Jump to content

can anyone explain what directx 12 tiers are?

gilang01

All I know is that DIRECTX 12 will have 3 tiers. All amd gcn cards will.support tier 3, whereas nvidia's maxwell cards can only support tier 2. What's the difference?

Link to comment
Share on other sites

Link to post
Share on other sites

Essentially, DX12 will be able to take advantage of GCN more than Kepler/Maxwell.

 

Least that's what I understand.

Link to comment
Share on other sites

Link to post
Share on other sites

DirectX has tiers now...? What!? I must have missed something.

 

That being said, I'll almost guarantee that my first gen GCN (7XXX series) cards won't be the same as the RX 2XX series cards.

Link to comment
Share on other sites

Link to post
Share on other sites

It's really complicated, but the short story is, it doesn't make any appreciable difference in performance in dx12. 

Something to do with resource binding. 

 

That's good to hear - I'll be on the lookout for the final version of DX12 and LTT's explanation of it.

 

What of all of the perfect scaling for multi-GPU setups and aggregated VRAM news I heard? Is any of that actually a thing?

Link to comment
Share on other sites

Link to post
Share on other sites

That's good to hear - I'll be on the lookout for the final version of DX12 and LTT's explanation of it.

 

What of all of the perfect scaling for multi-GPU setups and aggregated VRAM news I heard? Is any of that actually a thing?

bear in mind that all of this is still somewhat of a rumor. 

It seems like it is no longer required for the two cards to have exact copies of the same resources, so VRAM would not be strictly limited to that of a single card. However, perfect scaling would never exist (Says the Engineering Major), but it would give you, say, instead of roughly 150% performance for two cards, it would have 180%. 

Me: Computer Engineer. Geek. Nerd.

[Educational] Computer Architecture: Computer Memory Hierarchy

[Educational] Computer Architecture:  What is SSE/AVX? (SIMD)

Link to comment
Share on other sites

Link to post
Share on other sites

bear in mind that all of this is still somewhat of a rumor. 

It seems like it is no longer required for the two cards to have exact copies of the same resources, so VRAM would not be strictly limited to that of a single card. However, perfect scaling would never exist (Says the Engineering Major), but it would give you, say, instead of roughly 150% performance for two cards, it would have 180%. 

 

That'd be nice. And instead of 180% (if even that sometimes) for three cards, I might see 220%? Haha! I'll probably have upgraded by the time DX12 is out; I have a 4K monitor coming home and occasionally struggle with VRAM as it is on my 3x1080P setup. Three 7970s with 3GB each, if you're wondering.

 

Well, we can all hope for the best - but it's Microsoft so it's probably best not to.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×