3 ms·

> I agree with Unity's long term vision (move to .net, package manager, rendering pipelines, UI toolkit, dots). The real question for me though is if they are g

> I agree with Unity's long term vision (move to .net, package manager, rendering pipelines, UI toolkit, dots). The real question for me though is if they are going to pull it off, or get derailed along the way. I've had more time to think about this. Unity effectively has a golden goose of an ecosystem right now and they're about to murder it with boardroom bullshit. How much energy was spent on this merger that could have been redirected to doubling-down on the tech stack? Multiplayer could really use some more attention, IMO. It's definitely not too late for them to correct course, but I've seen this path so many times I do not reserve any hope. I took a loss on my entire Unity position this morning to get out from under any future bad decisions.

squeaky-clean 4y ago
The real issue IMO isn't that they aren't doubling down on the tech stack or paying attention to multiplayer. It's that are doing that but have gone full Google and kill these projects after they enter alpha or beta. For multiplayer there was the original UNet, then the FPS Multiplayer Demo stack, then HLAPI/MLAPI, now we're getting "Netcode for GameObjects". That also means their new tech stack rewrite for ECS won't get multiplayer support until they release "Netcode for ECS" at some later date. There's stuff like buying Bolt and then depreciating it months later for Unity Visual Scripting. It's like they have a ton of teams working on rewriting the engine but none of them communicate and often find out they've been rewriting the same feature as another team.
- throwuxiytayq 4y ago
  > For multiplayer there was the original UNet, then the FPS Multiplayer Demo stack, then HLAPI/MLAPI, now we're getting "Netcode for GameObjects". This doesn't seem like the correct timeline of events. Anyway, if you want a clearer picture why they decided to throw the packages away, I recommend reading the source and/or trying to use them. For your reference: https://github.com/needle-mirror/com.unity.multiplayer-hlapi https://github.com/needle-mirror/com.unity.multiplayer-hlapi > That also means their new tech stack rewrite for ECS won't get multiplayer support until they release "Netcode for ECS" at some later date. Uhhhh? The package is literally right there. It's one of the very earliest ECS packages they published, I think. They also reaffirmed that the package is in active development internally. https://docs.unity3d.com/Packages/com.unity.netcode@0.51/manual/index.html https://docs.unity3d.com/Packages/com.unity.netcode@0.51/man... > There's stuff like buying Bolt and then depreciating it months later for Unity Visual Scripting. Bolt is Unity Visual Scripting. I really don't know what you find so offending here (maybe except Unity's inability to develop their own tools in-house). > It's like they have a ton of teams working on rewriting the engine but none of them communicate and often find out they've been rewriting the same feature as another team. I can't come up with a single example of this. Unity occasionally has a tendency to develop a "vnext" tool while still supporting a "legacy" tool, which seems great for backwards compatibility (and for long-running projects), but it just keeps confusing people endlessly. I don't like Unity, but hell, I don't envy them for constantly having to deal with this shit either.
  - squeaky-clean 4y ago
    > Anyway, if you want a clearer picture why they decided to throw the packages away, I recommend reading the source and/or trying to use them. This isn't really a good excuse for why core features have been rewritten 4x without a production release and with no legacy version available for the latest engine. You left out MLAPI from your link and the FPS Multiplayer Sample NetCode https://docs-multiplayer.unity3d.com/netcode/0.1.0/getting-started/about/index.html https://docs-multiplayer.unity3d.com/netcode/0.1.0/getting-s... https://github.com/Unity-Technologies/FPSSample https://github.com/Unity-Technologies/FPSSample > Uhhhh? The package is literally right there. I mean actually release it in a way that's stable for teams working on a game. Not "This package is available as a preview, so it is not ready for production use." The current roadmap plan is for Netcode for ECS to reach production way after Netcode for GameObjects. Bolt is not Unity Visual Scripting. It's extremely similar but that only makes it more confusing. Bolt is a separate asset which is still maintained, the last update was a month ago. > develop a "vnext" tool while still supporting a "legacy" tool Except their only production level multiplayer library, UNet is not supported anymore. There is no official way to make multiplayer Unity games right now that isn't a preview package or deprecated.
thsbrown 4y ago
> I've had more time to think about this. Unity effectively has a golden goose of an ecosystem right now and they're about to murder it with boardroom bullshit. How much energy was spent on this merger that could have been redirected to doubling-down on the tech stack? Multiplayer could really use some more attention, IMO. I see your point, but in a way I think this merger makes sense to prioritize. I would argue the majority of their user base is using the engine in its current form for mobile development. Unfortunately Unity doesn't have a golden goose like fortnight to draw revenue while they are internally improving the engine. It stands to reason if they can draw in more revenue from ads from their pre-existing user base they have more money to keep them afloat when times are tough while they continue engine improvements. To be honest I don't like it. I would much rather they double down on fixing the engine and solidifying preexisting solutions. But I do think there might be a method to their madness. You might be right though, it might just be boardroom bullshit haha. I'm still holding out hope. And in the meantime I'm going to diversify my engine knowledge.

squeaky-clean 4y ago

The real issue IMO isn't that they aren't doubling down on the tech stack or paying attention to multiplayer. It's that are doing that but have gone full Google and kill these projects after they enter alpha or beta. For multiplayer there was the original UNet, then the FPS Multiplayer Demo stack, then HLAPI/MLAPI, now we're getting "Netcode for GameObjects". That also means their new tech stack rewrite for ECS won't get multiplayer support until they release "Netcode for ECS" at some later date. There's stuff like buying Bolt and then depreciating it months later for Unity Visual Scripting. It's like they have a ton of teams working on rewriting the engine but none of them communicate and often find out they've been rewriting the same feature as another team.

throwuxiytayq 4y ago
> For multiplayer there was the original UNet, then the FPS Multiplayer Demo stack, then HLAPI/MLAPI, now we're getting "Netcode for GameObjects". This doesn't seem like the correct timeline of events. Anyway, if you want a clearer picture why they decided to throw the packages away, I recommend reading the source and/or trying to use them. For your reference: https://github.com/needle-mirror/com.unity.multiplayer-hlapi https://github.com/needle-mirror/com.unity.multiplayer-hlapi > That also means their new tech stack rewrite for ECS won't get multiplayer support until they release "Netcode for ECS" at some later date. Uhhhh? The package is literally right there. It's one of the very earliest ECS packages they published, I think. They also reaffirmed that the package is in active development internally. https://docs.unity3d.com/Packages/com.unity.netcode@0.51/manual/index.html https://docs.unity3d.com/Packages/com.unity.netcode@0.51/man... > There's stuff like buying Bolt and then depreciating it months later for Unity Visual Scripting. Bolt is Unity Visual Scripting. I really don't know what you find so offending here (maybe except Unity's inability to develop their own tools in-house). > It's like they have a ton of teams working on rewriting the engine but none of them communicate and often find out they've been rewriting the same feature as another team. I can't come up with a single example of this. Unity occasionally has a tendency to develop a "vnext" tool while still supporting a "legacy" tool, which seems great for backwards compatibility (and for long-running projects), but it just keeps confusing people endlessly. I don't like Unity, but hell, I don't envy them for constantly having to deal with this shit either.
- squeaky-clean 4y ago
  > Anyway, if you want a clearer picture why they decided to throw the packages away, I recommend reading the source and/or trying to use them. This isn't really a good excuse for why core features have been rewritten 4x without a production release and with no legacy version available for the latest engine. You left out MLAPI from your link and the FPS Multiplayer Sample NetCode https://docs-multiplayer.unity3d.com/netcode/0.1.0/getting-started/about/index.html https://docs-multiplayer.unity3d.com/netcode/0.1.0/getting-s... https://github.com/Unity-Technologies/FPSSample https://github.com/Unity-Technologies/FPSSample > Uhhhh? The package is literally right there. I mean actually release it in a way that's stable for teams working on a game. Not "This package is available as a preview, so it is not ready for production use." The current roadmap plan is for Netcode for ECS to reach production way after Netcode for GameObjects. Bolt is not Unity Visual Scripting. It's extremely similar but that only makes it more confusing. Bolt is a separate asset which is still maintained, the last update was a month ago. > develop a "vnext" tool while still supporting a "legacy" tool Except their only production level multiplayer library, UNet is not supported anymore. There is no official way to make multiplayer Unity games right now that isn't a preview package or deprecated.

thsbrown 4y ago

> I've had more time to think about this. Unity effectively has a golden goose of an ecosystem right now and they're about to murder it with boardroom bullshit. How much energy was spent on this merger that could have been redirected to doubling-down on the tech stack? Multiplayer could really use some more attention, IMO. I see your point, but in a way I think this merger makes sense to prioritize. I would argue the majority of their user base is using the engine in its current form for mobile development. Unfortunately Unity doesn't have a golden goose like fortnight to draw revenue while they are internally improving the engine. It stands to reason if they can draw in more revenue from ads from their pre-existing user base they have more money to keep them afloat when times are tough while they continue engine improvements. To be honest I don't like it. I would much rather they double down on fixing the engine and solidifying preexisting solutions. But I do think there might be a method to their madness. You might be right though, it might just be boardroom bullshit haha. I'm still holding out hope. And in the meantime I'm going to diversify my engine knowledge.