As demonstrated in the fish demo?
No sir, it’s not included in release note page.
What is the ETA on this exactly? Kind of feels like it is being swept under a rug! 
There is no public ETA but there is actual people working on both the transform improvements and C# compute stuff.
Once we have something shareable it might also start with an extra preview build.
3 Likes
The compute stuff sounds fun!
1 Like
Sounds awesome 
In other words, it takes a couple of years before it’s available in the main build. Really hope I’m wrong!
1 Like
That sounds like cheating giving all the work to the GPU, which then uses the results to draw the scene. No it will never take off, our CPU’s would be left with next to nothing to do!
Suspect you are joking! But anyway interesting subject, the division of labour between CPU and GPU. The answer is “best tool for the job” - and GPUs do a lot of jobs these days. I’m not sure it should continue to be called a GPU in the years to come
It’s common practise and very desirable with this generation of consoles which have stronger GPUs than CPUs. You will find most current gen titles do things like this.
GPUs operate on vastly bigger data sets, consider that your gpu (regardless of what it is) should be able to sustain 60fps with a simple shader filling 4k resolution, which is 8,294,400 million pixels for a single pass. That’s a lot of number crunching. A typical game will end up filling 1080p res a few times over.
GPUs are really ideal for a lot of workflows where you might not need results instantly or results can remain on GPU.
1 Like
Being able to r/w Transforms safely via threads will make a massive difference. I’m hoping for thread safe Texture2D r/w next, which will help things like UMA (and possibly Morph3D), which make/use textures at runtime.
1 Like
Agree. Go on Unity, CPU/CPU performances are never sufficient! We need more horsepower!!
I thought these multithreaded transform performance improvements were supposed to be included in Unity 5.6, but that did not happen. Now it sounds like these improvements won’t be in Unity 2017 for a while. What is main holdup/challenge? I realize things take time. I am just really curious what the main integration challenge is.
I’ve locked this thread because it’s veering all over and against the intent of the 2017 forum which is to thrash out existing issues in 2017 beta. I am to blame for some of the veering, so I’ll stop myself too. Sorry.
If a developer wants to add a reply below please feel free, otherwise the answer is “when it’s in the release notes.”