• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

News Microsoft Reveals DirectStorage API Details at Game Stack Live Event

Rigg

Senior member
Microsoft says it's looking to implement DMA (direct memory access) between SSD and GPU using DirectStorage: "We are looking into supporting this in the future, but there's some more work we need to do in order for this functionality to be robust enough to include in Windows."

May eventually support external storage

PCIe 3.0 drives supported

DirecStorage uses its own compression algorithm, new class of compression tech

Does not require GDK on Windows

DirectCompute - Allows DirectStorage to run on existing GPUs


 
I guess my million dollar question here is... What does this help accomplish in the real world? Even more "open" open world games?

We already have lots of absurdly fast non-volitile men, lots of absurdly fast volatile system mem, plenty of outrageously fast volatile GPU mem...

This will basically allow the GPU to fetch data directly from the non-volitile men without the data having to go through system Ram, yes?
 
I guess my million dollar question here is... What does this help accomplish in the real world? Even more "open" open world games?
This enables asset streaming done right.

No asset pop-in as developers can predict when something will be visible and know it can be streamed in just-in-time. This means much higher quality textures/models overall for a given RAM size (as only the highest-quality assets in say ~130 degree FOV need to be kept in GPU memory).

Also no tight spots, long coridors or elevator rides between large open-world areas to hide the time it takes to load models/textures
 
This will basically allow the GPU to fetch data directly from the non-volitile men without the data having to go through system Ram, yes?
I believe this is the ultimate goal but the initial implementation will utilize system RAM. It will give the GPU direct access to system memory and it will do decompression on the GPU rather than the CPU. This should more efficiently utilize VRAM and increase effective bandwidth by keeping the data compressed until it hits the GPU.

 
I'm kind of curious if this has an implications for mining, especially on older cards that have 4 GB or less of memory. ETH mining scales based on memory bandwidth and this effectively opens up a lot more of it, especially for cards that didn't have enough VRAM to be able to mine since I think that ETH basically needs more than 4 GB.
 
I'm kind of curious if this has an implications for mining, especially on older cards that have 4 GB or less of memory. ETH mining scales based on memory bandwidth and this effectively opens up a lot more of it, especially for cards that didn't have enough VRAM to be able to mine since I think that ETH basically needs more than 4 GB.

-Wonder if it ties into the news about the price of storage potentially going through the roof thanks to "Proof of Storage" style mining...

Still proof of work, but now not even sub 6gb cards will be safe...
 
Back
Top