Announcement

Collapse
No announcement yet.

NVIDIA GeForce GTX 1050 Ti Linux Benchmarks

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • NVIDIA GeForce GTX 1050 Ti Linux Benchmarks

    Phoronix: NVIDIA GeForce GTX 1050 Ti Linux Benchmarks

    Yesterday I published the first GeForce GTX 1050 Linux benchmarks with OpenGL, OpenCL, and Vulkan coverage. With now having my hands on the EVGA GeForce GTX 1050 Ti SC GAMING graphics card, here are the first Linux benchmarks of the GTX 1050 Ti graphics card that can be fetched for less than $150 USD.

    http://www.phoronix.com/vr.php?view=23688

  • #2
    I would probably get one of these to hold me over until my next build... if nvidia had Mac drivers. I need drivers for the hackintosh and there are none supporting Pascal cards yet.

    I wonder if this card can boost to 1.9 GHz in Linux like it does on Windows?

    Comment


    • #3
      Originally posted by johnc View Post
      I wonder if this card can boost to 1.9 GHz in Linux like it does on Windows?
      They do, even though it's not visualized in nvidia-settings.

      Comment


      • #4
        So this card is factory overclocked? It seems like either save your money and just get a standard 1050 or spend a bit more and get an AMD 470.

        Comment


        • #5
          Now this test is really interesting. With only 2.3 TFLOPS vs. 4.9 TFLOPS for the RX 470 the results show how bad the current AMDGPU drivers still perform on Linux in several games. David vs. Goliath.

          Comment


          • #6
            Originally posted by oooverclocker View Post
            Now this test is really interesting. With only 2.3 TFLOPS vs. 4.9 TFLOPS for the RX 470 the results show how bad the current AMDGPU drivers still perform on Linux in several games. David vs. Goliath.
            The problem consists of two things:
            - Architectural efficiency - Generally Polaris need ~50% computational performance and ~50% memory capacity/bandwidth to perform comparable to a Pascal GPU.
            - Driver quality - AMD only perform well in Direct3D, all other APIs perform worse to some or large degree.

            Comment


            • #7
              Originally posted by efikkan View Post
              - Driver quality - AMD only perform well in Direct3D, all other APIs perform worse to some or large degree.
              NV is better for DirectX 11 and OpenGL, but AMD for DirectX 12 and Vulkan.

              Comment


              • #8
                Good tests Michael

                But need more tests case bioshock infinite, grid autosports, dirt showdown, f1 2015, tomb raider, shadow of mordor, metro and if possible others

                Without forget at 1080p max because this hardware are too weak for more than 1080p

                Comment


                • #9
                  Originally posted by efikkan View Post
                  The problem consists of two things:
                  - Architectural efficiency - Generally Polaris need ~50% computational performance and ~50% memory capacity/bandwidth to perform comparable to a Pascal GPU.
                  - Driver quality - AMD only perform well in Direct3D, all other APIs perform worse to some or large degree.
                  You're a complete idiot guy.

                  Comment


                  • #10
                    Originally posted by nadro View Post
                    NV is better for DirectX 11 and OpenGL, but AMD for DirectX 12 and Vulkan.
                    That is a common misconception, and is not true at all.

                    In addition to "low-level" features, Direct3D 12 also required several driver side changes, in terms of threading, state changes and memory management. AMD decided to only implement this for the new APIs, while Nvidia tried to bring as many improvements as possible to all APIs: (1)(2). This gave Nvidia an extra edge in Direct3D 11 and OpenGL games (for Kepler and newer). But since they have brought a significant portion of the Direct3D/Vulkan improvements to all APIs, the relative gain will of course be smaller, which in turn has lead many fans to claim that AMD is better for the new APIs because they get larger proportional gains. This is obviously a lack of technical and mathematical knowledge among users.

                    The other factor is AMD optimized games; several of the first games have been AMD exclusives ported to PC. These games will obviously favor AMD hardware to some degree, and when you take of the red glasses combine this with the awareness that Nvidia optimized all APIs, you'll see that there is actually no evidence that AMD is generally faster with the new APIs.

                    In fact, the "low-level" features of the new APIs is one of the reasons why games can easily be more biased now. When developers are given more control over memory structures, they would also need to explicitly benchmark to check if their implementation does not favor one class of hardware, if they disregard this they risk favoring one. This does not mean the API itself is better for one class of hardware, but it's really up to the developers to not make crappy code. This difference will in all likelihood dissipate over time for top PC games, if the developers are serious.

                    It's very dangerous to predict how the current hardware will perform on future games. We have yet not seen really good Direct3D 12 / Vulkan games yet, and as with every new API it will take a couple of years before we see the good stuff. If anything, the new features might actually create an even longer transition person this time around. So people, please be serious; do not use a couple of AMD games to predict the future you know very little about!

                    Comment

                    Working...
                    X