Dolphin Emulator Drops D3D12 Backend, Focuses On Vulkan

[labyrinth153]

Non-existant because they want to push their own bullshit API.
For that matter because the DX9 and DX11 backends worked awesomely on Apple.

OpenGL backend still works afaik.

Apple nut here. Dolphin versions 5 and newer do not work suitable under Mac OS. They require opengl 4.5, Mac OS is version 4.2. I get 60fps solid on dolphin 3.5/4 and 5 and newer aee unplayable. The hardware is sufficiently powerful enough, it just lacks features needed in opengl 4.5. I would love a metal backend. I think a project exists to support vulkan over metal but don't quote me on that.

[Hi-Angel]

Then there's Apple... What's the state of Vulkan on Apple?

I think it's alike to Ubuntu situation with Mir vs Wayland. They would support it at some point of time.

[geearf]

I would love a metal backend. I think a project exists to support vulkan over metal but don't quote me on that.

MoltenVK.

When I talked about it with a Dolphin dev (Phire I think), he wasn't impressed and didn't think it'd work out.

[Guest]

Ok. Let's put it this way: Since Apple pushes Metal has Vulkan seen any light on Apple? Does there exist any applications using Vulkan on Apple?

I don't care. Apple should die and their shit OS.

[Guest]

A recent RPCS3 bug report showed the Qt build to be a lot faster than the Wx one, though no-one understood why

Every Wx application I ever used was terribly slow and broken. eDonkey (or equivalent), Spring lobby client etc. Qt seems to be very lightweight when comes to end user experience.

[Xelix]

Seriously? Dolphin has always been a very multi-platform application. Despite this, and people's pleads for Vulkan implementation instead (or at least to be prioritized), they chose to do DX12 anyway. Now they find out the hard way that DX12 wasn't actually the best choice after all. Gee, good job listening to the community. Could've saved a lot of time and effort just doing things the right way the first time around.

The DX12 backend predates Vulkan itself, as others have mentioned.

You're very deluded if you think you have anything to teach to the Dolphin team about how to write an emulator.

[schmidtbag]

The DX12 backend predates Vulkan itself, as others have mentioned.

Your statement is meaningless. If you're referring to what was implemented in Dolphin first, my first comment pointed this out, so you didn't contribute anything new. If you're referring to the availability of DX12 and Vulkan, Dolphin's DX12 backend was implemented long after Vulkan was established. So, that point is moot.

You're very deluded if you think you have anything to teach to the Dolphin team about how to write an emulator.

The Dolphin team does a fantastic job at what they do. I don't need to "teach" them how to do anything. My point was that Vulkan support has always been a greater interest, and yet it wasn't anybody's top priority until after DX12 was implemented.

As for the others who responded to me - I get that it's an open community project and therefore nobody can rightfully tell them what to do. But, it is not within one's interest to ignore the interests of many others. To put it in another perspective, let's say you own a soup kitchen where you give free soup for the poor. But, you over-salt the soup, and nobody wants to eat it. If you're doing something for the community but the community wants it done a different way, what's the point of doing it?

[andrei_me]

Also, don't both low-level graphics API's have more benefit when you have a fast GPU and slow CPU?

On iGPU, it might release thermal and voltage usage on CPU to be used on it's iGPU, there's an old video of intel promoting vulkan demonstrating this

Edit:
Found it

[Helios747]

Your statement is meaningless. If you're referring to what was implemented in Dolphin first, my first comment pointed this out, so you didn't contribute anything new. If you're referring to the availability of DX12 and Vulkan, Dolphin's DX12 backend was implemented long after Vulkan was established. So, that point is moot.

The Dolphin team does a fantastic job at what they do. I don't need to "teach" them how to do anything. My point was that Vulkan support has always been a greater interest, and yet it wasn't anybody's top priority until after DX12 was implemented.

As for the others who responded to me - I get that it's an open community project and therefore nobody can rightfully tell them what to do. But, it is not within one's interest to ignore the interests of many others. To put it in another perspective, let's say you own a soup kitchen where you give free soup for the poor. But, you over-salt the soup, and nobody wants to eat it. If you're doing something for the community but the community wants it done a different way, what's the point of doing it?

We got a D3D12 backend because somebody wrote one.

We dropped the D3D12 backend because that person stopped maintaining it.

We got a Vulkan backend because somebody wrote one.


We're not developers aiming to serve users. We're developers that work on whats interesting to us. This is the same for most unpaid OSS projects. What the community wants is entirely irrelevant if nobody wants to work on it. Most OSS users don't seem to understand that.