they're probably not supposed to. In the end it's all about someone making something with Lumberyard that does what Garys Mod did but for the streaming marketplace. Something like "Twitch plays feed my baby" ;)

kidult

@dislekcia - "stream-bait fumblecore" ...that made me laugh quite a bit. Great name for the genre