I’m Saulty, the new Community Manager for Lumberyard. My background is in the games industry with 13 years of work experience at CCP (creators of EVE Online), Crytek, Virtuous and other studios.
I’m very excited to join the Lumberyard Community and my goal is to bring more awareness of our developers’ issues and needs to the Lumberyard team and drive engagement between the two. One of the first steps towards this is to get to know your needs and issues when working with Lumberyard, so expect a survey coming soon so you can tell the Lumberyard team what are the most pressing issues for you while using Lumberyard.
I’m also on the Lumberyard Discord channel so feel free to chat with me any time!
I using Lumberyard since 2016 , I make great video tutorials/Projects for Script Canvas lumberyard…I designed/prograpging/programming projects with a lot of game engines and since 2016 my main goal has been only Lumberyard.
hi @lmbr_Saulty! it’s great to see our new CM i’m just curious who was previous CM?
In continue of your speaking about connection between ly-devs and indie game devs …can you ping someone from ly-dev team and ask to make properties of AZ::Vector2/3: such like - x,y,z, make them public in c++) Why so simple types should uses setters and getters for class fields(
also it’s will be nice if Lua Editor can save user’s preferences, currently it’s just startup with defaults everytime, and do not take in account setting what you saved before.
Ray tracing in addition to updating the renderer since one of the main attractions of lumberyard is the legacy cryengine renderer and in the future when more features are stable create a new complete game template using new tools such as dynamic vegetation, script canvas, PhysX, Game State, you can not miss a basic configuration to establish connection with AWS servers (lobby, chat, etc.) and the future features that include, I know there are gems with examples of all of them but it would be incredible and we would learn a lot, seeing a Template with all that integrated.
Another great priority feature, make the gems have a behavior similar to Unity packages, just import the one you need or eliminate it without having to build the entire project completely.
Related to the above I do not see the need to have a camera, input, audio or physical gem, any genre of videogames uses these components.
And finally, as they had mentioned before, making a public roadmap this interests me a lot since I have been with this engine for some years and I would like to do tutorials about it for the Hispanic community, I don’t do it because I would hate to spend a lot of time learning something to then show him spending a lot of time editing the video preparing what I will do etc and that the next week they delete it because I had no idea that that was what was still on his road map.
I will end this with some questions.
Will they replace Lua with another programming language or will they also eliminate Lua and only ScriptCanvas will exist?
I ask this since I see that your documentation focuses more on ScriptCanvas than on lua, besides they have replaced everything that came with CryEngine
Why did you eliminate the boid gem?
Will they replace it with any other or will it no longer exist?
The next question goes a bit with the first thing I wrote
Will they update or replace the Cryengine renderer in the near future or will they focus first on the tools and then on the graphics?
YES, I agree with you (Only 50%) , I’m sad that lumberyard team eliminate the boid gem! if you see my bugs report & other bugs that lumberyard community reported , It shows that lumberyard team are at the beginning of the path as they say , because they have repeatedly said in the release of each new version release , I’m glad I’m in lumberyard.
I developed a project in Lumberyard 1.18 with name : Robot Islands but I can not read my project in Lumberyard 1.23 , I have no problem in Lumberyard 1.19
This is video Tutorials about a section of Robot Islands , Playlist link
I want just like you , make video tutorials for other languages (Kurdish & Persian) , I think , it is a necessary for optimize render algorithm in lumberyard like Vulkan API/DirectX12/13 or new Metal library, plus, Ray tracing , etc in lumberyard, but I/you/we know that develop a game engine is very hard & It is time consuming… so we must waiting for lumberyard 2.0 final or Lumberyard 1.x final, Your criticism is positive and true, I think amazon & Lumberyard team have two roadmap, first a public roadmap in near future( new features in new tools or exist tools) & second , a private roadmap in far future(Includes strategies, tactics and techniques for years to come)
I think SCRIPT CANVAS IS GREAT, also here I requested a public roadmap in 2016
of course I have a wish list like lumberyard community like Full Vehicle Script Canvas System in Open world game, Optimize Tools, Seasons system(Spring ,summer, … , this system exist in SpeedTree cinematic), Big radar map with script canvas/entities video tutorials, delete wwise and add a new audio system , at end Behavior Tree/AI Cover Points /AI Friends & Enemies video tutorials , all top topics & top goals is time consuming… so I’ll wait.
Yes, this is one of the key issues - Lumberyard rolls from one beta version to the next, without the community having any real idea what’s going on, or any sense of security. There’s no way of knowing if or when there will be breaking changes. That’s not really acceptable, and many developers won’t look twice at the engine until that’s sorted out.
I do understand that knowing the development direction Lumberyard will be taking is important to the community. To that end I will endeavor to bring as much transparency on it as possible, however, there are other considerations such as proprietary (secret sauce) work for Lumberyard that cannot be disclosed just yet. So complete insight may not be possible.
In the mean time I’ll see if I can start posting upcoming changes for future updates before a release so the community gets a heads up, and let me know what else you might be interested in (such as a feature list showing what is being worked on) and I’ll do my best to make it happen.
Well, I don’t think it’s necessary to reveal all the “secret sauce”, and I don’t think it’s all about upcoming features.
The problem stems from not really knowing where we are with the project. Unreal, for example, doesn’t have a public roadmap, but it does have a reliable release cadence. There is a numbered release every six months or so, each with several point updates to patch bugs, before moving on to the next version. The common practice is for a studio to freeze on a mature version as their project approaches completion.
With Lumberyard, it’s a rolling beta - we don’t know where we are, what might be changed or removed, when it might be ready to leave beta, or indeed if it ever will. I don’t think too many commercial projects are going to want to put themselves in that situation.
So, I think a roadmap to leaving beta, with the tasks that are required for that milestone, would be a good start.
With regards to the vector data question, I’ve contacted the dev team and they are looking into possible solutions for it.
For some context into the reason for the currently setup - Lumberyard vectors on x86 use Intel 128-bit SSE registers. Whereas floating point on x86 still uses the legacy x87 80 bit internal 0-operand stack based floating point that Intel used a long time ago. Frequently moving hot data between SSE registers and floating point can be enormously expensive and can easily cause critical performance issues for pseudo real time software like a game engine. For that reason, the accessors currently available use hand-written intrinsics to attempt to avoid any legacy floating point usage.
Hope that helps, and I’ll keep you posted on further developments on this.