Announcement

Collapse
No announcement yet.

15-Way Radeon/GeForce Source Engine Comparison On Ubuntu 14.04 LTS

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

  • #11
    Originally posted by dungeon View Post
    I can't be sure who blows away who because there is CPU bottleneck in some cases here, so with different kind of CPU things maybe different . To me results with 120 fps and up just shows card/driver is OK for those titles, but when benchmark results all goes up that margin it starts to be boring .
    With a Intel Core i7-4770K, it's hard to be a CPU bottleneck. Nvidia in some tests shows to have twice the performance of AMD. Particularly Half-Life 2 Lost Coast and Portal. Keep in mind these games run on an engine made in 2004, so it's not going to stress them that much.

    AMD's drivers clearly need work, even though the games are playable. I wouldn't be shocked if the open source drivers work better in those benchmarks.

    Comment


    • #12
      Originally posted by Dukenukemx View Post
      Nvidia in some tests shows to have twice the performance of AMD.
      I can't see that twice performance anywhere if you compare only compete cards Best card from AMD benchmarked here is Radeon r9 270x , that is comparabile only with Nvidia Geforce GTX 760 and no more than that. Those two performs just as expected - very similar in these games .

      There is no Radeon cards benchmarked here which is comparabile with GTX 770 i 780 so those must play better then any other... Micheal must have R9 280 and 290 to compare with those . R9 290 with GTX 780, R9 290X with GTX 780Ti, R9 280X with GTX 770, R9 270X with GTX 760 etc.

      Comment


      • #13
        Originally posted by Dukenukemx View Post
        With a Intel Core i7-4770K, it's hard to be a CPU bottleneck. Nvidia in some tests shows to have twice the performance of AMD. Particularly Half-Life 2 Lost Coast and Portal. Keep in mind these games run on an engine made in 2004, so it's not going to stress them that much.

        AMD's drivers clearly need work, even though the games are playable. I wouldn't be shocked if the open source drivers work better in those benchmarks.
        This has more to do with the hardware than the drivers, comparing a 780ti to R9 270X you are taking the best Nvidia card ($699 RRP) against a midrange AMD card ($199 RRP).

        Even taking into account the extra $50 Newegg are charging on top of RRP for the R9 270X, it is only a third of the price the 780ti.

        Comment


        • #14
          I am always amused when someone says x beats y when x costs 3 times more than y.
          If you want to compare, you have to group them by price!

          Comment


          • #15
            As a owner of Gigabyte R9 270X 2G, I'm glad it performs well against its competitor within similar price, GeForce GTX760.
            Their results are very close in most of benchmarked games. Only R9 270X wins in Portal while GeForce GTX760 leads a big step ahead in TF2.

            I guess it's not always true that Catalyst OpenGL sxxks.

            Comment


            • #16
              Some lower end Radeon hardware would be nice, like those hugely popular igpus in apus, such as Trinity and Kaveri. The batch of cards tested already have enough muscle to play source titles and high fps at ridiculous resolutions. It is nice to see though how terrible catalyst performs with source. Sadly input lag doesn't show up in the benchmarks making catalyst even worse than what benchmarks indicate.

              Comment


              • #17
                Originally posted by molecule-eye View Post
                It is nice to see though how terrible catalyst performs with source.
                The lowest performing AMD cards were still able to hit around 120 fps even in their worst performing benchmarks, I'd hardly call that terrible.

                Comment


                • #18
                  Originally posted by Herem View Post
                  The lowest performing AMD cards were still able to hit around 120 fps even in their worst performing benchmarks, I'd hardly call that terrible.
                  Dude, the game engine is 10 years old and the games are OL' SCHOOL! 120fps is dismal for cards of that calibre. Just look at how the nvidia cards are performing in comparison. If you're running lower end hardware (compared to what was benchmarked, but not low-end by any means) and want all the bells and whistles turned on (AA, etc.) then good luck with Catalyst, even with games that are 10 years old. That's sad!

                  Comment


                  • #19
                    Originally posted by molecule-eye View Post
                    Dude, the game engine is 10 years old and the games are OL' SCHOOL! 120fps is dismal for cards of that calibre. Just look at how the nvidia cards are performing in comparison. If you're running lower end hardware (compared to what was benchmarked, but not low-end by any means) and want all the bells and whistles turned on (AA, etc.) then good luck with Catalyst, even with games that are 10 years old. That's sad!
                    Cards of what calibre? There aren't any top end cards here from AMD as mentioned in the comments above. The most expensive card tested is slightly cheaper than a GTX 760 (even with the profiteering from resellers) and performs in line with this pricing.

                    I must have missed the tests in the article with all the bells and whistles turned on which proved "how terrible catalyst performs with source."

                    Comment

                    Working...
                    X