Official MLAPI Experimental Release 0.1.0

Hey Everyone!

Today we are finally announcing the release of our first experimental MLAPI package.

Since this is an experimental release, it is not available in the Unity Package Manager yet. Instead, you have to manually add it from Github. To learn how to install MLAPI into your project follow installation guide.

Version 0.1.0
With version 0.1.0 of MLAPI, we focused on integrating MLAPI in our ecosystem and introduced changes to improve performance and make the API easier to use and maintain. You can learn about what we changed in our release notes.

Our next round of releases will focus on adding new and exciting features to MLAPI. You can get an overview of what we have planned on our roadmap. We encourage everyone to actively participate in the roadmap and submit suggestions and feature requests to us.

What does experimental mean?
We are still working on a lot of future-facing functionality which means that some features are currently a work in progress and you may experience minor issues. Please note that Experimental packages are not meant for use in Production scenarios – instead, you could use experimental packages to preview and test-drive what we are working on as an early look.

There are a few things to be aware of when using this Experimental Unity package:

• Expect breaking API changes - using this early has risks!
• Expect no SLA on the timeline to ship patches/fixes to releases, as we are still establishing our test and release cadence.
• Expect this will be moving towards a fuller Release (aka we don't plan to throw this away)
• Expect us to answer questions of early adopters on Discord and Forums, and we welcome all feedback and issues logged!

Upgrading

For all existing MLAPI users, we encourage you to upgrade to our new version. We introduced many bug fixes in version 0.1.0 and will be better able to support you if you upgrade. MLAPI is now distributed as a Unity package which means upgrading isn’t straightforward. We have created a guide for upgrading to the new version.

If you have an existing UNet project or have used UNet previously checkout our UNet upgrade guide to learn some of the key differences between UNet and MLAPI.

Where to go for help?

Let us know what you think in this forum thread and MLAPI’s discord! This is the place where you can meet our Multiplayer Solution Architect team - engagement engineers helping to support you in building multiplayer games.

Together with the Unity Package, we are also releasing our new documentation site. Our documentation is open source and contributions are welcome (repository)!

Learn more on our Netcode landing page and save the date for Boss Room to actively learn with the upcoming sample game project.

 

You should pin this thread

 

Hi @luke-unity,

Thanks for the update. I'm a little bit confusing that,

1) If I have a product in production based on the existing MLAPI, should I migrate to the new API, Or should I just play with it and wait for it to get mature (reach 1.0.0)?

2) If I have a new product just get started at the moment and aim to be released later this year, Should I start with the new API or the old one?

I know that Unity has a history to make an inaccurate estimation for new features and sometimes just eventually lost track. But as a member work directly on the MLAPI team, your opinion will help manage our expectations.

3) I also have a suggestion for the Git tag naming: https://github.com/Unity-Technologies/com.unity.multiplayer.mlapi/tags

Now all old releases are prefixed with "v", but not for the new API releases. It could be confusing over time. Could you change it to a more readable tag pattern, like `new-mlapi/0.1.0` or `develop/0.1.0`.

 

Thank you for the valuable feedback.

My point is to narrow down to two classic scenarios the audiences may face and listen to advice from the internal guys to migrate or wait.

- Devs with a shipped product and may get some benefits from new API, i.e. better performance (new RPC implementation) or bug fixes.
- Devs with a chance to start a new product and aim to release it later this year.

From my understanding, the feedback is
- Wait if have a shipped product that running well.
- Start with caution, if you want to start a new product. Maybe it's wise to wait until the demo gets released, so at least there's one "product" that is shipped with the new API. Or you can research other 3rd party solutions with better expectation management.

As I said, it's not the first time developers are facing such choices. Because the way Unity is running - rolling out new exciting marketing material (video, slides, demo at conference), and publish preview features with ETA. And it usually takes a much longer time to get to the point that something is stable enough for everything one to pick it up. Sometimes the preview features are just hanging and less cared for. The early adoptions will pay significate energy to involve in the preview feature for bug fixes. Of course, it's not just Unity, relying on any new tech will likely running into the same scenario.

Don't get me wrong, I don't think ship early is bad. You never get a mature point if you don't ship a product/feature to the public first. It just about expectation management and how many resources (Dev/QA/Automated Testing) Unity keeps investing in the feature. Anyone with years of experience working with exciting new features from Unity may share the same concern.

Anyway, you provided some good information and thank you.

https://github.com/Unity-Technologies/com.unity.multiplayer.mlapi/issues/662

 

Can you guys add playmaker actions/support for us PM users ?

 

I think the actions should be developed by third party or by the Playmaker developer.

 

I am a bit curious about the release window for MLAPI, I mean when will it have the 1.0 release?
Because it has been two months since the first 0.1 release and I haven't seen any new big updates till now

 

I posted the following a while ago on our Discord about why we are not doing updates:

I only know about our plans between 0.1.0 and our next release. Long term our strategy will most likely change. Right now creating a release is a very manual process for us and takes a lot of time which we could spend on development instead. This is the case because we lack some of the infrastructure which other Unity packages have for testing. So our plan is to keep patch releases to a minimum for now. If a critical issue arises where there is no easy workaround we can definitely do a patch but so far it looks like there aren't any issues like that.

No patch releases does not mean that we are not fixing bugs. Some of the bugs which were raised are already fixed on the develop branch. It is just really hard for us to get them out to you guys. You always have the option to either use the develop branch or create your own branch and merge some of the bug fixes we did into it.

tl;dr No patch releases will be made unless there is a critical issue

 

Would love to use MLAPI with VIVOX on MULTIPLAY servers. I want the integrations. I showed @LeonhardP our Sony game in progress and would really love to work closer with Unity.

 

Please Unity, post updates on forums too. Many of us do not visit or use Discord because we try to limit the impact of social media on our lives. Forums aren't as engaging, and consume less time / stress.

I do visit Discord but it's likely to be only very rarely if I'm stuck or something.

For example I had no idea MLAPI was due to change it's name until someone said they heard about it on discord.

Cheers!

 

Fair point, we'll look into posting the announcements which we make in the Discord also in this sub forum. Would people prefer a certain format?
1. 1 closed thread where we just post announcements without discussions.
2. We create a new thread for each announcement and anyone can discuss it in that thread (like this thread)
3. A single thread where we do all announcements and discussions.
4. Both 1. and 2.

 

Whichever format you find easiest to maintain without letting it fall behind, I guess!

 

Vote for this + please pin it as well. Like this.

We can open new threads for discussion in the topic.

 

I created MLAPI Announcements for this. From now on we'll post announcements and updates in that thread.

 

Brilliant! Guess what, I didn't even know about the webinar! This is great, thank you.

 

Didn't know either ! Wanted to look at it but then clicked and stumbled on some orwellian pages with 9 prerequisites including but not limited to brain ablation & soul offering:
https://create.unity3d.com/netcode-for-gameobjects-and-message-ordering
https://create.unity3d.com/Multiplayer-Networking-Series-Webinar

Since when was Unity bought by China? can we have links for the free World please?

 

Well I won't get into politics, but it is a bit much, in exchange for some innocent just wanting to learn a bit more about MLAPI, an open source based networking library!

You probably inadvertently did already agree to similar elsewhere on Unity at some point.

 

So many wrong things in such a little sentence that it baffles my mind.

You mention open source yet the condition to access a talk about it isn't quite open, don't you want as many people as possible to have a look at it?
Can't you make a link between the type of person that would be interested in having a look and participate in such a project and what you're demanding by gating those videos with the exact opposite of an open community?
Then you assume that i might have been dumb/lazy to accept it by the past so now it's okay, get wronged by the past makes the wrong okay in the present...

The only good point is my message not being sensored (unless it's shadowed looking from the none response of any other users, or there were little to no interest in the subject from the get go ?).

 

Steady on, mods aren't Unity staff. We just try to help out, FYI. We do not speak for Unity and often criticise quite heavily if you look at mod posting history

But your complaint is valid and accepted.