Advertisement

Latest AMD Activity

Pagur said:
But this is clearly not the number provided by AMD which is just 304.8 GTexel/s (1905 MHz x 160 TMU).

A RX 5700 has memory bandwidth of 448 GB/s, and assuming a texel is one byte, this number matches better.

Pagur said:
it seems legit to compute the texture fill rate as follow: Texture Rate…

2,669 views
Advertisement
khawk
July 15, 2021 06:11 PM
AMD FidelityFX Super Resolution available for developers on GPUOpen

Following the recent announcement of AMD FidelityFX Super Resolution (FSR), AMD is today making the source code for its powerful spatial upscaling technology broadly available to game developers via GPUOp…

7,384 views

@ddlox Thank you. I will check it out for my issue. 

Glad you replied!

5,728 views
  • The driver is an interface to the hardware, so it's not always perfect. New games (especially AAA cutting edge ones) use new GPU features which often require fixes to ensure they are working properly
  • Optimizations that help the game run better make it into the new driver. This happens by seeing the …
  • 4,287 views

    Solved! The problem was that AMD required array of samplers to be filled. I had implemented workaround earlier, but because of the typo that I made when renaming macros, everything broke. Code for workaround:

    #define pointLampsCount 1
    #define dirLampsCount 1
    #define pointLightsLimit 8
    #define dirLi…
    12,414 views
    khawk
    December 13, 2019 04:55 PM
    AMD Releases

    AMD has released the "Vega" 7nm Instruction Set Architecture in PDF form. They have been releasing GPU ISA documentation for some time now, and now they've added the 7nm architecture to the list.

    In the ISA documentation you will find:

    • Chapter 1 begins the document begins with an overview of the AMD …
    3,133 views
    Advertisement
    Advertisement